* [PATCH v4 00/12] Add LoongArch support.
@ 2021-12-24 9:28 chenglulu
2021-12-24 9:28 ` [PATCH v4 01/12] LoongArch Port: Regenerate configure chenglulu
` (12 more replies)
0 siblings, 13 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
The LoongArch architecture (LoongArch) is an Instruction Set
Architecture (ISA) that has a Reduced Instruction Set Computer (RISC)
style.
The documents are on
https://loongson.github.io/LoongArch-Documentation/README-EN.html
The ELF ABI Documents are on:
https://loongson.github.io/LoongArch-Documentation/LoongArch-ELF-ABI-EN.html
The binutils has been merged into trunk:
https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=560b3fe208255ae909b4b1c88ba9c28b09043307
Note: We split -mabi= into -mabi=lp64d/f/s, the new options not support by upstream binutils yet,
this GCC port requires the following patch applied to binutils to build.
https://github.com/loongson/binutils-gdb/commit/aacb0bf860f02aa5a7dcb76dd0e392bf871c7586
(will be submitted to upstream after gcc side comfirmed)
changelog:
v1 -> v2
1. Split patch set.
2. Change some code style.
3. Add -mabi=lp64d/f/s options.
4. Change GLIBC_DYNAMIC_LINKER_LP64 name.
v2 -> v3
1. Change some code style.
2. Bug fix.
v3 -> v4
1. Change some code style.
2. Bug fix.
3. Delete some builtin macros.
Add LoongArch support.
chenglulu (12):
LoongArch Port: Regenerate configure
LoongArch Port: gcc build
LoongArch Port: Regenerate gcc/configure.
LoongArch Port: Machine Decsription files.
LoongArch Port: Machine description C files and .h files.
LoongArch Port: Builtin functions.
LoongArch Port: Builtin macros.
LoongArch Port: libgcc
LoongArch Port: Regenerate libgcc/configure.
LoongArch Port: libgomp
LoongArch Port: gcc/testsuite
LoongArch Port: Add doc.
config/picflag.m4 | 3 +
configure | 10 +-
configure.ac | 10 +-
contrib/config-list.mk | 5 +-
.../config/loongarch/loongarch-common.c | 63 +
gcc/config.gcc | 400 +-
gcc/config/host-linux.c | 2 +
gcc/config/loongarch/constraints.md | 212 +
gcc/config/loongarch/generic.md | 132 +
gcc/config/loongarch/genopts/genstr.sh | 91 +
.../loongarch/genopts/loongarch-strings | 58 +
gcc/config/loongarch/genopts/loongarch.opt.in | 189 +
gcc/config/loongarch/gnu-user.h | 78 +
gcc/config/loongarch/la464.md | 132 +
gcc/config/loongarch/larchintrin.h | 413 ++
gcc/config/loongarch/linux.h | 50 +
gcc/config/loongarch/loongarch-builtins.c | 511 ++
gcc/config/loongarch/loongarch-c.c | 111 +
gcc/config/loongarch/loongarch-cpu.c | 206 +
gcc/config/loongarch/loongarch-cpu.h | 30 +
gcc/config/loongarch/loongarch-def.c | 164 +
gcc/config/loongarch/loongarch-def.h | 151 +
gcc/config/loongarch/loongarch-driver.c | 187 +
gcc/config/loongarch/loongarch-driver.h | 69 +
gcc/config/loongarch/loongarch-ftypes.def | 106 +
gcc/config/loongarch/loongarch-modes.def | 29 +
gcc/config/loongarch/loongarch-opts.c | 582 ++
gcc/config/loongarch/loongarch-opts.h | 86 +
gcc/config/loongarch/loongarch-protos.h | 242 +
gcc/config/loongarch/loongarch-str.h | 57 +
gcc/config/loongarch/loongarch-tune.h | 72 +
gcc/config/loongarch/loongarch.c | 6461 +++++++++++++++++
gcc/config/loongarch/loongarch.h | 1291 ++++
gcc/config/loongarch/loongarch.md | 3829 ++++++++++
gcc/config/loongarch/loongarch.opt | 189 +
gcc/config/loongarch/predicates.md | 553 ++
gcc/config/loongarch/sync.md | 574 ++
gcc/config/loongarch/t-linux | 53 +
gcc/config/loongarch/t-loongarch | 59 +
gcc/configure | 85 +-
gcc/configure.ac | 33 +-
gcc/doc/install.texi | 47 +-
gcc/doc/invoke.texi | 201 +
gcc/doc/md.texi | 55 +
gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C | 2 +-
gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C | 2 +-
gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C | 2 +-
gcc/testsuite/gcc.dg/20020312-2.c | 2 +
gcc/testsuite/gcc.dg/loop-8.c | 2 +-
.../torture/stackalign/builtin-apply-2.c | 2 +-
gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c | 2 +-
.../gcc.target/loongarch/loongarch.exp | 40 +
.../gcc.target/loongarch/tst-asm-const.c | 16 +
gcc/testsuite/go.test/go-test.exp | 3 +
gcc/testsuite/lib/target-supports.exp | 14 +
libgcc/config.host | 28 +-
libgcc/config/loongarch/crtfastmath.c | 52 +
libgcc/config/loongarch/crti.S | 43 +
libgcc/config/loongarch/crtn.S | 39 +
libgcc/config/loongarch/lib2funcs.c | 0
libgcc/config/loongarch/linux-unwind.h | 80 +
libgcc/config/loongarch/sfp-machine.h | 152 +
libgcc/config/loongarch/t-crtstuff | 2 +
libgcc/config/loongarch/t-elf | 3 +
libgcc/config/loongarch/t-loongarch | 9 +
libgcc/config/loongarch/t-loongarch64 | 1 +
libgcc/config/loongarch/t-softfp-tf | 3 +
libgcc/configure | 5 +-
libgcc/configure.ac | 2 +-
libgomp/configure.tgt | 4 +
70 files changed, 18363 insertions(+), 28 deletions(-)
create mode 100644 gcc/common/config/loongarch/loongarch-common.c
create mode 100644 gcc/config/loongarch/constraints.md
create mode 100644 gcc/config/loongarch/generic.md
create mode 100755 gcc/config/loongarch/genopts/genstr.sh
create mode 100644 gcc/config/loongarch/genopts/loongarch-strings
create mode 100644 gcc/config/loongarch/genopts/loongarch.opt.in
create mode 100644 gcc/config/loongarch/gnu-user.h
create mode 100644 gcc/config/loongarch/la464.md
create mode 100644 gcc/config/loongarch/larchintrin.h
create mode 100644 gcc/config/loongarch/linux.h
create mode 100644 gcc/config/loongarch/loongarch-builtins.c
create mode 100644 gcc/config/loongarch/loongarch-c.c
create mode 100644 gcc/config/loongarch/loongarch-cpu.c
create mode 100644 gcc/config/loongarch/loongarch-cpu.h
create mode 100644 gcc/config/loongarch/loongarch-def.c
create mode 100644 gcc/config/loongarch/loongarch-def.h
create mode 100644 gcc/config/loongarch/loongarch-driver.c
create mode 100644 gcc/config/loongarch/loongarch-driver.h
create mode 100644 gcc/config/loongarch/loongarch-ftypes.def
create mode 100644 gcc/config/loongarch/loongarch-modes.def
create mode 100644 gcc/config/loongarch/loongarch-opts.c
create mode 100644 gcc/config/loongarch/loongarch-opts.h
create mode 100644 gcc/config/loongarch/loongarch-protos.h
create mode 100644 gcc/config/loongarch/loongarch-str.h
create mode 100644 gcc/config/loongarch/loongarch-tune.h
create mode 100644 gcc/config/loongarch/loongarch.c
create mode 100644 gcc/config/loongarch/loongarch.h
create mode 100644 gcc/config/loongarch/loongarch.md
create mode 100644 gcc/config/loongarch/loongarch.opt
create mode 100644 gcc/config/loongarch/predicates.md
create mode 100644 gcc/config/loongarch/sync.md
create mode 100644 gcc/config/loongarch/t-linux
create mode 100644 gcc/config/loongarch/t-loongarch
create mode 100644 gcc/testsuite/gcc.target/loongarch/loongarch.exp
create mode 100644 gcc/testsuite/gcc.target/loongarch/tst-asm-const.c
create mode 100644 libgcc/config/loongarch/crtfastmath.c
create mode 100644 libgcc/config/loongarch/crti.S
create mode 100644 libgcc/config/loongarch/crtn.S
create mode 100644 libgcc/config/loongarch/lib2funcs.c
create mode 100644 libgcc/config/loongarch/linux-unwind.h
create mode 100644 libgcc/config/loongarch/sfp-machine.h
create mode 100644 libgcc/config/loongarch/t-crtstuff
create mode 100644 libgcc/config/loongarch/t-elf
create mode 100644 libgcc/config/loongarch/t-loongarch
create mode 100644 libgcc/config/loongarch/t-loongarch64
create mode 100644 libgcc/config/loongarch/t-softfp-tf
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 01/12] LoongArch Port: Regenerate configure
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 02/12] LoongArch Port: gcc build chenglulu
` (11 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
* config/picflag.m4: Default add build option '-fpic' for LoongArch.
* configure: Add LoongArch tuples.
* configure.ac: Like wise.
---
config/picflag.m4 | 3 +++
configure | 10 +++++++++-
configure.ac | 10 +++++++++-
3 files changed, 21 insertions(+), 2 deletions(-)
diff --git a/config/picflag.m4 b/config/picflag.m4
index 8b106f9af88..0aefcf619bf 100644
--- a/config/picflag.m4
+++ b/config/picflag.m4
@@ -44,6 +44,9 @@ case "${$2}" in
# sets the default TLS model and affects inlining.
$1=-fPIC
;;
+ loongarch*-*-*)
+ $1=-fpic
+ ;;
mips-sgi-irix6*)
# PIC is the default.
;;
diff --git a/configure b/configure
index 9c2d7df1bb2..87548f0da96 100755
--- a/configure
+++ b/configure
@@ -3060,7 +3060,7 @@ case "${ENABLE_GOLD}" in
# Check for target supported by gold.
case "${target}" in
i?86-*-* | x86_64-*-* | sparc*-*-* | powerpc*-*-* | arm*-*-* \
- | aarch64*-*-* | tilegx*-*-* | mips*-*-* | s390*-*-*)
+ | aarch64*-*-* | tilegx*-*-* | mips*-*-* | s390*-*-* | loongarch*-*-*)
configdirs="$configdirs gold"
if test x${ENABLE_GOLD} = xdefault; then
default_ld=gold
@@ -3646,6 +3646,9 @@ case "${target}" in
i[3456789]86-*-*)
libgloss_dir=i386
;;
+ loongarch*-*-*)
+ libgloss_dir=loongarch
+ ;;
m68hc11-*-*|m6811-*-*|m68hc12-*-*|m6812-*-*)
libgloss_dir=m68hc11
;;
@@ -4030,6 +4033,11 @@ case "${target}" in
wasm32-*-*)
noconfigdirs="$noconfigdirs ld"
;;
+ loongarch*-*-linux*)
+ ;;
+ loongarch*-*-*)
+ noconfigdirs="$noconfigdirs gprof"
+ ;;
esac
# If we aren't building newlib, then don't build libgloss, since libgloss
diff --git a/configure.ac b/configure.ac
index 68cc5cc31fe..55362afeeae 100644
--- a/configure.ac
+++ b/configure.ac
@@ -353,7 +353,7 @@ case "${ENABLE_GOLD}" in
# Check for target supported by gold.
case "${target}" in
i?86-*-* | x86_64-*-* | sparc*-*-* | powerpc*-*-* | arm*-*-* \
- | aarch64*-*-* | tilegx*-*-* | mips*-*-* | s390*-*-*)
+ | aarch64*-*-* | tilegx*-*-* | mips*-*-* | s390*-*-* | loongarch*-*-*)
configdirs="$configdirs gold"
if test x${ENABLE_GOLD} = xdefault; then
default_ld=gold
@@ -899,6 +899,9 @@ case "${target}" in
i[[3456789]]86-*-*)
libgloss_dir=i386
;;
+ loongarch*-*-*)
+ libgloss_dir=loongarch
+ ;;
m68hc11-*-*|m6811-*-*|m68hc12-*-*|m6812-*-*)
libgloss_dir=m68hc11
;;
@@ -1283,6 +1286,11 @@ case "${target}" in
wasm32-*-*)
noconfigdirs="$noconfigdirs ld"
;;
+ loongarch*-*-linux*)
+ ;;
+ loongarch*-*-*)
+ noconfigdirs="$noconfigdirs gprof"
+ ;;
esac
# If we aren't building newlib, then don't build libgloss, since libgloss
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 02/12] LoongArch Port: gcc build
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
2021-12-24 9:28 ` [PATCH v4 01/12] LoongArch Port: Regenerate configure chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 03/12] LoongArch Port: Regenerate gcc/configure chenglulu
` (10 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/
* common/config/loongarch/loongarch-common.c: New file.
* config/loongarch/genopts/genstr.sh: New file.
* config/loongarch/genopts/loongarch-strings: New file.
* config/loongarch/genopts/loongarch.opt.in: New file.
* config/loongarch/gnu-user.h: New file.
* config/loongarch/linux.h: New file.
* config/loongarch/loongarch-cpu.c: New file.
* config/loongarch/loongarch-cpu.h: New file.
* config/loongarch/loongarch-driver.c: New file.
* config/loongarch/loongarch-driver.h: New file.
* config/loongarch/loongarch-opts.c: New file.
* config/loongarch/loongarch-opts.h: New file.
* config/loongarch/loongarch.opt: New file.
* config/loongarch/t-linux: New file.
* config/loongarch/t-loongarch: New file.
* config.gcc: Add LoongArch support.
* configure.ac: Add LoongArch support.
---
.../config/loongarch/loongarch-common.c | 63 ++
gcc/config.gcc | 400 +++++++++++-
gcc/config/loongarch/genopts/genstr.sh | 91 +++
.../loongarch/genopts/loongarch-strings | 58 ++
gcc/config/loongarch/genopts/loongarch.opt.in | 189 ++++++
gcc/config/loongarch/gnu-user.h | 78 +++
gcc/config/loongarch/linux.h | 50 ++
gcc/config/loongarch/loongarch-cpu.c | 206 +++++++
gcc/config/loongarch/loongarch-cpu.h | 30 +
gcc/config/loongarch/loongarch-def.c | 164 +++++
gcc/config/loongarch/loongarch-def.h | 151 +++++
gcc/config/loongarch/loongarch-driver.c | 187 ++++++
gcc/config/loongarch/loongarch-driver.h | 69 +++
gcc/config/loongarch/loongarch-opts.c | 582 ++++++++++++++++++
gcc/config/loongarch/loongarch-opts.h | 86 +++
gcc/config/loongarch/loongarch-str.h | 57 ++
gcc/config/loongarch/loongarch.opt | 189 ++++++
gcc/config/loongarch/t-linux | 53 ++
gcc/config/loongarch/t-loongarch | 59 ++
gcc/configure.ac | 33 +-
20 files changed, 2790 insertions(+), 5 deletions(-)
create mode 100644 gcc/common/config/loongarch/loongarch-common.c
create mode 100755 gcc/config/loongarch/genopts/genstr.sh
create mode 100644 gcc/config/loongarch/genopts/loongarch-strings
create mode 100644 gcc/config/loongarch/genopts/loongarch.opt.in
create mode 100644 gcc/config/loongarch/gnu-user.h
create mode 100644 gcc/config/loongarch/linux.h
create mode 100644 gcc/config/loongarch/loongarch-cpu.c
create mode 100644 gcc/config/loongarch/loongarch-cpu.h
create mode 100644 gcc/config/loongarch/loongarch-def.c
create mode 100644 gcc/config/loongarch/loongarch-def.h
create mode 100644 gcc/config/loongarch/loongarch-driver.c
create mode 100644 gcc/config/loongarch/loongarch-driver.h
create mode 100644 gcc/config/loongarch/loongarch-opts.c
create mode 100644 gcc/config/loongarch/loongarch-opts.h
create mode 100644 gcc/config/loongarch/loongarch-str.h
create mode 100644 gcc/config/loongarch/loongarch.opt
create mode 100644 gcc/config/loongarch/t-linux
create mode 100644 gcc/config/loongarch/t-loongarch
diff --git a/gcc/common/config/loongarch/loongarch-common.c b/gcc/common/config/loongarch/loongarch-common.c
new file mode 100644
index 00000000000..3f440aef1e7
--- /dev/null
+++ b/gcc/common/config/loongarch/loongarch-common.c
@@ -0,0 +1,63 @@
+/* Common hooks for LoongArch.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "common/common-target.h"
+#include "common/common-target-def.h"
+#include "opts.h"
+#include "flags.h"
+#include "diagnostic-core.h"
+
+/* Implement TARGET_HANDLE_OPTION. */
+
+static bool
+loongarch_handle_option (struct gcc_options *opts,
+ struct gcc_options *opts_set ATTRIBUTE_UNUSED,
+ const struct cl_decoded_option *decoded,
+ location_t loc ATTRIBUTE_UNUSED)
+{
+ size_t code = decoded->opt_index;
+ int value = decoded->value;
+
+ switch (code)
+ {
+ case OPT_mmemcpy:
+ if (value)
+ {
+ if (opts->x_optimize_size)
+ opts->x_target_flags |= MASK_MEMCPY;
+ }
+ else
+ opts->x_target_flags &= ~MASK_MEMCPY;
+ return true;
+
+ default:
+ return true;
+ }
+}
+
+#undef TARGET_DEFAULT_TARGET_FLAGS
+#define TARGET_DEFAULT_TARGET_FLAGS MASK_CHECK_ZERO_DIV
+#undef TARGET_HANDLE_OPTION
+#define TARGET_HANDLE_OPTION loongarch_handle_option
+
+struct gcc_targetm_common targetm_common = TARGETM_COMMON_INITIALIZER;
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 91806aa706f..7058a454b1b 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -454,6 +454,13 @@ mips*-*-*)
extra_objs="frame-header-opt.o"
extra_options="${extra_options} g.opt fused-madd.opt mips/mips-tables.opt"
;;
+loongarch*-*-*)
+ cpu_type=loongarch
+ extra_headers="larchintrin.h"
+ extra_objs="loongarch-c.o loongarch-builtins.o loongarch-cpu.o loongarch-opts.o loongarch-def.o"
+ extra_gcc_objs="loongarch-driver.o loongarch-cpu.o loongarch-opts.o loongarch-def.o"
+ extra_options="${extra_options} g.opt fused-madd.opt"
+ ;;
nds32*)
cpu_type=nds32
extra_headers="nds32_intrinsic.h nds32_isr.h nds32_init.inc"
@@ -2480,6 +2487,20 @@ riscv*-*-freebsd*)
# automatically detect that GAS supports it, yet we require it.
gcc_cv_initfini_array=yes
;;
+
+loongarch*-*-linux-gnu*)
+ tm_file="dbxelf.h elfos.h gnu-user.h linux.h linux-android.h glibc-stdint.h ${tm_file}"
+ tm_file="${tm_file} loongarch/gnu-user.h loongarch/linux.h"
+ extra_options="${extra_options} linux-android.opt"
+ tmake_file="${tmake_file} loongarch/t-linux"
+ gnu_ld=yes
+ gas=yes
+
+ # Force .init_array support. The configure script cannot always
+ # automatically detect that GAS supports it, yet we require it.
+ gcc_cv_initfini_array=yes
+ ;;
+
mips*-*-netbsd*) # NetBSD/mips, either endian.
target_cpu_default="MASK_ABICALLS"
tm_file="elfos.h ${tm_file} mips/elf.h ${nbsd_tm_file} mips/netbsd.h"
@@ -3586,7 +3607,7 @@ case ${target} in
;;
*-*-linux* | *-*-gnu*)
case ${target} in
- aarch64*-* | arm*-* | i[34567]86-* | powerpc*-* | s390*-* | sparc*-* | x86_64-*)
+ aarch64*-* | arm*-* | i[34567]86-* | powerpc*-* | s390*-* | sparc*-* | x86_64-* | loongarch*-*)
default_gnu_indirect_function=yes
;;
esac
@@ -4916,6 +4937,338 @@ case "${target}" in
esac
;;
+ loongarch*-*-*)
+ supported_defaults="abi arch tune fpu"
+
+ # Local variables
+ unset \
+ abi_pattern abi_default \
+ abiext_pattern abiext_default \
+ arch_pattern arch_default \
+ fpu_pattern fpu_default \
+ tune_pattern tune_default
+
+ # Inferring ABI from the triplet: pass 1
+ case ${target} in
+ loongarch64-*)
+ abiext_pattern="*"
+ abiext_default="base"
+ ;;
+ esac
+
+ # Inferring ABI from the triplet: pass 2
+ case ${target} in
+ loongarch64-*-linux-gnuf64)
+ la_canonical_triplet="loongarch64-linux-gnuf64"
+ abi_pattern="lp64d"
+ ;;
+ loongarch64-*-linux-gnuf32)
+ la_canonical_triplet="loongarch64-linux-gnuf32"
+ abi_pattern="lp64f"
+ ;;
+ loongarch64-*-linux-gnusf)
+ la_canonical_triplet="loongarch64-linux-gnusf"
+ abi_pattern="lp64s"
+ ;;
+ loongarch64-*-linux-gnu)
+ la_canonical_triplet="loongarch64-linux-gnuf64"
+ abi_pattern="lp64[dfs]"
+ abi_default="lp64d"
+ ;;
+ *)
+ echo "Unsupported target ${target}." 1>&2
+ exit 1
+ esac
+
+ ## Setting default value for with_abi.
+ case ${with_abi} in
+ "")
+ if test x${abi_default} != x; then
+ with_abi=${abi_default}
+ else
+ with_abi=${abi_pattern}
+ fi
+ ;;
+
+ *)
+ if echo "${with_abi}" | grep -E "^${abi_pattern}$"; then
+ : # OK
+ else
+ echo "Incompatible options:" \
+ "--with-abi=${with_abi} and --target=${target}." 1>&2
+ exit 1
+ fi
+ ;;
+ esac
+
+ ## Setting default value for with_abiext (internal)
+ case ${with_abiext} in
+ "")
+ if test x${abiext_default} != x; then
+ with_abiext=${abiext_default}
+ else
+ with_abiext=${abiext_pattern}
+ fi
+ ;;
+
+ *)
+ if echo "${with_abiext}" | grep -E "^${abiext_pattern}$"; then
+ : # OK
+ else
+ echo "The ABI extension type \"${with_abiext}\"" \
+ "is incompatible with --target=${target}." 1>&2
+ exit 1
+ fi
+
+ ;;
+ esac
+
+
+ # Inferring ISA-related default options from the ABI: pass 1
+ case ${with_abi}/${with_abiext} in
+ lp64*/base)
+ # architectures that support lp64* ABI
+ arch_pattern="native|loongarch64|la464"
+ # default architecture for lp64* ABI
+ arch_default="loongarch64"
+ ;;
+ *)
+ echo "Unsupported ABI type ${with_abi}/${with_abiext}." 1>&2
+ exit 1
+ ;;
+ esac
+
+ # Inferring ISA-related default options from the ABI: pass 2
+ case ${with_abi}/${with_abiext} in
+ lp64d/base)
+ fpu_pattern="64"
+ ;;
+ lp64f/base)
+ fpu_pattern="32|64"
+ fpu_default="32"
+ ;;
+ lp64s/base)
+ fpu_pattern="none|32|64"
+ fpu_default="none"
+ ;;
+ *)
+ echo "Unsupported ABI type ${with_abi}/${with_abiext}." 1>&2
+ exit 1
+ ;;
+ esac
+
+ ## Setting default value for with_arch.
+ case ${with_arch} in
+ "")
+ if test x${arch_default} != x; then
+ with_arch=${arch_default}
+ else
+ with_arch=${arch_pattern}
+ fi
+ ;;
+
+ *)
+ if echo "${with_arch}" | grep -E "^${arch_pattern}$"; then
+ : # OK
+ else
+ echo "${with_abi}/${with_abiext} ABI cannot be implemented with" \
+ "--with-arch=${with_arch}." 1>&2
+ exit 1
+ fi
+ ;;
+ esac
+
+ ## Setting default value for with_fpu.
+ if test x${with_fpu} == xnone; then
+ with_fpu="0"
+ fi
+
+ case ${with_fpu} in
+ "")
+ if test x${fpu_default} != x; then
+ with_fpu=${fpu_default}
+ else
+ with_fpu=${fpu_pattern}
+ fi
+ ;;
+
+ *)
+ if echo "${with_fpu}" | grep -E "^${fpu_pattern}$"; then
+ : # OK
+ else
+ echo "${with_abi}/${with_abiext} ABI cannot be implemented with" \
+ "--with-fpu=${with_fpu}." 1>&2
+ exit 1
+ fi
+ ;;
+ esac
+
+
+ # Inferring default with_tune from with_arch: pass 1
+ case ${with_arch} in
+ native)
+ tune_pattern="*"
+ tune_default="native"
+ ;;
+ loongarch64)
+ tune_pattern="loongarch64 | la464"
+ tune_default="la464"
+ ;;
+ *)
+ # By default, $with_tune == $with_arch
+ tune_pattern="$with_arch"
+ ;;
+ esac
+
+ ## Setting default value for with_tune.
+ case ${with_tune} in
+ "")
+ if test x${tune_default} != x; then
+ with_tune=${tune_default}
+ else
+ with_tune=${tune_pattern}
+ fi
+ ;;
+
+ *)
+ if echo "${with_tune}" | grep -E "^${tune_pattern}$"; then
+ : # OK
+ else
+ echo "Incompatible options: --with-tune=${with_tune}" \
+ "and --with-arch=${with_arch}." 1>&2
+ exit 1
+ fi
+ ;;
+ esac
+
+
+ # Perform final sanity checks.
+ case ${with_arch} in
+ loongarch64 | la464) ;; # OK, append here.
+ native)
+ if test x${host} != x${target}; then
+ echo "--with-arch=native is illegal for cross-compiler." 1>&2
+ exit 1
+ fi
+ ;;
+ "")
+ echo "Please set a default value for \${with_arch}" \
+ "according to your target triplet \"${target}\"." 1>&2
+ exit 1
+ ;;
+ *)
+ echo "Unknown arch in --with-arch=$with_arch" 1>&2
+ exit 1
+ ;;
+ esac
+
+ case ${with_abi} in
+ lp64d | lp64f | lp64s) ;; # OK, append here.
+ *)
+ echo "Unsupported ABI given in --with-abi=$with_abi" 1>&2
+ exit 1
+ ;;
+ esac
+
+ case ${with_abiext} in
+ base) ;; # OK, append here.
+ *)
+ echo "Unsupported ABI extention type $with_abiext" 1>&2
+ exit 1
+ ;;
+ esac
+
+ case ${with_fpu} in
+ none | 32 | 64) ;; # OK, append here.
+ *)
+ echo "Unknown fpu type in --with-fpu=$with_fpu" 1>&2
+ exit 1
+ ;;
+ esac
+
+ # Handle --with-multilib-list.
+ if test x${with_multilib_list} == x \
+ || test x${with_multilib_list} == xno \
+ || test x${with_multilib_list} == xdefault \
+ || test x${enable_multilib} != xyes; then
+
+ with_multilib_list="${with_abi}/${with_abiext}"
+ fi
+
+ # Check if the configured default ABI combination is included in
+ # ${with_multilib_list}.
+ loongarch_multilib_list_sane=no
+
+ # This one goes to TM_MULTILIB_CONFIG, for use in t-linux.
+ loongarch_multilib_list_make=""
+
+ # This one goes to tm_defines, for use in loongarch-driver.c.
+ loongarch_multilib_list_c=""
+
+ for e in $(tr ',' ' ' <<< "${with_multilib_list}"); do
+ e=($(tr '/' ' ' <<< "$e"))
+
+ # Base ABI type
+ case ${e[0]} in
+ lp64d) loongarch_multilib_list_c+="ABI_BASE_LP64D,";;
+ lp64f) loongarch_multilib_list_c+="ABI_BASE_LP64F,";;
+ lp64s) loongarch_multilib_list_c+="ABI_BASE_LP64S,";;
+ *)
+ echo "Unknown base ABI \"${e[0]}\" in --with-multilib-list." 1>&2
+ exit 1
+ ;;
+ esac
+ loongarch_multilib_list_make+="mabi=${e[0]}"
+
+ # ABI extension type
+ case ${e[1]} in
+ "" | base)
+ loongarch_multilib_list_make+=""
+ loongarch_multilib_list_c+="ABI_EXT_BASE,"
+ e[1]="base"
+ ;;
+ *)
+ echo "Unknown ABI extension \"${e[1]}\" in --with-multilib-list." 1>&2
+ exit 1
+ ;;
+ esac
+
+ case ${e[2]} in
+ "") ;; # OK
+ *)
+ echo "Unknown ABI in --with-multilib-list." 1>&2
+ exit 1
+ ;;
+ esac
+
+ if test x${with_abi} != x && test x${with_abiext} != x; then
+ if test x${e[0]} == x${with_abi} \
+ && test x${e[1]} == x${with_abiext}; then
+ loongarch_multilib_list_sane=yes
+ fi
+ fi
+
+ loongarch_multilib_list_make+=","
+ done
+
+ # Check if the default ABI combination is in the default list.
+ if test x${loongarch_multilib_list_sane} == xno; then
+ if test x${with_abiext} == xbase; then
+ with_abiext=""
+ else
+ with_abiext="/${with_abiext}"
+ fi
+
+ echo "Default ABI combination (${with_abi}${with_abiext})" \
+ "not found in --with-multilib-list." 1>&2
+ exit 1
+ fi
+
+ # Remove the excessive appending comma.
+ loongarch_multilib_list_c=${loongarch_multilib_list_c:0:-1}
+ loongarch_multilib_list_make=${loongarch_multilib_list_make:0:-1}
+ ;;
+
nds32*-*-*)
supported_defaults="arch cpu nds32_lib float fpu_config"
@@ -5353,6 +5706,51 @@ case ${target} in
tmake_file="mips/t-mips $tmake_file"
;;
+ loongarch*-*-*)
+ # Export canonical triplet.
+ tm_defines+=" LA_MULTIARCH_TRIPLET=${la_canonical_triplet}"
+
+ # Define macro __DISABLE_MULTILIB if --disable-multilib
+ tm_defines+=" TM_MULTILIB_LIST=${loongarch_multilib_list_c}"
+ if test x$enable_multilib == xyes; then
+ TM_MULTILIB_CONFIG="${loongarch_multilib_list_make}"
+ else
+ tm_defines+=" __DISABLE_MULTILIB"
+ fi
+
+ # Let --with- flags initialize the enum variables from loongarch.opt.
+ # See macro definitions from loongarch-opts.h and loongarch-cpu.h.
+ case ${with_arch} in
+ native) tm_defines+=" DEFAULT_CPU_ARCH=CPU_NATIVE" ;;
+ la464) tm_defines+=" DEFAULT_CPU_ARCH=CPU_LA464" ;;
+ loongarch64) tm_defines+=" DEFAULT_CPU_ARCH=CPU_LOONGARCH64" ;;
+ esac
+
+ case ${with_tune} in
+ native) tm_defines+=" DEFAULT_CPU_TUNE=CPU_NATIVE" ;;
+ la464) tm_defines+=" DEFAULT_CPU_TUNE=CPU_LA464" ;;
+ loongarch64) tm_defines+=" DEFAULT_CPU_TUNE=CPU_LOONGARCH64" ;;
+ esac
+
+ case ${with_abi} in
+ lp64d) tm_defines+=" DEFAULT_ABI_BASE=ABI_BASE_LP64D" ;;
+ lp64f) tm_defines+=" DEFAULT_ABI_BASE=ABI_BASE_LP64F" ;;
+ lp64s) tm_defines+=" DEFAULT_ABI_BASE=ABI_BASE_LP64S" ;;
+ esac
+
+ case ${with_abiext} in
+ base) tm_defines+=" DEFAULT_ABI_EXT=ABI_EXT_BASE" ;;
+ esac
+
+ case ${with_fpu} in
+ none) tm_defines="$tm_defines DEFAULT_ISA_EXT_FPU=ISA_EXT_NOFPU" ;;
+ 32) tm_defines="$tm_defines DEFAULT_ISA_EXT_FPU=ISA_EXT_FPU32" ;;
+ 64) tm_defines="$tm_defines DEFAULT_ISA_EXT_FPU=ISA_EXT_FPU64" ;;
+ esac
+
+ tmake_file="loongarch/t-loongarch $tmake_file"
+ ;;
+
powerpc*-*-* | rs6000-*-*)
# FIXME: The PowerPC port uses the value set at compile time,
# although it's only cosmetic.
diff --git a/gcc/config/loongarch/genopts/genstr.sh b/gcc/config/loongarch/genopts/genstr.sh
new file mode 100755
index 00000000000..62a0a0770a2
--- /dev/null
+++ b/gcc/config/loongarch/genopts/genstr.sh
@@ -0,0 +1,91 @@
+#!/bin/sh
+# A simple script that generates loongarch-str.h and loongarch.opt
+# from genopt/loongarch-optstr.
+#
+# Copyright (C) 2021 Free Software Foundation, Inc.
+#
+# This file is part of GCC.
+#
+# GCC is free software; you can redistribute it and/or modify it under
+# the terms of the GNU General Public License as published by the Free
+# Software Foundation; either version 3, or (at your option) any later
+# version.
+#
+# GCC is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+# License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3. If not see
+# <http://www.gnu.org/licenses/>.
+
+cd "$(dirname "$0")"
+
+# Generate a header containing definitions from the string table.
+gen_defines() {
+ cat <<EOF
+/* Generated automatically by "genstr" from "loongarch-strings".
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_STR_H
+#define LOONGARCH_STR_H
+EOF
+
+ sed -e '/^$/n' -e 's@#.*$@@' -e '/^$/d' \
+ -e 's@^\([^ \t]\+\)[ \t]*\([^ \t]*\)@#define \1 "\2"@' \
+ loongarch-strings
+
+ echo
+ echo "#endif /* LOONGARCH_STR_H */"
+}
+
+
+# Substitute all "@@<KEY>@@" to "<VALUE>" in loongarch.opt.in
+# according to the key-value pairs defined in loongarch-strings.
+
+gen_options() {
+
+ sed -e '/^$/n' -e 's@#.*$@@' -e '/^$/d' \
+ -e 's@^\([^ \t]\+\)[ \t]*\([^ \t]*\)@\1="\2"@' \
+ loongarch-strings | { \
+
+ # read the definitions
+ while read -r line; do
+ eval "$line"
+ done
+
+ # make the substitutions
+ sed -e 's@"@\\"@g' -e 's/@@\([^@]\+\)@@/${\1}/g' loongarch.opt.in | \
+ while read -r line; do
+ eval "echo \"$line\""
+ done
+ }
+}
+
+main() {
+ case "$1" in
+ header) gen_defines;;
+ opt) gen_options;;
+ *) echo "Unknown Command: \"$1\". Available: header, opt"; exit 1;;
+ esac
+}
+
+main "$@"
diff --git a/gcc/config/loongarch/genopts/loongarch-strings b/gcc/config/loongarch/genopts/loongarch-strings
new file mode 100644
index 00000000000..553304f9b83
--- /dev/null
+++ b/gcc/config/loongarch/genopts/loongarch-strings
@@ -0,0 +1,58 @@
+# Defines the key strings for LoongArch compiler options.
+#
+# Copyright (C) 2021 Free Software Foundation, Inc.
+#
+# This file is part of GCC.
+#
+# GCC is free software; you can redistribute it and/or modify it under
+# the terms of the GNU General Public License as published by the Free
+# Software Foundation; either version 3, or (at your option) any later
+# version.
+#
+# GCC is distributed in the hope that it will be useful, but WITHOUT
+# ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+# or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+# License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3. If not see
+# <http://www.gnu.org/licenses/>.
+
+# -march= / -mtune=
+OPTSTR_ARCH arch
+OPTSTR_TUNE tune
+
+STR_CPU_NATIVE native
+STR_CPU_LOONGARCH64 loongarch64
+STR_CPU_LA464 la464
+
+# Base architecture
+STR_ISA_BASE_LA64V100 la64
+
+# -mfpu
+OPTSTR_ISA_EXT_FPU fpu
+STR_ISA_EXT_NOFPU none
+STR_ISA_EXT_FPU0 0
+STR_ISA_EXT_FPU32 32
+STR_ISA_EXT_FPU64 64
+
+OPTSTR_SOFT_FLOAT soft-float
+OPTSTR_SINGLE_FLOAT single-float
+OPTSTR_DOUBLE_FLOAT double-float
+
+# -mabi=
+OPTSTR_ABI_BASE abi
+STR_ABI_BASE_LP64D lp64d
+STR_ABI_BASE_LP64F lp64f
+STR_ABI_BASE_LP64S lp64s
+
+# ABI extension types
+STR_ABI_EXT_BASE base
+
+# -mcmodel=
+OPTSTR_CMODEL cmodel
+STR_CMODEL_NORMAL normal
+STR_CMODEL_TINY tiny
+STR_CMODEL_TS tiny-static
+STR_CMODEL_LARGE large
+STR_CMODEL_EXTREME extreme
diff --git a/gcc/config/loongarch/genopts/loongarch.opt.in b/gcc/config/loongarch/genopts/loongarch.opt.in
new file mode 100644
index 00000000000..08950a32179
--- /dev/null
+++ b/gcc/config/loongarch/genopts/loongarch.opt.in
@@ -0,0 +1,189 @@
+; Generated by "genstr" from the template "loongarch.opt.in"
+; and definitions from "loongarch-strings".
+;
+; Copyright (C) 2021 Free Software Foundation, Inc.
+;
+; This file is part of GCC.
+;
+; GCC is free software; you can redistribute it and/or modify it under
+; the terms of the GNU General Public License as published by the Free
+; Software Foundation; either version 3, or (at your option) any later
+; version.
+;
+; GCC is distributed in the hope that it will be useful, but WITHOUT
+; ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+; License for more details.
+;
+; You should have received a copy of the GNU General Public License
+; along with GCC; see the file COPYING3. If not see
+; <http://www.gnu.org/licenses/>.
+;
+
+; Variables (macros) that should be exported by loongarch.opt:
+; la_opt_switches,
+; la_opt_abi_base, la_opt_abi_ext,
+; la_opt_cpu_arch, la_opt_cpu_tune,
+; la_opt_fpu,
+; la_cmodel.
+
+HeaderInclude
+config/loongarch/loongarch-opts.h
+
+HeaderInclude
+config/loongarch/loongarch-str.h
+
+Variable
+HOST_WIDE_INT la_opt_switches = 0
+
+; ISA related options
+;; Base ISA
+Enum
+Name(isa_base) Type(int)
+Basic ISAs of LoongArch:
+
+EnumValue
+Enum(isa_base) String(@@STR_ISA_BASE_LA64V100@@) Value(ISA_BASE_LA64V100)
+
+
+;; ISA extensions / adjustments
+Enum
+Name(isa_ext_fpu) Type(int)
+FPU types of LoongArch:
+
+EnumValue
+Enum(isa_ext_fpu) String(@@STR_ISA_EXT_NOFPU@@) Value(ISA_EXT_NOFPU)
+
+EnumValue
+Enum(isa_ext_fpu) String(@@STR_ISA_EXT_FPU32@@) Value(ISA_EXT_FPU32)
+
+EnumValue
+Enum(isa_ext_fpu) String(@@STR_ISA_EXT_FPU64@@) Value(ISA_EXT_FPU64)
+
+m@@OPTSTR_ISA_EXT_FPU@@=
+Target RejectNegative Joined ToLower Enum(isa_ext_fpu) Var(la_opt_fpu) Init(M_OPTION_NOT_SEEN)
+-m@@OPTSTR_ISA_EXT_FPU@@=FPU Generate code for the given FPU.
+
+m@@OPTSTR_ISA_EXT_FPU@@=@@STR_ISA_EXT_FPU0@@
+Target RejectNegative Alias(m@@OPTSTR_ISA_EXT_FPU@@=,@@STR_ISA_EXT_NOFPU@@)
+
+m@@OPTSTR_SOFT_FLOAT@@
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_SOFTF) Negative(m@@OPTSTR_SINGLE_FLOAT@@)
+Prevent the use of all hardware floating-point instructions.
+
+m@@OPTSTR_SINGLE_FLOAT@@
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_F32) Negative(m@@OPTSTR_DOUBLE_FLOAT@@)
+Restrict the use of hardware floating-point instructions to 32-bit operations.
+
+m@@OPTSTR_DOUBLE_FLOAT@@
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_F64) Negative(m@@OPTSTR_SOFT_FLOAT@@)
+Allow hardware floating-point instructions to cover both 32-bit and 64-bit operations.
+
+
+;; Base target models (implies ISA & tune parameters)
+Enum
+Name(cpu_type) Type(int)
+LoongArch CPU types:
+
+EnumValue
+Enum(cpu_type) String(@@STR_CPU_NATIVE@@) Value(CPU_NATIVE)
+
+EnumValue
+Enum(cpu_type) String(@@STR_CPU_LOONGARCH64@@) Value(CPU_LOONGARCH64)
+
+EnumValue
+Enum(cpu_type) String(@@STR_CPU_LA464@@) Value(CPU_LA464)
+
+m@@OPTSTR_ARCH@@=
+Target RejectNegative Joined Enum(cpu_type) Var(la_opt_cpu_arch) Init(M_OPTION_NOT_SEEN)
+-m@@OPTSTR_ARCH@@=PROCESSOR Generate code for the given PROCESSOR ISA.
+
+m@@OPTSTR_TUNE@@=
+Target RejectNegative Joined Enum(cpu_type) Var(la_opt_cpu_tune) Init(M_OPTION_NOT_SEEN)
+-m@@OPTSTR_TUNE@@=PROCESSOR Generate optimized code for PROCESSOR.
+
+
+; ABI related options
+; (ISA constraints on ABI are handled dynamically)
+
+;; Base ABI
+Enum
+Name(abi_base) Type(int)
+Base ABI types for LoongArch:
+
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_LP64D@@) Value(ABI_BASE_LP64D)
+
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_LP64F@@) Value(ABI_BASE_LP64F)
+
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_LP64S@@) Value(ABI_BASE_LP64S)
+
+m@@OPTSTR_ABI_BASE@@=
+Target RejectNegative Joined ToLower Enum(abi_base) Var(la_opt_abi_base) Init(M_OPTION_NOT_SEEN)
+-m@@OPTSTR_ABI_BASE@@=BASEABI Generate code that conforms to the given BASEABI.
+
+;; ABI Extension
+Variable
+int la_opt_abi_ext = M_OPTION_NOT_SEEN
+
+
+mbranch-cost=
+Target RejectNegative Joined UInteger Var(loongarch_branch_cost)
+-mbranch-cost=COST Set the cost of branches to roughly COST instructions.
+
+mcheck-zero-division
+Target Mask(CHECK_ZERO_DIV)
+Trap on integer divide by zero.
+
+mcond-move-int
+Target Var(TARGET_COND_MOVE_INT) Init(1)
+Conditional moves for integral are enabled.
+
+mcond-move-float
+Target Var(TARGET_COND_MOVE_FLOAT) Init(1)
+Conditional moves for float are enabled.
+
+mmemcpy
+Target Mask(MEMCPY)
+Don't optimize block moves.
+
+mlra
+Target Var(loongarch_lra_flag) Init(1) Save
+Use LRA instead of reload.
+
+noasmopt
+Driver
+
+mstrict-align
+Target Mask(STRICT_ALIGN) Save
+Do not generate unaligned memory accesses.
+
+mmax-inline-memcpy-size=
+Target Joined RejectNegative UInteger Var(loongarch_max_inline_memcpy_size) Init(1024)
+-mmax-inline-memcpy-size=SIZE Set the max size of memcpy to inline, default is 1024.
+
+; The code model option names for -mcmodel.
+Enum
+Name(cmodel) Type(int)
+The code model option names for -mcmodel:
+
+EnumValue
+Enum(cmodel) String(@@STR_CMODEL_NORMAL@@) Value(CMODEL_NORMAL)
+
+EnumValue
+Enum(cmodel) String(@@STR_CMODEL_TINY@@) Value(CMODEL_TINY)
+
+EnumValue
+Enum(cmodel) String(@@STR_CMODEL_TS@@) Value(CMODEL_TINY_STATIC)
+
+EnumValue
+Enum(cmodel) String(@@STR_CMODEL_LARGE@@) Value(CMODEL_LARGE)
+
+EnumValue
+Enum(cmodel) String(@@STR_CMODEL_EXTREME@@) Value(CMODEL_EXTREME)
+
+mcmodel=
+Target RejectNegative Joined Enum(cmodel) Var(la_opt_cmodel) Init(CMODEL_NORMAL)
+Specify the code model.
diff --git a/gcc/config/loongarch/gnu-user.h b/gcc/config/loongarch/gnu-user.h
new file mode 100644
index 00000000000..6f3137f7eef
--- /dev/null
+++ b/gcc/config/loongarch/gnu-user.h
@@ -0,0 +1,78 @@
+/* Definitions for LoongArch systems using GNU userspace.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+/* Define the size of the wide character type. */
+#undef WCHAR_TYPE
+#define WCHAR_TYPE "int"
+
+#undef WCHAR_TYPE_SIZE
+#define WCHAR_TYPE_SIZE 32
+
+
+/* GNU-specific SPEC definitions. */
+#define GNU_USER_LINK_EMULATION "elf" ABI_GRLEN_SPEC "loongarch"
+
+#define GLIBC_DYNAMIC_LINKER \
+ "/lib" ABI_GRLEN_SPEC "/ld-linux-loongarch-" ABI_SPEC ".so.1"
+
+#undef GNU_USER_TARGET_LINK_SPEC
+#define GNU_USER_TARGET_LINK_SPEC \
+ "%{G*} %{shared} -m " GNU_USER_LINK_EMULATION \
+ "%{!shared: %{static} %{!static: %{rdynamic:-export-dynamic} " \
+ "-dynamic-linker " GLIBC_DYNAMIC_LINKER "}}"
+
+
+/* Similar to standard Linux, but adding -ffast-math support. */
+#undef GNU_USER_TARGET_MATHFILE_SPEC
+#define GNU_USER_TARGET_MATHFILE_SPEC \
+ "%{Ofast|ffast-math|funsafe-math-optimizations:crtfastmath.o%s}"
+
+#undef LIB_SPEC
+#define LIB_SPEC GNU_USER_TARGET_LIB_SPEC
+
+#undef LINK_SPEC
+#define LINK_SPEC GNU_USER_TARGET_LINK_SPEC
+
+#undef ENDFILE_SPEC
+#define ENDFILE_SPEC \
+ GNU_USER_TARGET_MATHFILE_SPEC " " \
+ GNU_USER_TARGET_ENDFILE_SPEC
+
+#undef SUBTARGET_CPP_SPEC
+#define SUBTARGET_CPP_SPEC "%{posix:-D_POSIX_SOURCE} %{pthread:-D_REENTRANT}"
+
+/* A standard GNU/Linux mapping. On most targets, it is included in
+ CC1_SPEC itself by config/linux.h, but loongarch.h overrides CC1_SPEC
+ and provides this hook instead. */
+#undef SUBTARGET_CC1_SPEC
+#define SUBTARGET_CC1_SPEC GNU_USER_TARGET_CC1_SPEC
+
+#define TARGET_OS_CPP_BUILTINS() \
+ do \
+ { \
+ GNU_USER_TARGET_OS_CPP_BUILTINS (); \
+ /* The GNU C++ standard library requires this. */ \
+ if (c_dialect_cxx ()) \
+ builtin_define ("_GNU_SOURCE"); \
+ } \
+ while (0)
+
+
+#undef ASM_DECLARE_OBJECT_NAME
+#define ASM_DECLARE_OBJECT_NAME loongarch_declare_object_name
diff --git a/gcc/config/loongarch/linux.h b/gcc/config/loongarch/linux.h
new file mode 100644
index 00000000000..453429fda97
--- /dev/null
+++ b/gcc/config/loongarch/linux.h
@@ -0,0 +1,50 @@
+/* Definitions for Linux-based systems with libraries in ELF format.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+/* Default system library search paths.
+ * This ensures that a compiler configured with --disable-multilib
+ * can work in an multilib environment. */
+
+#if defined(__DISABLE_MULTILIB) && defined(__DISABLE_MULTIARCH)
+
+ #if DEFAULT_ABI_INT == ABI_BASE_LP64D
+ #define ABI_LIBDIR "lib64"
+ #elif DEFAULT_ABI_INT == ABI_BASE_LP64F
+ #define ABI_LIBDIR "lib64/f32"
+ #elif DEFAULT_ABI_INT == ABI_BASE_LP64S
+ #define ABI_LIBDIR "lib64/sf"
+ #endif
+
+#endif
+
+#ifndef ABI_LIBDIR
+#define ABI_LIBDIR "lib"
+#endif
+
+#define STANDARD_STARTFILE_PREFIX_1 "/" ABI_LIBDIR "/"
+#define STANDARD_STARTFILE_PREFIX_2 "/usr/" ABI_LIBDIR "/"
+
+
+/* Define this to be nonzero if static stack checking is supported. */
+#define STACK_CHECK_STATIC_BUILTIN 1
+
+/* The default value isn't sufficient in 64-bit mode. */
+#define STACK_CHECK_PROTECT (TARGET_64BIT ? 16 * 1024 : 12 * 1024)
+
+#define TARGET_ASM_FILE_END file_end_indicate_exec_stack
diff --git a/gcc/config/loongarch/loongarch-cpu.c b/gcc/config/loongarch/loongarch-cpu.c
new file mode 100644
index 00000000000..384fb54392c
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-cpu.c
@@ -0,0 +1,206 @@
+/* Definitions for LoongArch CPU properties.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "diagnostic-core.h"
+
+#include "loongarch-opts.h"
+#include "loongarch-cpu.h"
+#include "loongarch-str.h"
+
+/* Native CPU detection with "cpucfg" */
+#define N_CPUCFG_WORDS 0x15
+static uint32_t cpucfg_cache[N_CPUCFG_WORDS] = { 0 };
+static const int cpucfg_useful_idx[] = {0, 1, 2, 16, 17, 18, 19};
+
+static uint32_t
+read_cpucfg_word (int wordno)
+{
+ /* To make cross-compiler shut up. */
+ (void) wordno;
+ uint32_t ret = 0;
+
+ #ifdef __loongarch__
+ __asm__ __volatile__ ("cpucfg %0,%1\n\t"
+ :"=r"(ret)
+ :"r"(wordno)
+ :);
+ #endif
+
+ return ret;
+}
+
+void
+cache_cpucfg (void)
+{
+ for (unsigned int i = 0; i < sizeof (cpucfg_useful_idx) / sizeof (int); i++)
+ {
+ cpucfg_cache[cpucfg_useful_idx[i]]
+ = read_cpucfg_word (cpucfg_useful_idx[i]);
+ }
+}
+
+uint32_t
+get_native_prid (void)
+{
+ /* Fill loongarch_cpu_default_config[CPU_NATIVE] with cpucfg data,
+ see "Loongson Architecture Reference Manual"
+ (Volume 1, Section 2.2.10.5) */
+ return cpucfg_cache[0];
+}
+
+const char*
+get_native_prid_str (void)
+{
+ static char prid_str[9];
+ sprintf (prid_str, "%08x", cpucfg_cache[0]);
+ return (const char*) prid_str;
+}
+
+/* Fill property tables for CPU_NATIVE. */
+unsigned int
+fill_native_cpu_config (int p_arch_native, int p_tune_native)
+{
+ int ret_cpu_type;
+
+ /* Nothing needs to be done unless "-march/tune=native"
+ is given or implied. */
+ if (!(p_arch_native || p_tune_native))
+ return CPU_NATIVE;
+
+ /* Fill cpucfg_cache with the "cpucfg" instruction. */
+ cache_cpucfg ();
+
+
+ /* Fill: loongarch_cpu_default_isa[CPU_NATIVE].base
+ With: base architecture (ARCH)
+ At: cpucfg_words[1][1:0] */
+
+ #define NATIVE_BASE_ISA (loongarch_cpu_default_isa[CPU_NATIVE].base)
+ switch (cpucfg_cache[1] & 0x3)
+ {
+ case 0x02:
+ NATIVE_BASE_ISA = ISA_BASE_LA64V100;
+ break;
+
+ default:
+ if (p_arch_native)
+ fatal_error (UNKNOWN_LOCATION,
+ "unknown base architecture %<0x%x%>, %qs failed",
+ (unsigned int) (cpucfg_cache[1] & 0x3),
+ "-m" OPTSTR_ARCH "=" STR_CPU_NATIVE);
+ }
+
+ /* Fill: loongarch_cpu_default_isa[CPU_NATIVE].fpu
+ With: FPU type (FP, FP_SP, FP_DP)
+ At: cpucfg_words[2][2:0] */
+
+ #define NATIVE_FPU (loongarch_cpu_default_isa[CPU_NATIVE].fpu)
+ switch (cpucfg_cache[2] & 0x7)
+ {
+ case 0x07:
+ NATIVE_FPU = ISA_EXT_FPU64;
+ break;
+
+ case 0x03:
+ NATIVE_FPU = ISA_EXT_FPU32;
+ break;
+
+ case 0x00:
+ NATIVE_FPU = ISA_EXT_NOFPU;
+ break;
+
+ default:
+ if (p_arch_native)
+ fatal_error (UNKNOWN_LOCATION,
+ "unknown FPU type %<0x%x%>, %qs failed",
+ (unsigned int) (cpucfg_cache[2] & 0x7),
+ "-m" OPTSTR_ARCH "=" STR_CPU_NATIVE);
+ }
+
+ /* Fill: loongarch_cpu_cache[CPU_NATIVE]
+ With: cache size info
+ At: cpucfg_words[16:20][31:0] */
+
+ int l1d_present = 0, l1u_present = 0;
+ int l2d_present = 0;
+ uint32_t l1_szword, l2_szword;
+
+ l1u_present |= cpucfg_cache[16] & 3; /* bit[1:0]: unified l1 cache */
+ l1d_present |= cpucfg_cache[16] & 4; /* bit[2:2]: l1 dcache */
+ l1_szword = l1d_present ? 18 : (l1u_present ? 17 : 0);
+ l1_szword = l1_szword ? cpucfg_cache[l1_szword]: 0;
+
+ l2d_present |= cpucfg_cache[16] & 24; /* bit[4:3]: unified l2 cache */
+ l2d_present |= cpucfg_cache[16] & 128; /* bit[7:7]: l2 dcache */
+ l2_szword = l2d_present ? cpucfg_cache[19]: 0;
+
+ loongarch_cpu_cache[CPU_NATIVE].l1d_line_size
+ = 1 << ((l1_szword & 0x7f000000) >> 24); /* bit[30:24]: log2(linesize) */
+
+ loongarch_cpu_cache[CPU_NATIVE].l1d_size
+ = (1 << ((l1_szword & 0x00ff0000) >> 16)) /* bit[23:16]: log2(idx) */
+ * ((l1_szword & 0x0000ffff) + 1) /* bit[15:0]: sets - 1 */
+ * (1 << ((l1_szword & 0x7f000000) >> 24)) /* bit[30:24]: log2(linesize) */
+ >> 10; /* in kilobytes */
+
+ loongarch_cpu_cache[CPU_NATIVE].l2d_size
+ = (1 << ((l2_szword & 0x00ff0000) >> 16)) /* bit[23:16]: log2(idx) */
+ * ((l2_szword & 0x0000ffff) + 1) /* bit[15:0]: sets - 1 */
+ * (1 << ((l2_szword & 0x7f000000) >> 24)) /* bit[30:24]: log2(linesize) */
+ >> 10; /* in kilobytes */
+
+ /* Fill: ret_cpu_type
+ With: processor ID (PRID)
+ At: cpucfg_words[0][31:0] */
+
+ switch (cpucfg_cache[0] & 0x00ffff00)
+ {
+ case 0x0014c000: /* LA464 */
+ ret_cpu_type = CPU_LA464;
+ break;
+
+ default:
+ /* Unknown PRID. This is generally harmless as long as
+ the properties above can be obtained via "cpucfg". */
+ if (p_tune_native)
+ inform (UNKNOWN_LOCATION, "unknown processor ID %<0x%x%>, "
+ "some tuning parameters will fall back to default",
+ cpucfg_cache[0]);
+ break;
+ }
+
+ /* Properties that cannot be looked up directly using cpucfg. */
+ loongarch_cpu_issue_rate[CPU_NATIVE]
+ = loongarch_cpu_issue_rate[ret_cpu_type];
+
+ loongarch_cpu_multipass_dfa_lookahead[CPU_NATIVE]
+ = loongarch_cpu_multipass_dfa_lookahead[ret_cpu_type];
+
+ loongarch_cpu_rtx_cost_data[CPU_NATIVE]
+ = loongarch_cpu_rtx_cost_data[ret_cpu_type];
+
+ return ret_cpu_type;
+}
diff --git a/gcc/config/loongarch/loongarch-cpu.h b/gcc/config/loongarch/loongarch-cpu.h
new file mode 100644
index 00000000000..93d656f70f5
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-cpu.h
@@ -0,0 +1,30 @@
+/* Definitions for loongarch native cpu property detection routines.
+ Copyright (C) 2020-2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_CPU_H
+#define LOONGARCH_CPU_H
+
+#include "system.h"
+
+void cache_cpucfg (void);
+unsigned int fill_native_cpu_config (int p_arch_native, int p_tune_native);
+uint32_t get_native_prid (void);
+const char* get_native_prid_str (void);
+
+#endif /* LOONGARCH_CPU_H */
diff --git a/gcc/config/loongarch/loongarch-def.c b/gcc/config/loongarch/loongarch-def.c
new file mode 100644
index 00000000000..bc22a55edb0
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-def.c
@@ -0,0 +1,164 @@
+/* LoongArch static properties.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#include "loongarch-def.h"
+#include "loongarch-str.h"
+
+/* CPU property tables. */
+const char*
+loongarch_cpu_strings[N_TUNE_TYPES] = {
+ [CPU_NATIVE] = STR_CPU_NATIVE,
+ [CPU_LOONGARCH64] = STR_CPU_LOONGARCH64,
+ [CPU_LA464] = STR_CPU_LA464,
+};
+
+struct loongarch_isa
+loongarch_cpu_default_isa[N_ARCH_TYPES] = {
+ [CPU_LOONGARCH64] = {
+ .base = ISA_BASE_LA64V100,
+ .fpu = ISA_EXT_FPU64,
+ },
+ [CPU_LA464] = {
+ .base = ISA_BASE_LA64V100,
+ .fpu = ISA_EXT_FPU64,
+ },
+};
+
+struct loongarch_cache
+loongarch_cpu_cache[N_TUNE_TYPES] = {
+ [CPU_LOONGARCH64] = {
+ .l1d_line_size = 64,
+ .l1d_size = 64,
+ .l2d_size = 256,
+ },
+ [CPU_LA464] = {
+ .l1d_line_size = 64,
+ .l1d_size = 64,
+ .l2d_size = 256,
+ },
+};
+
+/* The following properties cannot be looked up directly using "cpucfg".
+ So it is necessary to provide a default value for "unknown native"
+ tune targets (i.e. -mtune=native while PRID does not correspond to
+ any known "-mtune" type). */
+
+struct loongarch_rtx_cost_data
+loongarch_cpu_rtx_cost_data[N_TUNE_TYPES] = {
+ [CPU_NATIVE] = {
+ DEFAULT_COSTS
+ },
+ [CPU_LOONGARCH64] = {
+ DEFAULT_COSTS
+ },
+ [CPU_LA464] = {
+ DEFAULT_COSTS
+ },
+};
+
+/* RTX costs to use when optimizing for size. */
+extern const struct loongarch_rtx_cost_data
+loongarch_rtx_cost_optimize_size = {
+ .fp_add = 4,
+ .fp_mult_sf = 4,
+ .fp_mult_df = 4,
+ .fp_div_sf = 4,
+ .fp_div_df = 4,
+ .int_mult_si = 4,
+ .int_mult_di = 4,
+ .int_div_si = 4,
+ .int_div_di = 4,
+ .branch_cost = 2,
+ .memory_latency = 4,
+};
+
+int
+loongarch_cpu_issue_rate[N_TUNE_TYPES] = {
+ [CPU_NATIVE] = 4,
+ [CPU_LOONGARCH64] = 4,
+ [CPU_LA464] = 4,
+};
+
+int
+loongarch_cpu_multipass_dfa_lookahead[N_TUNE_TYPES] = {
+ [CPU_NATIVE] = 4,
+ [CPU_LOONGARCH64] = 4,
+ [CPU_LA464] = 4,
+};
+
+/* Wiring string definitions from loongarch-str.h to global arrays
+ with standard index values from loongarch-opts.h, so we can
+ print config-related messages and do ABI self-spec filtering
+ from the driver in a self-consistent manner. */
+
+const char*
+loongarch_isa_base_strings[N_ISA_BASE_TYPES] = {
+ [ISA_BASE_LA64V100] = STR_ISA_BASE_LA64V100,
+};
+
+const char*
+loongarch_isa_ext_strings[N_ISA_EXT_TYPES] = {
+ [ISA_EXT_FPU64] = STR_ISA_EXT_FPU64,
+ [ISA_EXT_FPU32] = STR_ISA_EXT_FPU32,
+ [ISA_EXT_NOFPU] = STR_ISA_EXT_NOFPU,
+};
+
+const char*
+loongarch_abi_base_strings[N_ABI_BASE_TYPES] = {
+ [ABI_BASE_LP64D] = STR_ABI_BASE_LP64D,
+ [ABI_BASE_LP64F] = STR_ABI_BASE_LP64F,
+ [ABI_BASE_LP64S] = STR_ABI_BASE_LP64S,
+};
+
+const char*
+loongarch_abi_ext_strings[N_ABI_EXT_TYPES] = {
+ [ABI_EXT_BASE] = STR_ABI_EXT_BASE,
+};
+
+const char*
+loongarch_cmodel_strings[] = {
+ [CMODEL_NORMAL] = STR_CMODEL_NORMAL,
+ [CMODEL_TINY] = STR_CMODEL_TINY,
+ [CMODEL_TINY_STATIC] = STR_CMODEL_TS,
+ [CMODEL_LARGE] = STR_CMODEL_LARGE,
+ [CMODEL_EXTREME] = STR_CMODEL_EXTREME,
+};
+
+const char*
+loongarch_switch_strings[] = {
+ [SW_SOFT_FLOAT] = OPTSTR_SOFT_FLOAT,
+ [SW_SINGLE_FLOAT] = OPTSTR_SINGLE_FLOAT,
+ [SW_DOUBLE_FLOAT] = OPTSTR_DOUBLE_FLOAT,
+};
+
+
+/* ABI-related definitions. */
+const struct loongarch_isa
+abi_minimal_isa[N_ABI_BASE_TYPES][N_ABI_EXT_TYPES] = {
+ [ABI_BASE_LP64D] = {
+ [ABI_EXT_BASE] = {.base = ISA_BASE_LA64V100, .fpu = ISA_EXT_FPU64},
+ },
+ [ABI_BASE_LP64F] = {
+ [ABI_EXT_BASE] = {.base = ISA_BASE_LA64V100, .fpu = ISA_EXT_FPU32},
+ },
+ [ABI_BASE_LP64S] = {
+ [ABI_EXT_BASE] = {.base = ISA_BASE_LA64V100, .fpu = ISA_EXT_NOFPU},
+ },
+};
diff --git a/gcc/config/loongarch/loongarch-def.h b/gcc/config/loongarch/loongarch-def.h
new file mode 100644
index 00000000000..84c27271a69
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-def.h
@@ -0,0 +1,151 @@
+/* LoongArch definitions.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+/* Definition of standard codes for:
+ - base architecture types (isa_base),
+ - ISA extensions (isa_ext),
+ - base ABI types (abi_base),
+ - ABI extension types (abi_ext).
+
+ - code models (cmodel)
+ - other command-line switches (switch)
+
+ These values are primarily used for implementing option handling
+ logic in "loongarch.opt", "loongarch-driver.c" and "loongarch-opt.c".
+
+ As for the result of this option handling process, the following
+ scheme is adopted to represent the final configuration:
+
+ - The target ABI is encoded with a tuple (abi_base, abi_ext)
+ using the code defined below.
+
+ - The target ISA is encoded with a "struct loongarch_isa" defined
+ in loongarch-cpu.h.
+
+ - The target microarchitecture is represented with a cpu model
+ index defined in loongarch-cpu.h.
+*/
+
+#ifndef LOONGARCH_DEF_H
+#define LOONGARCH_DEF_H
+
+#include "loongarch-tune.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* enum isa_base */
+extern const char* loongarch_isa_base_strings[];
+#define ISA_BASE_LA64V100 0
+#define N_ISA_BASE_TYPES 1
+
+/* enum isa_ext_* */
+extern const char* loongarch_isa_ext_strings[];
+#define ISA_EXT_NOFPU 0
+#define ISA_EXT_FPU32 1
+#define ISA_EXT_FPU64 2
+#define N_ISA_EXT_FPU_TYPES 3
+#define N_ISA_EXT_TYPES 3
+
+/* enum abi_base */
+extern const char* loongarch_abi_base_strings[];
+#define ABI_BASE_LP64D 0
+#define ABI_BASE_LP64F 1
+#define ABI_BASE_LP64S 2
+#define N_ABI_BASE_TYPES 3
+
+/* enum abi_ext */
+extern const char* loongarch_abi_ext_strings[];
+#define ABI_EXT_BASE 0
+#define N_ABI_EXT_TYPES 1
+
+/* enum cmodel */
+extern const char* loongarch_cmodel_strings[];
+#define CMODEL_NORMAL 0
+#define CMODEL_TINY 1
+#define CMODEL_TINY_STATIC 2
+#define CMODEL_LARGE 3
+#define CMODEL_EXTREME 4
+#define N_CMODEL_TYPES 5
+
+/* enum switches */
+/* The "SW_" codes represent command-line switches (options that
+ accept no parameters). Definition for other switches that affects
+ the target ISA / ABI configuration will also be appended here
+ in the future. */
+
+extern const char* loongarch_switch_strings[];
+#define SW_SOFT_FLOAT 0
+#define SW_SINGLE_FLOAT 1
+#define SW_DOUBLE_FLOAT 2
+#define N_SWITCH_TYPES 3
+
+/* The common default value for variables whose assignments
+ are triggered by command-line options. */
+
+#define M_OPTION_NOT_SEEN -1
+#define M_OPT_ABSENT(opt_enum) ((opt_enum) == M_OPTION_NOT_SEEN)
+
+
+/* Internal representation of the target. */
+struct loongarch_isa
+{
+ unsigned char base; /* ISA_BASE_ */
+ unsigned char fpu; /* ISA_EXT_FPU_ */
+};
+
+struct loongarch_abi
+{
+ unsigned char base; /* ABI_BASE_ */
+ unsigned char ext; /* ABI_EXT_ */
+};
+
+struct loongarch_target
+{
+ struct loongarch_isa isa;
+ struct loongarch_abi abi;
+ unsigned char cpu_arch; /* CPU_ */
+ unsigned char cpu_tune; /* same */
+ unsigned char cpu_native; /* same */
+ unsigned char cmodel; /* CMODEL_ */
+};
+
+/* CPU properties. */
+/* index */
+#define CPU_NATIVE 0
+#define CPU_LOONGARCH64 1
+#define CPU_LA464 2
+#define N_ARCH_TYPES 3
+#define N_TUNE_TYPES 3
+
+/* parallel tables. */
+extern const char* loongarch_cpu_strings[];
+extern struct loongarch_isa loongarch_cpu_default_isa[];
+extern int loongarch_cpu_issue_rate[];
+extern int loongarch_cpu_multipass_dfa_lookahead[];
+
+extern struct loongarch_cache loongarch_cpu_cache[];
+extern struct loongarch_rtx_cost_data loongarch_cpu_rtx_cost_data[];
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* LOONGARCH_DEF_H */
diff --git a/gcc/config/loongarch/loongarch-driver.c b/gcc/config/loongarch/loongarch-driver.c
new file mode 100644
index 00000000000..8d78e2eb98b
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-driver.c
@@ -0,0 +1,187 @@
+/* Subroutines for the gcc driver.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "obstack.h"
+#include "diagnostic-core.h"
+
+#include "loongarch-opts.h"
+#include "loongarch-driver.h"
+
+static int
+ opt_arch_driver = M_OPTION_NOT_SEEN,
+ opt_tune_driver = M_OPTION_NOT_SEEN,
+ opt_fpu_driver = M_OPTION_NOT_SEEN,
+ opt_abi_base_driver = M_OPTION_NOT_SEEN,
+ opt_abi_ext_driver = M_OPTION_NOT_SEEN,
+ opt_cmodel_driver = M_OPTION_NOT_SEEN;
+
+int opt_switches = 0;
+
+/* This flag is set to 1 if we believe that the user might be avoiding
+ linking (implicitly) against something from the startfile search paths. */
+static int no_link = 0;
+
+#define LARCH_DRIVER_SET_M_FLAG(OPTS_ARRAY, N_OPTS, FLAG, STR) \
+ for (int i = 0; i < (N_OPTS); i++) \
+ { \
+ if ((OPTS_ARRAY)[i] != 0) \
+ if (strcmp ((STR), (OPTS_ARRAY)[i]) == 0) \
+ (FLAG) = i; \
+ }
+
+/* Use the public obstack from the gcc driver (defined in gcc.c).
+ This is for allocating space for the returned string. */
+extern struct obstack opts_obstack;
+
+#define APPEND_LTR(S) \
+ obstack_grow (&opts_obstack, (const void*) (S), \
+ sizeof ((S)) / sizeof (char) -1)
+
+#define APPEND_VAL(S) \
+ obstack_grow (&opts_obstack, (const void*) (S), strlen ((S)))
+
+
+const char*
+driver_set_m_flag (int argc, const char **argv)
+{
+ int parm_off = 0;
+
+ if (argc != 1)
+ return "%eset_m_flag requires exactly 1 argument.";
+
+#undef PARM
+#define PARM (argv[0] + parm_off)
+
+/* Note: sizeof (OPTSTR_##NAME) equals the length of "<option>=". */
+#undef MATCH_OPT
+#define MATCH_OPT(NAME) \
+ (strncmp (argv[0], OPTSTR_##NAME "=", \
+ (parm_off = sizeof (OPTSTR_##NAME))) == 0)
+
+ if (strcmp (argv[0], "no_link") == 0)
+ {
+ no_link = 1;
+ }
+ else if (MATCH_OPT (ABI_BASE))
+ {
+ LARCH_DRIVER_SET_M_FLAG (
+ loongarch_abi_base_strings, N_ABI_BASE_TYPES,
+ opt_abi_base_driver, PARM)
+ }
+ else if (MATCH_OPT (ISA_EXT_FPU))
+ {
+ LARCH_DRIVER_SET_M_FLAG (loongarch_isa_ext_strings, N_ISA_EXT_FPU_TYPES,
+ opt_fpu_driver, PARM)
+ }
+ else if (MATCH_OPT (ARCH))
+ {
+ LARCH_DRIVER_SET_M_FLAG (loongarch_cpu_strings, N_ARCH_TYPES,
+ opt_arch_driver, PARM)
+ }
+ else if (MATCH_OPT (TUNE))
+ {
+ LARCH_DRIVER_SET_M_FLAG (loongarch_cpu_strings, N_TUNE_TYPES,
+ opt_tune_driver, PARM)
+ }
+ else if (MATCH_OPT (CMODEL))
+ {
+ LARCH_DRIVER_SET_M_FLAG (loongarch_cmodel_strings, N_CMODEL_TYPES,
+ opt_cmodel_driver, PARM)
+ }
+ else /* switches */
+ {
+ int switch_idx = M_OPTION_NOT_SEEN;
+
+ LARCH_DRIVER_SET_M_FLAG (loongarch_switch_strings, N_SWITCH_TYPES,
+ switch_idx, argv[0])
+
+ if (switch_idx != M_OPTION_NOT_SEEN)
+ opt_switches |= loongarch_switch_mask[switch_idx];
+ }
+ return "";
+}
+
+const char*
+driver_get_normalized_m_opts (int argc, const char **argv)
+{
+ if (argc != 0)
+ {
+ (void) argv; /* to make compiler shut up about unused argument */
+ return " %eget_normalized_m_opts requires no argument.\n";
+ }
+
+ loongarch_config_target (& la_target,
+ opt_switches,
+ opt_arch_driver,
+ opt_tune_driver,
+ opt_fpu_driver,
+ opt_abi_base_driver,
+ opt_abi_ext_driver,
+ opt_cmodel_driver,
+ !no_link /* follow_multilib_list */);
+
+ /* Output normalized option strings. */
+ obstack_blank (&opts_obstack, 0);
+
+#undef APPEND_LTR
+#define APPEND_LTR(S) \
+ obstack_grow (&opts_obstack, (const void*) (S), \
+ sizeof ((S)) / sizeof (char) -1)
+
+#undef APPEND_VAL
+#define APPEND_VAL(S) \
+ obstack_grow (&opts_obstack, (const void*) (S), strlen ((S)))
+
+#undef APPEND_OPT
+#define APPEND_OPT(NAME) \
+ APPEND_LTR (" %<m" OPTSTR_##NAME "=* " \
+ " -m" OPTSTR_##NAME "=")
+
+ for (int i = 0; i < N_SWITCH_TYPES; i++)
+ {
+ APPEND_LTR (" %<m");
+ APPEND_VAL (loongarch_switch_strings[i]);
+ }
+
+ APPEND_OPT (ABI_BASE);
+ APPEND_VAL (loongarch_abi_base_strings[la_target.abi.base]);
+
+ APPEND_OPT (ARCH);
+ APPEND_VAL (loongarch_cpu_strings[la_target.cpu_arch]);
+
+ APPEND_OPT (ISA_EXT_FPU);
+ APPEND_VAL (loongarch_isa_ext_strings[la_target.isa.fpu]);
+
+ APPEND_OPT (CMODEL);
+ APPEND_VAL (loongarch_cmodel_strings[la_target.cmodel]);
+
+ APPEND_OPT (TUNE);
+ APPEND_VAL (loongarch_cpu_strings[la_target.cpu_tune]);
+
+ obstack_1grow (&opts_obstack, '\0');
+
+ return XOBFINISH (&opts_obstack, const char *);
+}
diff --git a/gcc/config/loongarch/loongarch-driver.h b/gcc/config/loongarch/loongarch-driver.h
new file mode 100644
index 00000000000..ce0222748dd
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-driver.h
@@ -0,0 +1,69 @@
+/* Subroutine headers for the gcc driver.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_DRIVER_H
+#define LOONGARCH_DRIVER_H
+
+#include "loongarch-str.h"
+
+extern const char*
+driver_set_m_flag (int argc, const char **argv);
+
+extern const char*
+driver_get_normalized_m_opts (int argc, const char **argv);
+
+#define EXTRA_SPEC_FUNCTIONS \
+ { "set_m_flag", driver_set_m_flag }, \
+ { "get_normalized_m_opts", driver_get_normalized_m_opts },
+
+/* Pre-process ABI-related options. */
+#define LA_SET_PARM_SPEC(NAME) \
+ " %{m" OPTSTR_##NAME "=*: %:set_m_flag(" OPTSTR_##NAME "=%*)}" \
+
+#define LA_SET_FLAG_SPEC(NAME) \
+ " %{m" OPTSTR_##NAME ": %:set_m_flag(" OPTSTR_##NAME ")}" \
+
+#define DRIVER_HANDLE_MACHINE_OPTIONS \
+ " %{c|S|E|nostdlib: %:set_m_flag(no_link)}" \
+ " %{nostartfiles: %{nodefaultlibs: %:set_m_flag(no_link)}}" \
+ LA_SET_PARM_SPEC (ABI_BASE) \
+ LA_SET_PARM_SPEC (ARCH) \
+ LA_SET_PARM_SPEC (TUNE) \
+ LA_SET_PARM_SPEC (ISA_EXT_FPU) \
+ LA_SET_PARM_SPEC (CMODEL) \
+ LA_SET_FLAG_SPEC (SOFT_FLOAT) \
+ LA_SET_FLAG_SPEC (SINGLE_FLOAT) \
+ LA_SET_FLAG_SPEC (DOUBLE_FLOAT) \
+ " %:get_normalized_m_opts()"
+
+#define DRIVER_SELF_SPECS \
+ DRIVER_HANDLE_MACHINE_OPTIONS
+
+/* ABI spec strings. */
+#define ABI_GRLEN_SPEC \
+ "%{mabi=lp64*:64}" \
+
+#define ABI_SPEC \
+ "%{mabi=lp64d:lp64d}" \
+ "%{mabi=lp64f:lp64f}" \
+ "%{mabi=lp64s:lp64s}" \
+
+#endif /* LOONGARCH_DRIVER_H */
+
diff --git a/gcc/config/loongarch/loongarch-opts.c b/gcc/config/loongarch/loongarch-opts.c
new file mode 100644
index 00000000000..fe9846b0cfb
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-opts.c
@@ -0,0 +1,582 @@
+/* Subroutines for loongarch-specific option handling.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "obstack.h"
+#include "diagnostic-core.h"
+
+#include "loongarch-cpu.h"
+#include "loongarch-opts.h"
+#include "loongarch-str.h"
+
+struct loongarch_target la_target;
+
+/* ABI-related configuration. */
+#define ABI_COUNT (sizeof(abi_priority_list)/sizeof(struct loongarch_abi))
+static const struct loongarch_abi
+abi_priority_list[] = {
+ {ABI_BASE_LP64D, ABI_EXT_BASE},
+ {ABI_BASE_LP64F, ABI_EXT_BASE},
+ {ABI_BASE_LP64S, ABI_EXT_BASE},
+};
+
+/* Initialize enabled_abi_types from TM_MULTILIB_LIST. */
+#ifdef __DISABLE_MULTILIB
+#define MULTILIB_LIST_LEN 1
+#else
+#define MULTILIB_LIST_LEN (sizeof (tm_multilib_list) / sizeof (int) / 2)
+static const int tm_multilib_list[] = { TM_MULTILIB_LIST };
+#endif
+static int enabled_abi_types[N_ABI_BASE_TYPES][N_ABI_EXT_TYPES] = { 0 };
+
+#define isa_required(ABI) (abi_minimal_isa[(ABI).base][(ABI).ext])
+extern "C" const struct loongarch_isa
+abi_minimal_isa[N_ABI_BASE_TYPES][N_ABI_EXT_TYPES];
+
+static inline int
+is_multilib_enabled (struct loongarch_abi abi)
+{
+ return enabled_abi_types[abi.base][abi.ext];
+}
+
+static void
+init_enabled_abi_types ()
+{
+#ifdef __DISABLE_MULTILIB
+ enabled_abi_types[DEFAULT_ABI_BASE][DEFAULT_ABI_EXT] = 1;
+#else
+ int abi_base, abi_ext;
+ for (unsigned int i = 0; i < MULTILIB_LIST_LEN; i++)
+ {
+ abi_base = tm_multilib_list[i << 1];
+ abi_ext = tm_multilib_list[(i << 1) + 1];
+ enabled_abi_types[abi_base][abi_ext] = 1;
+ }
+#endif
+}
+
+/* Switch masks. */
+#undef M
+#define M(NAME) OPTION_MASK_##NAME
+const int loongarch_switch_mask[N_SWITCH_TYPES] = {
+ /* SW_SOFT_FLOAT */ M(FORCE_SOFTF),
+ /* SW_SINGLE_FLOAT */ M(FORCE_F32),
+ /* SW_DOUBLE_FLOAT */ M(FORCE_F64),
+};
+#undef M
+
+/* String processing. */
+static struct obstack msg_obstack;
+#define APPEND_STRING(STR) obstack_grow (&msg_obstack, STR, strlen(STR));
+#define APPEND1(CH) obstack_1grow(&msg_obstack, CH);
+
+static const char* abi_str (struct loongarch_abi abi);
+static const char* isa_str (const struct loongarch_isa *isa, char separator);
+static const char* arch_str (const struct loongarch_target *target);
+static const char* multilib_enabled_abi_list ();
+
+/* Misc */
+static struct loongarch_abi isa_default_abi (const struct loongarch_isa *isa);
+static int isa_base_compat_p (const struct loongarch_isa *set1,
+ const struct loongarch_isa *set2);
+static int isa_fpu_compat_p (const struct loongarch_isa *set1,
+ const struct loongarch_isa *set2);
+static int abi_compat_p (const struct loongarch_isa *isa,
+ struct loongarch_abi abi);
+static int abi_default_cpu_arch (struct loongarch_abi abi);
+
+/* Checking configure-time defaults. */
+#ifndef DEFAULT_ABI_BASE
+#error missing definition of DEFAULT_ABI_BASE in ${tm_defines}.
+#endif
+
+#ifndef DEFAULT_ABI_EXT
+#error missing definition of DEFAULT_ABI_EXT in ${tm_defines}.
+#endif
+
+#ifndef DEFAULT_CPU_ARCH
+#error missing definition of DEFAULT_CPU_ARCH in ${tm_defines}.
+#endif
+
+#ifndef DEFAULT_ISA_EXT_FPU
+#error missing definition of DEFAULT_ISA_EXT_FPU in ${tm_defines}.
+#endif
+
+/* Handle combinations of -m machine option values
+ (see loongarch.opt and loongarch-opts.h). */
+void
+loongarch_config_target (struct loongarch_target *target,
+ HOST_WIDE_INT opt_switches,
+ int opt_arch, int opt_tune, int opt_fpu,
+ int opt_abi_base, int opt_abi_ext,
+ int opt_cmodel, int follow_multilib_list)
+{
+ struct loongarch_target t;
+
+ if (!target)
+ return;
+
+ /* Initialization */
+ init_enabled_abi_types ();
+ obstack_init (&msg_obstack);
+
+ struct {
+ int arch, tune, fpu, abi_base, abi_ext, cmodel;
+ } constrained = {
+ M_OPT_ABSENT(opt_arch) ? 0 : 1,
+ M_OPT_ABSENT(opt_tune) ? 0 : 1,
+ M_OPT_ABSENT(opt_fpu) ? 0 : 1,
+ M_OPT_ABSENT(opt_abi_base) ? 0 : 1,
+ M_OPT_ABSENT(opt_abi_ext) ? 0 : 1,
+ M_OPT_ABSENT(opt_cmodel) ? 0 : 1,
+ };
+
+#define on(NAME) ((loongarch_switch_mask[(SW_##NAME)] & opt_switches) \
+ && (on_switch = (SW_##NAME), 1))
+ int on_switch;
+
+ /* 1. Target ABI */
+ t.abi.base = constrained.abi_base ? opt_abi_base : DEFAULT_ABI_BASE;
+
+ t.abi.ext = constrained.abi_ext ? opt_abi_ext : DEFAULT_ABI_EXT;
+
+ /* Extra switch handling. */
+ if (on (SOFT_FLOAT) || on (SINGLE_FLOAT) || on (DOUBLE_FLOAT))
+ {
+ switch (on_switch)
+ {
+ case SW_SOFT_FLOAT:
+ opt_fpu = ISA_EXT_NOFPU;
+ break;
+
+ case SW_SINGLE_FLOAT:
+ opt_fpu = ISA_EXT_FPU32;
+ break;
+
+ case SW_DOUBLE_FLOAT:
+ opt_fpu = ISA_EXT_FPU64;
+ break;
+
+ default:
+ gcc_unreachable();
+ }
+ constrained.fpu = 1;
+
+ /* The target ISA is not ready yet, but (isa_required (t.abi)
+ + forced fpu) is enough for computing the forced base ABI. */
+ struct loongarch_isa default_isa = isa_required (t.abi);
+ struct loongarch_isa force_isa = default_isa;
+ struct loongarch_abi force_abi = t.abi;
+ force_isa.fpu = opt_fpu;
+ force_abi.base = isa_default_abi (&force_isa).base;
+
+ if (isa_fpu_compat_p (&default_isa, &force_isa));
+ /* Keep quiet if -m*-float does not promote the FP ABI. */
+ else if (constrained.abi_base && (t.abi.base != force_abi.base))
+ inform (UNKNOWN_LOCATION,
+ "%<-m%s%> overrides %<-m%s=%s%>, promoting ABI to %qs",
+ loongarch_switch_strings[on_switch],
+ OPTSTR_ABI_BASE, loongarch_abi_base_strings[t.abi.base],
+ abi_str (force_abi));
+
+ t.abi.base = force_abi.base;
+ }
+
+#ifdef __DISABLE_MULTILIB
+ if (follow_multilib_list)
+ if (t.abi.base != DEFAULT_ABI_BASE || t.abi.ext != DEFAULT_ABI_EXT)
+ {
+ static const struct loongarch_abi default_abi
+ = {DEFAULT_ABI_BASE, DEFAULT_ABI_EXT};
+
+ warning (0, "ABI changed (%qs -> %qs) while multilib is disabled",
+ abi_str (default_abi), abi_str (t.abi));
+ }
+#endif
+
+ /* 2. Target CPU */
+ t.cpu_arch = constrained.arch ? opt_arch : DEFAULT_CPU_ARCH;
+
+ t.cpu_tune = constrained.tune ? opt_tune
+ : (constrained.arch ? DEFAULT_CPU_ARCH : DEFAULT_CPU_TUNE);
+
+#ifdef __loongarch__
+ /* For native compilers, gather local CPU information
+ and fill the "CPU_NATIVE" index of arrays defined in
+ loongarch-cpu.c. */
+
+ t.cpu_native = fill_native_cpu_config (t.cpu_arch == CPU_NATIVE,
+ t.cpu_tune == CPU_NATIVE);
+
+#else
+ if (t.cpu_arch == CPU_NATIVE)
+ fatal_error (UNKNOWN_LOCATION,
+ "%qs does not work on a cross compiler",
+ "-m" OPTSTR_ARCH "=" STR_CPU_NATIVE);
+
+ else if (t.cpu_tune == CPU_NATIVE)
+ fatal_error (UNKNOWN_LOCATION,
+ "%qs does not work on a cross compiler",
+ "-m" OPTSTR_TUNE "=" STR_CPU_NATIVE);
+#endif
+
+ /* 3. Target ISA */
+config_target_isa:
+
+ /* Get default ISA from "-march" or its default value. */
+ t.isa = loongarch_cpu_default_isa[__ACTUAL_ARCH];
+
+ /* Apply incremental changes. */
+ /* "-march=native" overrides the default FPU type. */
+ t.isa.fpu = constrained.fpu ? opt_fpu :
+ ((t.cpu_arch == CPU_NATIVE && constrained.arch) ?
+ t.isa.fpu : DEFAULT_ISA_EXT_FPU);
+
+
+ /* 4. ABI-ISA compatibility */
+ /* Note:
+ - There IS a unique default -march value for each ABI type
+ (config.gcc: triplet -> abi -> default arch).
+
+ - If the base ABI is incompatible with the default arch,
+ try using the default -march it implies (and mark it
+ as "constrained" this time), then re-apply step 3.
+ */
+ struct loongarch_abi abi_tmp;
+ const struct loongarch_isa* isa_min;
+ struct loongarch_isa isa_tmp;
+
+ abi_tmp = t.abi;
+ isa_min = &isa_required (abi_tmp);
+ isa_tmp = t.isa;
+
+ if (isa_base_compat_p (&isa_tmp, isa_min)); /* OK */
+ else if (!constrained.arch)
+ {
+ /* Base architecture can only be implied by -march,
+ so we adjust that first if it is not constrained. */
+ int fallback_arch = abi_default_cpu_arch (t.abi);
+
+ if (t.cpu_arch == CPU_NATIVE)
+ warning (0, "your native CPU architecture (%qs) "
+ "does not support %qs ABI, falling back to %<-m%s=%s%>",
+ arch_str (&t), abi_str (t.abi), OPTSTR_ARCH,
+ loongarch_cpu_strings[fallback_arch]);
+ else
+ warning (0, "default CPU architecture (%qs) "
+ "does not support %qs ABI, falling back to %<-m%s=%s%>",
+ arch_str (&t), abi_str (t.abi), OPTSTR_ARCH,
+ loongarch_cpu_strings[fallback_arch]);
+
+ t.cpu_arch = fallback_arch;
+ constrained.arch = 1;
+ goto config_target_isa;
+ }
+ else if (!constrained.abi_base)
+ {
+ /* If -march is given while -mabi is not,
+ try selecting another base ABI type. */
+ abi_tmp.base = isa_default_abi (&isa_tmp).base;
+ }
+ else
+ goto fatal;
+
+ if (isa_fpu_compat_p (&isa_tmp, isa_min)); /* OK */
+ else if (!constrained.fpu)
+ isa_tmp.fpu = isa_min->fpu;
+ else if (!constrained.abi_base)
+ /* If -march is compatible with the default ABI
+ while -mfpu is not. */
+ abi_tmp.base = isa_default_abi (&isa_tmp).base;
+ else
+ goto fatal;
+
+ if (0)
+fatal:
+ fatal_error (UNKNOWN_LOCATION,
+ "unable to implement ABI %qs with instruction set %qs",
+ abi_str (t.abi), isa_str (&t.isa, '/'));
+
+
+ /* Using the fallback ABI. */
+ if (abi_tmp.base != t.abi.base || abi_tmp.ext != t.abi.ext)
+ {
+ /* This flag is only set in the GCC driver. */
+ if (follow_multilib_list)
+ {
+
+ /* Continue falling back until we find a feasible ABI type
+ enabled by TM_MULTILIB_LIST. */
+ if (!is_multilib_enabled (abi_tmp))
+ {
+ for (unsigned int i = 0; i < ABI_COUNT; i++)
+ {
+ if (is_multilib_enabled (abi_priority_list[i])
+ && abi_compat_p (&isa_tmp, abi_priority_list[i]))
+ {
+ abi_tmp = abi_priority_list[i];
+
+ warning (0, "ABI %qs cannot be implemented due to "
+ "limited instruction set %qs, "
+ "falling back to %qs", abi_str (t.abi),
+ isa_str (&t.isa, '/'), abi_str (abi_tmp));
+
+ goto fallback;
+ }
+ }
+
+ /* Otherwise, keep using abi_tmp with a warning. */
+#ifdef __DISABLE_MULTILIB
+ warning (0, "instruction set %qs cannot implement "
+ "default ABI %qs, falling back to %qs",
+ isa_str (&t.isa, '/'), abi_str (t.abi),
+ abi_str (abi_tmp));
+#else
+ warning (0, "no multilib-enabled ABI (%qs) can be implemented "
+ "with instruction set %qs, falling back to %qs",
+ multilib_enabled_abi_list (),
+ isa_str (&t.isa, '/'), abi_str (abi_tmp));
+#endif
+ }
+ }
+
+fallback:
+ t.abi = abi_tmp;
+ }
+ else if (follow_multilib_list)
+ {
+ if (!is_multilib_enabled (t.abi))
+ {
+ inform (UNKNOWN_LOCATION,
+ "ABI %qs is not enabled at configure-time, "
+ "the linker might report an error", abi_str (t.abi));
+
+ inform (UNKNOWN_LOCATION, "ABI with startfiles: %s",
+ multilib_enabled_abi_list ());
+ }
+ }
+
+
+ /* 5. Target code model */
+ t.cmodel = constrained.cmodel ? opt_cmodel : CMODEL_NORMAL;
+
+ /* cleanup and return */
+ obstack_free (&msg_obstack, NULL);
+ *target = t;
+}
+
+/* Returns the default ABI for the given instruction set. */
+static inline struct loongarch_abi
+isa_default_abi (const struct loongarch_isa *isa)
+{
+ struct loongarch_abi abi;
+
+ switch (isa->fpu)
+ {
+ case ISA_EXT_FPU64:
+ if (isa->base == ISA_BASE_LA64V100)
+ abi.base = ABI_BASE_LP64D;
+ break;
+
+ case ISA_EXT_FPU32:
+ if (isa->base == ISA_BASE_LA64V100)
+ abi.base = ABI_BASE_LP64F;
+ break;
+
+ case ISA_EXT_NOFPU:
+ if (isa->base == ISA_BASE_LA64V100)
+ abi.base = ABI_BASE_LP64S;
+ break;
+
+ default:
+ gcc_unreachable ();
+ }
+
+ abi.ext = ABI_EXT_BASE;
+ return abi;
+}
+
+/* Check if set2 is a subset of set1. */
+static inline int
+isa_base_compat_p (const struct loongarch_isa *set1,
+ const struct loongarch_isa *set2)
+{
+ switch (set2->base)
+ {
+ case ISA_BASE_LA64V100:
+ return (set1->base == ISA_BASE_LA64V100);
+
+ default:
+ gcc_unreachable ();
+ }
+}
+
+static inline int
+isa_fpu_compat_p (const struct loongarch_isa *set1,
+ const struct loongarch_isa *set2)
+{
+ switch (set2->fpu)
+ {
+ case ISA_EXT_FPU64:
+ return set1->fpu == ISA_EXT_FPU64;
+
+ case ISA_EXT_FPU32:
+ return set1->fpu == ISA_EXT_FPU32 || set1->fpu == ISA_EXT_FPU64;
+
+ case ISA_EXT_NOFPU:
+ return 1;
+
+ default:
+ gcc_unreachable ();
+ }
+
+}
+
+static inline int
+abi_compat_p (const struct loongarch_isa *isa, struct loongarch_abi abi)
+{
+ int compatible = 1;
+ const struct loongarch_isa *isa2 = &isa_required (abi);
+
+ /* Append conditionals for new ISA components below. */
+ compatible = compatible && isa_base_compat_p (isa, isa2);
+ compatible = compatible && isa_fpu_compat_p (isa, isa2);
+ return compatible;
+}
+
+/* The behavior of this function should be consistent
+ with config.gcc. */
+static inline int
+abi_default_cpu_arch (struct loongarch_abi abi)
+{
+ switch (abi.base)
+ {
+ case ABI_BASE_LP64D:
+ case ABI_BASE_LP64F:
+ case ABI_BASE_LP64S:
+ if (abi.ext == ABI_EXT_BASE)
+ return CPU_LOONGARCH64;
+ }
+ gcc_unreachable ();
+}
+
+static const char*
+abi_str (struct loongarch_abi abi)
+{
+ /* "/base" can be omitted */
+ if (abi.ext == ABI_EXT_BASE)
+ return (const char*)
+ obstack_copy0 (&msg_obstack, loongarch_abi_base_strings[abi.base],
+ strlen (loongarch_abi_base_strings[abi.base]));
+ else
+ {
+ APPEND_STRING (loongarch_abi_base_strings[abi.base])
+ APPEND1 ('/')
+ APPEND_STRING (loongarch_abi_ext_strings[abi.ext])
+ APPEND1 ('\0')
+
+ return XOBFINISH (&msg_obstack, const char *);
+ }
+}
+
+static const char*
+isa_str (const struct loongarch_isa *isa, char separator)
+{
+ APPEND_STRING (loongarch_isa_base_strings[isa->base])
+ APPEND1 (separator)
+
+ if (isa->fpu == ISA_EXT_NOFPU)
+ {
+ APPEND_STRING ("no" OPTSTR_ISA_EXT_FPU)
+ }
+ else
+ {
+ APPEND_STRING (OPTSTR_ISA_EXT_FPU)
+ APPEND_STRING (loongarch_isa_ext_strings[isa->fpu])
+ }
+ APPEND1 ('\0')
+
+ /* Add more here. */
+
+ return XOBFINISH (&msg_obstack, const char *);
+}
+
+static const char*
+arch_str (const struct loongarch_target *target)
+{
+ if (target->cpu_arch == CPU_NATIVE)
+ {
+ if (target->cpu_native == CPU_NATIVE)
+ {
+ /* Describe a native CPU with unknown PRID. */
+ const char* isa_string = isa_str (&target->isa, ',');
+ APPEND_STRING ("PRID: 0x")
+ APPEND_STRING (get_native_prid_str ())
+ APPEND_STRING (", ISA features: ")
+ APPEND_STRING (isa_string)
+ APPEND1 ('\0')
+ }
+ else
+ APPEND_STRING (loongarch_cpu_strings[target->cpu_native]);
+ }
+ else
+ APPEND_STRING (loongarch_cpu_strings[target->cpu_arch]);
+
+ APPEND1 ('\0')
+ return XOBFINISH (&msg_obstack, const char *);
+}
+
+static const char*
+multilib_enabled_abi_list ()
+{
+ int enabled_abi_idx[MULTILIB_LIST_LEN] = { 0 };
+ const char* enabled_abi_str[MULTILIB_LIST_LEN] = { NULL };
+ unsigned int j = 0;
+
+ for (unsigned int i = 0; i < ABI_COUNT && j < MULTILIB_LIST_LEN; i++)
+ {
+ if (enabled_abi_types[abi_priority_list[i].base]
+ [abi_priority_list[i].ext])
+ {
+ enabled_abi_idx[j++] = i;
+ }
+ }
+
+ for (unsigned int k = 0; k < j; k++)
+ {
+ enabled_abi_str[k] = abi_str (abi_priority_list[enabled_abi_idx[k]]);
+ }
+
+ for (unsigned int k = 0; k < j - 1; k++)
+ {
+ APPEND_STRING (enabled_abi_str[k])
+ APPEND1 (',')
+ APPEND1 (' ')
+ }
+ APPEND_STRING (enabled_abi_str[j - 1])
+ APPEND1 ('\0')
+
+ return XOBFINISH (&msg_obstack, const char *);
+}
diff --git a/gcc/config/loongarch/loongarch-opts.h b/gcc/config/loongarch/loongarch-opts.h
new file mode 100644
index 00000000000..9c205e01980
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-opts.h
@@ -0,0 +1,86 @@
+/* Definitions for loongarch-specific option handling.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_OPTS_H
+#define LOONGARCH_OPTS_H
+
+
+/* Target configuration */
+extern struct loongarch_target la_target;
+
+/* Switch masks */
+extern const int loongarch_switch_mask[];
+
+#include "loongarch-def.h"
+
+#if !defined(IN_LIBGCC2) && !defined(IN_TARGET_LIBS) && !defined(IN_RTS)
+/* Handler for "-m" option combinations,
+ shared by the driver and the compiler proper. */
+void
+loongarch_config_target (struct loongarch_target *target,
+ HOST_WIDE_INT opt_switches,
+ int opt_arch, int opt_tune, int opt_fpu,
+ int opt_abi_base, int opt_abi_ext,
+ int opt_cmodel, int follow_multilib_list);
+#endif
+
+
+/* Macros for common conditional expressions used in loongarch.{c,h,md} */
+#define TARGET_CMODEL_NORMAL (la_target.cmodel == CMODEL_NORMAL)
+#define TARGET_CMODEL_TINY (la_target.cmodel == CMODEL_TINY)
+#define TARGET_CMODEL_TINY_STATIC (la_target.cmodel == CMODEL_TINY_STATIC)
+#define TARGET_CMODEL_LARGE (la_target.cmodel == CMODEL_LARGE)
+#define TARGET_CMODEL_EXTREME (la_target.cmodel == CMODEL_EXTREME)
+
+#define TARGET_HARD_FLOAT (la_target.isa.fpu != ISA_EXT_NOFPU)
+#define TARGET_HARD_FLOAT_ABI (la_target.abi.base == ABI_BASE_LP64D \
+ || la_target.abi.base == ABI_BASE_LP64F)
+
+#define TARGET_SOFT_FLOAT (la_target.isa.fpu == ISA_EXT_NOFPU)
+#define TARGET_SOFT_FLOAT_ABI (la_target.abi.base == ABI_BASE_LP64S)
+#define TARGET_SINGLE_FLOAT (la_target.isa.fpu == ISA_EXT_FPU32)
+#define TARGET_SINGLE_FLOAT_ABI (la_target.abi.base == ABI_BASE_LP64F)
+#define TARGET_DOUBLE_FLOAT (la_target.isa.fpu == ISA_EXT_FPU64)
+#define TARGET_DOUBLE_FLOAT_ABI (la_target.abi.base == ABI_BASE_LP64D)
+
+#define TARGET_64BIT (la_target.isa.base == ISA_BASE_LA64V100)
+#define TARGET_ABI_LP64 (la_target.abi.base == ABI_BASE_LP64D \
+ || la_target.abi.base == ABI_BASE_LP64F \
+ || la_target.abi.base == ABI_BASE_LP64S)
+
+#define TARGET_ARCH_NATIVE (la_target.cpu_arch == CPU_NATIVE)
+#define __ACTUAL_ARCH (TARGET_ARCH_NATIVE \
+ ? (la_target.cpu_native < N_ARCH_TYPES \
+ ? (la_target.cpu_native) : (CPU_NATIVE)) \
+ : (la_target.cpu_arch))
+
+#define TARGET_TUNE_NATIVE (la_target.cpu_tune == CPU_NATIVE)
+#define __ACTUAL_TUNE (TARGET_TUNE_NATIVE \
+ ? (la_target.cpu_native < N_TUNE_TYPES \
+ ? (la_target.cpu_native) : (CPU_NATIVE)) \
+ : (la_target.cpu_tune))
+
+#define TARGET_ARCH_LOONGARCH64 (__ACTUAL_ARCH == CPU_LOONGARCH64)
+#define TARGET_ARCH_LA464 (__ACTUAL_ARCH == CPU_LA464)
+
+#define TARGET_TUNE_LOONGARCH64 (__ACTUAL_TUNE == CPU_LOONGARCH64)
+#define TARGET_TUNE_LA464 (__ACTUAL_TUNE == CPU_LA464)
+
+#endif /* LOONGARCH_OPTS_H */
diff --git a/gcc/config/loongarch/loongarch-str.h b/gcc/config/loongarch/loongarch-str.h
new file mode 100644
index 00000000000..4f0a76cdc28
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-str.h
@@ -0,0 +1,57 @@
+/* Generated automatically by "genstr" from "loongarch-strings".
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_STR_H
+#define LOONGARCH_STR_H
+
+#define OPTSTR_ARCH "arch"
+#define OPTSTR_TUNE "tune"
+
+#define STR_CPU_NATIVE "native"
+#define STR_CPU_LOONGARCH64 "loongarch64"
+#define STR_CPU_LA464 "la464"
+
+#define STR_ISA_BASE_LA64V100 "la64"
+
+#define OPTSTR_ISA_EXT_FPU "fpu"
+#define STR_ISA_EXT_NOFPU "none"
+#define STR_ISA_EXT_FPU0 "0"
+#define STR_ISA_EXT_FPU32 "32"
+#define STR_ISA_EXT_FPU64 "64"
+
+#define OPTSTR_SOFT_FLOAT "soft-float"
+#define OPTSTR_SINGLE_FLOAT "single-float"
+#define OPTSTR_DOUBLE_FLOAT "double-float"
+
+#define OPTSTR_ABI_BASE "abi"
+#define STR_ABI_BASE_LP64D "lp64d"
+#define STR_ABI_BASE_LP64F "lp64f"
+#define STR_ABI_BASE_LP64S "lp64s"
+
+#define STR_ABI_EXT_BASE "base"
+
+#define OPTSTR_CMODEL "cmodel"
+#define STR_CMODEL_NORMAL "normal"
+#define STR_CMODEL_TINY "tiny"
+#define STR_CMODEL_TS "tiny-static"
+#define STR_CMODEL_LARGE "large"
+#define STR_CMODEL_EXTREME "extreme"
+
+#endif /* LOONGARCH_STR_H */
diff --git a/gcc/config/loongarch/loongarch.opt b/gcc/config/loongarch/loongarch.opt
new file mode 100644
index 00000000000..9e0ea68f538
--- /dev/null
+++ b/gcc/config/loongarch/loongarch.opt
@@ -0,0 +1,189 @@
+; Generated by "genstr" from the template "loongarch.opt.in"
+; and definitions from "loongarch-strings".
+;
+; Copyright (C) 2021 Free Software Foundation, Inc.
+;
+; This file is part of GCC.
+;
+; GCC is free software; you can redistribute it and/or modify it under
+; the terms of the GNU General Public License as published by the Free
+; Software Foundation; either version 3, or (at your option) any later
+; version.
+;
+; GCC is distributed in the hope that it will be useful, but WITHOUT
+; ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+; License for more details.
+;
+; You should have received a copy of the GNU General Public License
+; along with GCC; see the file COPYING3. If not see
+; <http://www.gnu.org/licenses/>.
+;
+
+; Variables (macros) that should be exported by loongarch.opt:
+; la_opt_switches,
+; la_opt_abi_base, la_opt_abi_ext,
+; la_opt_cpu_arch, la_opt_cpu_tune,
+; la_opt_fpu,
+; la_cmodel.
+
+HeaderInclude
+config/loongarch/loongarch-opts.h
+
+HeaderInclude
+config/loongarch/loongarch-str.h
+
+Variable
+HOST_WIDE_INT la_opt_switches = 0
+
+; ISA related options
+;; Base ISA
+Enum
+Name(isa_base) Type(int)
+Basic ISAs of LoongArch:
+
+EnumValue
+Enum(isa_base) String(la64) Value(ISA_BASE_LA64V100)
+
+
+;; ISA extensions / adjustments
+Enum
+Name(isa_ext_fpu) Type(int)
+FPU types of LoongArch:
+
+EnumValue
+Enum(isa_ext_fpu) String(none) Value(ISA_EXT_NOFPU)
+
+EnumValue
+Enum(isa_ext_fpu) String(32) Value(ISA_EXT_FPU32)
+
+EnumValue
+Enum(isa_ext_fpu) String(64) Value(ISA_EXT_FPU64)
+
+mfpu=
+Target RejectNegative Joined ToLower Enum(isa_ext_fpu) Var(la_opt_fpu) Init(M_OPTION_NOT_SEEN)
+-mfpu=FPU Generate code for the given FPU.
+
+mfpu=0
+Target RejectNegative Alias(mfpu=,none)
+
+msoft-float
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_SOFTF) Negative(msingle-float)
+Prevent the use of all hardware floating-point instructions.
+
+msingle-float
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_F32) Negative(mdouble-float)
+Restrict the use of hardware floating-point instructions to 32-bit operations.
+
+mdouble-float
+Target Driver RejectNegative Var(la_opt_switches) Mask(FORCE_F64) Negative(msoft-float)
+Allow hardware floating-point instructions to cover both 32-bit and 64-bit operations.
+
+
+;; Base target models (implies ISA & tune parameters)
+Enum
+Name(cpu_type) Type(int)
+LoongArch CPU types:
+
+EnumValue
+Enum(cpu_type) String(native) Value(CPU_NATIVE)
+
+EnumValue
+Enum(cpu_type) String(loongarch64) Value(CPU_LOONGARCH64)
+
+EnumValue
+Enum(cpu_type) String(la464) Value(CPU_LA464)
+
+march=
+Target RejectNegative Joined Enum(cpu_type) Var(la_opt_cpu_arch) Init(M_OPTION_NOT_SEEN)
+-march=PROCESSOR Generate code for the given PROCESSOR ISA.
+
+mtune=
+Target RejectNegative Joined Enum(cpu_type) Var(la_opt_cpu_tune) Init(M_OPTION_NOT_SEEN)
+-mtune=PROCESSOR Generate optimized code for PROCESSOR.
+
+
+; ABI related options
+; (ISA constraints on ABI are handled dynamically)
+
+;; Base ABI
+Enum
+Name(abi_base) Type(int)
+Base ABI types for LoongArch:
+
+EnumValue
+Enum(abi_base) String(lp64d) Value(ABI_BASE_LP64D)
+
+EnumValue
+Enum(abi_base) String(lp64f) Value(ABI_BASE_LP64F)
+
+EnumValue
+Enum(abi_base) String(lp64s) Value(ABI_BASE_LP64S)
+
+mabi=
+Target RejectNegative Joined ToLower Enum(abi_base) Var(la_opt_abi_base) Init(M_OPTION_NOT_SEEN)
+-mabi=BASEABI Generate code that conforms to the given BASEABI.
+
+;; ABI Extension
+Variable
+int la_opt_abi_ext = M_OPTION_NOT_SEEN
+
+
+mbranch-cost=
+Target RejectNegative Joined UInteger Var(loongarch_branch_cost)
+-mbranch-cost=COST Set the cost of branches to roughly COST instructions.
+
+mcheck-zero-division
+Target Mask(CHECK_ZERO_DIV)
+Trap on integer divide by zero.
+
+mcond-move-int
+Target Var(TARGET_COND_MOVE_INT) Init(1)
+Conditional moves for integral are enabled.
+
+mcond-move-float
+Target Var(TARGET_COND_MOVE_FLOAT) Init(1)
+Conditional moves for float are enabled.
+
+mmemcpy
+Target Mask(MEMCPY)
+Don't optimize block moves.
+
+mlra
+Target Var(loongarch_lra_flag) Init(1) Save
+Use LRA instead of reload.
+
+noasmopt
+Driver
+
+mstrict-align
+Target Var(TARGET_STRICT_ALIGN) Init(1)
+Do not generate unaligned memory accesses.
+
+mmax-inline-memcpy-size=
+Target Joined RejectNegative UInteger Var(loongarch_max_inline_memcpy_size) Init(1024)
+-mmax-inline-memcpy-size=SIZE Set the max size of memcpy to inline, default is 1024.
+
+; The code model option names for -mcmodel.
+Enum
+Name(cmodel) Type(int)
+The code model option names for -mcmodel:
+
+EnumValue
+Enum(cmodel) String(normal) Value(CMODEL_NORMAL)
+
+EnumValue
+Enum(cmodel) String(tiny) Value(CMODEL_TINY)
+
+EnumValue
+Enum(cmodel) String(tiny-static) Value(CMODEL_TINY_STATIC)
+
+EnumValue
+Enum(cmodel) String(large) Value(CMODEL_LARGE)
+
+EnumValue
+Enum(cmodel) String(extreme) Value(CMODEL_EXTREME)
+
+mcmodel=
+Target RejectNegative Joined Enum(cmodel) Var(la_opt_cmodel) Init(CMODEL_NORMAL)
+Specify the code model.
diff --git a/gcc/config/loongarch/t-linux b/gcc/config/loongarch/t-linux
new file mode 100644
index 00000000000..7be867d2351
--- /dev/null
+++ b/gcc/config/loongarch/t-linux
@@ -0,0 +1,53 @@
+# Copyright (C) 2021 Free Software Foundation, Inc.
+#
+# This file is part of GCC.
+#
+# GCC is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3, or (at your option)
+# any later version.
+#
+# GCC is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3. If not see
+# <http://www.gnu.org/licenses/>.
+
+# Multilib
+MULTILIB_OPTIONS = mabi=lp64d/mabi=lp64f/mabi=lp64s
+MULTILIB_DIRNAMES = base/lp64d base/lp64f base/lp64s
+
+# The GCC driver always gets all abi-related options on the command line.
+# (see loongarch-driver.c:driver_get_normalized_m_opts)
+comma=,
+MULTILIB_REQUIRED = $(subst $(comma), ,$(TM_MULTILIB_CONFIG))
+
+# Multiarch
+ifneq ($(call if_multiarch,yes),yes)
+ # Define __DISABLE_MULTIARCH if multiarch is disabled.
+ tm_defines += __DISABLE_MULTIARCH
+else
+ # Only define MULTIARCH_DIRNAME when multiarch is enabled,
+ # or it would always introduce ${target} into the search path.
+ MULTIARCH_DIRNAME = $(LA_MULTIARCH_TRIPLET)
+endif
+
+# Don't define MULTILIB_OSDIRNAMES if multilib is disabled.
+ifeq ($(filter __DISABLE_MULTILIB,$(tm_defines)),)
+
+ MULTILIB_OSDIRNAMES = \
+ mabi.lp64d=../lib64$\
+ $(call if_multiarch,:loongarch64-linux-gnuf64)
+
+ MULTILIB_OSDIRNAMES += \
+ mabi.lp64f=../lib64/f32$\
+ $(call if_multiarch,:loongarch64-linux-gnuf32)
+
+ MULTILIB_OSDIRNAMES += \
+ mabi.lp64s=../lib64/sf$\
+ $(call if_multiarch,:loongarch64-linux-gnusf)
+
+endif
diff --git a/gcc/config/loongarch/t-loongarch b/gcc/config/loongarch/t-loongarch
new file mode 100644
index 00000000000..0396a3a6d7c
--- /dev/null
+++ b/gcc/config/loongarch/t-loongarch
@@ -0,0 +1,59 @@
+# Copyright (C) 2021 Free Software Foundation, Inc.
+#
+# This file is part of GCC.
+#
+# GCC is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3, or (at your option)
+# any later version.
+#
+# GCC is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3. If not see
+# <http://www.gnu.org/licenses/>.
+
+# Canonical target triplet from config.gcc
+LA_MULTIARCH_TRIPLET = $(patsubst LA_MULTIARCH_TRIPLET=%,%,$\
+$(filter LA_MULTIARCH_TRIPLET=%,$(tm_defines)))
+
+# String definition header
+LA_STR_H = $(srcdir)/config/loongarch/loongarch-str.h
+
+loongarch-c.o: $(srcdir)/config/loongarch/loongarch-c.c $(CONFIG_H) $(SYSTEM_H) \
+ coretypes.h $(TM_H) $(TREE_H) output.h $(C_COMMON_H) $(TARGET_H)
+ $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) \
+ $(srcdir)/config/loongarch/loongarch-c.c
+
+loongarch-builtins.o: $(srcdir)/config/loongarch/loongarch-builtins.c $(CONFIG_H) \
+ $(SYSTEM_H) coretypes.h $(TM_H) $(RTL_H) $(TREE_H) $(RECOG_H) langhooks.h \
+ $(DIAGNOSTIC_CORE_H) $(OPTABS_H) $(srcdir)/config/loongarch/loongarch-ftypes.def \
+ $(srcdir)/config/loongarch/loongarch-modes.def
+ $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) \
+ $(srcdir)/config/loongarch/loongarch-builtins.c
+
+loongarch-driver.o : $(srcdir)/config/loongarch/loongarch-driver.c $(LA_STR_H) \
+ $(CONFIG_H) $(SYSTEM_H)
+ $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) $<
+
+loongarch-opts.o: $(srcdir)/config/loongarch/loongarch-opts.c $(LA_STR_H)
+ $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) $<
+
+loongarch-cpu.o: $(srcdir)/config/loongarch/loongarch-cpu.c $(LA_STR_H)
+ $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) $<
+
+loongarch-def.o: $(srcdir)/config/loongarch/loongarch-def.c $(LA_STR_H)
+ $(CC) -c $(ALL_CFLAGS) $(ALL_CPPFLAGS) $(INCLUDES) $<
+
+$(srcdir)/config/loongarch/loongarch.opt: \
+ $(srcdir)/config/loongarch/genopts/genstr.sh \
+ $(srcdir)/config/loongarch/genopts/loongarch.opt.in
+ $(SHELL) $< opt > $@
+
+$(LA_STR_H): \
+ $(srcdir)/config/loongarch/genopts/genstr.sh \
+ $(srcdir)/config/loongarch/genopts/loongarch-strings
+ $(SHELL) $< header > $@
diff --git a/gcc/configure.ac b/gcc/configure.ac
index 0db9c4cb027..b206c604d47 100644
--- a/gcc/configure.ac
+++ b/gcc/configure.ac
@@ -937,6 +937,9 @@ AC_ARG_ENABLE(fixed-point,
mips*-*-*)
enable_fixed_point=yes
;;
+ loongarch*-*-*)
+ enable_fixed_point=yes
+ ;;
*)
AC_MSG_WARN([fixed-point is not supported for this target, ignored])
enable_fixed_point=no
@@ -3800,6 +3803,17 @@ foo: data8 25
movl r24 = @tprel(foo#)'
tls_as_opt=--fatal-warnings
;;
+ loongarch*-*-*)
+ conftest_s='
+ .section .tdata,"awT",@progbits
+x: .word 2
+ .text
+ la.tls.gd $a0,x
+ bl __tls_get_addr'
+ tls_first_major=0
+ tls_first_minor=0
+ tls_as_opt='--fatal-warnings'
+ ;;
microblaze*-*-*)
conftest_s='
.section .tdata,"awT",@progbits
@@ -5260,6 +5274,17 @@ configured with --enable-newlib-nano-formatted-io.])
[AC_DEFINE(HAVE_AS_MARCH_ZIFENCEI, 1,
[Define if the assembler understands -march=rv*_zifencei.])])
;;
+ loongarch*-*-*)
+ gcc_GAS_CHECK_FEATURE([.dtprelword support],
+ gcc_cv_as_loongarch_dtprelword, [2,18,0],,
+ [.section .tdata,"awT",@progbits
+x:
+ .word 2
+ .text
+ .dtprelword x+0x8000],,
+ [AC_DEFINE(HAVE_AS_DTPRELWORD, 1,
+ [Define if your assembler supports .dtprelword.])])
+ ;;
s390*-*-*)
gcc_GAS_CHECK_FEATURE([.gnu_attribute support],
gcc_cv_as_s390_gnu_attribute,,
@@ -5293,11 +5318,11 @@ configured with --enable-newlib-nano-formatted-io.])
;;
esac
-# Mips and HP-UX need the GNU assembler.
+# Mips, LoongArch and HP-UX need the GNU assembler.
# Linux on IA64 might be able to use the Intel assembler.
case "$target" in
- mips*-*-* | *-*-hpux* )
+ mips*-*-* | loongarch*-*-* | *-*-hpux* )
if test x$gas_flag = xyes \
|| test x"$host" != x"$build" \
|| test ! -x "$gcc_cv_as" \
@@ -5454,8 +5479,8 @@ esac
# ??? Once 2.11 is released, probably need to add first known working
# version to the per-target configury.
case "$cpu_type" in
- aarch64 | alpha | arc | arm | avr | bfin | cris | csky | i386 | m32c | m68k \
- | microblaze | mips | nds32 | nios2 | pa | riscv | rs6000 | score | sparc \
+ aarch64 | alpha | arc | arm | avr | bfin | cris | csky | i386 | loongarch | m32c \
+ | m68k | microblaze | mips | nds32 | nios2 | pa | riscv | rs6000 | score | sparc \
| tilegx | tilepro | visium | xstormy16 | xtensa)
insn="nop"
;;
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 03/12] LoongArch Port: Regenerate gcc/configure.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
2021-12-24 9:28 ` [PATCH v4 01/12] LoongArch Port: Regenerate configure chenglulu
2021-12-24 9:28 ` [PATCH v4 02/12] LoongArch Port: gcc build chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 04/12] LoongArch Port: Machine Decsription files chenglulu
` (9 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
---
gcc/configure | 85 +++++++++++++++++++++++++++++++++++++++++++++++----
1 file changed, 79 insertions(+), 6 deletions(-)
diff --git a/gcc/configure b/gcc/configure
index 992a9d70092..999557a3b51 100755
--- a/gcc/configure
+++ b/gcc/configure
@@ -5352,7 +5352,26 @@ else
GDC="$ac_cv_prog_GDC"
fi
+{ $as_echo "$as_me:${as_lineno-$LINENO}: checking whether the D compiler works" >&5
+$as_echo_n "checking whether the D compiler works... " >&6; }
+if ${acx_cv_d_compiler_works+:} false; then :
+ $as_echo_n "(cached) " >&6
+else
+ cat >conftest.d <<EOF
+module conftest; int main() { return 0; }
+EOF
+acx_cv_d_compiler_works=no
if test "x$GDC" != xno; then
+ errors=`(${GDC} -I"$srcdir"/d -c conftest.d) 2>&1 || echo failure`
+ if test x"$errors" = x && test -f conftest.$ac_objext; then
+ acx_cv_d_compiler_works=yes
+ fi
+ rm -f conftest.*
+fi
+fi
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $acx_cv_d_compiler_works" >&5
+$as_echo "$acx_cv_d_compiler_works" >&6; }
+if test "x$GDC" != xno && test x$acx_cv_d_compiler_works != xno; then
have_gdc=yes
else
have_gdc=no
@@ -5423,6 +5442,9 @@ case "${target}" in
# sets the default TLS model and affects inlining.
PICFLAG_FOR_TARGET=-fPIC
;;
+ loongarch*-*-*)
+ PICFLAG_FOR_TARGET=-fpic
+ ;;
mips-sgi-irix6*)
# PIC is the default.
;;
@@ -7865,6 +7887,9 @@ else
mips*-*-*)
enable_fixed_point=yes
;;
+ loongarch*-*-*)
+ enable_fixed_point=yes
+ ;;
*)
{ $as_echo "$as_me:${as_lineno-$LINENO}: WARNING: fixed-point is not supported for this target, ignored" >&5
$as_echo "$as_me: WARNING: fixed-point is not supported for this target, ignored" >&2;}
@@ -19561,7 +19586,7 @@ else
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
lt_status=$lt_dlunknown
cat > conftest.$ac_ext <<_LT_EOF
-#line 19564 "configure"
+#line 19589 "configure"
#include "confdefs.h"
#if HAVE_DLFCN_H
@@ -19667,7 +19692,7 @@ else
lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2
lt_status=$lt_dlunknown
cat > conftest.$ac_ext <<_LT_EOF
-#line 19670 "configure"
+#line 19695 "configure"
#include "confdefs.h"
#if HAVE_DLFCN_H
@@ -25432,6 +25457,17 @@ foo: data8 25
movl r24 = @tprel(foo#)'
tls_as_opt=--fatal-warnings
;;
+ loongarch*-*-*)
+ conftest_s='
+ .section .tdata,"awT",@progbits
+x: .word 2
+ .text
+ la.tls.gd $a0,x
+ bl __tls_get_addr'
+ tls_first_major=0
+ tls_first_minor=0
+ tls_as_opt='--fatal-warnings'
+ ;;
microblaze*-*-*)
conftest_s='
.section .tdata,"awT",@progbits
@@ -28654,6 +28690,43 @@ $as_echo "#define HAVE_AS_MARCH_ZIFENCEI 1" >>confdefs.h
fi
;;
+ loongarch*-*-*)
+ { $as_echo "$as_me:${as_lineno-$LINENO}: checking assembler for .dtprelword support" >&5
+$as_echo_n "checking assembler for .dtprelword support... " >&6; }
+if ${gcc_cv_as_loongarch_dtprelword+:} false; then :
+ $as_echo_n "(cached) " >&6
+else
+ gcc_cv_as_loongarch_dtprelword=no
+ if test x$gcc_cv_as != x; then
+ $as_echo '' > conftest.s
+ if { ac_try='$gcc_cv_as $gcc_cv_as_flags 2,18,0 -o conftest.o conftest.s >&5'
+ { { eval echo "\"\$as_me\":${as_lineno-$LINENO}: \"$ac_try\""; } >&5
+ (eval $ac_try) 2>&5
+ ac_status=$?
+ $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
+ test $ac_status = 0; }; }
+ then
+ .section .tdata,"awT",@progbits
+x:
+ .word 2
+ .text
+ .dtprelword x+0x8000
+ else
+ echo "configure: failed program was" >&5
+ cat conftest.s >&5
+ fi
+ rm -f conftest.o conftest.s
+ fi
+fi
+{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $gcc_cv_as_loongarch_dtprelword" >&5
+$as_echo "$gcc_cv_as_loongarch_dtprelword" >&6; }
+
+if test $gcc_cv_as_loongarch_dtprelword != yes; then
+
+$as_echo "#define HAVE_AS_DTPRELWORD 1" >>confdefs.h
+
+fi
+ ;;
s390*-*-*)
{ $as_echo "$as_me:${as_lineno-$LINENO}: checking assembler for .gnu_attribute support" >&5
$as_echo_n "checking assembler for .gnu_attribute support... " >&6; }
@@ -28817,11 +28890,11 @@ fi
;;
esac
-# Mips and HP-UX need the GNU assembler.
+# Mips, LoongArch and HP-UX need the GNU assembler.
# Linux on IA64 might be able to use the Intel assembler.
case "$target" in
- mips*-*-* | *-*-hpux* )
+ mips*-*-* | loongarch*-*-* | *-*-hpux* )
if test x$gas_flag = xyes \
|| test x"$host" != x"$build" \
|| test ! -x "$gcc_cv_as" \
@@ -29258,8 +29331,8 @@ esac
# ??? Once 2.11 is released, probably need to add first known working
# version to the per-target configury.
case "$cpu_type" in
- aarch64 | alpha | arc | arm | avr | bfin | cris | csky | i386 | m32c | m68k \
- | microblaze | mips | nds32 | nios2 | pa | riscv | rs6000 | score | sparc \
+ aarch64 | alpha | arc | arm | avr | bfin | cris | csky | i386 | loongarch | m32c \
+ | m68k | microblaze | mips | nds32 | nios2 | pa | riscv | rs6000 | score | sparc \
| tilegx | tilepro | visium | xstormy16 | xtensa)
insn="nop"
;;
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 04/12] LoongArch Port: Machine Decsription files.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (2 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 03/12] LoongArch Port: Regenerate gcc/configure chenglulu
@ 2021-12-24 9:28 ` chenglulu
2022-01-06 17:54 ` Xi Ruoyao
2021-12-24 9:28 ` [PATCH v4 05/12] LoongArch Port: Machine description C files and .h files chenglulu
` (8 subsequent siblings)
12 siblings, 1 reply; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/
* config/loongarch/constraints.md: New file.
* config/loongarch/generic.md: New file.
* config/loongarch/la464.md: New file.
* config/loongarch/loongarch-ftypes.def: New file.
* config/loongarch/loongarch-modes.def: New file.
* config/loongarch/loongarch.md: New file.
* config/loongarch/predicates.md: New file.
* config/loongarch/sync.md: New file.
---
gcc/config/loongarch/constraints.md | 212 ++
gcc/config/loongarch/generic.md | 132 +
gcc/config/loongarch/la464.md | 132 +
gcc/config/loongarch/loongarch-ftypes.def | 106 +
gcc/config/loongarch/loongarch-modes.def | 29 +
gcc/config/loongarch/loongarch.md | 3829 +++++++++++++++++++++
gcc/config/loongarch/predicates.md | 553 +++
gcc/config/loongarch/sync.md | 574 +++
8 files changed, 5567 insertions(+)
create mode 100644 gcc/config/loongarch/constraints.md
create mode 100644 gcc/config/loongarch/generic.md
create mode 100644 gcc/config/loongarch/la464.md
create mode 100644 gcc/config/loongarch/loongarch-ftypes.def
create mode 100644 gcc/config/loongarch/loongarch-modes.def
create mode 100644 gcc/config/loongarch/loongarch.md
create mode 100644 gcc/config/loongarch/predicates.md
create mode 100644 gcc/config/loongarch/sync.md
diff --git a/gcc/config/loongarch/constraints.md b/gcc/config/loongarch/constraints.md
new file mode 100644
index 00000000000..db88a042c58
--- /dev/null
+++ b/gcc/config/loongarch/constraints.md
@@ -0,0 +1,212 @@
+;; Constraint definitions for LoongArch.
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+;;
+;; This file is part of GCC.
+;;
+;; GCC is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation; either version 3, or (at your option)
+;; any later version.
+;;
+;; GCC is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+;;
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+;; Register constraints
+
+;; "a" "A constant call global and noplt address."
+;; "b" <-----unused
+;; "c" "A constant call local address."
+;; "d" <-----unused
+;; "e" JIRL_REGS
+;; "f" FP_REGS
+;; "g" <-----unused
+;; "h" "A constant call plt address."
+;; "i" "Matches a general integer constant." (Global non-architectural)
+;; "j" SIBCALL_REGS
+;; "k" <-----unused
+;; "l" "A signed 16-bit constant."
+;; "m" "A memory operand whose address is formed by a base register and offset
+;; that is suitable for use in instructions with the same addressing mode
+;; as @code{st.w} and @code{ld.w}."
+;; "n" "Matches a non-symbolic integer constant." (Global non-architectural)
+;; "o" "Matches an offsettable memory reference." (Global non-architectural)
+;; "p" "Matches a general address." (Global non-architectural)
+;; "q" CSR_REGS
+;; "r" GENERAL_REGS (Global non-architectural)
+;; "s" "Matches a symbolic integer constant." (Global non-architectural)
+;; "t" "A constant call weak address"
+;; "u" "A signed 52bit constant and low 32-bit is zero (for logic instructions)"
+;; "v" "A signed 64-bit constant and low 44-bit is zero (for logic instructions)."
+;; "w" "Matches any valid memory."
+;; "x" <-----unused
+;; "y" <-----unused
+;; "z" FCC_REGS
+;; "A" <-----unused
+;; "B" <-----unused
+;; "C" <-----unused
+;; "D" <-----unused
+;; "E" "Matches a floating-point constant." (Global non-architectural)
+;; "F" "Matches a floating-point constant." (Global non-architectural)
+;; "G" "Floating-point zero."
+;; "H" <-----unused
+;; "I" "A signed 12-bit constant (for arithmetic instructions)."
+;; "J" "Integer zero."
+;; "K" "An unsigned 12-bit constant (for logic instructions)."
+;; "L" <-----unused
+;; "M" <-----unused
+;; "N" <-----unused
+;; "O" <-----unused
+;; "P" <-----unused
+;; "Q" "A signed 12-bit constant"
+;; "R" <-----unused
+;; "S" <-----unused
+;; "T" <-----unused
+;; "U" <-----unused
+;; "V" "Matches a non-offsettable memory reference." (Global non-architectural)
+;; "W" "A memory address based on a member of @code{BASE_REG_CLASS}. This is
+;; true for all references."
+;; "X" "Matches anything." (Global non-architectural)
+;; "Y" -
+;; "Yd"
+;; "A constant @code{move_operand} that can be safely loaded using
+;; @code{la}."
+;; "Yx"
+;; "Z" -
+;; "ZC"
+;; "A memory operand whose address is formed by a base register and offset
+;; that is suitable for use in instructions with the same addressing mode
+;; as @code{ll.w} and @code{sc.w}."
+;; "ZB"
+;; "An address that is held in a general-purpose register.
+;; The offset is zero"
+;; "<" "Matches a pre-dec or post-dec operand." (Global non-architectural)
+;; ">" "Matches a pre-inc or post-inc operand." (Global non-architectural)
+
+(define_constraint "a"
+ "@internal
+ A constant call global and noplt address."
+ (match_operand 0 "is_const_call_global_noplt_symbol"))
+
+(define_constraint "c"
+ "@internal
+ A constant call local address."
+ (match_operand 0 "is_const_call_local_symbol"))
+
+(define_register_constraint "e" "JIRL_REGS"
+ "@internal")
+
+(define_register_constraint "f" "TARGET_HARD_FLOAT ? FP_REGS : NO_REGS"
+ "A floating-point register (if available).")
+
+(define_constraint "h"
+ "@internal
+ A constant call plt address."
+ (match_operand 0 "is_const_call_plt_symbol"))
+
+(define_register_constraint "j" "SIBCALL_REGS"
+ "@internal")
+
+(define_constraint "l"
+"A signed 16-bit constant."
+(and (match_code "const_int")
+ (match_test "IMM16_OPERAND (ival)")))
+
+(define_memory_constraint "m"
+ "A memory operand whose address is formed by a base register and offset
+ that is suitable for use in instructions with the same addressing mode
+ as @code{st.w} and @code{ld.w}."
+ (and (match_code "mem")
+ (match_test "loongarch_12bit_offset_address_p (XEXP (op, 0), mode)")))
+
+(define_register_constraint "q" "CSR_REGS"
+ "A general-purpose register except for $r0 and $r1 for lcsr.")
+
+(define_constraint "t"
+ "@internal
+ A constant call weak address."
+ (match_operand 0 "is_const_call_weak_symbol"))
+
+(define_constraint "u"
+ "A signed 52bit constant and low 32-bit is zero (for logic instructions)."
+ (and (match_code "const_int")
+ (match_test "LU32I_OPERAND (ival)")))
+
+(define_constraint "v"
+ "A nsigned 64-bit constant and low 44-bit is zero (for logic instructions)."
+ (and (match_code "const_int")
+ (match_test "LU52I_OPERAND (ival)")))
+
+(define_register_constraint "z" "FCC_REGS"
+ "A floating-point condition code register.")
+
+;; Floating-point constraints
+
+(define_constraint "G"
+ "Floating-point zero."
+ (and (match_code "const_double")
+ (match_test "op == CONST0_RTX (mode)")))
+
+;; Integer constraints
+
+(define_constraint "I"
+ "A signed 12-bit constant (for arithmetic instructions)."
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND (ival)")))
+
+(define_constraint "J"
+ "Integer zero."
+ (and (match_code "const_int")
+ (match_test "ival == 0")))
+
+(define_constraint "K"
+ "An unsigned 12-bit constant (for logic instructions)."
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND_UNSIGNED (ival)")))
+
+;; General constraints
+
+(define_constraint "Q"
+ "@internal"
+ (match_operand 0 "const_arith_operand"))
+
+(define_memory_constraint "W"
+ "@internal
+ A memory address based on a member of @code{BASE_REG_CLASS}. This is
+ true for allreferences."
+ (and (match_code "mem")
+ (match_operand 0 "memory_operand")
+ (and (not (match_operand 0 "stack_operand"))
+ (not (match_test "CONSTANT_P (XEXP (op, 0))")))))
+
+(define_constraint "Yd"
+ "@internal
+ A constant @code{move_operand} that can be safely loaded using
+ @code{la}."
+ (and (match_operand 0 "move_operand")
+ (match_test "CONSTANT_P (op)")))
+
+(define_constraint "Yx"
+ "@internal"
+ (match_operand 0 "low_bitmask_operand"))
+
+(define_memory_constraint "ZC"
+ "A memory operand whose address is formed by a base register and offset
+ that is suitable for use in instructions with the same addressing mode
+ as @code{ll.w} and @code{sc.w}."
+ (and (match_code "mem")
+ (match_test "loongarch_14bit_shifted_offset_address_p (XEXP (op, 0), mode)")))
+
+(define_memory_constraint "ZB"
+ "@internal
+ An address that is held in a general-purpose register.
+ The offset is zero"
+ (and (match_code "mem")
+ (match_test "GET_CODE (XEXP (op,0)) == REG")))
+
diff --git a/gcc/config/loongarch/generic.md b/gcc/config/loongarch/generic.md
new file mode 100644
index 00000000000..d1dc2d7fa0c
--- /dev/null
+++ b/gcc/config/loongarch/generic.md
@@ -0,0 +1,132 @@
+;; Generic DFA-based pipeline description for LoongArch targets
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+;; Based on MIPS target for GNU compiler.
+
+;; This file is part of GCC.
+
+;; GCC is free software; you can redistribute it and/or modify it
+;; under the terms of the GNU General Public License as published
+;; by the Free Software Foundation; either version 3, or (at your
+;; option) any later version.
+
+;; GCC is distributed in the hope that it will be useful, but WITHOUT
+;; ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+;; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+;; License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+
+;; Pipeline descriptions.
+;;
+;; generic.md provides a fallback for processors without a specific
+;; pipeline description. It is derived from the old define_function_unit
+;; version and uses the "alu" and "imuldiv" units declared below.
+;;
+;; Some of the processor-specific files are also derived from old
+;; define_function_unit descriptions and simply override the parts of
+;; generic.md that don't apply. The other processor-specific files
+;; are self-contained.
+(define_automaton "alu,imuldiv")
+
+(define_cpu_unit "alu" "alu")
+(define_cpu_unit "imuldiv" "imuldiv")
+
+;; Ghost instructions produce no real code and introduce no hazards.
+;; They exist purely to express an effect on dataflow.
+(define_insn_reservation "ghost" 0
+ (eq_attr "type" "ghost")
+ "nothing")
+
+;; This file is derived from the old define_function_unit description.
+;; Each reservation can be overridden on a processor-by-processor basis.
+
+(define_insn_reservation "generic_alu" 1
+ (eq_attr "type" "unknown,prefetch,prefetchx,condmove,const,arith,
+ shift,slt,clz,trap,multi,nop,logical,signext,move")
+ "alu")
+
+(define_insn_reservation "generic_load" 3
+ (eq_attr "type" "load,fpload,fpidxload")
+ "alu")
+
+(define_insn_reservation "generic_store" 1
+ (eq_attr "type" "store,fpstore,fpidxstore")
+ "alu")
+
+(define_insn_reservation "generic_xfer" 2
+ (eq_attr "type" "mftg,mgtf")
+ "alu")
+
+(define_insn_reservation "generic_branch" 1
+ (eq_attr "type" "branch,jump,call")
+ "alu")
+
+(define_insn_reservation "generic_imul" 17
+ (eq_attr "type" "imul")
+ "imuldiv*17")
+
+(define_insn_reservation "generic_fcvt" 1
+ (eq_attr "type" "fcvt")
+ "alu")
+
+(define_insn_reservation "generic_fmove" 2
+ (eq_attr "type" "fabs,fneg,fmove")
+ "alu")
+
+(define_insn_reservation "generic_fcmp" 3
+ (eq_attr "type" "fcmp")
+ "alu")
+
+(define_insn_reservation "generic_fadd" 4
+ (eq_attr "type" "fadd")
+ "alu")
+
+(define_insn_reservation "generic_fmul_single" 7
+ (and (eq_attr "type" "fmul,fmadd")
+ (eq_attr "mode" "SF"))
+ "alu")
+
+(define_insn_reservation "generic_fmul_double" 8
+ (and (eq_attr "type" "fmul,fmadd")
+ (eq_attr "mode" "DF"))
+ "alu")
+
+(define_insn_reservation "generic_fdiv_single" 23
+ (and (eq_attr "type" "fdiv,frdiv")
+ (eq_attr "mode" "SF"))
+ "alu")
+
+(define_insn_reservation "generic_fdiv_double" 36
+ (and (eq_attr "type" "fdiv,frdiv")
+ (eq_attr "mode" "DF"))
+ "alu")
+
+(define_insn_reservation "generic_fsqrt_single" 54
+ (and (eq_attr "type" "fsqrt,frsqrt")
+ (eq_attr "mode" "SF"))
+ "alu")
+
+(define_insn_reservation "generic_fsqrt_double" 112
+ (and (eq_attr "type" "fsqrt,frsqrt")
+ (eq_attr "mode" "DF"))
+ "alu")
+
+(define_insn_reservation "generic_atomic" 10
+ (eq_attr "type" "atomic")
+ "alu")
+
+;; Sync loop consists of (in order)
+;; (1) optional sync,
+;; (2) LL instruction,
+;; (3) branch and 1-2 ALU instructions,
+;; (4) SC instruction,
+;; (5) branch and ALU instruction.
+;; The net result of this reservation is a big delay with a flush of
+;; ALU pipeline.
+(define_insn_reservation "generic_sync_loop" 40
+ (eq_attr "type" "syncloop")
+ "alu*39")
diff --git a/gcc/config/loongarch/la464.md b/gcc/config/loongarch/la464.md
new file mode 100644
index 00000000000..3a8cc6051ff
--- /dev/null
+++ b/gcc/config/loongarch/la464.md
@@ -0,0 +1,132 @@
+;; Pipeline model for LoongArch la464 cores.
+
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+
+;; This file is part of GCC.
+;;
+;; GCC is free software; you can redistribute it and/or modify it
+;; under the terms of the GNU General Public License as published
+;; by the Free Software Foundation; either version 3, or (at your
+;; option) any later version.
+;;
+;; GCC is distributed in the hope that it will be useful, but WITHOUT
+;; ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+;; or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+;; License for more details.
+;;
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+;; Uncomment the following line to output automata for debugging.
+;; (automata_option "v")
+
+;; Automaton for integer instructions.
+(define_automaton "la464_a_alu")
+
+;; Automaton for floating-point instructions.
+(define_automaton "la464_a_falu")
+
+;; Automaton for memory operations.
+(define_automaton "la464_a_mem")
+
+;; Describe the resources.
+
+(define_cpu_unit "la464_alu1" "la464_a_alu")
+(define_cpu_unit "la464_alu2" "la464_a_alu")
+(define_cpu_unit "la464_mem1" "la464_a_mem")
+(define_cpu_unit "la464_mem2" "la464_a_mem")
+(define_cpu_unit "la464_falu1" "la464_a_falu")
+(define_cpu_unit "la464_falu2" "la464_a_falu")
+
+;; Describe instruction reservations.
+
+(define_insn_reservation "la464_arith" 1
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "arith,clz,const,logical,
+ move,nop,shift,signext,slt"))
+ "la464_alu1 | la464_alu2")
+
+(define_insn_reservation "la464_branch" 1
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "branch,jump,call,condmove,trap"))
+ "la464_alu1 | la464_alu2")
+
+(define_insn_reservation "la464_imul" 7
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "imul"))
+ "la464_alu1 | la464_alu2")
+
+(define_insn_reservation "la464_idiv_si" 12
+ (and (match_test "TARGET_ARCH_LA464")
+ (and (eq_attr "type" "idiv")
+ (eq_attr "mode" "SI")))
+ "la464_alu1 | la464_alu2")
+
+(define_insn_reservation "la464_idiv_di" 25
+ (and (match_test "TARGET_ARCH_LA464")
+ (and (eq_attr "type" "idiv")
+ (eq_attr "mode" "DI")))
+ "la464_alu1 | la464_alu2")
+
+(define_insn_reservation "la464_load" 4
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "load"))
+ "la464_mem1 | la464_mem2")
+
+(define_insn_reservation "la464_gpr_fp" 16
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "mftg,mgtf"))
+ "la464_mem1")
+
+(define_insn_reservation "la464_fpload" 4
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "fpload"))
+ "la464_mem1 | la464_mem2")
+
+(define_insn_reservation "la464_prefetch" 0
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "prefetch,prefetchx"))
+ "la464_mem1 | la464_mem2")
+
+(define_insn_reservation "la464_store" 0
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "store,fpstore,fpidxstore"))
+ "la464_mem1 | la464_mem2")
+
+(define_insn_reservation "la464_fadd" 4
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "fadd,fmul,fmadd"))
+ "la464_falu1 | la464_falu2")
+
+(define_insn_reservation "la464_fcmp" 2
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "fabs,fcmp,fmove,fneg"))
+ "la464_falu1 | la464_falu2")
+
+(define_insn_reservation "la464_fcvt" 4
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "fcvt"))
+ "la464_falu1 | la464_falu2")
+
+(define_insn_reservation "la464_fdiv_sf" 12
+ (and (match_test "TARGET_ARCH_LA464")
+ (and (eq_attr "type" "fdiv,frdiv,fsqrt,frsqrt")
+ (eq_attr "mode" "SF")))
+ "la464_falu1 | la464_falu2")
+
+(define_insn_reservation "la464_fdiv_df" 19
+ (and (match_test "TARGET_ARCH_LA464")
+ (and (eq_attr "type" "fdiv,frdiv,fsqrt,frsqrt")
+ (eq_attr "mode" "DF")))
+ "la464_falu1 | la464_falu2")
+
+;; Force single-dispatch for unknown or multi.
+(define_insn_reservation "la464_unknown" 1
+ (and (match_test "TARGET_ARCH_LA464")
+ (eq_attr "type" "unknown,multi,atomic,syncloop"))
+ "la464_alu1 + la464_alu2 + la464_falu1
+ + la464_falu2 + la464_mem1 + la464_mem2")
+
+;; End of DFA-based pipeline description for la464
diff --git a/gcc/config/loongarch/loongarch-ftypes.def b/gcc/config/loongarch/loongarch-ftypes.def
new file mode 100644
index 00000000000..e370337df1d
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-ftypes.def
@@ -0,0 +1,106 @@
+/* Definitions of prototypes for LoongArch built-in functions.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS target for GNU ckompiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+/* Invoke DEF_LARCH_FTYPE (NARGS, LIST) for each prototype used by
+ LoongArch built-in functions, where:
+
+ NARGS is the number of arguments.
+ LIST contains the return-type code followed by the codes for each
+ argument type.
+
+ Argument- and return-type codes are either modes or one of the following:
+
+ VOID for void_type_node
+ INT for integer_type_node
+ POINTER for ptr_type_node
+
+ (we don't use PTR because that's a ANSI-compatibillity macro).
+
+ Please keep this list lexicographically sorted by the LIST argument. */
+
+DEF_LARCH_FTYPE (1, (DF, DF))
+DEF_LARCH_FTYPE (2, (DF, DF, DF))
+
+DEF_LARCH_FTYPE (1, (DI, DI))
+DEF_LARCH_FTYPE (2, (DI, DI, DI))
+DEF_LARCH_FTYPE (3, (DI, DI, DI, QI))
+DEF_LARCH_FTYPE (2, (DI, DI, SI))
+DEF_LARCH_FTYPE (3, (DI, DI, SI, SI))
+DEF_LARCH_FTYPE (2, (DI, DI, UQI))
+DEF_LARCH_FTYPE (3, (DI, DI, USI, USI))
+DEF_LARCH_FTYPE (2, (DI, POINTER, SI))
+DEF_LARCH_FTYPE (1, (DI, SI))
+DEF_LARCH_FTYPE (2, (DI, SI, SI))
+DEF_LARCH_FTYPE (1, (DI, UQI))
+DEF_LARCH_FTYPE (2, (DI, USI, USI))
+
+DEF_LARCH_FTYPE (2, (HI, HI, HI))
+
+DEF_LARCH_FTYPE (2, (INT, DF, DF))
+DEF_LARCH_FTYPE (2, (INT, SF, SF))
+
+DEF_LARCH_FTYPE (2, (QI, QI, QI))
+
+DEF_LARCH_FTYPE (1, (SF, SF))
+DEF_LARCH_FTYPE (2, (SF, SF, SF))
+
+DEF_LARCH_FTYPE (2, (SI, DI, SI))
+DEF_LARCH_FTYPE (2, (SI, HI, SI))
+DEF_LARCH_FTYPE (2, (SI, POINTER, SI))
+DEF_LARCH_FTYPE (2, (SI, QI, SI))
+DEF_LARCH_FTYPE (1, (SI, SI))
+DEF_LARCH_FTYPE (2, (SI, SI, SI))
+DEF_LARCH_FTYPE (3, (SI, SI, SI, QI))
+DEF_LARCH_FTYPE (3, (SI, SI, SI, SI))
+DEF_LARCH_FTYPE (2, (SI, SI, UQI))
+DEF_LARCH_FTYPE (1, (SI, UDI))
+DEF_LARCH_FTYPE (1, (SI, UQI))
+DEF_LARCH_FTYPE (1, (SI, VOID))
+
+DEF_LARCH_FTYPE (2, (UDI, UDI, UDI))
+DEF_LARCH_FTYPE (3, (UDI, UDI, UDI, USI))
+DEF_LARCH_FTYPE (2, (UDI, UDI, USI))
+DEF_LARCH_FTYPE (1, (UDI, USI))
+
+DEF_LARCH_FTYPE (1, (UHI, USI))
+
+DEF_LARCH_FTYPE (1, (UQI, USI))
+
+DEF_LARCH_FTYPE (1, (USI, UQI))
+DEF_LARCH_FTYPE (1, (USI, USI))
+DEF_LARCH_FTYPE (2, (USI, USI, USI))
+DEF_LARCH_FTYPE (3, (USI, USI, USI, USI))
+DEF_LARCH_FTYPE (1, (USI, VOID))
+
+DEF_LARCH_FTYPE (2, (VOID, DI, DI))
+DEF_LARCH_FTYPE (2, (VOID, DI, UQI))
+DEF_LARCH_FTYPE (2, (VOID, SI, CVPOINTER))
+DEF_LARCH_FTYPE (2, (VOID, SI, SI))
+DEF_LARCH_FTYPE (2, (VOID, SI, UQI))
+DEF_LARCH_FTYPE (2, (VOID, UDI, USI))
+DEF_LARCH_FTYPE (2, (VOID, UHI, USI))
+DEF_LARCH_FTYPE (2, (VOID, UQI, SI))
+DEF_LARCH_FTYPE (2, (VOID, UQI, USI))
+DEF_LARCH_FTYPE (1, (VOID, USI))
+DEF_LARCH_FTYPE (3, (VOID, USI, UDI, SI))
+DEF_LARCH_FTYPE (2, (VOID, USI, UQI))
+DEF_LARCH_FTYPE (2, (VOID, USI, USI))
+DEF_LARCH_FTYPE (3, (VOID, USI, USI, SI))
diff --git a/gcc/config/loongarch/loongarch-modes.def b/gcc/config/loongarch/loongarch-modes.def
new file mode 100644
index 00000000000..b33e94504f1
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-modes.def
@@ -0,0 +1,29 @@
+/* LoongArch extra machine modes.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS target for GNU compiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+FLOAT_MODE (TF, 16, ieee_quad_format);
+
+VECTOR_MODES (FLOAT, 8); /* V4HF V2SF */
+
+/* For floating point conditions in FCC registers. */
+CC_MODE (FCC);
+
+INT_MODE (OI, 32);
diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
new file mode 100644
index 00000000000..15611d60686
--- /dev/null
+++ b/gcc/config/loongarch/loongarch.md
@@ -0,0 +1,3829 @@
+;; Machine Description for LoongArch for GNU compiler.
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+;; Based on MIPS target for GNU compiler.
+
+;; This file is part of GCC.
+
+;; GCC is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation; either version 3, or (at your option)
+;; any later version.
+
+;; GCC is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+(define_c_enum "unspec" [
+ ;; Integer operations that are too cumbersome to describe directly.
+ UNSPEC_REVB_2H
+ UNSPEC_REVB_4H
+ UNSPEC_REVH_D
+
+ ;; Floating-point moves.
+ UNSPEC_LOAD_LOW
+ UNSPEC_LOAD_HIGH
+ UNSPEC_STORE_WORD
+ UNSPEC_MOVGR2FRH
+ UNSPEC_MOVFRH2GR
+
+ ;; Floating-point environment.
+ UNSPEC_MOVFCSR2GR
+ UNSPEC_MOVGR2FCSR
+
+ ;; GP manipulation.
+ UNSPEC_EH_RETURN
+
+ ;;
+ UNSPEC_FRINT
+ UNSPEC_FCLASS
+ UNSPEC_BYTEPICK_W
+ UNSPEC_BYTEPICK_D
+ UNSPEC_BITREV_4B
+ UNSPEC_BITREV_8B
+
+ ;; Symbolic accesses.
+ UNSPEC_LOAD_CALL
+
+ ;; Blockage and synchronisation.
+ UNSPEC_BLOCKAGE
+ UNSPEC_DBAR
+ UNSPEC_IBAR
+
+ ;; CPUCFG
+ UNSPEC_CPUCFG
+ UNSPEC_ASRTLE_D
+ UNSPEC_ASRTGT_D
+
+ UNSPEC_CSRRD
+ UNSPEC_CSRWR
+ UNSPEC_CSRXCHG
+ UNSPEC_IOCSRRD
+ UNSPEC_IOCSRWR
+
+ ;; CACOP
+ UNSPEC_CACOP
+
+ ;; PTE
+ UNSPEC_LDDIR
+ UNSPEC_LDPTE
+
+ ;; Interrupt handling.
+ UNSPEC_ERTN
+
+ ;; Used in a call expression in place of args_size. It's present for PIC
+ ;; indirect calls where it contains args_size and the function symbol.
+ UNSPEC_CALL_ATTR
+
+ ;; Stack checking.
+ UNSPEC_PROBE_STACK_RANGE
+
+ ;; TLS
+ UNSPEC_TLS_GD
+ UNSPEC_TLS_LD
+ UNSPEC_TLS_LE
+ UNSPEC_TLS_IE
+
+ ;; load immediate
+ UNSPEC_LU52I_D
+
+ UNSPEC_TIE
+
+ ;; Register save and restore.
+ UNSPECV_GPR_SAVE
+ UNSPECV_GPR_RESTORE
+
+ ;; CRC
+ UNSPEC_CRC
+ UNSPEC_CRCC
+])
+
+(define_constants
+ [(RETURN_ADDR_REGNUM 1)
+ (T0_REGNUM 12)
+ (T1_REGNUM 13)
+ (S0_REGNUM 23)
+
+ ;; PIC long branch sequences are never longer than 100 bytes.
+ (MAX_PIC_BRANCH_LENGTH 100)
+])
+
+(include "predicates.md")
+(include "constraints.md")
+\f
+;; ....................
+;;
+;; Attributes
+;;
+;; ....................
+
+(define_attr "got" "unset,load"
+ (const_string "unset"))
+
+;; For jirl instructions, this attribute is DIRECT when the target address
+;; is symbolic and INDIRECT when it is a register.
+(define_attr "jirl" "unset,direct,indirect"
+ (const_string "unset"))
+
+
+;; Classification of moves, extensions and truncations. Most values
+;; are as for "type" (see below) but there are also the following
+;; move-specific values:
+;;
+;; sll0 "slli.w DEST,SRC,0", which on 64-bit targets is guaranteed
+;; to produce a sign-extended DEST, even if SRC is not
+;; properly sign-extended
+;; pick_ins BSTRPICK.W, BSTRPICK.D, BSTRINS.W or BSTRINS.D instruction
+;; andi a single ANDI instruction
+;; shift_shift a shift left followed by a shift right
+;;
+;; This attribute is used to determine the instruction's length and
+;; scheduling type. For doubleword moves, the attribute always describes
+;; the split instructions; in some cases, it is more appropriate for the
+;; scheduling type to be "multi" instead.
+(define_attr "move_type"
+ "unknown,load,fpload,store,fpstore,mgtf,mftg,imul,move,fmove,
+ const,signext,pick_ins,logical,arith,sll0,andi,shift_shift"
+ (const_string "unknown"))
+
+(define_attr "alu_type" "unknown,add,sub,not,nor,and,or,xor"
+ (const_string "unknown"))
+
+;; Main data type used by the insn
+(define_attr "mode" "unknown,none,QI,HI,SI,DI,TI,OI,SF,DF,TF,FCC"
+ (const_string "unknown"))
+
+;; True if the main data type is twice the size of a word.
+(define_attr "dword_mode" "no,yes"
+ (cond [(and (eq_attr "mode" "DI,DF")
+ (not (match_test "TARGET_64BIT")))
+ (const_string "yes")
+
+ (and (eq_attr "mode" "TI,TF")
+ (match_test "TARGET_64BIT"))
+ (const_string "yes")]
+ (const_string "no")))
+
+;; True if the main data type is four times of the size of a word.
+(define_attr "qword_mode" "no,yes"
+ (cond [(and (eq_attr "mode" "TI,TF")
+ (not (match_test "TARGET_64BIT")))
+ (const_string "yes")]
+ (const_string "no")))
+
+;; True if the main data type is eight times of the size of a word.
+(define_attr "oword_mode" "no,yes"
+ (cond [(and (eq_attr "mode" "OI")
+ (not (match_test "TARGET_64BIT")))
+ (const_string "yes")]
+ (const_string "no")))
+
+;; Classification of each insn.
+;; branch conditional branch
+;; jump unconditional jump
+;; call unconditional call
+;; load load instruction(s)
+;; fpload floating point load
+;; fpidxload floating point indexed load
+;; store store instruction(s)
+;; fpstore floating point store
+;; fpidxstore floating point indexed store
+;; prefetch memory prefetch (register + offset)
+;; prefetchx memory indexed prefetch (register + register)
+;; condmove conditional moves
+;; mgtf move generate register to float register
+;; mftg move float register to generate register
+;; const load constant
+;; arith integer arithmetic instructions
+;; logical integer logical instructions
+;; shift integer shift instructions
+;; slt set less than instructions
+;; signext sign extend instructions
+;; clz the clz and clo instructions
+;; trap trap if instructions
+;; imul integer multiply
+;; idiv integer divide
+;; move integer register move ({,D}ADD{,U} with rt = 0)
+;; fmove floating point register move
+;; fadd floating point add/subtract
+;; fmul floating point multiply
+;; fmadd floating point multiply-add
+;; fdiv floating point divide
+;; frdiv floating point reciprocal divide
+;; fabs floating point absolute value
+;; fneg floating point negation
+;; fcmp floating point compare
+;; fcvt floating point convert
+;; fsqrt floating point square root
+;; frsqrt floating point reciprocal square root
+;; multi multiword sequence (or user asm statements)
+;; atomic atomic memory update instruction
+;; syncloop memory atomic operation implemented as a sync loop
+;; nop no operation
+;; ghost an instruction that produces no real code
+(define_attr "type"
+ "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxstore,
+ prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
+ shift,slt,signext,clz,trap,imul,idiv,move,
+ fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcvt,fsqrt,
+ frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
+ (cond [(eq_attr "jirl" "!unset") (const_string "call")
+ (eq_attr "got" "load") (const_string "load")
+
+ (eq_attr "alu_type" "add,sub") (const_string "arith")
+
+ (eq_attr "alu_type" "not,nor,and,or,xor") (const_string "logical")
+
+ ;; If a doubleword move uses these expensive instructions,
+ ;; it is usually better to schedule them in the same way
+ ;; as the singleword form, rather than as "multi".
+ (eq_attr "move_type" "load") (const_string "load")
+ (eq_attr "move_type" "fpload") (const_string "fpload")
+ (eq_attr "move_type" "store") (const_string "store")
+ (eq_attr "move_type" "fpstore") (const_string "fpstore")
+ (eq_attr "move_type" "mgtf") (const_string "mgtf")
+ (eq_attr "move_type" "mftg") (const_string "mftg")
+
+ ;; These types of move are always single insns.
+ (eq_attr "move_type" "imul") (const_string "imul")
+ (eq_attr "move_type" "fmove") (const_string "fmove")
+ (eq_attr "move_type" "signext") (const_string "signext")
+ (eq_attr "move_type" "pick_ins") (const_string "arith")
+ (eq_attr "move_type" "arith") (const_string "arith")
+ (eq_attr "move_type" "logical") (const_string "logical")
+ (eq_attr "move_type" "sll0") (const_string "shift")
+ (eq_attr "move_type" "andi") (const_string "logical")
+
+ ;; These types of move are always split.
+ (eq_attr "move_type" "shift_shift")
+ (const_string "multi")
+
+ ;; These types of move are split for octaword modes only.
+ (and (eq_attr "move_type" "move,const")
+ (eq_attr "oword_mode" "yes"))
+ (const_string "multi")
+
+ ;; These types of move are split for quadword modes only.
+ (and (eq_attr "move_type" "move,const")
+ (eq_attr "qword_mode" "yes"))
+ (const_string "multi")
+
+ ;; These types of move are split for doubleword modes only.
+ (and (eq_attr "move_type" "move,const")
+ (eq_attr "dword_mode" "yes"))
+ (const_string "multi")
+ (eq_attr "move_type" "move") (const_string "move")
+ (eq_attr "move_type" "const") (const_string "const")]
+ (const_string "unknown")))
+
+;; Mode for conversion types (fcvt)
+;; I2S integer to float single (SI/DI to SF)
+;; I2D integer to float double (SI/DI to DF)
+;; S2I float to integer (SF to SI/DI)
+;; D2I float to integer (DF to SI/DI)
+;; D2S double to float single
+;; S2D float single to double
+
+(define_attr "cnv_mode" "unknown,I2S,I2D,S2I,D2I,D2S,S2D"
+ (const_string "unknown"))
+
+(define_attr "compression" "none,all"
+ (const_string "none"))
+
+;; The number of individual instructions that a non-branch pattern generates
+(define_attr "insn_count" ""
+ (cond [;; "Ghost" instructions occupy no space.
+ (eq_attr "type" "ghost")
+ (const_int 0)
+
+ ;; Check for doubleword moves that are decomposed into two
+ ;; instructions.
+ (and (eq_attr "move_type" "mgtf,mftg,move")
+ (eq_attr "dword_mode" "yes"))
+ (const_int 2)
+
+ ;; Check for quadword moves that are decomposed into four
+ ;; instructions.
+ (and (eq_attr "move_type" "mgtf,mftg,move")
+ (eq_attr "qword_mode" "yes"))
+ (const_int 4)
+
+ ;; Check for Octaword moves that are decomposed into eight
+ ;; instructions.
+ (and (eq_attr "move_type" "mgtf,mftg,move")
+ (eq_attr "oword_mode" "yes"))
+ (const_int 8)
+
+ ;; Constants, loads and stores are handled by external routines.
+ (and (eq_attr "move_type" "const")
+ (eq_attr "dword_mode" "yes"))
+ (symbol_ref "loongarch_split_const_insns (operands[1])")
+ (eq_attr "move_type" "const")
+ (symbol_ref "loongarch_const_insns (operands[1])")
+ (eq_attr "move_type" "load,fpload")
+ (symbol_ref "loongarch_load_store_insns (operands[1], insn)")
+ (eq_attr "move_type" "store,fpstore")
+ (symbol_ref "loongarch_load_store_insns (operands[0], insn)")
+
+ (eq_attr "type" "idiv")
+ (symbol_ref "loongarch_idiv_insns (GET_MODE (PATTERN (insn)))")]
+(const_int 1)))
+
+;; Length of instruction in bytes.
+(define_attr "length" ""
+ (cond [
+ ;; Branch instructions have a range of [-0x20000,0x1fffc].
+ ;; If a branch is outside this range, we have a choice of two
+ ;; sequences.
+ ;;
+ ;; For PIC, an out-of-range branch like:
+ ;;
+ ;; bne r1,r2,target
+ ;;
+ ;; becomes the equivalent of:
+ ;;
+ ;; beq r1,r2,1f
+ ;; la rd,target
+ ;; jirl $r0,rd,0
+ ;; 1:
+ ;;
+ ;; The non-PIC case is similar except that we use a direct
+ ;; jump instead of an la/jirl pair. Since the target of this
+ ;; jump is an absolute 28-bit bit address (the other bits
+ ;; coming from the address of the delay slot) this form cannot
+ ;; cross a 256MB boundary. We could provide the option of
+ ;; using la/jirl in this case too, but we do not do so at
+ ;; present.
+ ;;
+ ;; from the shorten_branches reference address.
+ (eq_attr "type" "branch")
+ (cond [;; Any variant can handle the 17-bit range.
+ (and (le (minus (match_dup 0) (pc)) (const_int 65532))
+ (le (minus (pc) (match_dup 0)) (const_int 65534)))
+ (const_int 4)
+
+ ;; The non-PIC case: branch, and J.
+ (match_test "!flag_pic")
+ (const_int 8)]
+
+ ;; Use MAX_PIC_BRANCH_LENGTH as a (gross) overestimate.
+ ;; loongarch_adjust_insn_length substitutes the correct length.
+ ;;
+ ;; Note that we can't simply use (symbol_ref ...) here
+ ;; because genattrtab needs to know the maximum length
+ ;; of an insn.
+ (const_int MAX_PIC_BRANCH_LENGTH))]
+ (symbol_ref "get_attr_insn_count (insn) * 4")))
+
+;; Attribute describing the processor.
+;; The type of hardware hazard associated with this instruction.
+;; DELAY means that the next instruction cannot read the result
+;; of this one.
+(define_attr "hazard" "none,delay,forbidden_slot"
+ (const_string "none"))
+
+;; Can the instruction be put into a delay slot?
+(define_attr "can_delay" "no,yes"
+ (if_then_else (and (eq_attr "type" "!branch,call,jump")
+ (eq_attr "hazard" "none")
+ (match_test "get_attr_insn_count (insn) == 1"))
+ (const_string "yes")
+ (const_string "no")))
+
+;; Describe a user's asm statement.
+(define_asm_attributes
+ [(set_attr "type" "multi")
+ (set_attr "can_delay" "no")])
+\f
+;; This mode iterator allows 32-bit and 64-bit GPR patterns to be generated
+;; from the same template.
+(define_mode_iterator GPR [SI (DI "TARGET_64BIT")])
+
+;; A copy of GPR that can be used when a pattern has two independent
+;; modes.
+(define_mode_iterator GPR2 [SI (DI "TARGET_64BIT")])
+
+;; Likewise, but for XLEN-sized quantities.
+(define_mode_iterator X [(SI "!TARGET_64BIT") (DI "TARGET_64BIT")])
+
+;; This mode iterator allows 16-bit and 32-bit GPR patterns and 32-bit 64-bit
+;; FPR patterns to be generated from the same template.
+(define_mode_iterator JOIN_MODE [HI
+ SI
+ (SF "TARGET_HARD_FLOAT")
+ (DF "TARGET_DOUBLE_FLOAT")])
+
+;; This mode iterator allows :P to be used for patterns that operate on
+;; pointer-sized quantities. Exactly one of the two alternatives will match.
+(define_mode_iterator P [(SI "Pmode == SImode") (DI "Pmode == DImode")])
+
+;; 64-bit modes for which we provide move patterns.
+(define_mode_iterator MOVE64 [DI DF])
+
+;; 128-bit modes for which we provide move patterns on 64-bit targets.
+(define_mode_iterator MOVE128 [TI TF])
+
+;; Iterator for sub-32-bit integer modes.
+(define_mode_iterator SHORT [QI HI])
+
+;; Likewise the 64-bit truncate-and-shift patterns.
+(define_mode_iterator SUBDI [QI HI SI])
+
+;; This mode iterator allows the QI HI SI and DI extension patterns to be
+(define_mode_iterator QHWD [QI HI SI (DI "TARGET_64BIT")])
+
+;; Iterator for hardware-supported floating-point modes.
+(define_mode_iterator ANYF [(SF "TARGET_HARD_FLOAT")
+ (DF "TARGET_DOUBLE_FLOAT")])
+
+;; A floating-point mode for which moves involving FPRs may need to be split.
+(define_mode_iterator SPLITF
+ [(DF "!TARGET_64BIT && TARGET_DOUBLE_FLOAT")
+ (DI "!TARGET_64BIT && TARGET_DOUBLE_FLOAT")
+ (TF "TARGET_64BIT && TARGET_DOUBLE_FLOAT")])
+
+;; In GPR templates, a string like "mul.<d>" will expand to "mul" in the
+;; 32-bit "mul.w" and "mul.d" in the 64-bit version.
+(define_mode_attr d [(SI "w") (DI "d")])
+
+;; This attribute gives the length suffix for a load or store instruction.
+;; The same suffixes work for zero and sign extensions.
+(define_mode_attr size [(QI "b") (HI "h") (SI "w") (DI "d")])
+(define_mode_attr SIZE [(QI "B") (HI "H") (SI "W") (DI "D")])
+
+;; This attributes gives the mode mask of a SHORT.
+(define_mode_attr mask [(QI "0x00ff") (HI "0xffff")])
+
+;; This attributes gives the size (bits) of a SHORT.
+(define_mode_attr qi_hi [(QI "7") (HI "15")])
+
+;; Instruction names for stores.
+(define_mode_attr store [(QI "sb") (HI "sh") (SI "sw") (DI "sd")])
+
+;; Similarly for LoongArch indexed FPR loads and stores.
+(define_mode_attr floadx [(SF "fldx.s") (DF "fldx.d") (V2SF "fldx.d")])
+(define_mode_attr fstorex [(SF "fstx.s") (DF "fstx.d") (V2SF "fstx.d")])
+
+;; Similarly for LoongArch indexed GPR loads and stores.
+(define_mode_attr loadx [(QI "ldx.b")
+ (HI "ldx.h")
+ (SI "ldx.w")
+ (DI "ldx.d")])
+(define_mode_attr storex [(QI "stx.b")
+ (HI "stx.h")
+ (SI "stx.w")
+ (DI "stx.d")])
+
+;; This attribute gives the format suffix for floating-point operations.
+(define_mode_attr fmt [(SF "s") (DF "d")])
+
+;; This attribute gives the upper-case mode name for one unit of a
+;; floating-point mode or vector mode.
+(define_mode_attr UNITMODE [(SF "SF") (DF "DF") (V2SF "SF")])
+
+;; This attribute gives the integer mode that has half the size of
+;; the controlling mode.
+(define_mode_attr HALFMODE [(DF "SI") (DI "SI") (V2SF "SI") (TF "DI")])
+
+;; This attribute gives the integer prefix for some instructions templates.
+(define_mode_attr p [(SI "") (DI "d")])
+
+;; This code iterator allows signed and unsigned widening multiplications
+;; to use the same template.
+(define_code_iterator any_extend [sign_extend zero_extend])
+
+;; This code iterator allows the two right shift instructions to be
+;; generated from the same template.
+(define_code_iterator any_shiftrt [ashiftrt lshiftrt])
+
+;; This code iterator allows the three shift instructions to be generated
+;; from the same template.
+(define_code_iterator any_shift [ashift ashiftrt lshiftrt])
+
+;; This code iterator allows the three bitwise instructions to be generated
+;; from the same template.
+(define_code_iterator any_bitwise [and ior xor])
+
+;; This code iterator allows unsigned and signed division to be generated
+;; from the same template.
+(define_code_iterator any_div [div udiv mod umod])
+
+;; This code iterator allows all native floating-point comparisons to be
+;; generated from the same template.
+(define_code_iterator fcond [unordered uneq unlt unle eq lt le
+ ordered ltgt ne ge gt unge ungt])
+
+;; Equality operators.
+(define_code_iterator equality_op [eq ne])
+
+;; These code iterators allow the signed and unsigned scc operations to use
+;; the same template.
+(define_code_iterator any_gt [gt gtu])
+(define_code_iterator any_ge [ge geu])
+(define_code_iterator any_lt [lt ltu])
+(define_code_iterator any_le [le leu])
+
+(define_code_iterator any_return [return simple_return])
+
+;; <u> expands to an empty string when doing a signed operation and
+;; "u" when doing an unsigned operation.
+(define_code_attr u [(sign_extend "") (zero_extend "u")
+ (div "") (udiv "u")
+ (mod "") (umod "u")
+ (gt "") (gtu "u")
+ (ge "") (geu "u")
+ (lt "") (ltu "u")
+ (le "") (leu "u")])
+
+;; <U> is like <u> except uppercase.
+(define_code_attr U [(sign_extend "") (zero_extend "U")])
+
+;; <su> is like <u>, but the signed form expands to "s" rather than "".
+(define_code_attr su [(sign_extend "s") (zero_extend "u")])
+
+;; <optab> expands to the name of the optab for a particular code.
+(define_code_attr optab [(ashift "ashl")
+ (ashiftrt "ashr")
+ (lshiftrt "lshr")
+ (ior "ior")
+ (xor "xor")
+ (and "and")
+ (plus "add")
+ (minus "sub")
+ (mult "mul")
+ (div "div")
+ (udiv "udiv")
+ (mod "mod")
+ (umod "umod")
+ (return "return")
+ (simple_return "simple_return")])
+
+;; <insn> expands to the name of the insn that implements a particular code.
+(define_code_attr insn [(ashift "sll")
+ (ashiftrt "sra")
+ (lshiftrt "srl")
+ (ior "or")
+ (xor "xor")
+ (and "and")
+ (plus "addu")
+ (minus "subu")
+ (div "div")
+ (udiv "div")
+ (mod "mod")
+ (umod "mod")])
+
+;; <fcond> is the c.cond.fmt condition associated with a particular code.
+(define_code_attr fcond [(unordered "cun")
+ (uneq "cueq")
+ (unlt "cult")
+ (unle "cule")
+ (eq "ceq")
+ (lt "slt")
+ (le "sle")
+ (ordered "cor")
+ (ltgt "sne")
+ (ne "cune")
+ (ge "sge")
+ (gt "sgt")
+ (unge "cuge")
+ (ungt "cugt")])
+
+;; The sel mnemonic to use depending on the condition test.
+(define_code_attr sel [(eq "masknez") (ne "maskeqz")])
+(define_code_attr selinv [(eq "maskeqz") (ne "masknez")])
+
+;;
+;; ....................
+;;
+;; CONDITIONAL TRAPS
+;;
+;; ....................
+;;
+
+(define_insn "trap"
+ [(trap_if (const_int 1) (const_int 0))]
+ ""
+{
+ return "break\t0";
+}
+ [(set_attr "type" "trap")])
+
+
+\f
+;;
+;; ....................
+;;
+;; ADDITION
+;;
+;; ....................
+;;
+
+(define_insn "add<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (plus:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ ""
+ "fadd.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+(define_insn "add<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (plus:GPR (match_operand:GPR 1 "register_operand" "r,r")
+ (match_operand:GPR 2 "arith_operand" "r,Q")))]
+ ""
+ "add%i2.<d>\t%0,%1,%2";
+ [(set_attr "alu_type" "add")
+ (set_attr "compression" "*,*")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "*addsi3_extended"
+ [(set (match_operand:DI 0 "register_operand" "=r,r")
+ (sign_extend:DI
+ (plus:SI (match_operand:SI 1 "register_operand" "r,r")
+ (match_operand:SI 2 "arith_operand" "r,Q"))))]
+ "TARGET_64BIT"
+ "add%i2.w\t%0,%1,%2"
+ [(set_attr "alu_type" "add")
+ (set_attr "mode" "SI")])
+
+(define_insn "*addsi3_extended2"
+ [(set (match_operand:DI 0 "register_operand" "=r,r")
+ (sign_extend:DI
+ (subreg:SI (plus:DI (match_operand:DI 1 "register_operand" "r,r")
+ (match_operand:DI 2 "arith_operand" "r,Q"))
+ 0)))]
+ "TARGET_64BIT"
+ "add%i2.w\t%0,%1,%2"
+ [(set_attr "alu_type" "add")
+ (set_attr "mode" "SI")])
+
+\f
+;;
+;; ....................
+;;
+;; SUBTRACTION
+;;
+;; ....................
+;;
+
+(define_insn "sub<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (minus:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ ""
+ "fsub.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+(define_insn "sub<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (minus:GPR (match_operand:GPR 1 "register_operand" "rJ")
+ (match_operand:GPR 2 "register_operand" "r")))]
+ ""
+ "sub.<d>\t%0,%z1,%2"
+ [(set_attr "alu_type" "sub")
+ (set_attr "compression" "*")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "*subsi3_extended"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (minus:SI (match_operand:SI 1 "register_operand" "rJ")
+ (match_operand:SI 2 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "sub.w\t%0,%z1,%2"
+ [(set_attr "alu_type" "sub")
+ (set_attr "mode" "DI")])
+
+(define_insn "*subsi3_extended2"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (subreg:SI (minus:DI (match_operand:DI 1 "reg_or_0_operand" "rJ")
+ (match_operand:DI 2 "register_operand" "r"))
+ 0)))]
+ "TARGET_64BIT"
+ "sub.w\t%0,%z1,%2"
+ [(set_attr "alu_type" "sub")
+ (set_attr "mode" "SI")])
+
+\f
+;;
+;; ....................
+;;
+;; MULTIPLICATION
+;;
+;; ....................
+;;
+
+(define_insn "mul<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (mult:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fmul.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fmul")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "mul<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (mult:GPR (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:GPR 2 "register_operand" "r")))]
+ ""
+ "mul.<d>\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "mulsidi3_64bit"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (mult:DI (sign_extend:DI (match_operand:SI 1 "register_operand" "r"))
+ (sign_extend:DI (match_operand:SI 2 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "mul.d\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "DI")])
+
+(define_insn "*mulsi3_extended"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (mult:SI (match_operand:SI 1 "register_operand" "r")
+ (match_operand:SI 2 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "mul.w\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "SI")])
+
+(define_insn "*mulsi3_extended2"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (subreg:SI (mult:DI (match_operand:DI 1 "register_operand" "r")
+ (match_operand:DI 2 "register_operand" "r"))
+ 0)))]
+ "TARGET_64BIT"
+ "mul.w\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "SI")])
+
+
+;;
+;; ........................
+;;
+;; MULTIPLICATION HIGH-PART
+;;
+;; ........................
+;;
+
+
+(define_expand "<u>mulditi3"
+ [(set (match_operand:TI 0 "register_operand")
+ (mult:TI (any_extend:TI (match_operand:DI 1 "register_operand"))
+ (any_extend:TI (match_operand:DI 2 "register_operand"))))]
+ "TARGET_64BIT"
+{
+ rtx low = gen_reg_rtx (DImode);
+ emit_insn (gen_muldi3 (low, operands[1], operands[2]));
+
+ rtx high = gen_reg_rtx (DImode);
+ emit_insn (gen_<u>muldi3_highpart (high, operands[1], operands[2]));
+
+ emit_move_insn (gen_lowpart (DImode, operands[0]), low);
+ emit_move_insn (gen_highpart (DImode, operands[0]), high);
+ DONE;
+})
+
+(define_insn "<u>muldi3_highpart"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (truncate:DI
+ (lshiftrt:TI
+ (mult:TI (any_extend:TI
+ (match_operand:DI 1 "register_operand" " r"))
+ (any_extend:TI
+ (match_operand:DI 2 "register_operand" " r")))
+ (const_int 64))))]
+ "TARGET_64BIT"
+ "mulh.d<u>\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "DI")])
+
+(define_expand "<u>mulsidi3"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (mult:DI (any_extend:DI
+ (match_operand:SI 1 "register_operand" " r"))
+ (any_extend:DI
+ (match_operand:SI 2 "register_operand" " r"))))]
+ "!TARGET_64BIT"
+{
+ rtx temp = gen_reg_rtx (SImode);
+ emit_insn (gen_mulsi3 (temp, operands[1], operands[2]));
+ emit_insn (gen_<u>mulsi3_highpart (loongarch_subword (operands[0], true),
+ operands[1], operands[2]));
+ emit_insn (gen_movsi (loongarch_subword (operands[0], false), temp));
+ DONE;
+})
+
+(define_insn "<u>mulsi3_highpart"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (truncate:SI
+ (lshiftrt:DI
+ (mult:DI (any_extend:DI
+ (match_operand:SI 1 "register_operand" " r"))
+ (any_extend:DI
+ (match_operand:SI 2 "register_operand" " r")))
+ (const_int 32))))]
+ "!TARGET_64BIT"
+ "mulh.w<u>\t%0,%1,%2"
+ [(set_attr "type" "imul")
+ (set_attr "mode" "SI")])
+
+;;
+;; ....................
+;;
+;; DIVISION and REMAINDER
+;;
+;; ....................
+;;
+
+;; Float division and modulus.
+(define_expand "div<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand")
+ (div:ANYF (match_operand:ANYF 1 "reg_or_1_operand")
+ (match_operand:ANYF 2 "register_operand")))]
+ "TARGET_HARD_FLOAT"
+{})
+
+(define_insn "*div<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (div:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fdiv.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fdiv")
+ (set_attr "mode" "<UNITMODE>")
+ (set_attr "insn_count" "1")])
+
+;; In 3A5000, the reciprocal operation is the same as the division operation.
+
+(define_insn "*recip<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (div:ANYF (match_operand:ANYF 1 "const_1_operand" "")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "frecip.<fmt>\t%0,%2"
+ [(set_attr "type" "frdiv")
+ (set_attr "mode" "<UNITMODE>")
+ (set_attr "insn_count" "1")])
+
+;; Integer division and modulus.
+
+(define_insn "<optab><mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=&r")
+ (any_div:GPR (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:GPR 2 "register_operand" "r")))]
+ ""
+ {
+ return loongarch_output_division ("<insn>.<d><u>\t%0,%1,%2", operands);
+ }
+ [(set_attr "type" "idiv")
+ (set_attr "mode" "<MODE>")])
+
+
+;; Floating point multiply accumulate instructions.
+
+;; a * b + c
+(define_expand "fma<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand")
+ (fma:ANYF (match_operand:ANYF 1 "register_operand")
+ (match_operand:ANYF 2 "register_operand")
+ (match_operand:ANYF 3 "register_operand")))]
+ "TARGET_HARD_FLOAT")
+
+(define_insn "*fma<mode>4_madd4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (fma:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")
+ (match_operand:ANYF 3 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fmadd.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; a * b - c
+(define_insn "fms<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (fma:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")
+ (neg:ANYF (match_operand:ANYF 3 "register_operand" "f"))))]
+ "TARGET_HARD_FLOAT"
+ "fmsub.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; fnma is defined in GCC as (fma (neg op1) op2 op3)
+;; (-op1 * op2) + op3 ==> -(op1 * op2) + op3 ==> -((op1 * op2) - op3)
+;; The loongarch nmsub instructions implement -((op1 * op2) - op3)
+;; This transformation means we may return the wrong signed zero
+;; so we check HONOR_SIGNED_ZEROS.
+
+;; -a * b + c
+(define_insn "fnma<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (fma:ANYF (neg:ANYF (match_operand:ANYF 1 "register_operand" "f"))
+ (match_operand:ANYF 2 "register_operand" "f")
+ (match_operand:ANYF 3 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT && !HONOR_SIGNED_ZEROS (<MODE>mode)"
+ "fnmsub.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; fnms is defined as: (fma (neg op1) op2 (neg op3))
+;; ((-op1) * op2) - op3 ==> -(op1 * op2) - op3 ==> -((op1 * op2) + op3)
+;; The loongarch nmadd instructions implement -((op1 * op2) + op3)
+;; This transformation means we may return the wrong signed zero
+;; so we check HONOR_SIGNED_ZEROS.
+
+;; -a * b - c
+(define_insn "fnms<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (fma:ANYF
+ (neg:ANYF (match_operand:ANYF 1 "register_operand" "f"))
+ (match_operand:ANYF 2 "register_operand" "f")
+ (neg:ANYF (match_operand:ANYF 3 "register_operand" "f"))))]
+ "TARGET_HARD_FLOAT && !HONOR_SIGNED_ZEROS (<MODE>mode)"
+ "fnmadd.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; -(-a * b - c), modulo signed zeros
+(define_insn "*fma<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (neg:ANYF
+ (fma:ANYF
+ (neg:ANYF (match_operand:ANYF 1 "register_operand" " f"))
+ (match_operand:ANYF 2 "register_operand" " f")
+ (neg:ANYF (match_operand:ANYF 3 "register_operand" " f")))))]
+ "TARGET_HARD_FLOAT && !HONOR_SIGNED_ZEROS (<MODE>mode)"
+ "fmadd.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; -(-a * b + c), modulo signed zeros
+(define_insn "*fms<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (neg:ANYF
+ (fma:ANYF
+ (neg:ANYF (match_operand:ANYF 1 "register_operand" " f"))
+ (match_operand:ANYF 2 "register_operand" " f")
+ (match_operand:ANYF 3 "register_operand" " f"))))]
+ "TARGET_HARD_FLOAT && !HONOR_SIGNED_ZEROS (<MODE>mode)"
+ "fmsub.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; -(a * b + c)
+(define_insn "*fnms<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (neg:ANYF
+ (fma:ANYF
+ (match_operand:ANYF 1 "register_operand" " f")
+ (match_operand:ANYF 2 "register_operand" " f")
+ (match_operand:ANYF 3 "register_operand" " f"))))]
+ "TARGET_HARD_FLOAT"
+ "fnmadd.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+
+;; -(a * b - c)
+(define_insn "*fnma<mode>4"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (neg:ANYF
+ (fma:ANYF
+ (match_operand:ANYF 1 "register_operand" " f")
+ (match_operand:ANYF 2 "register_operand" " f")
+ (neg:ANYF (match_operand:ANYF 3 "register_operand" " f")))))]
+ "TARGET_HARD_FLOAT"
+ "fnmsub.<fmt>\t%0,%1,%2,%3"
+ [(set_attr "type" "fmadd")
+ (set_attr "mode" "<UNITMODE>")])
+\f
+;;
+;; ....................
+;;
+;; SQUARE ROOT
+;;
+;; ....................
+
+(define_insn "sqrt<mode>2"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (sqrt:ANYF (match_operand:ANYF 1 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fsqrt.<fmt>\t%0,%1"
+ [(set_attr "type" "fsqrt")
+ (set_attr "mode" "<UNITMODE>")
+ (set_attr "insn_count" "1")])
+
+(define_insn "*rsqrt<mode>a"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (div:ANYF (match_operand:ANYF 1 "const_1_operand" "")
+ (sqrt:ANYF (match_operand:ANYF 2 "register_operand" "f"))))]
+ "TARGET_HARD_FLOAT && flag_unsafe_math_optimizations"
+ "frsqrt.<fmt>\t%0,%2"
+ [(set_attr "type" "frsqrt")
+ (set_attr "mode" "<UNITMODE>")
+ (set_attr "insn_count" "1")])
+
+(define_insn "*rsqrt<mode>b"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (sqrt:ANYF (div:ANYF (match_operand:ANYF 1 "const_1_operand" "")
+ (match_operand:ANYF 2 "register_operand" "f"))))]
+ "TARGET_HARD_FLOAT && flag_unsafe_math_optimizations"
+ "frsqrt.<fmt>\t%0,%2"
+ [(set_attr "type" "frsqrt")
+ (set_attr "mode" "<UNITMODE>")
+ (set_attr "insn_count" "1")])
+\f
+;;
+;; ....................
+;;
+;; ABSOLUTE VALUE
+;;
+;; ....................
+
+(define_insn "abs<mode>2"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (abs:ANYF (match_operand:ANYF 1 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fabs.<fmt>\t%0,%1"
+ [(set_attr "type" "fabs")
+ (set_attr "mode" "<UNITMODE>")])
+\f
+;;
+;; ...................
+;;
+;; Count leading zeroes.
+;;
+;; ...................
+;;
+
+(define_insn "clz<mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (clz:GPR (match_operand:GPR 1 "register_operand" "r")))]
+ ""
+ "clz.<d>\t%0,%1"
+ [(set_attr "type" "clz")
+ (set_attr "mode" "<MODE>")])
+
+;;
+;; ...................
+;;
+;; Count trailing zeroes.
+;;
+;; ...................
+;;
+
+(define_insn "ctz<mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (ctz:GPR (match_operand:GPR 1 "register_operand" "r")))]
+ ""
+ "ctz.<d>\t%0,%1"
+ [(set_attr "type" "clz")
+ (set_attr "mode" "<MODE>")])
+
+;;
+;; ....................
+;;
+;; MIN/MAX
+;;
+;; ....................
+
+(define_insn "smax<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (smax:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fmax.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fmove")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "smin<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (smin:ANYF (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fmin.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fmove")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "smaxa<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (if_then_else:ANYF
+ (gt (abs:ANYF (match_operand:ANYF 1 "register_operand" "f"))
+ (abs:ANYF (match_operand:ANYF 2 "register_operand" "f")))
+ (match_dup 1)
+ (match_dup 2)))]
+ "TARGET_HARD_FLOAT"
+ "fmaxa.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fmove")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "smina<mode>3"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (if_then_else:ANYF
+ (lt (abs:ANYF (match_operand:ANYF 1 "register_operand" "f"))
+ (abs:ANYF (match_operand:ANYF 2 "register_operand" "f")))
+ (match_dup 1)
+ (match_dup 2)))]
+ "TARGET_HARD_FLOAT"
+ "fmina.<fmt>\t%0,%1,%2"
+ [(set_attr "type" "fmove")
+ (set_attr "mode" "<MODE>")])
+\f
+;;
+;; ....................
+;;
+;; NEGATION and ONE'S COMPLEMENT
+;;
+;; ....................
+
+(define_insn "neg<mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (neg:GPR (match_operand:GPR 1 "register_operand" "r")))]
+ ""
+ "sub.<d>\t%0,%.,%1"
+ [(set_attr "alu_type" "sub")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "one_cmpl<mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (not:GPR (match_operand:GPR 1 "register_operand" "r")))]
+ ""
+ "nor\t%0,%.,%1"
+ [(set_attr "alu_type" "not")
+ (set_attr "compression" "*")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "neg<mode>2"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (neg:ANYF (match_operand:ANYF 1 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fneg.<fmt>\t%0,%1"
+ [(set_attr "type" "fneg")
+ (set_attr "mode" "<UNITMODE>")])
+\f
+
+;;
+;; ....................
+;;
+;; LOGICAL
+;;
+;; ....................
+;;
+
+(define_insn "<optab><mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (any_bitwise:GPR (match_operand:GPR 1 "register_operand" "r,r")
+ (match_operand:GPR 2 "uns_arith_operand" "r,K")))]
+ ""
+ "<insn>%i2\t%0,%1,%2"
+ [(set_attr "type" "logical")
+ (set_attr "compression" "*,*")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "and<mode>3_extended"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (and:GPR (match_operand:GPR 1 "nonimmediate_operand" "r")
+ (match_operand:GPR 2 "low_bitmask_operand" "Yx")))]
+ ""
+{
+ int len;
+
+ len = low_bitmask_len (<MODE>mode, INTVAL (operands[2]));
+ operands[2] = GEN_INT (len-1);
+ return "bstrpick.<d>\t%0,%1,%2,0";
+}
+ [(set_attr "move_type" "pick_ins")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "*iorhi3"
+ [(set (match_operand:HI 0 "register_operand" "=r,r")
+ (ior:HI (match_operand:HI 1 "register_operand" "r,r")
+ (match_operand:HI 2 "uns_arith_operand" "r,K")))]
+ ""
+ "or%i2\t%0,%1,%2"
+ [(set_attr "type" "logical")
+ (set_attr "mode" "HI")])
+
+(define_insn "*nor<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (and:GPR (not:GPR (match_operand:GPR 1 "register_operand" "r"))
+ (not:GPR (match_operand:GPR 2 "register_operand" "r"))))]
+ ""
+ "nor\t%0,%1,%2"
+ [(set_attr "type" "logical")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "andn<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (and:GPR
+ (not:GPR (match_operand:GPR 1 "register_operand" "r"))
+ (match_operand:GPR 2 "register_operand" "r")))]
+ ""
+ "andn\t%0,%2,%1"
+ [(set_attr "type" "logical")])
+
+(define_insn "orn<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (ior:GPR
+ (not:GPR (match_operand:GPR 1 "register_operand" "r"))
+ (match_operand:GPR 2 "register_operand" "r")))]
+ ""
+ "orn\t%0,%2,%1"
+ [(set_attr "type" "logical")])
+
+\f
+;;
+;; ....................
+;;
+;; TRUNCATION
+;;
+;; ....................
+
+(define_insn "truncdfsf2"
+ [(set (match_operand:SF 0 "register_operand" "=f")
+ (float_truncate:SF (match_operand:DF 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "fcvt.s.d\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "cnv_mode" "D2S")
+ (set_attr "mode" "SF")])
+
+;; Integer truncation patterns. Truncating SImode values to smaller
+;; modes is a no-op, as it is for most other GCC ports. Truncating
+;; DImode values to SImode is not a no-op for TARGET_64BIT since we
+;; need to make sure that the lower 32 bits are properly sign-extended
+;; (see TARGET_TRULY_NOOP_TRUNCATION). Truncating DImode values into modes
+;; smaller than SImode is equivalent to two separate truncations:
+;;
+;; A B
+;; DI ---> HI == DI ---> SI ---> HI
+;; DI ---> QI == DI ---> SI ---> QI
+;;
+;; Step A needs a real instruction but step B does not.
+
+(define_insn "truncdi<mode>2"
+ [(set (match_operand:SUBDI 0 "nonimmediate_operand" "=r,m")
+ (truncate:SUBDI (match_operand:DI 1 "register_operand" "r,r")))]
+ "TARGET_64BIT"
+ "@
+ slli.w\t%0,%1,0
+ st.<size>\t%1,%0"
+ [(set_attr "move_type" "sll0,store")
+ (set_attr "mode" "SI")])
+
+(define_insn "truncdisi2_extended"
+ [(set (match_operand:SI 0 "nonimmediate_operand" "=ZC")
+ (truncate:SI (match_operand:DI 1 "register_operand" "r")))]
+ "TARGET_64BIT"
+ "stptr.w\t%1,%0"
+ [(set_attr "move_type" "store")
+ (set_attr "mode" "SI")])
+
+;; Combiner patterns to optimize shift/truncate combinations.
+
+(define_insn "*ashr_trunc<mode>"
+ [(set (match_operand:SUBDI 0 "register_operand" "=r")
+ (truncate:SUBDI
+ (ashiftrt:DI (match_operand:DI 1 "register_operand" "r")
+ (match_operand:DI 2 "const_arith_operand" ""))))]
+ "TARGET_64BIT && IN_RANGE (INTVAL (operands[2]), 32, 63)"
+ "srai.d\t%0,%1,%2"
+ [(set_attr "type" "shift")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "*lshr32_trunc<mode>"
+ [(set (match_operand:SUBDI 0 "register_operand" "=r")
+ (truncate:SUBDI
+ (lshiftrt:DI (match_operand:DI 1 "register_operand" "r")
+ (const_int 32))))]
+ "TARGET_64BIT"
+ "srai.d\t%0,%1,32"
+ [(set_attr "type" "shift")
+ (set_attr "mode" "<MODE>")])
+\f
+;;
+;; ....................
+;;
+;; ZERO EXTENSION
+;;
+;; ....................
+
+(define_insn "zero_extendsidi2"
+ [(set (match_operand:DI 0 "register_operand" "=r,r,r")
+ (zero_extend:DI (match_operand:SI 1 "nonimmediate_operand" "r,ZC,W")))]
+ "TARGET_64BIT"
+ "@
+ bstrpick.d\t%0,%1,31,0
+ ldptr.w\t%0,%1\n\tlu32i.d\t%0,0
+ ld.wu\t%0,%1"
+ [(set_attr "move_type" "arith,load,load")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "1,2,1")])
+
+(define_insn "*zero_extendsidi2_internal"
+ [(set (match_operand:DI 0 "register_operand" "=r,r,r")
+ (subreg:DI (match_operand:SI 1 "nonimmediate_operand" "r,ZC,W") 0))]
+ "TARGET_64BIT"
+ "@
+ bstrpick.d\t%0,%1,31,0
+ ldptr.w\t%0,%1\n\tlu32i.d\t%0,0
+ ld.wu\t%0,%1"
+ [(set_attr "move_type" "arith,load,load")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "1,2,1")])
+
+;; See the comment before the *and<mode>3 pattern why this is generated by
+;; combine.
+
+(define_insn "zero_extend<SHORT:mode><GPR:mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (zero_extend:GPR
+ (match_operand:SHORT 1 "nonimmediate_operand" "r,m")))]
+ ""
+ "@
+ bstrpick.<d>\t%0,%1,<SHORT:qi_hi>,0
+ ld.<SHORT:size>u\t%0,%1"
+ [(set_attr "move_type" "pick_ins,load")
+ (set_attr "compression" "*,*")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "zero_extendqihi2"
+ [(set (match_operand:HI 0 "register_operand" "=r,r")
+ (zero_extend:HI (match_operand:QI 1 "nonimmediate_operand" "r,m")))]
+ ""
+ "@
+ andi\t%0,%1,0xff
+ ld.bu\t%0,%1"
+ [(set_attr "move_type" "andi,load")
+ (set_attr "mode" "HI")])
+
+;; Combiner patterns to optimize truncate/zero_extend combinations.
+
+(define_insn "*zero_extend<GPR:mode>_trunc<SHORT:mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (zero_extend:GPR
+ (truncate:SHORT (match_operand:DI 1 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "bstrpick.<d>\t%0,%1,<SHORT:qi_hi>,0"
+ [(set_attr "move_type" "pick_ins")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*zero_extendhi_truncqi"
+ [(set (match_operand:HI 0 "register_operand" "=r")
+ (zero_extend:HI
+ (truncate:QI (match_operand:DI 1 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "andi\t%0,%1,0xff"
+ [(set_attr "alu_type" "and")
+ (set_attr "mode" "HI")])
+\f
+;;
+;; ....................
+;;
+;; SIGN EXTENSION
+;;
+;; ....................
+
+;; Extension insns.
+;; Those for integer source operand are ordered widest source type first.
+
+;; When TARGET_64BIT, all SImode integer and accumulator registers
+;; should already be in sign-extended form (see TARGET_TRULY_NOOP_TRUNCATION
+;; and truncdisi2). We can therefore get rid of register->register
+;; instructions if we constrain the source to be in the same register as
+;; the destination.
+;;
+;; Only the pre-reload scheduler sees the type of the register alternatives;
+;; we split them into nothing before the post-reload scheduler runs.
+;; These alternatives therefore have type "move" in order to reflect
+;; what happens if the two pre-reload operands cannot be tied, and are
+;; instead allocated two separate GPRs. We don't distinguish between
+;; the GPR and LO cases because we don't usually know during pre-reload
+;; scheduling whether an operand will be LO or not.
+(define_insn_and_split "extendsidi2"
+ [(set (match_operand:DI 0 "register_operand" "=r,r,r")
+ (sign_extend:DI (match_operand:SI 1 "nonimmediate_operand" "0,ZC,m")))]
+ "TARGET_64BIT"
+ "@
+ #
+ ldptr.w\t%0,%1
+ ld.w\t%0,%1"
+ "&& reload_completed && register_operand (operands[1], VOIDmode)"
+ [(const_int 0)]
+{
+ emit_note (NOTE_INSN_DELETED);
+ DONE;
+}
+ [(set_attr "move_type" "move,load,load")
+ (set_attr "mode" "DI")])
+
+(define_insn "extend<SHORT:mode><GPR:mode>2"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (sign_extend:GPR
+ (match_operand:SHORT 1 "nonimmediate_operand" "r,m")))]
+ ""
+ "@
+ ext.w.<SHORT:size>\t%0,%1
+ ld.<SHORT:size>\t%0,%1"
+ [(set_attr "move_type" "signext,load")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "extendqihi2"
+ [(set (match_operand:HI 0 "register_operand" "=r,r")
+ (sign_extend:HI
+ (match_operand:QI 1 "nonimmediate_operand" "r,m")))]
+ ""
+ "@
+ ext.w.b\t%0,%1
+ ld.b\t%0,%1"
+ [(set_attr "move_type" "signext,load")
+ (set_attr "mode" "SI")])
+
+;; Combiner patterns for truncate/sign_extend combinations. The SI versions
+;; use the shift/truncate patterns.
+
+(define_insn_and_split "*extenddi_truncate<mode>"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (truncate:SHORT (match_operand:DI 1 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "#"
+ "&& reload_completed"
+ [(set (match_dup 2)
+ (ashift:DI (match_dup 1)
+ (match_dup 3)))
+ (set (match_dup 0)
+ (ashiftrt:DI (match_dup 2)
+ (match_dup 3)))]
+{
+ operands[2] = gen_lowpart (DImode, operands[0]);
+ operands[3] = GEN_INT (BITS_PER_WORD - GET_MODE_BITSIZE (<MODE>mode));
+}
+ [(set_attr "move_type" "shift_shift")
+ (set_attr "mode" "DI")])
+
+(define_insn_and_split "*extendsi_truncate<mode>"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (sign_extend:SI
+ (truncate:SHORT (match_operand:DI 1 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "#"
+ "&& reload_completed"
+ [(set (match_dup 2)
+ (ashift:DI (match_dup 1)
+ (match_dup 3)))
+ (set (match_dup 0)
+ (truncate:SI (ashiftrt:DI (match_dup 2)
+ (match_dup 3))))]
+{
+ operands[2] = gen_lowpart (DImode, operands[0]);
+ operands[3] = GEN_INT (BITS_PER_WORD - GET_MODE_BITSIZE (<MODE>mode));
+}
+ [(set_attr "move_type" "shift_shift")
+ (set_attr "mode" "SI")])
+
+(define_insn_and_split "*extendhi_truncateqi"
+ [(set (match_operand:HI 0 "register_operand" "=r")
+ (sign_extend:HI
+ (truncate:QI (match_operand:DI 1 "register_operand" "r"))))]
+ "TARGET_64BIT"
+ "#"
+ "&& reload_completed"
+ [(set (match_dup 2)
+ (ashift:DI (match_dup 1)
+ (const_int 56)))
+ (set (match_dup 0)
+ (truncate:HI (ashiftrt:DI (match_dup 2)
+ (const_int 56))))]
+{
+ operands[2] = gen_lowpart (DImode, operands[0]);
+}
+ [(set_attr "move_type" "shift_shift")
+ (set_attr "mode" "SI")])
+
+(define_insn "extendsfdf2"
+ [(set (match_operand:DF 0 "register_operand" "=f")
+ (float_extend:DF (match_operand:SF 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "fcvt.d.s\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "cnv_mode" "S2D")
+ (set_attr "mode" "DF")])
+\f
+;;
+;; ....................
+;;
+;; CONVERSIONS
+;;
+;; ....................
+
+;; conversion of a floating-point value to a integer
+
+(define_insn "fix_truncdfsi2"
+ [(set (match_operand:SI 0 "register_operand" "=f")
+ (fix:SI (match_operand:DF 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ftintrz.w.d %0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "DF")
+ (set_attr "cnv_mode" "D2I")])
+
+(define_insn "fix_truncsfsi2"
+ [(set (match_operand:SI 0 "register_operand" "=f")
+ (fix:SI (match_operand:SF 1 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "ftintrz.w.s %0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "SF")
+ (set_attr "cnv_mode" "S2I")])
+
+
+(define_insn "fix_truncdfdi2"
+ [(set (match_operand:DI 0 "register_operand" "=f")
+ (fix:DI (match_operand:DF 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ftintrz.l.d %0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "DF")
+ (set_attr "cnv_mode" "D2I")])
+
+
+(define_insn "fix_truncsfdi2"
+ [(set (match_operand:DI 0 "register_operand" "=f")
+ (fix:DI (match_operand:SF 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ftintrz.l.s %0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "SF")
+ (set_attr "cnv_mode" "S2I")])
+
+;; conversion of an integeral (or boolean) value to a floating-point value
+
+(define_insn "floatsidf2"
+ [(set (match_operand:DF 0 "register_operand" "=f")
+ (float:DF (match_operand:SI 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ffint.d.w\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "DF")
+ (set_attr "cnv_mode" "I2D")])
+
+
+(define_insn "floatdidf2"
+ [(set (match_operand:DF 0 "register_operand" "=f")
+ (float:DF (match_operand:DI 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ffint.d.l\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "DF")
+ (set_attr "cnv_mode" "I2D")])
+
+
+(define_insn "floatsisf2"
+ [(set (match_operand:SF 0 "register_operand" "=f")
+ (float:SF (match_operand:SI 1 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "ffint.s.w\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "SF")
+ (set_attr "cnv_mode" "I2S")])
+
+
+(define_insn "floatdisf2"
+ [(set (match_operand:SF 0 "register_operand" "=f")
+ (float:SF (match_operand:DI 1 "register_operand" "f")))]
+ "TARGET_DOUBLE_FLOAT"
+ "ffint.s.l\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "SF")
+ (set_attr "cnv_mode" "I2S")])
+
+;; floating point value by converting to value to an unsigned integer
+
+(define_expand "fixuns_truncdfsi2"
+ [(set (match_operand:SI 0 "register_operand")
+ (unsigned_fix:SI (match_operand:DF 1 "register_operand")))]
+ "TARGET_DOUBLE_FLOAT"
+{
+ rtx reg1 = gen_reg_rtx (DFmode);
+ rtx reg2 = gen_reg_rtx (DFmode);
+ rtx reg3 = gen_reg_rtx (SImode);
+ rtx_code_label *label1 = gen_label_rtx ();
+ rtx_code_label *label2 = gen_label_rtx ();
+ rtx test;
+ REAL_VALUE_TYPE offset;
+
+ real_2expN (&offset, 31, DFmode);
+
+ if (reg1) /* Turn off complaints about unreached code. */
+ {
+ loongarch_emit_move (reg1,
+ const_double_from_real_value (offset, DFmode));
+ do_pending_stack_adjust ();
+
+ test = gen_rtx_GE (VOIDmode, operands[1], reg1);
+ emit_jump_insn (gen_cbranchdf4 (test, operands[1], reg1, label1));
+
+ emit_insn (gen_fix_truncdfsi2 (operands[0], operands[1]));
+ emit_jump_insn (gen_rtx_SET (pc_rtx,
+ gen_rtx_LABEL_REF (VOIDmode, label2)));
+ emit_barrier ();
+
+ emit_label (label1);
+ loongarch_emit_move (reg2, gen_rtx_MINUS (DFmode, operands[1], reg1));
+ loongarch_emit_move (reg3, GEN_INT (trunc_int_for_mode
+ (BITMASK_HIGH, SImode)));
+
+ emit_insn (gen_fix_truncdfsi2 (operands[0], reg2));
+ emit_insn (gen_iorsi3 (operands[0], operands[0], reg3));
+
+ emit_label (label2);
+
+ /* Allow REG_NOTES to be set on last insn (labels don't have enough
+ fields, and can't be used for REG_NOTES anyway). */
+ emit_use (stack_pointer_rtx);
+ DONE;
+ }
+})
+
+
+(define_expand "fixuns_truncdfdi2"
+ [(set (match_operand:DI 0 "register_operand")
+ (unsigned_fix:DI (match_operand:DF 1 "register_operand")))]
+ "TARGET_DOUBLE_FLOAT"
+{
+ rtx reg1 = gen_reg_rtx (DFmode);
+ rtx reg2 = gen_reg_rtx (DFmode);
+ rtx reg3 = gen_reg_rtx (DImode);
+ rtx_code_label *label1 = gen_label_rtx ();
+ rtx_code_label *label2 = gen_label_rtx ();
+ rtx test;
+ REAL_VALUE_TYPE offset;
+
+ real_2expN (&offset, 63, DFmode);
+
+ loongarch_emit_move (reg1, const_double_from_real_value (offset, DFmode));
+ do_pending_stack_adjust ();
+
+ test = gen_rtx_GE (VOIDmode, operands[1], reg1);
+ emit_jump_insn (gen_cbranchdf4 (test, operands[1], reg1, label1));
+
+ emit_insn (gen_fix_truncdfdi2 (operands[0], operands[1]));
+ emit_jump_insn (gen_rtx_SET (pc_rtx, gen_rtx_LABEL_REF (VOIDmode, label2)));
+ emit_barrier ();
+
+ emit_label (label1);
+ loongarch_emit_move (reg2, gen_rtx_MINUS (DFmode, operands[1], reg1));
+ loongarch_emit_move (reg3, GEN_INT (BITMASK_HIGH));
+ emit_insn (gen_ashldi3 (reg3, reg3, GEN_INT (32)));
+
+ emit_insn (gen_fix_truncdfdi2 (operands[0], reg2));
+ emit_insn (gen_iordi3 (operands[0], operands[0], reg3));
+
+ emit_label (label2);
+
+ /* Allow REG_NOTES to be set on last insn (labels don't have enough
+ fields, and can't be used for REG_NOTES anyway). */
+ emit_use (stack_pointer_rtx);
+ DONE;
+})
+
+
+(define_expand "fixuns_truncsfsi2"
+ [(set (match_operand:SI 0 "register_operand")
+ (unsigned_fix:SI (match_operand:SF 1 "register_operand")))]
+ "TARGET_HARD_FLOAT"
+{
+ rtx reg1 = gen_reg_rtx (SFmode);
+ rtx reg2 = gen_reg_rtx (SFmode);
+ rtx reg3 = gen_reg_rtx (SImode);
+ rtx_code_label *label1 = gen_label_rtx ();
+ rtx_code_label *label2 = gen_label_rtx ();
+ rtx test;
+ REAL_VALUE_TYPE offset;
+
+ real_2expN (&offset, 31, SFmode);
+
+ loongarch_emit_move (reg1, const_double_from_real_value (offset, SFmode));
+ do_pending_stack_adjust ();
+
+ test = gen_rtx_GE (VOIDmode, operands[1], reg1);
+ emit_jump_insn (gen_cbranchsf4 (test, operands[1], reg1, label1));
+
+ emit_insn (gen_fix_truncsfsi2 (operands[0], operands[1]));
+ emit_jump_insn (gen_rtx_SET (pc_rtx, gen_rtx_LABEL_REF (VOIDmode, label2)));
+ emit_barrier ();
+
+ emit_label (label1);
+ loongarch_emit_move (reg2, gen_rtx_MINUS (SFmode, operands[1], reg1));
+ loongarch_emit_move (reg3, GEN_INT (trunc_int_for_mode
+ (BITMASK_HIGH, SImode)));
+
+ emit_insn (gen_fix_truncsfsi2 (operands[0], reg2));
+ emit_insn (gen_iorsi3 (operands[0], operands[0], reg3));
+
+ emit_label (label2);
+
+ /* Allow REG_NOTES to be set on last insn (labels don't have enough
+ fields, and can't be used for REG_NOTES anyway). */
+ emit_use (stack_pointer_rtx);
+ DONE;
+})
+
+
+(define_expand "fixuns_truncsfdi2"
+ [(set (match_operand:DI 0 "register_operand")
+ (unsigned_fix:DI (match_operand:SF 1 "register_operand")))]
+ "TARGET_DOUBLE_FLOAT"
+{
+ rtx reg1 = gen_reg_rtx (SFmode);
+ rtx reg2 = gen_reg_rtx (SFmode);
+ rtx reg3 = gen_reg_rtx (DImode);
+ rtx_code_label *label1 = gen_label_rtx ();
+ rtx_code_label *label2 = gen_label_rtx ();
+ rtx test;
+ REAL_VALUE_TYPE offset;
+
+ real_2expN (&offset, 63, SFmode);
+
+ loongarch_emit_move (reg1, const_double_from_real_value (offset, SFmode));
+ do_pending_stack_adjust ();
+
+ test = gen_rtx_GE (VOIDmode, operands[1], reg1);
+ emit_jump_insn (gen_cbranchsf4 (test, operands[1], reg1, label1));
+
+ emit_insn (gen_fix_truncsfdi2 (operands[0], operands[1]));
+ emit_jump_insn (gen_rtx_SET (pc_rtx, gen_rtx_LABEL_REF (VOIDmode, label2)));
+ emit_barrier ();
+
+ emit_label (label1);
+ loongarch_emit_move (reg2, gen_rtx_MINUS (SFmode, operands[1], reg1));
+ loongarch_emit_move (reg3, GEN_INT (BITMASK_HIGH));
+ emit_insn (gen_ashldi3 (reg3, reg3, GEN_INT (32)));
+
+ emit_insn (gen_fix_truncsfdi2 (operands[0], reg2));
+ emit_insn (gen_iordi3 (operands[0], operands[0], reg3));
+
+ emit_label (label2);
+
+ /* Allow REG_NOTES to be set on last insn (labels don't have enough
+ fields, and can't be used for REG_NOTES anyway). */
+ emit_use (stack_pointer_rtx);
+ DONE;
+})
+\f
+;;
+;; ....................
+;;
+;; EXTRACT AND INSERT
+;;
+;; ....................
+
+(define_expand "extzv<mode>"
+ [(set (match_operand:GPR 0 "register_operand")
+ (zero_extract:GPR (match_operand:GPR 1 "register_operand")
+ (match_operand 2 "const_int_operand")
+ (match_operand 3 "const_int_operand")))]
+ ""
+{
+ if (!loongarch_use_ins_ext_p (operands[1], INTVAL (operands[2]),
+ INTVAL (operands[3])))
+ FAIL;
+})
+
+(define_insn "*extzv<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (zero_extract:GPR (match_operand:GPR 1 "register_operand" "r")
+ (match_operand 2 "const_int_operand" "")
+ (match_operand 3 "const_int_operand" "")))]
+ "loongarch_use_ins_ext_p (operands[1], INTVAL (operands[2]),
+ INTVAL (operands[3]))"
+{
+ operands[2] = GEN_INT (INTVAL (operands[2]) + INTVAL (operands[3]) - 1);
+ return "bstrpick.<d>\t%0,%1,%2,%3";
+}
+ [(set_attr "type" "arith")
+ (set_attr "mode" "<MODE>")])
+
+(define_expand "insv<mode>"
+ [(set (zero_extract:GPR (match_operand:GPR 0 "register_operand")
+ (match_operand 1 "const_int_operand")
+ (match_operand 2 "const_int_operand"))
+ (match_operand:GPR 3 "reg_or_0_operand"))]
+ ""
+{
+ if (!loongarch_use_ins_ext_p (operands[0], INTVAL (operands[1]),
+ INTVAL (operands[2])))
+ FAIL;
+})
+
+(define_insn "*insv<mode>"
+ [(set (zero_extract:GPR (match_operand:GPR 0 "register_operand" "+r")
+ (match_operand:SI 1 "const_int_operand" "")
+ (match_operand:SI 2 "const_int_operand" ""))
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ"))]
+ "loongarch_use_ins_ext_p (operands[0], INTVAL (operands[1]),
+ INTVAL (operands[2]))"
+{
+ operands[1] = GEN_INT (INTVAL (operands[1]) + INTVAL (operands[2]) - 1);
+ return "bstrins.<d>\t%0,%z3,%1,%2";
+}
+ [(set_attr "type" "arith")
+ (set_attr "mode" "<MODE>")])
+
+\f
+;;
+;; ....................
+;;
+;; DATA MOVEMENT
+;;
+;; ....................
+
+;; Allow combine to split complex const_int load sequences, using operand 2
+;; to store the intermediate results. See move_operand for details.
+(define_split
+ [(set (match_operand:GPR 0 "register_operand")
+ (match_operand:GPR 1 "splittable_const_int_operand"))
+ (clobber (match_operand:GPR 2 "register_operand"))]
+ ""
+ [(const_int 0)]
+{
+ loongarch_move_integer (operands[2], operands[0], INTVAL (operands[1]));
+ DONE;
+})
+
+;; 64-bit integer moves
+
+;; Unlike most other insns, the move insns can't be split with
+;; different predicates, because register spilling and other parts of
+;; the compiler, have memoized the insn number already.
+
+(define_expand "movdi"
+ [(set (match_operand:DI 0 "")
+ (match_operand:DI 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (DImode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movdi_32bit"
+ [(set (match_operand:DI 0 "nonimmediate_operand" "=r,r,r,ZC,r,m,*f,*f,*r,*m")
+ (match_operand:DI 1 "move_operand" "r,i,ZC,r,m,r,*J*r,*m,*f,*f"))]
+ "!TARGET_64BIT
+ && (register_operand (operands[0], DImode)
+ || reg_or_0_operand (operands[1], DImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,load,store,load,store,mgtf,fpload,mftg,fpstore")
+ (set_attr "mode" "DI")])
+
+(define_insn "*movdi_64bit"
+ [(set (match_operand:DI 0 "nonimmediate_operand" "=r,r,r,ZC,r,m,*f,*f,*r,*m")
+ (match_operand:DI 1 "move_operand" "r,Yd,ZC,rJ,m,rJ,*r*J,*m,*f,*f"))]
+ "TARGET_64BIT
+ && (register_operand (operands[0], DImode)
+ || reg_or_0_operand (operands[1], DImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,load,store,load,store,mgtf,fpload,mftg,fpstore")
+ (set_attr "mode" "DI")])
+
+;; 32-bit Integer moves
+
+(define_expand "movsi"
+ [(set (match_operand:SI 0 "")
+ (match_operand:SI 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (SImode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movsi_internal"
+ [(set (match_operand:SI 0 "nonimmediate_operand" "=r,r,r,ZC,r,m,*f,*f,*r,*m,*r,*z")
+ (match_operand:SI 1 "move_operand" "r,Yd,ZC,rJ,m,rJ,*r*J,*m,*f,*f,*z,*r"))]
+ "(register_operand (operands[0], SImode)
+ || reg_or_0_operand (operands[1], SImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,load,store,load,store,mgtf,fpload,mftg,fpstore,mftg,mgtf")
+ (set_attr "compression" "all,*,*,*,*,*,*,*,*,*,*,*")
+ (set_attr "mode" "SI")])
+
+;; 16-bit Integer moves
+
+;; Unlike most other insns, the move insns can't be split with
+;; different predicates, because register spilling and other parts of
+;; the compiler, have memoized the insn number already.
+;; Unsigned loads are used because LOAD_EXTEND_OP returns ZERO_EXTEND.
+
+(define_expand "movhi"
+ [(set (match_operand:HI 0 "")
+ (match_operand:HI 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (HImode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movhi_internal"
+ [(set (match_operand:HI 0 "nonimmediate_operand" "=r,r,r,r,m")
+ (match_operand:HI 1 "move_operand" "r,Yd,I,m,rJ"))]
+ "(register_operand (operands[0], HImode)
+ || reg_or_0_operand (operands[1], HImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,const,load,store")
+ (set_attr "compression" "all,all,*,*,*")
+ (set_attr "mode" "HI")])
+
+;; 8-bit Integer moves
+
+;; Unlike most other insns, the move insns can't be split with
+;; different predicates, because register spilling and other parts of
+;; the compiler, have memoized the insn number already.
+;; Unsigned loads are used because LOAD_EXTEND_OP returns ZERO_EXTEND.
+
+(define_expand "movqi"
+ [(set (match_operand:QI 0 "")
+ (match_operand:QI 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (QImode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movqi_internal"
+ [(set (match_operand:QI 0 "nonimmediate_operand" "=r,r,r,m")
+ (match_operand:QI 1 "move_operand" "r,I,m,rJ"))]
+ "(register_operand (operands[0], QImode)
+ || reg_or_0_operand (operands[1], QImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,load,store")
+ (set_attr "compression" "all,*,*,*")
+ (set_attr "mode" "QI")])
+
+;; 32-bit floating point moves
+
+(define_expand "movsf"
+ [(set (match_operand:SF 0 "")
+ (match_operand:SF 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (SFmode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movsf_hardfloat"
+ [(set (match_operand:SF 0 "nonimmediate_operand" "=f,f,f,m,m,*f,*r,*r,*r,*m")
+ (match_operand:SF 1 "move_operand" "f,G,m,f,G,*r,*f,*G*r,*m,*r"))]
+ "TARGET_HARD_FLOAT
+ && (register_operand (operands[0], SFmode)
+ || reg_or_0_operand (operands[1], SFmode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "fmove,mgtf,fpload,fpstore,store,mgtf,mftg,move,load,store")
+ (set_attr "mode" "SF")])
+
+(define_insn "*movsf_softfloat"
+ [(set (match_operand:SF 0 "nonimmediate_operand" "=r,r,m")
+ (match_operand:SF 1 "move_operand" "Gr,m,r"))]
+ "TARGET_SOFT_FLOAT
+ && (register_operand (operands[0], SFmode)
+ || reg_or_0_operand (operands[1], SFmode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,load,store")
+ (set_attr "mode" "SF")])
+
+
+;; 64-bit floating point moves
+
+(define_expand "movdf"
+ [(set (match_operand:DF 0 "")
+ (match_operand:DF 1 ""))]
+ ""
+{
+ if (loongarch_legitimize_move (DFmode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movdf_hardfloat"
+ [(set (match_operand:DF 0 "nonimmediate_operand" "=f,f,f,m,m,*f,*r,*r,*r,*m")
+ (match_operand:DF 1 "move_operand" "f,G,m,f,G,*r,*f,*r*G,*m,*r"))]
+ "TARGET_DOUBLE_FLOAT
+ && (register_operand (operands[0], DFmode)
+ || reg_or_0_operand (operands[1], DFmode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "fmove,mgtf,fpload,fpstore,store,mgtf,mftg,move,load,store")
+ (set_attr "mode" "DF")])
+
+(define_insn "*movdf_softfloat"
+ [(set (match_operand:DF 0 "nonimmediate_operand" "=r,r,m")
+ (match_operand:DF 1 "move_operand" "rG,m,rG"))]
+ "(TARGET_SOFT_FLOAT || TARGET_SINGLE_FLOAT)
+ && (register_operand (operands[0], DFmode)
+ || reg_or_0_operand (operands[1], DFmode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,load,store")
+ (set_attr "mode" "DF")])
+
+
+;; 128-bit integer moves
+
+(define_expand "movti"
+ [(set (match_operand:TI 0)
+ (match_operand:TI 1))]
+ "TARGET_64BIT"
+{
+ if (loongarch_legitimize_move (TImode, operands[0], operands[1]))
+ DONE;
+})
+
+(define_insn "*movti"
+ [(set (match_operand:TI 0 "nonimmediate_operand" "=r,r,r,m")
+ (match_operand:TI 1 "move_operand" "r,i,m,rJ"))]
+ "TARGET_64BIT
+ && (register_operand (operands[0], TImode)
+ || reg_or_0_operand (operands[1], TImode))"
+ { return loongarch_output_move (operands[0], operands[1]); }
+ [(set_attr "move_type" "move,const,load,store")
+ (set (attr "mode")
+ (if_then_else (eq_attr "move_type" "imul")
+ (const_string "SI")
+ (const_string "TI")))])
+
+
+;; 128-bit floating point moves
+
+(define_expand "movtf"
+ [(set (match_operand:TF 0)
+ (match_operand:TF 1))]
+ "TARGET_64BIT"
+{
+ if (loongarch_legitimize_move (TFmode, operands[0], operands[1]))
+ DONE;
+})
+
+;; This pattern handles both hard- and soft-float cases.
+(define_insn "*movtf"
+ [(set (match_operand:TF 0 "nonimmediate_operand" "=r,r,m,f,r,f,m")
+ (match_operand:TF 1 "move_operand" "rG,m,rG,rG,f,m,f"))]
+ "TARGET_64BIT
+ && (register_operand (operands[0], TFmode)
+ || reg_or_0_operand (operands[1], TFmode))"
+ "#"
+ [(set_attr "move_type" "move,load,store,mgtf,mftg,fpload,fpstore")
+ (set_attr "mode" "TF")])
+
+
+(define_split
+ [(set (match_operand:MOVE64 0 "nonimmediate_operand")
+ (match_operand:MOVE64 1 "move_operand"))]
+ "reload_completed && loongarch_split_move_insn_p (operands[0], operands[1], insn)"
+ [(const_int 0)]
+{
+ loongarch_split_move_insn (operands[0], operands[1], curr_insn);
+ DONE;
+})
+
+(define_split
+ [(set (match_operand:MOVE128 0 "nonimmediate_operand")
+ (match_operand:MOVE128 1 "move_operand"))]
+ "reload_completed && loongarch_split_move_insn_p (operands[0], operands[1], insn)"
+ [(const_int 0)]
+{
+ loongarch_split_move_insn (operands[0], operands[1], curr_insn);
+ DONE;
+})
+
+;; Emit a doubleword move in which exactly one of the operands is
+;; a floating-point register. We can't just emit two normal moves
+;; because of the constraints imposed by the FPU register model;
+;; see loongarch_cannot_change_mode_class for details. Instead, we keep
+;; the FPR whole and use special patterns to refer to each word of
+;; the other operand.
+
+(define_expand "move_doubleword_fpr<mode>"
+ [(set (match_operand:SPLITF 0)
+ (match_operand:SPLITF 1))]
+ ""
+{
+ if (FP_REG_RTX_P (operands[0]))
+ {
+ rtx low = loongarch_subword (operands[1], 0);
+ rtx high = loongarch_subword (operands[1], 1);
+ emit_insn (gen_load_low<mode> (operands[0], low));
+ if (!TARGET_64BIT)
+ emit_insn (gen_movgr2frh<mode> (operands[0], high, operands[0]));
+ else
+ emit_insn (gen_load_high<mode> (operands[0], high, operands[0]));
+ }
+ else
+ {
+ rtx low = loongarch_subword (operands[0], 0);
+ rtx high = loongarch_subword (operands[0], 1);
+ emit_insn (gen_store_word<mode> (low, operands[1], const0_rtx));
+ if (!TARGET_64BIT)
+ emit_insn (gen_movfrh2gr<mode> (high, operands[1]));
+ else
+ emit_insn (gen_store_word<mode> (high, operands[1], const1_rtx));
+ }
+ DONE;
+})
+
+;; Conditional move instructions.
+
+(define_insn "*sel<code><GPR:mode>_using_<GPR2:mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (if_then_else:GPR
+ (equality_op:GPR2 (match_operand:GPR2 1 "register_operand" "r,r")
+ (const_int 0))
+ (match_operand:GPR 2 "reg_or_0_operand" "r,J")
+ (match_operand:GPR 3 "reg_or_0_operand" "J,r")))]
+ "register_operand (operands[2], <GPR:MODE>mode)
+ != register_operand (operands[3], <GPR:MODE>mode)"
+ "@
+ <sel>\t%0,%2,%1
+ <selinv>\t%0,%3,%1"
+ [(set_attr "type" "condmove")
+ (set_attr "mode" "<GPR:MODE>")])
+
+;; sel.fmt copies the 3rd argument when the 1st is non-zero and the 2nd
+;; argument if the 1st is zero. This means operand 2 and 3 are
+;; inverted in the instruction.
+
+(define_insn "*sel<mode>"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (if_then_else:ANYF
+ (ne:FCC (match_operand:FCC 1 "register_operand" "z")
+ (const_int 0))
+ (match_operand:ANYF 2 "reg_or_0_operand" "f")
+ (match_operand:ANYF 3 "reg_or_0_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fsel\t%0,%3,%2,%1"
+ [(set_attr "type" "condmove")
+ (set_attr "mode" "<ANYF:MODE>")])
+
+;; These are the main define_expand's used to make conditional moves.
+
+(define_expand "mov<mode>cc"
+ [(set (match_operand:GPR 0 "register_operand")
+ (if_then_else:GPR (match_operator 1 "comparison_operator"
+ [(match_operand:GPR 2 "reg_or_0_operand")
+ (match_operand:GPR 3 "reg_or_0_operand")])))]
+ "TARGET_COND_MOVE_INT"
+{
+ if (!INTEGRAL_MODE_P (GET_MODE (XEXP (operands[1], 0))))
+ FAIL;
+
+ loongarch_expand_conditional_move (operands);
+ DONE;
+})
+
+(define_expand "mov<mode>cc"
+ [(set (match_operand:ANYF 0 "register_operand")
+ (if_then_else:ANYF (match_operator 1 "comparison_operator"
+ [(match_operand:ANYF 2 "reg_or_0_operand")
+ (match_operand:ANYF 3 "reg_or_0_operand")])))]
+ "TARGET_COND_MOVE_FLOAT"
+{
+ if (!FLOAT_MODE_P (GET_MODE (XEXP (operands[1], 0))))
+ FAIL;
+
+ loongarch_expand_conditional_move (operands);
+ DONE;
+})
+;; lu32i.d
+(define_insn "lu32i_d"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (ior:DI
+ (zero_extend:DI
+ (subreg:SI (match_operand:DI 1 "register_operand" "0") 0))
+ (match_operand:DI 2 "const_lu32i_operand" "u")))]
+ "TARGET_64BIT"
+ "lu32i.d\t%0,%X2>>32"
+ [(set_attr "type" "arith")
+ (set_attr "mode" "DI")])
+
+;; lu52i.d
+
+(define_insn "lu52i_d"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (ior:DI
+ (and:DI (match_operand:DI 1 "register_operand" "r")
+ (match_operand 2 "lu52i_mask_operand"))
+ (match_operand 3 "const_lu52i_operand" "v")))]
+ "TARGET_64BIT"
+ "lu52i.d\t%0,%1,%X3>>52"
+ [(set_attr "type" "arith")
+ (set_attr "mode" "DI")])
+
+;; Convert floating-point numbers to integers
+
+(define_insn "frint_<fmt>"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
+ UNSPEC_FRINT))]
+ "TARGET_HARD_FLOAT"
+ "frint.<fmt>\t%0,%1"
+ [(set_attr "type" "fcvt")
+ (set_attr "mode" "<MODE>")])
+
+;; LoongArch supports loading and storing a floating point register from
+;; the sum of two general registers. We use two versions for each of
+;; these four instructions: one where the two general registers are
+;; SImode, and one where they are DImode. This is because general
+;; registers will be in SImode when they hold 32-bit values, but,
+;; since the 32-bit values are always sign extended, the f{ld/st}x.{s/d}
+;; instructions will still work correctly.
+
+;; ??? Perhaps it would be better to support these instructions by
+;; modifying TARGET_LEGITIMATE_ADDRESS_P and friends. However, since
+;; these instructions can only be used to load and store floating
+;; point registers, that would probably cause trouble in reload.
+
+(define_insn "*<ANYF:floadx>_<P:mode>"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (mem:ANYF (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r"))))]
+ "TARGET_HARD_FLOAT"
+ "<ANYF:floadx>\t%0,%1,%2"
+ [(set_attr "type" "fpidxload")
+ (set_attr "mode" "<ANYF:UNITMODE>")])
+
+(define_insn "*<ANYF:fstorex>_<P:mode>"
+ [(set (mem:ANYF (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r")))
+ (match_operand:ANYF 0 "register_operand" "f"))]
+ "TARGET_HARD_FLOAT"
+ "<ANYF:fstorex>\t%0,%1,%2"
+ [(set_attr "type" "fpidxstore")
+ (set_attr "mode" "<ANYF:UNITMODE>")])
+
+;; loading and storing a integer register from the sum of two general
+;; registers.
+
+(define_insn "*<GPR:loadx>_<P:mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (mem:GPR
+ (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r"))))]
+ ""
+ "<GPR:loadx>\t%0,%1,%2"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*<GPR:storex>_<P:mode>"
+ [(set (mem:GPR (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r")))
+ (match_operand:GPR 0 "register_operand" "r"))]
+ ""
+ "<GPR:storex>\t%0,%1,%2"
+ [(set_attr "type" "store")
+ (set_attr "mode" "<GPR:MODE>")])
+
+;; SHORT mode sign_extend.
+(define_insn "*extend_<SHORT:loadx>_<GPR:mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (sign_extend:GPR
+ (mem:SHORT
+ (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r")))))]
+ ""
+ "<SHORT:loadx>\t%0,%1,%2"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*extend_<SHORT:storex>"
+ [(set (mem:SHORT (plus:P (match_operand:P 1 "register_operand" "r")
+ (match_operand:P 2 "register_operand" "r")))
+ (match_operand:SHORT 0 "register_operand" "r"))]
+ ""
+ "<SHORT:storex>\t%0,%1,%2"
+ [(set_attr "type" "store")
+ (set_attr "mode" "SI")])
+
+;; Load the low word of operand 0 with operand 1.
+(define_insn "load_low<mode>"
+ [(set (match_operand:SPLITF 0 "register_operand" "=f,f")
+ (unspec:SPLITF [(match_operand:<HALFMODE> 1 "general_operand" "rJ,m")]
+ UNSPEC_LOAD_LOW))]
+ "TARGET_HARD_FLOAT"
+{
+ operands[0] = loongarch_subword (operands[0], 0);
+ return loongarch_output_move (operands[0], operands[1]);
+}
+ [(set_attr "move_type" "mgtf,fpload")
+ (set_attr "mode" "<HALFMODE>")])
+
+;; Load the high word of operand 0 from operand 1, preserving the value
+;; in the low word.
+(define_insn "load_high<mode>"
+ [(set (match_operand:SPLITF 0 "register_operand" "=f,f")
+ (unspec:SPLITF [(match_operand:<HALFMODE> 1 "general_operand" "rJ,m")
+ (match_operand:SPLITF 2 "register_operand" "0,0")]
+ UNSPEC_LOAD_HIGH))]
+ "TARGET_HARD_FLOAT"
+{
+ operands[0] = loongarch_subword (operands[0], 1);
+ return loongarch_output_move (operands[0], operands[1]);
+}
+ [(set_attr "move_type" "mgtf,fpload")
+ (set_attr "mode" "<HALFMODE>")])
+
+;; Store one word of operand 1 in operand 0. Operand 2 is 1 to store the
+;; high word and 0 to store the low word.
+(define_insn "store_word<mode>"
+ [(set (match_operand:<HALFMODE> 0 "nonimmediate_operand" "=r,m")
+ (unspec:<HALFMODE> [(match_operand:SPLITF 1 "register_operand" "f,f")
+ (match_operand 2 "const_int_operand")]
+ UNSPEC_STORE_WORD))]
+ "TARGET_HARD_FLOAT"
+{
+ operands[1] = loongarch_subword (operands[1], INTVAL (operands[2]));
+ return loongarch_output_move (operands[0], operands[1]);
+}
+ [(set_attr "move_type" "mftg,fpstore")
+ (set_attr "mode" "<HALFMODE>")])
+
+;; Thread-Local Storage
+
+(define_insn "got_load_tls_gd<mode>"
+ [(set (match_operand:P 0 "register_operand" "=r")
+ (unspec:P
+ [(match_operand:P 1 "symbolic_operand" "")]
+ UNSPEC_TLS_GD))]
+ ""
+ "la.tls.gd\t%0,%1"
+ [(set_attr "got" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "got_load_tls_ld<mode>"
+ [(set (match_operand:P 0 "register_operand" "=r")
+ (unspec:P
+ [(match_operand:P 1 "symbolic_operand" "")]
+ UNSPEC_TLS_LD))]
+ ""
+ "la.tls.ld\t%0,%1"
+ [(set_attr "got" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "got_load_tls_le<mode>"
+ [(set (match_operand:P 0 "register_operand" "=r")
+ (unspec:P
+ [(match_operand:P 1 "symbolic_operand" "")]
+ UNSPEC_TLS_LE))]
+ ""
+ "la.tls.le\t%0,%1"
+ [(set_attr "got" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "got_load_tls_ie<mode>"
+ [(set (match_operand:P 0 "register_operand" "=r")
+ (unspec:P
+ [(match_operand:P 1 "symbolic_operand" "")]
+ UNSPEC_TLS_IE))]
+ ""
+ "la.tls.ie\t%0,%1"
+ [(set_attr "got" "load")
+ (set_attr "mode" "<MODE>")])
+
+;; Move operand 1 to the high word of operand 0 using movgr2frh, preserving the
+;; value in the low word.
+(define_insn "movgr2frh<mode>"
+ [(set (match_operand:SPLITF 0 "register_operand" "=f")
+ (unspec:SPLITF [(match_operand:<HALFMODE> 1 "reg_or_0_operand" "rJ")
+ (match_operand:SPLITF 2 "register_operand" "0")]
+ UNSPEC_MOVGR2FRH))]
+ "TARGET_DOUBLE_FLOAT"
+ "movgr2frh.w\t%z1,%0"
+ [(set_attr "move_type" "mgtf")
+ (set_attr "mode" "<HALFMODE>")])
+
+;; Move high word of operand 1 to operand 0 using movfrh2gr.
+(define_insn "movfrh2gr<mode>"
+ [(set (match_operand:<HALFMODE> 0 "register_operand" "=r")
+ (unspec:<HALFMODE> [(match_operand:SPLITF 1 "register_operand" "f")]
+ UNSPEC_MOVFRH2GR))]
+ "TARGET_DOUBLE_FLOAT"
+ "movfrh2gr.s\t%0,%1"
+ [(set_attr "move_type" "mftg")
+ (set_attr "mode" "<HALFMODE>")])
+
+\f
+;; Expand in-line code to clear the instruction cache between operand[0] and
+;; operand[1].
+(define_expand "clear_cache"
+ [(match_operand 0 "pmode_register_operand")
+ (match_operand 1 "pmode_register_operand")]
+ ""
+ "
+{
+ emit_insn (gen_ibar (const0_rtx));
+ DONE;
+}")
+
+(define_insn "ibar"
+ [(unspec_volatile:SI [(match_operand 0 "const_uimm15_operand")] UNSPEC_IBAR)]
+ ""
+ "ibar\t%0")
+
+(define_insn "dbar"
+ [(unspec_volatile:SI [(match_operand 0 "const_uimm15_operand")] UNSPEC_DBAR)]
+ ""
+ "dbar\t%0")
+
+\f
+
+;; Privileged state instruction
+
+(define_insn "cpucfg"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec_volatile:SI [(match_operand:SI 1 "register_operand" "r")]
+ UNSPEC_CPUCFG))]
+ ""
+ "cpucfg\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "SI")])
+
+(define_insn "asrtle_d"
+ [(unspec_volatile:DI [(match_operand:DI 0 "register_operand" "r")
+ (match_operand:DI 1 "register_operand" "r")]
+ UNSPEC_ASRTLE_D)]
+ "TARGET_64BIT"
+ "asrtle.d\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "DI")])
+
+(define_insn "asrtgt_d"
+ [(unspec_volatile:DI [(match_operand:DI 0 "register_operand" "r")
+ (match_operand:DI 1 "register_operand" "r")]
+ UNSPEC_ASRTGT_D)]
+ "TARGET_64BIT"
+ "asrtgt.d\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "DI")])
+
+(define_insn "<p>csrrd"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (unspec_volatile:GPR [(match_operand 1 "const_uimm14_operand")]
+ UNSPEC_CSRRD))]
+ ""
+ "csrrd\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "<p>csrwr"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (unspec_volatile:GPR
+ [(match_operand:GPR 1 "register_operand" "0")
+ (match_operand 2 "const_uimm14_operand")]
+ UNSPEC_CSRWR))]
+ ""
+ "csrwr\t%0,%2"
+ [(set_attr "type" "store")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "<p>csrxchg"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (unspec_volatile:GPR
+ [(match_operand:GPR 1 "register_operand" "0")
+ (match_operand:GPR 2 "register_operand" "q")
+ (match_operand 3 "const_uimm14_operand")]
+ UNSPEC_CSRXCHG))]
+ ""
+ "csrxchg\t%0,%2,%3"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "iocsrrd_<size>"
+ [(set (match_operand:QHWD 0 "register_operand" "=r")
+ (unspec_volatile:QHWD [(match_operand:SI 1 "register_operand" "r")]
+ UNSPEC_IOCSRRD))]
+ ""
+ "iocsrrd.<size>\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "iocsrwr_<size>"
+ [(unspec_volatile:QHWD [(match_operand:QHWD 0 "register_operand" "r")
+ (match_operand:SI 1 "register_operand" "r")]
+ UNSPEC_IOCSRWR)]
+ ""
+ "iocsrwr.<size>\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "<p>cacop"
+ [(unspec_volatile:X [(match_operand 0 "const_uimm5_operand")
+ (match_operand:X 1 "register_operand" "r")
+ (match_operand 2 "const_imm12_operand")]
+ UNSPEC_CACOP)]
+ ""
+ "cacop\t%0,%1,%2"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "<p>lddir"
+ [(unspec_volatile:X [(match_operand:X 0 "register_operand" "r")
+ (match_operand:X 1 "register_operand" "r")
+ (match_operand 2 "const_uimm5_operand")]
+ UNSPEC_LDDIR)]
+ ""
+ "lddir\t%0,%1,%2"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "<p>ldpte"
+ [(unspec_volatile:X [(match_operand:X 0 "register_operand" "r")
+ (match_operand 1 "const_uimm5_operand")]
+ UNSPEC_LDPTE)]
+ ""
+ "ldpte\t%0,%1"
+ [(set_attr "type" "load")
+ (set_attr "mode" "<MODE>")])
+
+\f
+;; Block moves, see loongarch.c for more details.
+;; Argument 0 is the destination
+;; Argument 1 is the source
+;; Argument 2 is the length
+;; Argument 3 is the alignment
+
+(define_expand "cpymemsi"
+ [(parallel [(set (match_operand:BLK 0 "general_operand")
+ (match_operand:BLK 1 "general_operand"))
+ (use (match_operand:SI 2 ""))
+ (use (match_operand:SI 3 "const_int_operand"))])]
+ " !TARGET_MEMCPY"
+{
+ if (loongarch_expand_block_move (operands[0], operands[1], operands[2]))
+ DONE;
+ else
+ FAIL;
+})
+\f
+;;
+;; ....................
+;;
+;; SHIFTS
+;;
+;; ....................
+
+(define_insn "<optab><mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (any_shift:GPR (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:SI 2 "arith_operand" "rI")))]
+ ""
+{
+ if (CONST_INT_P (operands[2]))
+ operands[2] = GEN_INT (INTVAL (operands[2])
+ & (GET_MODE_BITSIZE (<MODE>mode) - 1));
+
+ return "<insn>%i2.<d>\t%0,%1,%2";
+}
+ [(set_attr "type" "shift")
+ (set_attr "compression" "none")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "*<optab>si3_extend"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (sign_extend:DI
+ (any_shift:SI (match_operand:SI 1 "register_operand" "r")
+ (match_operand:SI 2 "arith_operand" "rI"))))]
+ "TARGET_64BIT"
+{
+ if (CONST_INT_P (operands[2]))
+ operands[2] = GEN_INT (INTVAL (operands[2]) & 0x1f);
+
+ return "<insn>%i2.w\t%0,%1,%2";
+}
+ [(set_attr "type" "shift")
+ (set_attr "mode" "SI")])
+
+(define_insn "rotr<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r,r")
+ (rotatert:GPR (match_operand:GPR 1 "register_operand" "r,r")
+ (match_operand:SI 2 "arith_operand" "r,I")))]
+ ""
+ "rotr%i2.<d>\t%0,%1,%2"
+ [(set_attr "type" "shift,shift")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "zero_extend_ashift1"
+ [ (set (match_operand:DI 0 "register_operand" "=r")
+ (and:DI (ashift:DI (subreg:DI (match_operand:SI 1 "register_operand" "r") 0)
+ (match_operand 2 "const_immlsa_operand" ""))
+ (match_operand 3 "shift_mask_operand" "")))]
+"TARGET_64BIT"
+"bstrpick.d\t%0,%1,31,0\n\talsl.d\t%0,%0,$r0,%2"
+[(set_attr "type" "arith")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "2")])
+
+(define_insn "zero_extend_ashift2"
+ [ (set (match_operand:DI 0 "register_operand" "=r")
+ (and:DI (ashift:DI (match_operand:DI 1 "register_operand" "r")
+ (match_operand 2 "const_immlsa_operand" ""))
+ (match_operand 3 "shift_mask_operand" "")))]
+"TARGET_64BIT"
+"bstrpick.d\t%0,%1,31,0\n\talsl.d\t%0,%0,$r0,%2"
+[(set_attr "type" "arith")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "2")])
+
+(define_insn "alsl_paired1"
+ [(set (match_operand:DI 0 "register_operand" "=&r")
+ (plus:DI (and:DI (ashift:DI (subreg:DI (match_operand:SI 1 "register_operand" "r") 0)
+ (match_operand 2 "const_immlsa_operand" ""))
+ (match_operand 3 "shift_mask_operand" ""))
+ (match_operand:DI 4 "register_operand" "r")))]
+ "TARGET_64BIT"
+ "bstrpick.d\t%0,%1,31,0\n\talsl.d\t%0,%0,%4,%2"
+ [(set_attr "type" "arith")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "2")])
+
+(define_insn "alsl_paired2"
+ [(set (match_operand:DI 0 "register_operand" "=&r")
+ (plus:DI (match_operand:DI 1 "register_operand" "r")
+ (and:DI (ashift:DI (match_operand:DI 2 "register_operand" "r")
+ (match_operand 3 "const_immlsa_operand" ""))
+ (match_operand 4 "shift_mask_operand" ""))))]
+ "TARGET_64BIT"
+ "bstrpick.d\t%0,%2,31,0\n\talsl.d\t%0,%0,%1,%3"
+ [(set_attr "type" "arith")
+ (set_attr "mode" "DI")
+ (set_attr "insn_count" "2")])
+
+(define_insn "alsl<mode>3"
+ [(set (match_operand:GPR 0 "register_operand" "=r")
+ (plus:GPR (ashift:GPR (match_operand:GPR 1 "register_operand" "r")
+ (match_operand 2 "const_immlsa_operand" ""))
+ (match_operand:GPR 3 "register_operand" "r")))]
+ ""
+ "alsl.<d>\t%0,%1,%3,%2"
+ [(set_attr "type" "arith")
+ (set_attr "mode" "<MODE>")])
+
+\f
+
+;; Reverse the order of bytes of operand 1 and store the result in operand 0.
+
+(define_insn "bswaphi2"
+ [(set (match_operand:HI 0 "register_operand" "=r")
+ (bswap:HI (match_operand:HI 1 "register_operand" "r")))]
+ ""
+ "revb.2h\t%0,%1"
+ [(set_attr "type" "shift")])
+
+(define_insn_and_split "bswapsi2"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (bswap:SI (match_operand:SI 1 "register_operand" "r")))]
+ ""
+ "#"
+ ""
+ [(set (match_dup 0) (unspec:SI [(match_dup 1)] UNSPEC_REVB_2H))
+ (set (match_dup 0) (rotatert:SI (match_dup 0) (const_int 16)))]
+ ""
+ [(set_attr "insn_count" "2")])
+
+(define_insn_and_split "bswapdi2"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (bswap:DI (match_operand:DI 1 "register_operand" "r")))]
+ "TARGET_64BIT"
+ "#"
+ ""
+ [(set (match_dup 0) (unspec:DI [(match_dup 1)] UNSPEC_REVB_4H))
+ (set (match_dup 0) (unspec:DI [(match_dup 0)] UNSPEC_REVH_D))]
+ ""
+ [(set_attr "insn_count" "2")])
+
+(define_insn "revb_2h"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec:SI [(match_operand:SI 1 "register_operand" "r")] UNSPEC_REVB_2H))]
+ ""
+ "revb.2h\t%0,%1"
+ [(set_attr "type" "shift")])
+
+(define_insn "revb_4h"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (unspec:DI [(match_operand:DI 1 "register_operand" "r")] UNSPEC_REVB_4H))]
+ "TARGET_64BIT"
+ "revb.4h\t%0,%1"
+ [(set_attr "type" "shift")])
+
+(define_insn "revh_d"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (unspec:DI [(match_operand:DI 1 "register_operand" "r")] UNSPEC_REVH_D))]
+ "TARGET_64BIT"
+ "revh.d\t%0,%1"
+ [(set_attr "type" "shift")])
+\f
+;;
+;; ....................
+;;
+;; CONDITIONAL BRANCHES
+;;
+;; ....................
+
+;; Conditional branches
+
+(define_insn "*branch_fp_FCCmode"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "equality_operator"
+ [(match_operand:FCC 2 "register_operand" "z")
+ (const_int 0)])
+ (label_ref (match_operand 0 "" ""))
+ (pc)))]
+ "TARGET_HARD_FLOAT"
+{
+ return loongarch_output_conditional_branch (insn, operands,
+ LARCH_BRANCH ("b%F1", "%Z2%0"),
+ LARCH_BRANCH ("b%W1", "%Z2%0"));
+}
+ [(set_attr "type" "branch")])
+
+(define_insn "*branch_fp_inverted_FCCmode"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "equality_operator"
+ [(match_operand:FCC 2 "register_operand" "z")
+ (const_int 0)])
+ (pc)
+ (label_ref (match_operand 0 "" ""))))]
+ "TARGET_HARD_FLOAT"
+{
+ return loongarch_output_conditional_branch (insn, operands,
+ LARCH_BRANCH ("b%W1", "%Z2%0"),
+ LARCH_BRANCH ("b%F1", "%Z2%0"));
+}
+ [(set_attr "type" "branch")])
+
+;; Conditional branches on ordered comparisons with zero.
+
+(define_insn "*branch_order<mode>"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "order_operator"
+ [(match_operand:GPR 2 "register_operand" "r,r")
+ (match_operand:GPR 3 "reg_or_0_operand" "J,r")])
+ (label_ref (match_operand 0 "" ""))
+ (pc)))]
+ ""
+ { return loongarch_output_order_conditional_branch (insn, operands, false); }
+ [(set_attr "type" "branch")
+ (set_attr "hazard" "forbidden_slot")])
+
+(define_insn "*branch_order<mode>_inverted"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "order_operator"
+ [(match_operand:GPR 2 "register_operand" "r,r")
+ (match_operand:GPR 3 "reg_or_0_operand" "J,r")])
+ (pc)
+ (label_ref (match_operand 0 "" ""))))]
+ ""
+ { return loongarch_output_order_conditional_branch (insn, operands, true); }
+ [(set_attr "type" "branch")
+ (set_attr "hazard" "forbidden_slot")])
+
+;; Conditional branch on equality comparison.
+
+(define_insn "*branch_equality<mode>"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "equality_operator"
+ [(match_operand:GPR 2 "register_operand" "r")
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ")])
+ (label_ref (match_operand 0 "" ""))
+ (pc)))]
+ ""
+ { return loongarch_output_equal_conditional_branch (insn, operands, false); }
+ [(set_attr "type" "branch")
+ (set_attr "hazard" "forbidden_slot")])
+
+
+(define_insn "*branch_equality<mode>_inverted"
+ [(set (pc)
+ (if_then_else
+ (match_operator 1 "equality_operator"
+ [(match_operand:GPR 2 "register_operand" "r")
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ")])
+ (pc)
+ (label_ref (match_operand 0 "" ""))))]
+ ""
+ { return loongarch_output_equal_conditional_branch (insn, operands, true); }
+ [(set_attr "type" "branch")
+ (set_attr "hazard" "forbidden_slot")])
+
+
+(define_expand "cbranch<mode>4"
+ [(set (pc)
+ (if_then_else (match_operator 0 "comparison_operator"
+ [(match_operand:GPR 1 "register_operand")
+ (match_operand:GPR 2 "nonmemory_operand")])
+ (label_ref (match_operand 3 ""))
+ (pc)))]
+ ""
+{
+ loongarch_expand_conditional_branch (operands);
+ DONE;
+})
+
+(define_expand "cbranch<mode>4"
+ [(set (pc)
+ (if_then_else (match_operator 0 "comparison_operator"
+ [(match_operand:ANYF 1 "register_operand")
+ (match_operand:ANYF 2 "register_operand")])
+ (label_ref (match_operand 3 ""))
+ (pc)))]
+ ""
+{
+ loongarch_expand_conditional_branch (operands);
+ DONE;
+})
+
+;; Used to implement built-in functions.
+(define_expand "condjump"
+ [(set (pc)
+ (if_then_else (match_operand 0)
+ (label_ref (match_operand 1))
+ (pc)))])
+
+
+\f
+;;
+;; ....................
+;;
+;; SETTING A REGISTER FROM A COMPARISON
+;;
+;; ....................
+
+;; Destination is always set in SI mode.
+
+(define_expand "cstore<mode>4"
+ [(set (match_operand:SI 0 "register_operand")
+ (match_operator:SI 1 "loongarch_cstore_operator"
+ [(match_operand:GPR 2 "register_operand")
+ (match_operand:GPR 3 "nonmemory_operand")]))]
+ ""
+{
+ loongarch_expand_scc (operands);
+ DONE;
+})
+
+(define_insn "*seq_zero_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (eq:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (const_int 0)))]
+ ""
+ "sltui\t%0,%1,1"
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+
+(define_insn "*sne_zero_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (ne:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (const_int 0)))]
+ ""
+ "sltu\t%0,%.,%1"
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*sgt<u>_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (any_gt:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:GPR 2 "reg_or_0_operand" "rJ")))]
+ ""
+ "slt<u>\t%0,%z2,%1"
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*sge<u>_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (any_ge:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (const_int 1)))]
+ ""
+ "slt<u>i\t%0,%.,%1"
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*slt<u>_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (any_lt:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:GPR 2 "arith_operand" "rI")))]
+ ""
+ "slt<u>%i2\t%0,%1,%2";
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+(define_insn "*sle<u>_<GPR:mode><GPR2:mode>"
+ [(set (match_operand:GPR2 0 "register_operand" "=r")
+ (any_le:GPR2 (match_operand:GPR 1 "register_operand" "r")
+ (match_operand:GPR 2 "sle_operand" "")))]
+ ""
+{
+ operands[2] = GEN_INT (INTVAL (operands[2]) + 1);
+ return "slt<u>i\t%0,%1,%2";
+}
+ [(set_attr "type" "slt")
+ (set_attr "mode" "<GPR:MODE>")])
+
+\f
+;;
+;; ....................
+;;
+;; FLOATING POINT COMPARISONS
+;;
+;; ....................
+
+(define_insn "s<code>_<ANYF:mode>_using_FCCmode"
+ [(set (match_operand:FCC 0 "register_operand" "=z")
+ (fcond:FCC (match_operand:ANYF 1 "register_operand" "f")
+ (match_operand:ANYF 2 "register_operand" "f")))]
+ "TARGET_HARD_FLOAT"
+ "fcmp.<fcond>.<fmt>\t%Z0%1,%2"
+ [(set_attr "type" "fcmp")
+ (set_attr "mode" "FCC")])
+
+\f
+;;
+;; ....................
+;;
+;; UNCONDITIONAL BRANCHES
+;;
+;; ....................
+
+;; Unconditional branches.
+
+(define_expand "jump"
+ [(set (pc)
+ (label_ref (match_operand 0)))])
+
+(define_insn "*jump_absolute"
+ [(set (pc)
+ (label_ref (match_operand 0)))]
+ "!flag_pic"
+{
+ return "b\t%l0";
+}
+ [(set_attr "type" "branch")])
+
+(define_insn "*jump_pic"
+ [(set (pc)
+ (label_ref (match_operand 0)))]
+ "flag_pic"
+{
+ return "b\t%0";
+}
+ [(set_attr "type" "branch")])
+
+(define_expand "indirect_jump"
+ [(set (pc) (match_operand 0 "register_operand"))]
+ ""
+{
+ operands[0] = force_reg (Pmode, operands[0]);
+ emit_jump_insn (PMODE_INSN (gen_indirect_jump, (operands[0])));
+ DONE;
+})
+
+(define_insn "indirect_jump_<mode>"
+ [(set (pc) (match_operand:P 0 "register_operand" "r"))]
+ ""
+ {
+ return "jr\t%0";
+ }
+ [(set_attr "type" "jump")
+ (set_attr "mode" "none")])
+
+(define_expand "tablejump"
+ [(set (pc)
+ (match_operand 0 "register_operand"))
+ (use (label_ref (match_operand 1 "")))]
+ ""
+{
+ if (flag_pic)
+ operands[0] = expand_simple_binop (Pmode, PLUS, operands[0],
+ gen_rtx_LABEL_REF (Pmode,
+ operands[1]),
+ NULL_RTX, 0, OPTAB_DIRECT);
+ emit_jump_insn (PMODE_INSN (gen_tablejump, (operands[0], operands[1])));
+ DONE;
+})
+
+(define_insn "tablejump_<mode>"
+ [(set (pc)
+ (match_operand:P 0 "register_operand" "r"))
+ (use (label_ref (match_operand 1 "" "")))]
+ ""
+ {
+ return "jr\t%0";
+ }
+ [(set_attr "type" "jump")
+ (set_attr "mode" "none")])
+
+
+\f
+;;
+;; ....................
+;;
+;; Function prologue/epilogue
+;;
+;; ....................
+;;
+
+(define_expand "prologue"
+ [(const_int 1)]
+ ""
+{
+ loongarch_expand_prologue ();
+ DONE;
+})
+
+;; Block any insns from being moved before this point, since the
+;; profiling call to mcount can use various registers that aren't
+;; saved or used to pass arguments.
+
+(define_insn "blockage"
+ [(unspec_volatile [(const_int 0)] UNSPEC_BLOCKAGE)]
+ ""
+ ""
+ [(set_attr "type" "ghost")
+ (set_attr "mode" "none")])
+
+(define_insn "probe_stack_range_<P:mode>"
+ [(set (match_operand:P 0 "register_operand" "=r")
+ (unspec_volatile:P [(match_operand:P 1 "register_operand" "0")
+ (match_operand:P 2 "register_operand" "r")
+ (match_operand:P 3 "register_operand" "r")]
+ UNSPEC_PROBE_STACK_RANGE))]
+ ""
+{
+ return loongarch_output_probe_stack_range (operands[0],
+ operands[2],
+ operands[3]);
+}
+ [(set_attr "type" "unknown")
+ (set_attr "can_delay" "no")
+ (set_attr "mode" "<MODE>")])
+
+(define_expand "epilogue"
+ [(const_int 2)]
+ ""
+{
+ loongarch_expand_epilogue (false);
+ DONE;
+})
+
+(define_expand "sibcall_epilogue"
+ [(const_int 2)]
+ ""
+{
+ loongarch_expand_epilogue (true);
+ DONE;
+})
+
+;; Trivial return. Make it look like a normal return insn as that
+;; allows jump optimizations to work better.
+
+(define_expand "return"
+ [(simple_return)]
+ "loongarch_can_use_return_insn ()"
+ { })
+
+(define_expand "simple_return"
+ [(simple_return)]
+ ""
+ { })
+
+(define_insn "*<optab>"
+ [(any_return)]
+ ""
+ {
+ operands[0] = gen_rtx_REG (Pmode, RETURN_ADDR_REGNUM);
+ return "jr\t%0";
+ }
+ [(set_attr "type" "jump")
+ (set_attr "mode" "none")])
+
+;; Normal return.
+
+(define_insn "<optab>_internal"
+ [(any_return)
+ (use (match_operand 0 "pmode_register_operand" ""))]
+ ""
+ {
+ return "jr\t%0";
+ }
+ [(set_attr "type" "jump")
+ (set_attr "mode" "none")])
+
+;; Exception return.
+(define_insn "loongarch_ertn"
+ [(return)
+ (unspec_volatile [(const_int 0)] UNSPEC_ERTN)]
+ ""
+ "ertn"
+ [(set_attr "type" "trap")
+ (set_attr "mode" "none")])
+
+;; This is used in compiling the unwind routines.
+(define_expand "eh_return"
+ [(use (match_operand 0 "general_operand"))]
+ ""
+{
+ if (GET_MODE (operands[0]) != word_mode)
+ operands[0] = convert_to_mode (word_mode, operands[0], 0);
+ if (TARGET_64BIT)
+ emit_insn (gen_eh_set_ra_di (operands[0]));
+ else
+ emit_insn (gen_eh_set_ra_si (operands[0]));
+ DONE;
+})
+
+;; Clobber the return address on the stack. We can't expand this
+;; until we know where it will be put in the stack frame.
+
+(define_insn "eh_set_ra_si"
+ [(unspec [(match_operand:SI 0 "register_operand" "r")] UNSPEC_EH_RETURN)
+ (clobber (match_scratch:SI 1 "=&r"))]
+ "! TARGET_64BIT"
+ "#")
+
+(define_insn "eh_set_ra_di"
+ [(unspec [(match_operand:DI 0 "register_operand" "r")] UNSPEC_EH_RETURN)
+ (clobber (match_scratch:DI 1 "=&r"))]
+ "TARGET_64BIT"
+ "#")
+
+(define_split
+ [(unspec [(match_operand 0 "register_operand")] UNSPEC_EH_RETURN)
+ (clobber (match_scratch 1))]
+ "reload_completed"
+ [(const_int 0)]
+{
+ loongarch_set_return_address (operands[0], operands[1]);
+ DONE;
+})
+
+
+\f
+;;
+;; ....................
+;;
+;; FUNCTION CALLS
+;;
+;; ....................
+
+;; Sibling calls. All these patterns use jump instructions.
+
+(define_expand "sibcall"
+ [(parallel [(call (match_operand 0 "")
+ (match_operand 1 ""))
+ (use (match_operand 2 "")) ;; next_arg_reg
+ (use (match_operand 3 ""))])] ;; struct_value_size_rtx
+ ""
+{
+ rtx target = loongarch_legitimize_call_address (XEXP (operands[0], 0));
+
+ emit_call_insn (gen_sibcall_internal (target, operands[1]));
+ DONE;
+})
+
+(define_insn "sibcall_internal"
+ [(call (mem:SI (match_operand 0 "call_insn_operand" "j,c,a,t,h"))
+ (match_operand 1 "" ""))]
+ "SIBLING_CALL_P (insn)"
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jr\t%0";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,(%%pcrel(%0+0x20000))>>18\n\t"
+ "jirl\t$r0,$r12,%%pcrel(%0+4)-(%%pcrel(%0+4+0x20000)>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r12,$r13,%0\n\tjr\t$r12";
+ else
+ return "b\t%0";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "b\t%0";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%0\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%0\n\tjr\t$r12";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%0\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%0\n\tjr\t$r12";
+ case 4:
+ if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return "b\t%%plt(%0)";
+ else if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,(%%plt(%0)+0x20000)>>18\n\t"
+ "jirl\t$r0,$r12,%%plt(%0)+4-((%%plt(%0)+(4+0x20000))>>18<<18)";
+ else
+ {
+ sorry ("cmodel extreme and tiny static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")])
+
+(define_expand "sibcall_value"
+ [(parallel [(set (match_operand 0 "")
+ (call (match_operand 1 "")
+ (match_operand 2 "")))
+ (use (match_operand 3 ""))])] ;; next_arg_reg
+ ""
+{
+ rtx target = loongarch_legitimize_call_address (XEXP (operands[1], 0));
+
+ /* Handle return values created by loongarch_return_fpr_pair. */
+ if (GET_CODE (operands[0]) == PARALLEL && XVECLEN (operands[0], 0) == 2)
+ {
+ rtx arg1 = XEXP (XVECEXP (operands[0],0, 0), 0);
+ rtx arg2 = XEXP (XVECEXP (operands[0],0, 1), 0);
+
+ emit_call_insn (gen_sibcall_value_multiple_internal (arg1, target,
+ operands[2],
+ arg2));
+ }
+ else
+ {
+ /* Handle return values created by loongarch_return_fpr_single. */
+ if (GET_CODE (operands[0]) == PARALLEL && XVECLEN (operands[0], 0) == 1)
+ operands[0] = XEXP (XVECEXP (operands[0], 0, 0), 0);
+
+ emit_call_insn (gen_sibcall_value_internal (operands[0], target,
+ operands[2]));
+ }
+ DONE;
+})
+
+(define_insn "sibcall_value_internal"
+ [(set (match_operand 0 "register_operand" "")
+ (call (mem:SI (match_operand 1 "call_insn_operand" "j,c,a,t,h"))
+ (match_operand 2 "" "")))]
+ "SIBLING_CALL_P (insn)"
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jr\t%1";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,%%pcrel(%1+0x20000)>>18\n\t"
+ "jirl\t$r0,$r12,%%pcrel(%1+4)-((%%pcrel(%1+4+0x20000))>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "b\t%1";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "b\t%1";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%1\n\tjr\t$r12";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%1\n\tjr\t$r12";
+ case 4:
+ if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return " b\t%%plt(%1)";
+ else if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,(%%plt(%1)+0x20000)>>18\n\t"
+ "jirl\t$r0,$r12,%%plt(%1)+4-((%%plt(%1)+(4+0x20000))>>18<<18)";
+ else
+ {
+ sorry ("loongarch cmodel extreme and tiny-static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")])
+
+(define_insn "sibcall_value_multiple_internal"
+ [(set (match_operand 0 "register_operand" "")
+ (call (mem:SI (match_operand 1 "call_insn_operand" "j,c,a,t,h"))
+ (match_operand 2 "" "")))
+ (set (match_operand 3 "register_operand" "")
+ (call (mem:SI (match_dup 1))
+ (match_dup 2)))]
+ "SIBLING_CALL_P (insn)"
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jr\t%1";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,%%pcrel(%1+0x20000)>>18\n\t"
+ "jirl\t$r0,$r12,%%pcrel(%1+4)-(%%pcrel(%1+4+0x20000)>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "b\t%1";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "b\t%1";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%1\n\tjr\t$r12";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r12,$r13,%1\n\tjr\t$r12";
+ else
+ return "la.global\t$r12,%1\n\tjr\t$r12";
+ case 4:
+ if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return "b\t%%plt(%1)";
+ else if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r12,(%%plt(%1)+0x20000)>>18\n\t"
+ "jirl\t$r0,$r12,%%plt(%1)+4-((%%plt(%1)+(4+0x20000))>>18<<18)";
+ else
+ {
+ sorry ("loongarch cmodel extreme and tiny-static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")])
+
+(define_expand "call"
+ [(parallel [(call (match_operand 0 "")
+ (match_operand 1 ""))
+ (use (match_operand 2 "")) ;; next_arg_reg
+ (use (match_operand 3 ""))])] ;; struct_value_size_rtx
+ ""
+{
+ rtx target = loongarch_legitimize_call_address (XEXP (operands[0], 0));
+
+ emit_call_insn (gen_call_internal (target, operands[1]));
+ DONE;
+})
+
+(define_insn "call_internal"
+ [(call (mem:SI (match_operand 0 "call_insn_operand" "e,c,a,t,h"))
+ (match_operand 1 "" ""))
+ (clobber (reg:SI RETURN_ADDR_REGNUM))]
+ ""
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jirl\t$r1,%0,0";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,%%pcrel(%0+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%pcrel(%0+4)-(%%pcrel(%0+4+0x20000)>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r1,$r12,%0\n\tjirl\t$r1,$r1,0";
+ else
+ return "bl\t%0";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "bl\t%0";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%0\n\tjirl\t$r1,$r1,0";
+ else
+ return "la.global\t$r1,%0\n\tjirl\t$r1,$r1,0";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%0\n\tjirl\t$r1,$r1,0";
+ else
+ return "la.global\t$r1,%0\n\tjirl\t$r1,$r1,0";
+ case 4:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,(%%plt(%0)+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%plt(%0)+4-((%%plt(%0)+(4+0x20000))>>18<<18)";
+ else if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return "bl\t%%plt(%0)";
+ else
+ {
+ sorry ("cmodel extreme and tiny-static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")
+ (set_attr "insn_count" "1,2,3,3,2")])
+
+(define_expand "call_value"
+ [(parallel [(set (match_operand 0 "")
+ (call (match_operand 1 "")
+ (match_operand 2 "")))
+ (use (match_operand 3 ""))])] ;; next_arg_reg
+ ""
+{
+ rtx target = loongarch_legitimize_call_address (XEXP (operands[1], 0));
+ /* Handle return values created by loongarch_return_fpr_pair. */
+ if (GET_CODE (operands[0]) == PARALLEL && XVECLEN (operands[0], 0) == 2)
+ {
+ rtx arg1 = XEXP (XVECEXP (operands[0], 0, 0), 0);
+ rtx arg2 = XEXP (XVECEXP (operands[0], 0, 1), 0);
+
+ emit_call_insn (gen_call_value_multiple_internal (arg1, target,
+ operands[2], arg2));
+ }
+ else
+ {
+ /* Handle return values created by loongarch_return_fpr_single. */
+ if (GET_CODE (operands[0]) == PARALLEL && XVECLEN (operands[0], 0) == 1)
+ operands[0] = XEXP (XVECEXP (operands[0], 0, 0), 0);
+
+ emit_call_insn (gen_call_value_internal (operands[0], target,
+ operands[2]));
+ }
+ DONE;
+})
+
+;; See comment for call_internal.
+(define_insn "call_value_internal"
+ [(set (match_operand 0 "register_operand" "")
+ (call (mem:SI (match_operand 1 "call_insn_operand" "e,c,a,t,h"))
+ (match_operand 2 "" "")))
+ (clobber (reg:SI RETURN_ADDR_REGNUM))]
+ ""
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jirl\t$r1,%1,0";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,%%pcrel(%1+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%pcrel(%1+4)-(%%pcrel(%1+4+0x20000)>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0";
+ else
+ return "bl\t%1";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "bl\t%1";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0";
+ else
+ return "la.global\t$r1,%1\n\tjirl\t$r1,$r1,0";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0";
+ else
+ return "la.global\t$r1,%1\n\tjirl\t$r1,$r1,0";
+ case 4:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,(%%plt(%1)+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%plt(%1)+4-((%%plt(%1)+(4+0x20000))>>18<<18)";
+ else if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return "bl\t%%plt(%1)";
+ else
+ {
+ sorry ("loongarch cmodel extreme and tiny-static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")
+ (set_attr "insn_count" "1,2,3,3,2")])
+
+;; See comment for call_internal.
+(define_insn "call_value_multiple_internal"
+ [(set (match_operand 0 "register_operand" "")
+ (call (mem:SI (match_operand 1 "call_insn_operand" "e,c,a,t,h"))
+ (match_operand 2 "" "")))
+ (set (match_operand 3 "register_operand" "")
+ (call (mem:SI (match_dup 1))
+ (match_dup 2)))
+ (clobber (reg:SI RETURN_ADDR_REGNUM))]
+ ""
+{
+ switch (which_alternative)
+ {
+ case 0:
+ return "jirl\t$r1,%1,0";
+ case 1:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,%%pcrel(%1+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%pcrel(%1+4)-(%%pcrel(%1+4+0x20000)>>18<<18)";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.local\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0";
+ else
+ return "bl\t%1";
+ case 2:
+ if (TARGET_CMODEL_TINY_STATIC)
+ return "bl\t%1";
+ else if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0 ";
+ else
+ return "la.global\t$r1,%1\n\tjirl\t$r1,$r1,0";
+ case 3:
+ if (TARGET_CMODEL_EXTREME)
+ return "la.global\t$r1,$r12,%1\n\tjirl\t$r1,$r1,0";
+ else
+ return "la.global\t$r1,%1\n\tjirl\t$r1,$r1,0";
+ case 4:
+ if (TARGET_CMODEL_LARGE)
+ return "pcaddu18i\t$r1,(%%plt(%1)+0x20000)>>18\n\t"
+ "jirl\t$r1,$r1,%%plt(%1)+4-((%%plt(%1)+(4+0x20000))>>18<<18)";
+ else if (TARGET_CMODEL_NORMAL || TARGET_CMODEL_TINY)
+ return "bl\t%%plt(%1)";
+ else
+ {
+ sorry ("loongarch cmodel extreme and tiny-static not support plt");
+ return ""; /* GCC complains about may fall through. */
+ }
+ default:
+ gcc_unreachable ();
+ }
+}
+ [(set_attr "jirl" "indirect,direct,direct,direct,direct")
+ (set_attr "insn_count" "1,2,3,3,2")])
+
+
+;; Call subroutine returning any type.
+
+(define_expand "untyped_call"
+ [(parallel [(call (match_operand 0 "")
+ (const_int 0))
+ (match_operand 1 "")
+ (match_operand 2 "")])]
+ ""
+{
+ int i;
+
+ emit_call_insn (gen_call (operands[0], const0_rtx, NULL, const0_rtx));
+
+ for (i = 0; i < XVECLEN (operands[2], 0); i++)
+ {
+ rtx set = XVECEXP (operands[2], 0, i);
+ loongarch_emit_move (SET_DEST (set), SET_SRC (set));
+ }
+
+ emit_insn (gen_blockage ());
+ DONE;
+})
+\f
+;;
+;; ....................
+;;
+;; MISC.
+;;
+;; ....................
+;;
+
+(define_insn "nop"
+ [(const_int 0)]
+ ""
+ "nop"
+ [(set_attr "type" "nop")
+ (set_attr "mode" "none")])
+
+;; __builtin_loongarch_movfcsr2gr: move the FCSR into operand 0.
+(define_insn "loongarch_movfcsr2gr"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec_volatile:SI [(match_operand 1 "const_uimm5_operand")]
+ UNSPEC_MOVFCSR2GR))]
+ "TARGET_HARD_FLOAT"
+ "movfcsr2gr\t%0,$r%1")
+
+;; __builtin_loongarch_movgr2fcsr: move operand 0 into the FCSR.
+(define_insn "loongarch_movgr2fcsr"
+ [(unspec_volatile [(match_operand 0 "const_uimm5_operand")
+ (match_operand:SI 1 "register_operand" "r")]
+ UNSPEC_MOVGR2FCSR)]
+ "TARGET_HARD_FLOAT"
+ "movgr2fcsr\t$r%0,%1")
+
+(define_insn "fclass_<fmt>"
+ [(set (match_operand:ANYF 0 "register_operand" "=f")
+ (unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
+ UNSPEC_FCLASS))]
+ "TARGET_HARD_FLOAT"
+ "fclass.<fmt>\t%0,%1"
+ [(set_attr "type" "unknown")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "bytepick_w"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec:SI [(match_operand:SI 1 "register_operand" "r")
+ (match_operand:SI 2 "register_operand" "r")
+ (match_operand:SI 3 "const_0_to_3_operand" "n")]
+ UNSPEC_BYTEPICK_W))]
+ ""
+ "bytepick.w\t%0,%1,%2,%z3"
+ [(set_attr "mode" "SI")])
+
+(define_insn "bytepick_d"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (unspec:DI [(match_operand:DI 1 "register_operand" "r")
+ (match_operand:DI 2 "register_operand" "r")
+ (match_operand:DI 3 "const_0_to_7_operand" "n")]
+ UNSPEC_BYTEPICK_D))]
+ ""
+ "bytepick.d\t%0,%1,%2,%z3"
+ [(set_attr "mode" "DI")])
+
+(define_insn "bitrev_4b"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec:SI [(match_operand:SI 1 "register_operand" "r")]
+ UNSPEC_BITREV_4B))]
+ ""
+ "bitrev.4b\t%0,%1"
+ [(set_attr "type" "unknown")
+ (set_attr "mode" "SI")])
+
+(define_insn "bitrev_8b"
+ [(set (match_operand:DI 0 "register_operand" "=r")
+ (unspec:DI [(match_operand:DI 1 "register_operand" "r")]
+ UNSPEC_BITREV_8B))]
+ ""
+ "bitrev.8b\t%0,%1"
+ [(set_attr "type" "unknown")
+ (set_attr "mode" "DI")])
+
+(define_insn "stack_tie<mode>"
+ [(set (mem:BLK (scratch))
+ (unspec:BLK [(match_operand:GPR 0 "register_operand" "r")
+ (match_operand:GPR 1 "register_operand" "r")]
+ UNSPEC_TIE))]
+ ""
+ ""
+ [(set_attr "length" "0")]
+)
+
+(define_insn "gpr_restore_return"
+ [(return)
+ (use (match_operand 0 "pmode_register_operand" ""))
+ (const_int 0)]
+ ""
+ "")
+\f
+(define_split
+ [(match_operand 0 "small_data_pattern")]
+ "reload_completed"
+ [(match_dup 0)]
+ { operands[0] = loongarch_rewrite_small_data (operands[0]); })
+
+
+;; Match paired HI/SI/SF/DFmode load/stores.
+(define_insn "*join2_load_store<JOIN_MODE:mode>"
+ [(set (match_operand:JOIN_MODE 0 "nonimmediate_operand" "=r,f,m,m,r,ZC")
+ (match_operand:JOIN_MODE 1 "nonimmediate_operand" "m,m,r,f,ZC,r"))
+ (set (match_operand:JOIN_MODE 2 "nonimmediate_operand" "=r,f,m,m,r,ZC")
+ (match_operand:JOIN_MODE 3 "nonimmediate_operand" "m,m,r,f,ZC,r"))]
+ "reload_completed"
+ {
+ bool load_p = (which_alternative == 0 || which_alternative == 1);
+ /* Reg-renaming pass reuses base register if it is dead after bonded loads.
+ Hardware does not bond those loads, even when they are consecutive.
+ However, order of the loads need to be checked for correctness. */
+ if (!load_p || !reg_overlap_mentioned_p (operands[0], operands[1]))
+ {
+ output_asm_insn (loongarch_output_move (operands[0], operands[1]),
+ operands);
+ output_asm_insn (loongarch_output_move (operands[2], operands[3]),
+ &operands[2]);
+ }
+ else
+ {
+ output_asm_insn (loongarch_output_move (operands[2], operands[3]),
+ &operands[2]);
+ output_asm_insn (loongarch_output_move (operands[0], operands[1]),
+ operands);
+ }
+ return "";
+ }
+ [(set_attr "move_type" "load,fpload,store,fpstore,load,store")
+ (set_attr "insn_count" "2,2,2,2,2,2")])
+
+;; 2 HI/SI/SF/DF loads are bonded.
+(define_peephole2
+ [(set (match_operand:JOIN_MODE 0 "register_operand")
+ (match_operand:JOIN_MODE 1 "non_volatile_mem_operand"))
+ (set (match_operand:JOIN_MODE 2 "register_operand")
+ (match_operand:JOIN_MODE 3 "non_volatile_mem_operand"))]
+ "loongarch_load_store_bonding_p (operands, <JOIN_MODE:MODE>mode, true)"
+ [(parallel [(set (match_dup 0)
+ (match_dup 1))
+ (set (match_dup 2)
+ (match_dup 3))])]
+ "")
+
+;; 2 HI/SI/SF/DF stores are bonded.
+(define_peephole2
+ [(set (match_operand:JOIN_MODE 0 "memory_operand")
+ (match_operand:JOIN_MODE 1 "register_operand"))
+ (set (match_operand:JOIN_MODE 2 "memory_operand")
+ (match_operand:JOIN_MODE 3 "register_operand"))]
+ "loongarch_load_store_bonding_p (operands, <JOIN_MODE:MODE>mode, false)"
+ [(parallel [(set (match_dup 0)
+ (match_dup 1))
+ (set (match_dup 2)
+ (match_dup 3))])]
+ "")
+
+;; Match paired HImode loads.
+(define_insn "*join2_loadhi"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (any_extend:SI (match_operand:HI 1 "non_volatile_mem_operand" "m")))
+ (set (match_operand:SI 2 "register_operand" "=r")
+ (any_extend:SI (match_operand:HI 3 "non_volatile_mem_operand" "m")))]
+ "reload_completed"
+ {
+ /* Reg-renaming pass reuses base register if it is dead after bonded loads.
+ Hardware does not bond those loads, even when they are consecutive.
+ However, order of the loads need to be checked for correctness. */
+ if (!reg_overlap_mentioned_p (operands[0], operands[1]))
+ {
+ output_asm_insn ("ld.h<u>\t%0,%1", operands);
+ output_asm_insn ("ld.h<u>\t%2,%3", operands);
+ }
+ else
+ {
+ output_asm_insn ("ld.h<u>\t%2,%3", operands);
+ output_asm_insn ("ld.h<u>\t%0,%1", operands);
+ }
+
+ return "";
+ }
+ [(set_attr "move_type" "load")
+ (set_attr "insn_count" "2")])
+
+
+;; 2 HI loads are bonded.
+(define_peephole2
+ [(set (match_operand:SI 0 "register_operand")
+ (any_extend:SI (match_operand:HI 1 "non_volatile_mem_operand")))
+ (set (match_operand:SI 2 "register_operand")
+ (any_extend:SI (match_operand:HI 3 "non_volatile_mem_operand")))]
+ "loongarch_load_store_bonding_p (operands, HImode, true)"
+ [(parallel [(set (match_dup 0)
+ (any_extend:SI (match_dup 1)))
+ (set (match_dup 2)
+ (any_extend:SI (match_dup 3)))])]
+ "")
+
+\f
+
+(define_mode_iterator QHSD [QI HI SI DI])
+
+(define_insn "crc_w_<size>_w"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec:SI [(match_operand:QHSD 1 "register_operand" "r")
+ (match_operand:SI 2 "register_operand" "r")]
+ UNSPEC_CRC))]
+ ""
+ "crc.w.<size>.w\t%0,%1,%2"
+ [(set_attr "type" "unknown")
+ (set_attr "mode" "<MODE>")])
+
+(define_insn "crcc_w_<size>_w"
+ [(set (match_operand:SI 0 "register_operand" "=r")
+ (unspec:SI [(match_operand:QHSD 1 "register_operand" "r")
+ (match_operand:SI 2 "register_operand" "r")]
+ UNSPEC_CRCC))]
+ ""
+ "crcc.w.<size>.w\t%0,%1,%2"
+ [(set_attr "type" "unknown")
+ (set_attr "mode" "<MODE>")])
+
+;; Synchronization instructions.
+
+(include "sync.md")
+
+(include "generic.md")
+(include "la464.md")
+
+(define_c_enum "unspec" [
+ UNSPEC_ADDRESS_FIRST
+])
diff --git a/gcc/config/loongarch/predicates.md b/gcc/config/loongarch/predicates.md
new file mode 100644
index 00000000000..681ee93ec32
--- /dev/null
+++ b/gcc/config/loongarch/predicates.md
@@ -0,0 +1,553 @@
+;; Predicate definitions for LoongArch target.
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+;; Based on MIPS target for GNU compiler.
+;;
+;; This file is part of GCC.
+;;
+;; GCC is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation; either version 3, or (at your option)
+;; any later version.
+;;
+;; GCC is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+;;
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+(define_predicate "const_uns_arith_operand"
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND_UNSIGNED (INTVAL (op))")))
+
+(define_predicate "uns_arith_operand"
+ (ior (match_operand 0 "const_uns_arith_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "const_lu32i_operand"
+ (and (match_code "const_int")
+ (match_test "LU32I_OPERAND (INTVAL (op))")))
+
+(define_predicate "const_lu52i_operand"
+ (and (match_code "const_int")
+ (match_test "LU52I_OPERAND (INTVAL (op))")))
+
+(define_predicate "const_arith_operand"
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND (INTVAL (op))")))
+
+(define_predicate "const_imm16_operand"
+ (and (match_code "const_int")
+ (match_test "IMM16_OPERAND (INTVAL (op))")))
+
+(define_predicate "arith_operand"
+ (ior (match_operand 0 "const_arith_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "const_immlsa_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 1, 4)")))
+
+(define_predicate "const_uimm3_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 7)")))
+
+(define_predicate "const_uimm4_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 15)")))
+
+(define_predicate "const_uimm5_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 31)")))
+
+(define_predicate "const_uimm6_operand"
+ (and (match_code "const_int")
+ (match_test "UIMM6_OPERAND (INTVAL (op))")))
+
+(define_predicate "const_uimm7_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 127)")))
+
+(define_predicate "const_uimm8_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 255)")))
+
+(define_predicate "const_uimm14_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 16383)")))
+
+(define_predicate "const_uimm15_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 32767)")))
+
+(define_predicate "const_imm5_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), -16, 15)")))
+
+(define_predicate "const_imm10_operand"
+ (and (match_code "const_int")
+ (match_test "IMM10_OPERAND (INTVAL (op))")))
+
+(define_predicate "const_imm12_operand"
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND (INTVAL (op))")))
+
+(define_predicate "reg_imm10_operand"
+ (ior (match_operand 0 "const_imm10_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "aq8b_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 0)")))
+
+(define_predicate "aq8h_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 1)")))
+
+(define_predicate "aq8w_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 2)")))
+
+(define_predicate "aq8d_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 3)")))
+
+(define_predicate "aq10b_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 10, 0)")))
+
+(define_predicate "aq10h_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 10, 1)")))
+
+(define_predicate "aq10w_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 10, 2)")))
+
+(define_predicate "aq10d_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 10, 3)")))
+
+(define_predicate "aq12b_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 12, 0)")))
+
+(define_predicate "aq12h_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 11, 1)")))
+
+(define_predicate "aq12w_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 10, 2)")))
+
+(define_predicate "aq12d_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 9, 3)")))
+
+(define_predicate "sle_operand"
+ (and (match_code "const_int")
+ (match_test "IMM12_OPERAND (INTVAL (op) + 1)")))
+
+(define_predicate "sleu_operand"
+ (and (match_operand 0 "sle_operand")
+ (match_test "INTVAL (op) + 1 != 0")))
+
+(define_predicate "const_0_operand"
+ (and (match_code "const_int,const_double,const_vector")
+ (match_test "op == CONST0_RTX (GET_MODE (op))")))
+
+(define_predicate "const_m1_operand"
+ (and (match_code "const_int,const_double,const_vector")
+ (match_test "op == CONSTM1_RTX (GET_MODE (op))")))
+
+(define_predicate "reg_or_m1_operand"
+ (ior (match_operand 0 "const_m1_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "reg_or_0_operand"
+ (ior (match_operand 0 "const_0_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "const_1_operand"
+ (and (match_code "const_int,const_double,const_vector")
+ (match_test "op == CONST1_RTX (GET_MODE (op))")))
+
+(define_predicate "reg_or_1_operand"
+ (ior (match_operand 0 "const_1_operand")
+ (match_operand 0 "register_operand")))
+
+;; These are used in vec_merge, hence accept bitmask as const_int.
+(define_predicate "const_exp_2_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (exact_log2 (INTVAL (op)), 0, 1)")))
+
+(define_predicate "const_exp_4_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (exact_log2 (INTVAL (op)), 0, 3)")))
+
+(define_predicate "const_exp_8_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (exact_log2 (INTVAL (op)), 0, 7)")))
+
+(define_predicate "const_exp_16_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (exact_log2 (INTVAL (op)), 0, 15)")))
+
+(define_predicate "const_exp_32_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (exact_log2 (INTVAL (op)), 0, 31)")))
+
+;; This is used for indexing into vectors, and hence only accepts const_int.
+(define_predicate "const_0_or_1_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 1)")))
+
+(define_predicate "const_2_or_3_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 2, 3)")))
+
+(define_predicate "const_0_to_3_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 3)")))
+
+(define_predicate "const_0_to_7_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 7)")))
+
+(define_predicate "const_4_to_7_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 4, 7)")))
+
+(define_predicate "const_8_to_15_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 7)")))
+
+(define_predicate "const_16_to_31_operand"
+ (and (match_code "const_int")
+ (match_test "IN_RANGE (INTVAL (op), 0, 7)")))
+
+(define_predicate "qi_mask_operand"
+ (and (match_code "const_int")
+ (match_test "UINTVAL (op) == 0xff")))
+
+(define_predicate "hi_mask_operand"
+ (and (match_code "const_int")
+ (match_test "UINTVAL (op) == 0xffff")))
+
+(define_predicate "lu52i_mask_operand"
+ (and (match_code "const_int")
+ (match_test "UINTVAL (op) == 0xfffffffffffff")))
+
+(define_predicate "shift_mask_operand"
+ (and (match_code "const_int")
+ (ior (match_test "UINTVAL (op) == 0x3fffffffc")
+ (match_test "UINTVAL (op) == 0x1fffffffe")
+ (match_test "UINTVAL (op) == 0x7fffffff8")
+ (match_test "UINTVAL (op) == 0xffffffff0"))))
+
+(define_predicate "si_mask_operand"
+ (and (match_code "const_int")
+ (match_test "UINTVAL (op) == 0xffffffff")))
+
+(define_predicate "and_load_operand"
+ (ior (match_operand 0 "qi_mask_operand")
+ (match_operand 0 "hi_mask_operand")
+ (match_operand 0 "si_mask_operand")))
+
+(define_predicate "low_bitmask_operand"
+ (and (match_code "const_int")
+ (match_test "low_bitmask_len (mode, INTVAL (op)) > 12")))
+
+(define_predicate "and_reg_operand"
+ (ior (match_operand 0 "register_operand")
+ (match_operand 0 "const_uns_arith_operand")
+ (match_operand 0 "low_bitmask_operand")
+ (match_operand 0 "si_mask_operand")))
+
+(define_predicate "and_operand"
+ (ior (match_operand 0 "and_load_operand")
+ (match_operand 0 "and_reg_operand")))
+
+(define_predicate "d_operand"
+ (and (match_code "reg")
+ (match_test "GP_REG_P (REGNO (op))")))
+
+(define_predicate "db4_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op) + 1, 4, 0)")))
+
+(define_predicate "db7_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op) + 1, 7, 0)")))
+
+(define_predicate "db8_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op) + 1, 8, 0)")))
+
+(define_predicate "ib3_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op) - 1, 3, 0)")))
+
+(define_predicate "sb4_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 4, 0)")))
+
+(define_predicate "sb5_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 5, 0)")))
+
+(define_predicate "sb8_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 0)")))
+
+(define_predicate "sd8_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_signed_immediate_p (INTVAL (op), 8, 3)")))
+
+(define_predicate "ub4_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 4, 0)")))
+
+(define_predicate "ub8_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 8, 0)")))
+
+(define_predicate "uh4_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 4, 1)")))
+
+(define_predicate "uw4_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 4, 2)")))
+
+(define_predicate "uw5_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 5, 2)")))
+
+(define_predicate "uw6_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 6, 2)")))
+
+(define_predicate "uw8_operand"
+ (and (match_code "const_int")
+ (match_test "loongarch_unsigned_immediate_p (INTVAL (op), 8, 2)")))
+
+(define_predicate "addiur2_operand"
+ (and (match_code "const_int")
+ (ior (match_test "INTVAL (op) == -1")
+ (match_test "INTVAL (op) == 1")
+ (match_test "INTVAL (op) == 4")
+ (match_test "INTVAL (op) == 8")
+ (match_test "INTVAL (op) == 12")
+ (match_test "INTVAL (op) == 16")
+ (match_test "INTVAL (op) == 20")
+ (match_test "INTVAL (op) == 24"))))
+
+(define_predicate "addiusp_operand"
+ (and (match_code "const_int")
+ (ior (match_test "(IN_RANGE (INTVAL (op), 2, 257))")
+ (match_test "(IN_RANGE (INTVAL (op), -258, -3))"))))
+
+(define_predicate "andi16_operand"
+ (and (match_code "const_int")
+ (ior (match_test "IN_RANGE (INTVAL (op), 1, 4)")
+ (match_test "IN_RANGE (INTVAL (op), 7, 8)")
+ (match_test "IN_RANGE (INTVAL (op), 15, 16)")
+ (match_test "IN_RANGE (INTVAL (op), 31, 32)")
+ (match_test "IN_RANGE (INTVAL (op), 63, 64)")
+ (match_test "INTVAL (op) == 255")
+ (match_test "INTVAL (op) == 32768")
+ (match_test "INTVAL (op) == 65535"))))
+
+(define_predicate "movep_src_register"
+ (and (match_code "reg")
+ (ior (match_test ("IN_RANGE (REGNO (op), 2, 3)"))
+ (match_test ("IN_RANGE (REGNO (op), 16, 20)")))))
+
+(define_predicate "movep_src_operand"
+ (ior (match_operand 0 "const_0_operand")
+ (match_operand 0 "movep_src_register")))
+
+(define_predicate "fcc_reload_operand"
+ (and (match_code "reg,subreg")
+ (match_test "FCC_REG_P (true_regnum (op))")))
+
+(define_predicate "muldiv_target_operand"
+ (match_operand 0 "register_operand"))
+
+(define_predicate "const_call_insn_operand"
+ (match_code "const,symbol_ref,label_ref")
+{
+ enum loongarch_symbol_type symbol_type;
+
+ if (!loongarch_symbolic_constant_p (op, SYMBOL_CONTEXT_CALL, &symbol_type))
+ return false;
+
+ switch (symbol_type)
+ {
+ case SYMBOL_GOT_DISP:
+ /* Without explicit relocs, there is no special syntax for
+ loading the address of a call destination into a register.
+ Using "la.global JIRL_REGS,foo; jirl JIRL_REGS" would prevent the lazy
+ binding of "foo", so keep the address of global symbols with the jirl
+ macro. */
+ return 1;
+
+ default:
+ return false;
+ }
+})
+
+(define_predicate "call_insn_operand"
+ (ior (match_operand 0 "const_call_insn_operand")
+ (match_operand 0 "register_operand")))
+
+(define_predicate "is_const_call_local_symbol"
+ (and (match_operand 0 "const_call_insn_operand")
+ (ior (match_test "loongarch_global_symbol_p (op) == 0")
+ (match_test "loongarch_symbol_binds_local_p (op) != 0"))
+ (match_test "CONSTANT_P (op)")))
+
+(define_predicate "is_const_call_weak_symbol"
+ (and (match_operand 0 "const_call_insn_operand")
+ (not (match_operand 0 "is_const_call_local_symbol"))
+ (match_test "loongarch_weak_symbol_p (op) != 0")
+ (match_test "CONSTANT_P (op)")))
+
+(define_predicate "is_const_call_plt_symbol"
+ (and (match_operand 0 "const_call_insn_operand")
+ (match_test "flag_plt != 0")
+ (match_test "loongarch_global_symbol_noweak_p (op) != 0")
+ (match_test "CONSTANT_P (op)")))
+
+(define_predicate "is_const_call_global_noplt_symbol"
+ (and (match_operand 0 "const_call_insn_operand")
+ (match_test "flag_plt == 0")
+ (match_test "loongarch_global_symbol_noweak_p (op) != 0")
+ (match_test "CONSTANT_P (op)")))
+
+;; A legitimate CONST_INT operand that takes more than one instruction
+;; to load.
+(define_predicate "splittable_const_int_operand"
+ (match_code "const_int")
+{
+ /* Don't handle multi-word moves this way; we don't want to introduce
+ the individual word-mode moves until after reload. */
+ if (GET_MODE_SIZE (mode) > UNITS_PER_WORD)
+ return false;
+
+ /* Otherwise check whether the constant can be loaded in a single
+ instruction. */
+ return !LU12I_INT (op) && !IMM12_INT (op) && !IMM12_INT_UNSIGNED (op)
+ && !LU52I_INT (op);
+})
+
+(define_predicate "move_operand"
+ (match_operand 0 "general_operand")
+{
+ enum loongarch_symbol_type symbol_type;
+
+ /* The thinking here is as follows:
+
+ (1) The move expanders should split complex load sequences into
+ individual instructions. Those individual instructions can
+ then be optimized by all rtl passes.
+
+ (2) The target of pre-reload load sequences should not be used
+ to store temporary results. If the target register is only
+ assigned one value, reload can rematerialize that value
+ on demand, rather than spill it to the stack.
+
+ (3) If we allowed pre-reload passes like combine and cse to recreate
+ complex load sequences, we would want to be able to split the
+ sequences before reload as well, so that the pre-reload scheduler
+ can see the individual instructions. This falls foul of (2);
+ the splitter would be forced to reuse the target register for
+ intermediate results.
+
+ (4) We want to define complex load splitters for combine. These
+ splitters can request a temporary scratch register, which avoids
+ the problem in (2). They allow things like:
+
+ (set (reg T1) (high SYM))
+ (set (reg T2) (low (reg T1) SYM))
+ (set (reg X) (plus (reg T2) (const_int OFFSET)))
+
+ to be combined into:
+
+ (set (reg T3) (high SYM+OFFSET))
+ (set (reg X) (lo_sum (reg T3) SYM+OFFSET))
+
+ if T2 is only used this once. */
+ switch (GET_CODE (op))
+ {
+ case CONST_INT:
+ return !splittable_const_int_operand (op, mode);
+
+ case CONST:
+ case SYMBOL_REF:
+ case LABEL_REF:
+ return (loongarch_symbolic_constant_p (op, SYMBOL_CONTEXT_LEA,
+ &symbol_type));
+ default:
+ return true;
+ }
+})
+
+(define_predicate "consttable_operand"
+ (match_test "CONSTANT_P (op)"))
+
+(define_predicate "symbolic_operand"
+ (match_code "const,symbol_ref,label_ref")
+{
+ enum loongarch_symbol_type type;
+ return loongarch_symbolic_constant_p (op, SYMBOL_CONTEXT_LEA, &type);
+})
+
+(define_predicate "got_disp_operand"
+ (match_code "const,symbol_ref,label_ref")
+{
+ enum loongarch_symbol_type type;
+ return (loongarch_symbolic_constant_p (op, SYMBOL_CONTEXT_LEA, &type)
+ && type == SYMBOL_GOT_DISP);
+})
+
+(define_predicate "symbol_ref_operand"
+ (match_code "symbol_ref"))
+
+(define_predicate "stack_operand"
+ (and (match_code "mem")
+ (match_test "loongarch_stack_address_p (XEXP (op, 0), GET_MODE (op))")))
+
+(define_predicate "equality_operator"
+ (match_code "eq,ne"))
+
+(define_predicate "extend_operator"
+ (match_code "zero_extend,sign_extend"))
+
+(define_predicate "trap_comparison_operator"
+ (match_code "eq,ne,lt,ltu,ge,geu"))
+
+(define_predicate "order_operator"
+ (match_code "lt,ltu,le,leu,ge,geu,gt,gtu"))
+
+;; For NE, cstore uses sltu instructions in which the first operand is $0.
+
+(define_predicate "loongarch_cstore_operator"
+ (match_code "ne,eq,gt,gtu,ge,geu,lt,ltu,le,leu"))
+
+(define_predicate "small_data_pattern"
+ (and (match_code "set,parallel,unspec,unspec_volatile,prefetch")
+ (match_test "loongarch_small_data_pattern_p (op)")))
+
+(define_predicate "mem_noofs_operand"
+ (and (match_code "mem")
+ (match_code "reg" "0")))
+
+;; Return 1 if the operand is in non-volatile memory.
+(define_predicate "non_volatile_mem_operand"
+ (and (match_operand 0 "memory_operand")
+ (not (match_test "MEM_VOLATILE_P (op)"))))
diff --git a/gcc/config/loongarch/sync.md b/gcc/config/loongarch/sync.md
new file mode 100644
index 00000000000..69e9c137ac8
--- /dev/null
+++ b/gcc/config/loongarch/sync.md
@@ -0,0 +1,574 @@
+;; Machine description for LoongArch atomic operations.
+;; Copyright (C) 2021 Free Software Foundation, Inc.
+;; Contributed by Loongson Ltd.
+;; Based on MIPS and RISC-V target for GNU compiler.
+
+;; This file is part of GCC.
+
+;; GCC is free software; you can redistribute it and/or modify
+;; it under the terms of the GNU General Public License as published by
+;; the Free Software Foundation; either version 3, or (at your option)
+;; any later version.
+
+;; GCC is distributed in the hope that it will be useful,
+;; but WITHOUT ANY WARRANTY; without even the implied warranty of
+;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+;; GNU General Public License for more details.
+
+;; You should have received a copy of the GNU General Public License
+;; along with GCC; see the file COPYING3. If not see
+;; <http://www.gnu.org/licenses/>.
+
+(define_c_enum "unspec" [
+ UNSPEC_COMPARE_AND_SWAP
+ UNSPEC_COMPARE_AND_SWAP_ADD
+ UNSPEC_COMPARE_AND_SWAP_SUB
+ UNSPEC_COMPARE_AND_SWAP_AND
+ UNSPEC_COMPARE_AND_SWAP_XOR
+ UNSPEC_COMPARE_AND_SWAP_OR
+ UNSPEC_COMPARE_AND_SWAP_NAND
+ UNSPEC_SYNC_OLD_OP
+ UNSPEC_SYNC_EXCHANGE
+ UNSPEC_ATOMIC_STORE
+ UNSPEC_MEMORY_BARRIER
+])
+
+(define_code_iterator any_atomic [plus ior xor and])
+(define_code_attr atomic_optab
+ [(plus "add") (ior "or") (xor "xor") (and "and")])
+
+;; This attribute gives the format suffix for atomic memory operations.
+(define_mode_attr amo [(SI "w") (DI "d")])
+
+;; <amop> expands to the name of the atomic operand that implements a
+;; particular code.
+(define_code_attr amop [(ior "or") (xor "xor") (and "and") (plus "add")])
+
+;; Memory barriers.
+
+(define_expand "mem_thread_fence"
+ [(match_operand:SI 0 "const_int_operand" "")] ;; model
+ ""
+{
+ if (INTVAL (operands[0]) != MEMMODEL_RELAXED)
+ {
+ rtx mem = gen_rtx_MEM (BLKmode, gen_rtx_SCRATCH (Pmode));
+ MEM_VOLATILE_P (mem) = 1;
+ emit_insn (gen_mem_thread_fence_1 (mem, operands[0]));
+ }
+ DONE;
+})
+
+;; Until the LoongArch memory model (hence its mapping from C++) is finalized,
+;; conservatively emit a full FENCE.
+(define_insn "mem_thread_fence_1"
+ [(set (match_operand:BLK 0 "" "")
+ (unspec:BLK [(match_dup 0)] UNSPEC_MEMORY_BARRIER))
+ (match_operand:SI 1 "const_int_operand" "")] ;; model
+ ""
+ "dbar\t0")
+
+;; Atomic memory operations.
+
+;; Implement atomic stores with amoswap. Fall back to fences for atomic loads.
+(define_insn "atomic_store<mode>"
+ [(set (match_operand:GPR 0 "memory_operand" "+ZB")
+ (unspec_volatile:GPR
+ [(match_operand:GPR 1 "reg_or_0_operand" "rJ")
+ (match_operand:SI 2 "const_int_operand")] ;; model
+ UNSPEC_ATOMIC_STORE))]
+ ""
+ "amswap%A2.<amo>\t$zero,%z1,%0"
+ [(set (attr "length") (const_int 8))])
+
+(define_insn "atomic_<atomic_optab><mode>"
+ [(set (match_operand:GPR 0 "memory_operand" "+ZB")
+ (unspec_volatile:GPR
+ [(any_atomic:GPR (match_dup 0)
+ (match_operand:GPR 1 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 2 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+ "am<amop>%A2.<amo>\t$zero,%z1,%0"
+ [(set (attr "length") (const_int 8))])
+
+(define_insn "atomic_fetch_<atomic_optab><mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r")
+ (match_operand:GPR 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR
+ [(any_atomic:GPR (match_dup 1)
+ (match_operand:GPR 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+ "am<amop>%A3.<amo>\t%0,%z2,%1"
+ [(set (attr "length") (const_int 8))])
+
+(define_insn "atomic_exchange<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r")
+ (unspec_volatile:GPR
+ [(match_operand:GPR 1 "memory_operand" "+ZB")
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_EXCHANGE))
+ (set (match_dup 1)
+ (match_operand:GPR 2 "register_operand" "r"))]
+ ""
+ "amswap%A3.<amo>\t%0,%z2,%1"
+ [(set (attr "length") (const_int 8))])
+
+(define_insn "atomic_cas_value_strong<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r")
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ")
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ")
+ (match_operand:SI 4 "const_int_operand") ;; mod_s
+ (match_operand:SI 5 "const_int_operand")] ;; mod_f
+ UNSPEC_COMPARE_AND_SWAP))
+ (clobber (match_scratch:GPR 6 "=&r"))]
+ ""
+{
+ return "%G5\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "bne\\t%0,%z2,2f\\n\\t"
+ "or%i3\\t%6,$zero,%3\\n\\t"
+ "sc.<amo>\\t%6,%1\\n\\t"
+ "beq\\t$zero,%6,1b\\n\\t"
+ "b\\t3f\\n\\t"
+ "2:\\n\\t"
+ "dbar\\t0x700\\n\\t"
+ "3:\\n\\t";
+}
+ [(set (attr "length") (const_int 32))])
+
+(define_expand "atomic_compare_and_swap<mode>"
+ [(match_operand:SI 0 "register_operand" "") ;; bool output
+ (match_operand:GPR 1 "register_operand" "") ;; val output
+ (match_operand:GPR 2 "memory_operand" "") ;; memory
+ (match_operand:GPR 3 "reg_or_0_operand" "") ;; expected value
+ (match_operand:GPR 4 "reg_or_0_operand" "") ;; desired value
+ (match_operand:SI 5 "const_int_operand" "") ;; is_weak
+ (match_operand:SI 6 "const_int_operand" "") ;; mod_s
+ (match_operand:SI 7 "const_int_operand" "")] ;; mod_f
+ ""
+{
+ emit_insn (gen_atomic_cas_value_strong<mode> (operands[1], operands[2],
+ operands[3], operands[4],
+ operands[6], operands[7]));
+
+ rtx compare = operands[1];
+ if (operands[3] != const0_rtx)
+ {
+ rtx difference = gen_rtx_MINUS (<MODE>mode, operands[1], operands[3]);
+ compare = gen_reg_rtx (<MODE>mode);
+ emit_insn (gen_rtx_SET (compare, difference));
+ }
+
+ if (word_mode != <MODE>mode)
+ {
+ rtx reg = gen_reg_rtx (word_mode);
+ emit_insn (gen_rtx_SET (reg, gen_rtx_SIGN_EXTEND (word_mode, compare)));
+ compare = reg;
+ }
+
+ emit_insn (gen_rtx_SET (operands[0],
+ gen_rtx_EQ (SImode, compare, const0_rtx)));
+ DONE;
+})
+
+(define_expand "atomic_test_and_set"
+ [(match_operand:QI 0 "register_operand" "") ;; bool output
+ (match_operand:QI 1 "memory_operand" "+ZB") ;; memory
+ (match_operand:SI 2 "const_int_operand" "")] ;; model
+ ""
+{
+ /* We have no QImode atomics, so use the address LSBs to form a mask,
+ then use an aligned SImode atomic. */
+ rtx result = operands[0];
+ rtx mem = operands[1];
+ rtx model = operands[2];
+ rtx addr = force_reg (Pmode, XEXP (mem, 0));
+ rtx tmp_reg = gen_reg_rtx (Pmode);
+ rtx zero_reg = gen_rtx_REG (Pmode, 0);
+
+ rtx aligned_addr = gen_reg_rtx (Pmode);
+ emit_move_insn (tmp_reg, gen_rtx_PLUS (Pmode, zero_reg, GEN_INT (-4)));
+ emit_move_insn (aligned_addr, gen_rtx_AND (Pmode, addr, tmp_reg));
+
+ rtx aligned_mem = change_address (mem, SImode, aligned_addr);
+ set_mem_alias_set (aligned_mem, 0);
+
+ rtx offset = gen_reg_rtx (SImode);
+ emit_move_insn (offset, gen_rtx_AND (SImode, gen_lowpart (SImode, addr),
+ GEN_INT (3)));
+
+ rtx tmp = gen_reg_rtx (SImode);
+ emit_move_insn (tmp, GEN_INT (1));
+
+ rtx shmt = gen_reg_rtx (SImode);
+ emit_move_insn (shmt, gen_rtx_ASHIFT (SImode, offset, GEN_INT (3)));
+
+ rtx word = gen_reg_rtx (SImode);
+ emit_move_insn (word, gen_rtx_ASHIFT (SImode, tmp, shmt));
+
+ tmp = gen_reg_rtx (SImode);
+ emit_insn (gen_atomic_fetch_orsi (tmp, aligned_mem, word, model));
+
+ emit_move_insn (gen_lowpart (SImode, result),
+ gen_rtx_LSHIFTRT (SImode, tmp, shmt));
+ DONE;
+})
+
+(define_insn "atomic_cas_value_cmp_and_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r")
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ")
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ")
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ")
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ")
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP))
+ (clobber (match_scratch:GPR 7 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%2\\n\\t"
+ "bne\\t%7,%z4,2f\\n\\t"
+ "and\\t%7,%0,%z3\\n\\t"
+ "or%i5\\t%7,%7,%5\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b\\n\\t"
+ "b\\t3f\\n\\t"
+ "2:\\n\\t"
+ "dbar\\t0x700\\n\\t"
+ "3:\\n\\t";
+}
+ [(set (attr "length") (const_int 40))])
+
+(define_expand "atomic_compare_and_swap<mode>"
+ [(match_operand:SI 0 "register_operand" "") ;; bool output
+ (match_operand:SHORT 1 "register_operand" "") ;; val output
+ (match_operand:SHORT 2 "memory_operand" "") ;; memory
+ (match_operand:SHORT 3 "reg_or_0_operand" "") ;; expected value
+ (match_operand:SHORT 4 "reg_or_0_operand" "") ;; desired value
+ (match_operand:SI 5 "const_int_operand" "") ;; is_weak
+ (match_operand:SI 6 "const_int_operand" "") ;; mod_s
+ (match_operand:SI 7 "const_int_operand" "")] ;; mod_f
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_cmp_and_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[1], operands[2],
+ operands[3], operands[4], operands[7]);
+
+ rtx compare = operands[1];
+ if (operands[3] != const0_rtx)
+ {
+ machine_mode mode = GET_MODE (operands[3]);
+ rtx op1 = convert_modes (SImode, mode, operands[1], true);
+ rtx op3 = convert_modes (SImode, mode, operands[3], true);
+ rtx difference = gen_rtx_MINUS (SImode, op1, op3);
+ compare = gen_reg_rtx (SImode);
+ emit_insn (gen_rtx_SET (compare, difference));
+ }
+
+ if (word_mode != <MODE>mode)
+ {
+ rtx reg = gen_reg_rtx (word_mode);
+ emit_insn (gen_rtx_SET (reg, gen_rtx_SIGN_EXTEND (word_mode, compare)));
+ compare = reg;
+ }
+
+ emit_insn (gen_rtx_SET (operands[0],
+ gen_rtx_EQ (SImode, compare, const0_rtx)));
+ DONE;
+})
+
+(define_insn "atomic_cas_value_add_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_ADD))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "add.w\\t%8,%0,%z5\\n\\t"
+ "and\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+
+ [(set (attr "length") (const_int 32))])
+
+(define_insn "atomic_cas_value_sub_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_SUB))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "sub.w\\t%8,%0,%z5\\n\\t"
+ "and\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+ [(set (attr "length") (const_int 32))])
+
+(define_insn "atomic_cas_value_and_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_AND))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "and\\t%8,%0,%z5\\n\\t"
+ "and\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+ [(set (attr "length") (const_int 32))])
+
+(define_insn "atomic_cas_value_xor_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_XOR))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "xor\\t%8,%0,%z5\\n\\t"
+ "and\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+
+ [(set (attr "length") (const_int 32))])
+
+(define_insn "atomic_cas_value_or_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_OR))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "or\\t%8,%0,%z5\\n\\t"
+ "and\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+
+ [(set (attr "length") (const_int 32))])
+
+(define_insn "atomic_cas_value_nand_7_<mode>"
+ [(set (match_operand:GPR 0 "register_operand" "=&r") ;; res
+ (match_operand:GPR 1 "memory_operand" "+ZC"))
+ (set (match_dup 1)
+ (unspec_volatile:GPR [(match_operand:GPR 2 "reg_or_0_operand" "rJ") ;; mask
+ (match_operand:GPR 3 "reg_or_0_operand" "rJ") ;; inverted_mask
+ (match_operand:GPR 4 "reg_or_0_operand" "rJ") ;; old val
+ (match_operand:GPR 5 "reg_or_0_operand" "rJ") ;; new val
+ (match_operand:SI 6 "const_int_operand")] ;; model
+ UNSPEC_COMPARE_AND_SWAP_NAND))
+ (clobber (match_scratch:GPR 7 "=&r"))
+ (clobber (match_scratch:GPR 8 "=&r"))]
+ ""
+{
+ return "%G6\\n\\t"
+ "1:\\n\\t"
+ "ll.<amo>\\t%0,%1\\n\\t"
+ "and\\t%7,%0,%3\\n\\t"
+ "and\\t%8,%0,%z5\\n\\t"
+ "xor\\t%8,%8,%z2\\n\\t"
+ "or%i8\\t%7,%7,%8\\n\\t"
+ "sc.<amo>\\t%7,%1\\n\\t"
+ "beq\\t$zero,%7,1b";
+}
+ [(set (attr "length") (const_int 32))])
+
+(define_expand "atomic_exchange<mode>"
+ [(set (match_operand:SHORT 0 "register_operand")
+ (unspec_volatile:SHORT
+ [(match_operand:SHORT 1 "memory_operand")
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_EXCHANGE))
+ (set (match_dup 1)
+ (match_operand:SHORT 2 "register_operand"))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_cmp_and_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_add<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(plus:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_add_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_sub<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(minus:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_sub_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_and<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(and:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_and_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_xor<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(xor:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_xor_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_or<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(ior:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ"))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_or_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
+
+(define_expand "atomic_fetch_nand<mode>"
+ [(set (match_operand:SHORT 0 "register_operand" "=&r")
+ (match_operand:SHORT 1 "memory_operand" "+ZB"))
+ (set (match_dup 1)
+ (unspec_volatile:SHORT
+ [(not:SHORT (and:SHORT (match_dup 1)
+ (match_operand:SHORT 2 "reg_or_0_operand" "rJ")))
+ (match_operand:SI 3 "const_int_operand")] ;; model
+ UNSPEC_SYNC_OLD_OP))]
+ ""
+{
+ union loongarch_gen_fn_ptrs generator;
+ generator.fn_7 = gen_atomic_cas_value_nand_7_si;
+ loongarch_expand_atomic_qihi (generator, operands[0], operands[1],
+ operands[1], operands[2], operands[3]);
+ DONE;
+})
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 05/12] LoongArch Port: Machine description C files and .h files.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (3 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 04/12] LoongArch Port: Machine Decsription files chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 06/12] LoongArch Port: Builtin functions chenglulu
` (7 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/
* config/host-linux.c: Add LoongArch support.
* config/loongarch/loongarch-protos.h: New file.
* config/loongarch/loongarch-tune.h: Likewise.
* config/loongarch/loongarch.c: Likewise.
* config/loongarch/loongarch.h: Likewise.
---
gcc/config/host-linux.c | 2 +
gcc/config/loongarch/loongarch-protos.h | 242 +
gcc/config/loongarch/loongarch-tune.h | 72 +
gcc/config/loongarch/loongarch.c | 6461 +++++++++++++++++++++++
gcc/config/loongarch/loongarch.h | 1291 +++++
5 files changed, 8068 insertions(+)
create mode 100644 gcc/config/loongarch/loongarch-protos.h
create mode 100644 gcc/config/loongarch/loongarch-tune.h
create mode 100644 gcc/config/loongarch/loongarch.c
create mode 100644 gcc/config/loongarch/loongarch.h
diff --git a/gcc/config/host-linux.c b/gcc/config/host-linux.c
index 0481ecff3f2..cbbc8d13533 100644
--- a/gcc/config/host-linux.c
+++ b/gcc/config/host-linux.c
@@ -98,6 +98,8 @@
# define TRY_EMPTY_VM_SPACE 0x60000000
#elif defined(__riscv) && defined (__LP64__)
# define TRY_EMPTY_VM_SPACE 0x1000000000
+#elif defined(__loongarch__) && defined(__LP64__)
+# define TRY_EMPTY_VM_SPACE 0x8000000000
#else
# define TRY_EMPTY_VM_SPACE 0
#endif
diff --git a/gcc/config/loongarch/loongarch-protos.h b/gcc/config/loongarch/loongarch-protos.h
new file mode 100644
index 00000000000..58b5d72fc85
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-protos.h
@@ -0,0 +1,242 @@
+/* Prototypes of target machine for GNU compiler. LoongArch version.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS target for GNU compiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef GCC_LOONGARCH_PROTOS_H
+#define GCC_LOONGARCH_PROTOS_H
+
+/* Describes how a symbol is used.
+
+ SYMBOL_CONTEXT_CALL
+ The symbol is used as the target of a call instruction.
+
+ SYMBOL_CONTEXT_LEA
+ The symbol is used in a load-address operation.
+
+ SYMBOL_CONTEXT_MEM
+ The symbol is used as the address in a MEM. */
+enum loongarch_symbol_context {
+ SYMBOL_CONTEXT_CALL,
+ SYMBOL_CONTEXT_LEA,
+ SYMBOL_CONTEXT_MEM
+};
+
+/* Classifies a SYMBOL_REF, LABEL_REF or UNSPEC address.
+
+ SYMBOL_GOT_DISP
+ The symbol's value will be loaded directly from the GOT.
+
+ SYMBOL_TLS
+ A thread-local symbol.
+
+ SYMBOL_TLSGD
+ SYMBOL_TLSLDM
+ UNSPEC wrappers around SYMBOL_TLS, corresponding to the
+ thread-local storage relocation operators.
+ */
+enum loongarch_symbol_type {
+ SYMBOL_GOT_DISP,
+ SYMBOL_TLS,
+ SYMBOL_TLSGD,
+ SYMBOL_TLSLDM,
+};
+#define NUM_SYMBOL_TYPES (SYMBOL_TLSLDM + 1)
+
+/* Classifies a type of call.
+
+ LARCH_CALL_NORMAL
+ A normal call or call_value pattern.
+
+ LARCH_CALL_SIBCALL
+ A sibcall or sibcall_value pattern.
+
+ LARCH_CALL_EPILOGUE
+ A call inserted in the epilogue. */
+enum loongarch_call_type {
+ LARCH_CALL_NORMAL,
+ LARCH_CALL_SIBCALL,
+ LARCH_CALL_EPILOGUE
+};
+
+/* Controls the conditions under which certain instructions are split.
+
+ SPLIT_IF_NECESSARY
+ Only perform splits that are necessary for correctness
+ (because no unsplit version exists).
+
+ SPLIT_FOR_SPEED
+ Perform splits that are necessary for correctness or
+ beneficial for code speed.
+
+ SPLIT_FOR_SIZE
+ Perform splits that are necessary for correctness or
+ beneficial for code size. */
+enum loongarch_split_type {
+ SPLIT_IF_NECESSARY,
+ SPLIT_FOR_SPEED,
+ SPLIT_FOR_SIZE
+};
+
+extern const char *const loongarch_fp_conditions[16];
+
+/* Routines implemented in loongarch.c. */
+extern rtx loongarch_emit_move (rtx, rtx);
+extern HOST_WIDE_INT loongarch_initial_elimination_offset (int, int);
+extern void loongarch_expand_prologue (void);
+extern void loongarch_expand_epilogue (bool);
+extern bool loongarch_can_use_return_insn (void);
+\f
+extern bool loongarch_symbolic_constant_p (rtx, enum loongarch_symbol_context,
+ enum loongarch_symbol_type *);
+extern int loongarch_regno_mode_ok_for_base_p (int, machine_mode, bool);
+extern bool loongarch_stack_address_p (rtx, machine_mode);
+extern int loongarch_address_insns (rtx, machine_mode, bool);
+extern int loongarch_const_insns (rtx);
+extern int loongarch_split_const_insns (rtx);
+extern int loongarch_split_128bit_const_insns (rtx);
+extern int loongarch_load_store_insns (rtx, rtx_insn *);
+extern int loongarch_idiv_insns (machine_mode);
+#ifdef RTX_CODE
+extern void loongarch_emit_binary (enum rtx_code, rtx, rtx, rtx);
+#endif
+extern bool loongarch_split_symbol (rtx, rtx, machine_mode, rtx *);
+extern rtx loongarch_unspec_address (rtx, enum loongarch_symbol_type);
+extern rtx loongarch_strip_unspec_address (rtx);
+extern void loongarch_move_integer (rtx, rtx, unsigned HOST_WIDE_INT);
+extern bool loongarch_legitimize_move (machine_mode, rtx, rtx);
+extern rtx loongarch_legitimize_call_address (rtx);
+
+extern rtx loongarch_subword (rtx, bool);
+extern bool loongarch_split_move_p (rtx, rtx, enum loongarch_split_type);
+extern void loongarch_split_move (rtx, rtx, enum loongarch_split_type, rtx);
+extern bool loongarch_split_move_insn_p (rtx, rtx, rtx);
+extern void loongarch_split_move_insn (rtx, rtx, rtx);
+extern const char *loongarch_output_move (rtx, rtx);
+extern bool loongarch_cfun_has_cprestore_slot_p (void);
+#ifdef RTX_CODE
+extern void loongarch_expand_scc (rtx *);
+extern void loongarch_expand_conditional_branch (rtx *);
+extern void loongarch_expand_conditional_move (rtx *);
+extern void loongarch_expand_conditional_trap (rtx);
+#endif
+extern void loongarch_set_return_address (rtx, rtx);
+extern bool loongarch_move_by_pieces_p (unsigned HOST_WIDE_INT, unsigned int);
+extern bool loongarch_store_by_pieces_p (unsigned HOST_WIDE_INT, unsigned int);
+extern bool loongarch_expand_block_move (rtx, rtx, rtx);
+
+extern bool loongarch_expand_ext_as_unaligned_load (rtx, rtx, HOST_WIDE_INT,
+ HOST_WIDE_INT, bool);
+extern bool loongarch_expand_ins_as_unaligned_store (rtx, rtx, HOST_WIDE_INT,
+ HOST_WIDE_INT);
+extern HOST_WIDE_INT loongarch_debugger_offset (rtx, HOST_WIDE_INT);
+
+extern void loongarch_output_external (FILE *, tree, const char *);
+extern void loongarch_output_ascii (FILE *, const char *, size_t);
+extern void loongarch_output_aligned_decl_common (FILE *, tree, const char *,
+ unsigned HOST_WIDE_INT,
+ unsigned int);
+extern void loongarch_declare_common_object (FILE *, const char *,
+ const char *,
+ unsigned HOST_WIDE_INT,
+ unsigned int, bool);
+extern void loongarch_declare_object (FILE *, const char *, const char *,
+ const char *, ...) ATTRIBUTE_PRINTF_4;
+extern void loongarch_declare_object_name (FILE *, const char *, tree);
+extern void loongarch_finish_declare_object (FILE *, tree, int, int);
+extern void loongarch_set_text_contents_type (FILE *, const char *,
+ unsigned long, bool);
+
+extern bool loongarch_small_data_pattern_p (rtx);
+extern rtx loongarch_rewrite_small_data (rtx);
+extern rtx loongarch_return_addr (int, rtx);
+
+extern enum reg_class loongarch_secondary_reload_class (enum reg_class,
+ machine_mode,
+ rtx, bool);
+extern int loongarch_class_max_nregs (enum reg_class, machine_mode);
+
+extern machine_mode loongarch_hard_regno_caller_save_mode (unsigned int,
+ unsigned int,
+ machine_mode);
+extern int loongarch_adjust_insn_length (rtx_insn *, int);
+extern const char *loongarch_output_conditional_branch (rtx_insn *, rtx *,
+ const char *,
+ const char *);
+extern const char *loongarch_output_order_conditional_branch (rtx_insn *,
+ rtx *,
+ bool);
+extern const char *loongarch_output_equal_conditional_branch (rtx_insn *,
+ rtx *,
+ bool);
+extern const char *loongarch_output_division (const char *, rtx *);
+extern const char *loongarch_output_probe_stack_range (rtx, rtx, rtx);
+extern bool loongarch_hard_regno_rename_ok (unsigned int, unsigned int);
+extern int loongarch_dspalu_bypass_p (rtx, rtx);
+extern rtx loongarch_prefetch_cookie (rtx, rtx);
+
+extern bool loongarch_global_symbol_p (const_rtx);
+extern bool loongarch_global_symbol_noweak_p (const_rtx);
+extern bool loongarch_weak_symbol_p (const_rtx);
+extern bool loongarch_symbol_binds_local_p (const_rtx);
+
+extern const char *current_section_name (void);
+extern unsigned int current_section_flags (void);
+extern bool loongarch_use_ins_ext_p (rtx, HOST_WIDE_INT, HOST_WIDE_INT);
+
+union loongarch_gen_fn_ptrs
+{
+ rtx (*fn_8) (rtx, rtx, rtx, rtx, rtx, rtx, rtx, rtx);
+ rtx (*fn_7) (rtx, rtx, rtx, rtx, rtx, rtx, rtx);
+ rtx (*fn_6) (rtx, rtx, rtx, rtx, rtx, rtx);
+ rtx (*fn_5) (rtx, rtx, rtx, rtx, rtx);
+ rtx (*fn_4) (rtx, rtx, rtx, rtx);
+};
+
+extern void loongarch_expand_atomic_qihi (union loongarch_gen_fn_ptrs,
+ rtx, rtx, rtx, rtx, rtx);
+
+extern bool loongarch_signed_immediate_p (unsigned HOST_WIDE_INT, int, int);
+extern bool loongarch_unsigned_immediate_p (unsigned HOST_WIDE_INT, int, int);
+extern bool loongarch_12bit_offset_address_p (rtx, machine_mode);
+extern bool loongarch_14bit_shifted_offset_address_p (rtx, machine_mode);
+extern rtx loongarch_expand_thread_pointer (rtx);
+
+extern bool loongarch_eh_uses (unsigned int);
+extern bool loongarch_epilogue_uses (unsigned int);
+extern bool loongarch_load_store_bonding_p (rtx *, machine_mode, bool);
+extern bool loongarch_split_symbol_type (enum loongarch_symbol_type);
+
+typedef rtx (*mulsidi3_gen_fn) (rtx, rtx, rtx);
+
+extern void loongarch_register_frame_header_opt (void);
+
+extern void loongarch_declare_function_name (FILE *, const char *, tree);
+
+/* Routines implemented in loongarch-c.c. */
+void loongarch_cpu_cpp_builtins (cpp_reader *);
+
+extern void loongarch_init_builtins (void);
+extern void loongarch_atomic_assign_expand_fenv (tree *, tree *, tree *);
+extern tree loongarch_builtin_decl (unsigned int, bool);
+extern rtx loongarch_expand_builtin (tree, rtx, rtx subtarget ATTRIBUTE_UNUSED,
+ machine_mode, int);
+extern tree loongarch_build_builtin_va_list (void);
+
+#endif /* ! GCC_LOONGARCH_PROTOS_H */
diff --git a/gcc/config/loongarch/loongarch-tune.h b/gcc/config/loongarch/loongarch-tune.h
new file mode 100644
index 00000000000..1ff99e9c28b
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-tune.h
@@ -0,0 +1,72 @@
+/* Definitions for microarchitecture-related data structures.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef LOONGARCH_TUNE_H
+#define LOONGARCH_TUNE_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* RTX costs of various operations on the different architectures. */
+struct loongarch_rtx_cost_data
+{
+ unsigned short fp_add;
+ unsigned short fp_mult_sf;
+ unsigned short fp_mult_df;
+ unsigned short fp_div_sf;
+ unsigned short fp_div_df;
+ unsigned short int_mult_si;
+ unsigned short int_mult_di;
+ unsigned short int_div_si;
+ unsigned short int_div_di;
+ unsigned short branch_cost;
+ unsigned short memory_latency;
+};
+
+/* Default RTX cost initializer. */
+#define COSTS_N_INSNS(N) ((N) * 4)
+#define DEFAULT_COSTS \
+ .fp_add = COSTS_N_INSNS (1), \
+ .fp_mult_sf = COSTS_N_INSNS (2), \
+ .fp_mult_df = COSTS_N_INSNS (2), \
+ .fp_div_sf = COSTS_N_INSNS (4), \
+ .fp_div_df = COSTS_N_INSNS (4), \
+ .int_mult_si = COSTS_N_INSNS (1), \
+ .int_mult_di = COSTS_N_INSNS (1), \
+ .int_div_si = COSTS_N_INSNS (1), \
+ .int_div_di = COSTS_N_INSNS (1), \
+ .branch_cost = 2, \
+ .memory_latency = 4
+
+/* Costs to use when optimizing for size. */
+extern const struct loongarch_rtx_cost_data loongarch_rtx_cost_optimize_size;
+
+/* Cache size record of known processor models. */
+struct loongarch_cache {
+ int l1d_line_size; /* bytes */
+ int l1d_size; /* KiB */
+ int l2d_size; /* kiB */
+};
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* LOONGARCH_TUNE_H */
diff --git a/gcc/config/loongarch/loongarch.c b/gcc/config/loongarch/loongarch.c
new file mode 100644
index 00000000000..a3501dd505c
--- /dev/null
+++ b/gcc/config/loongarch/loongarch.c
@@ -0,0 +1,6461 @@
+/* Subroutines used for LoongArch code generation.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS and RISC-V target for GNU compiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "backend.h"
+#include "target.h"
+#include "rtl.h"
+#include "tree.h"
+#include "memmodel.h"
+#include "gimple.h"
+#include "cfghooks.h"
+#include "df.h"
+#include "tm_p.h"
+#include "stringpool.h"
+#include "attribs.h"
+#include "optabs.h"
+#include "regs.h"
+#include "emit-rtl.h"
+#include "recog.h"
+#include "cgraph.h"
+#include "diagnostic.h"
+#include "insn-attr.h"
+#include "output.h"
+#include "alias.h"
+#include "fold-const.h"
+#include "varasm.h"
+#include "stor-layout.h"
+#include "calls.h"
+#include "explow.h"
+#include "expr.h"
+#include "libfuncs.h"
+#include "reload.h"
+#include "common/common-target.h"
+#include "langhooks.h"
+#include "cfgrtl.h"
+#include "cfganal.h"
+#include "sched-int.h"
+#include "gimplify.h"
+#include "target-globals.h"
+#include "tree-pass.h"
+#include "context.h"
+#include "builtins.h"
+#include "rtl-iter.h"
+
+/* This file should be included last. */
+#include "target-def.h"
+
+/* True if X is an UNSPEC wrapper around a SYMBOL_REF or LABEL_REF. */
+#define UNSPEC_ADDRESS_P(X) \
+ (GET_CODE (X) == UNSPEC \
+ && XINT (X, 1) >= UNSPEC_ADDRESS_FIRST \
+ && XINT (X, 1) < UNSPEC_ADDRESS_FIRST + NUM_SYMBOL_TYPES)
+
+/* Extract the symbol or label from UNSPEC wrapper X. */
+#define UNSPEC_ADDRESS(X) XVECEXP (X, 0, 0)
+
+/* Extract the symbol type from UNSPEC wrapper X. */
+#define UNSPEC_ADDRESS_TYPE(X) \
+ ((enum loongarch_symbol_type) (XINT (X, 1) - UNSPEC_ADDRESS_FIRST))
+
+/* True if INSN is a loongarch.md pattern or asm statement. */
+/* ??? This test exists through the compiler, perhaps it should be
+ moved to rtl.h. */
+#define USEFUL_INSN_P(INSN) \
+ (NONDEBUG_INSN_P (INSN) \
+ && GET_CODE (PATTERN (INSN)) != USE \
+ && GET_CODE (PATTERN (INSN)) != CLOBBER)
+
+/* True if bit BIT is set in VALUE. */
+#define BITSET_P(VALUE, BIT) (((VALUE) & (1 << (BIT))) != 0)
+
+/* Classifies an address.
+
+ ADDRESS_REG
+ A natural register + offset address. The register satisfies
+ loongarch_valid_base_register_p and the offset is a const_arith_operand.
+
+ ADDRESS_CONST_INT
+ A signed 16-bit constant address.
+
+ ADDRESS_SYMBOLIC:
+ A constant symbolic address. */
+enum loongarch_address_type
+{
+ ADDRESS_REG,
+ ADDRESS_CONST_INT,
+ ADDRESS_SYMBOLIC
+};
+
+
+/* Information about an address described by loongarch_address_type.
+
+ ADDRESS_CONST_INT
+ No fields are used.
+
+ ADDRESS_REG
+ REG is the base register and OFFSET is the constant offset.
+
+ ADDRESS_SYMBOLIC
+ SYMBOL_TYPE is the type of symbol that the address references. */
+struct loongarch_address_info
+{
+ enum loongarch_address_type type;
+ rtx reg;
+ rtx offset;
+ enum loongarch_symbol_type symbol_type;
+};
+
+/* Method of loading instant numbers:
+
+ METHOD_NORMAL:
+ Load 0-31 bit of the immediate number.
+
+ METHOD_LU32I:
+ Load 32-51 bit of the immediate number.
+
+ METHOD_LU52I:
+ Load 52-63 bit of the immediate number.
+
+ METHOD_INSV:
+ immediate like 0xfff00000fffffxxx
+ */
+enum loongarch_load_imm_method
+{
+ METHOD_NORMAL,
+ METHOD_LU32I,
+ METHOD_LU52I,
+ METHOD_INSV
+};
+
+/* One stage in a constant building sequence. These sequences have
+ the form:
+
+ A = VALUE[0]
+ A = A CODE[1] VALUE[1]
+ A = A CODE[2] VALUE[2]
+ ...
+
+ where A is an accumulator, each CODE[i] is a binary rtl operation
+ and each VALUE[i] is a constant integer. CODE[0] is undefined. */
+struct loongarch_integer_op
+{
+ enum rtx_code code;
+ unsigned HOST_WIDE_INT value;
+ enum loongarch_load_imm_method method;
+};
+
+/* The largest number of operations needed to load an integer constant.
+ The worst accepted case for 64-bit constants is LU12I.W,LU32I.D,LU52I.D,ORI
+ or LU12I.W,LU32I.D,LU52I.D,ADDI.D DECL_ASSEMBLER_NAME. */
+#define LARCH_MAX_INTEGER_OPS 4
+
+/* Arrays that map GCC register numbers to debugger register numbers. */
+int loongarch_dwarf_regno[FIRST_PSEUDO_REGISTER];
+
+/* Index [M][R] is true if register R is allowed to hold a value of mode M. */
+static bool loongarch_hard_regno_mode_ok_p[MAX_MACHINE_MODE]
+ [FIRST_PSEUDO_REGISTER];
+
+/* Index C is true if character C is a valid PRINT_OPERAND punctation
+ character. */
+static bool loongarch_print_operand_punct[256];
+
+/* Cached value of can_issue_more. This is cached in loongarch_variable_issue
+ hook and returned from loongarch_sched_reorder2. */
+static int cached_can_issue_more;
+
+/* Index R is the smallest register class that contains register R. */
+const enum reg_class loongarch_regno_to_class[FIRST_PSEUDO_REGISTER] = {
+ GR_REGS, GR_REGS, GR_REGS, GR_REGS,
+ JIRL_REGS, JIRL_REGS, JIRL_REGS, JIRL_REGS,
+ JIRL_REGS, JIRL_REGS, JIRL_REGS, JIRL_REGS,
+ SIBCALL_REGS, SIBCALL_REGS, SIBCALL_REGS, SIBCALL_REGS,
+ SIBCALL_REGS, SIBCALL_REGS, SIBCALL_REGS, SIBCALL_REGS,
+ SIBCALL_REGS, GR_REGS, GR_REGS, JIRL_REGS,
+ JIRL_REGS, JIRL_REGS, JIRL_REGS, JIRL_REGS,
+ JIRL_REGS, JIRL_REGS, JIRL_REGS, JIRL_REGS,
+
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FP_REGS, FP_REGS, FP_REGS, FP_REGS,
+ FCC_REGS, FCC_REGS, FCC_REGS, FCC_REGS,
+ FCC_REGS, FCC_REGS, FCC_REGS, FCC_REGS,
+ FRAME_REGS, FRAME_REGS
+};
+
+/* The value of TARGET_ATTRIBUTE_TABLE. */
+static const struct attribute_spec loongarch_attribute_table[] = {
+ /* { name, min_len, max_len, decl_req, type_req, fn_type_req,
+ affects_type_identity, handler, exclude } */
+ { NULL, 0, 0, false, false, false, false, NULL, NULL }
+};
+
+/* Which cost information to use. */
+static const struct loongarch_rtx_cost_data *loongarch_cost;
+
+/* Information about a single argument. */
+struct loongarch_arg_info
+{
+ /* True if the argument is at least partially passed on the stack. */
+ bool stack_p;
+
+ /* The number of integer registers allocated to this argument. */
+ unsigned int num_gprs;
+
+ /* The offset of the first register used, provided num_gprs is nonzero.
+ If passed entirely on the stack, the value is MAX_ARGS_IN_REGISTERS. */
+ unsigned int gpr_offset;
+
+ /* The number of floating-point registers allocated to this argument. */
+ unsigned int num_fprs;
+
+ /* The offset of the first register used, provided num_fprs is nonzero. */
+ unsigned int fpr_offset;
+};
+
+/* Implement TARGET_FUNCTION_ARG_BOUNDARY. Every parameter gets at
+ least PARM_BOUNDARY bits of alignment, but will be given anything up
+ to PREFERRED_STACK_BOUNDARY bits if the type requires it. */
+
+static unsigned int
+loongarch_function_arg_boundary (machine_mode mode, const_tree type)
+{
+ unsigned int alignment;
+
+ /* Use natural alignment if the type is not aggregate data. */
+ if (type && !AGGREGATE_TYPE_P (type))
+ alignment = TYPE_ALIGN (TYPE_MAIN_VARIANT (type));
+ else
+ alignment = type ? TYPE_ALIGN (type) : GET_MODE_ALIGNMENT (mode);
+
+ return MIN (PREFERRED_STACK_BOUNDARY, MAX (PARM_BOUNDARY, alignment));
+}
+
+/* If MODE represents an argument that can be passed or returned in
+ floating-point registers, return the number of registers, else 0. */
+
+static unsigned
+loongarch_pass_mode_in_fpr_p (machine_mode mode)
+{
+ if (GET_MODE_UNIT_SIZE (mode) <= UNITS_PER_FP_ARG)
+ {
+ if (GET_MODE_CLASS (mode) == MODE_FLOAT)
+ return 1;
+
+ if (GET_MODE_CLASS (mode) == MODE_COMPLEX_FLOAT)
+ return 2;
+ }
+
+ return 0;
+}
+
+typedef struct
+{
+ const_tree type;
+ HOST_WIDE_INT offset;
+} loongarch_aggregate_field;
+
+/* Identify subfields of aggregates that are candidates for passing in
+ floating-point registers. */
+
+static int
+loongarch_flatten_aggregate_field (const_tree type,
+ loongarch_aggregate_field fields[2], int n,
+ HOST_WIDE_INT offset)
+{
+ switch (TREE_CODE (type))
+ {
+ case RECORD_TYPE:
+ /* Can't handle incomplete types nor sizes that are not fixed. */
+ if (!COMPLETE_TYPE_P (type)
+ || TREE_CODE (TYPE_SIZE (type)) != INTEGER_CST
+ || !tree_fits_uhwi_p (TYPE_SIZE (type)))
+ return -1;
+
+ for (tree f = TYPE_FIELDS (type); f; f = DECL_CHAIN (f))
+ if (TREE_CODE (f) == FIELD_DECL)
+ {
+ if (!TYPE_P (TREE_TYPE (f)))
+ return -1;
+
+ HOST_WIDE_INT pos = offset + int_byte_position (f);
+ n = loongarch_flatten_aggregate_field (TREE_TYPE (f), fields, n,
+ pos);
+ if (n < 0)
+ return -1;
+ }
+ return n;
+
+ case ARRAY_TYPE:
+ {
+ HOST_WIDE_INT n_elts;
+ loongarch_aggregate_field subfields[2];
+ tree index = TYPE_DOMAIN (type);
+ tree elt_size = TYPE_SIZE_UNIT (TREE_TYPE (type));
+ int n_subfields = loongarch_flatten_aggregate_field (TREE_TYPE (type),
+ subfields, 0,
+ offset);
+
+ /* Can't handle incomplete types nor sizes that are not fixed. */
+ if (n_subfields <= 0
+ || !COMPLETE_TYPE_P (type)
+ || TREE_CODE (TYPE_SIZE (type)) != INTEGER_CST
+ || !index
+ || !TYPE_MAX_VALUE (index)
+ || !tree_fits_uhwi_p (TYPE_MAX_VALUE (index))
+ || !TYPE_MIN_VALUE (index)
+ || !tree_fits_uhwi_p (TYPE_MIN_VALUE (index))
+ || !tree_fits_uhwi_p (elt_size))
+ return -1;
+
+ n_elts = 1 + tree_to_uhwi (TYPE_MAX_VALUE (index))
+ - tree_to_uhwi (TYPE_MIN_VALUE (index));
+ gcc_assert (n_elts >= 0);
+
+ for (HOST_WIDE_INT i = 0; i < n_elts; i++)
+ for (int j = 0; j < n_subfields; j++)
+ {
+ if (n >= 2)
+ return -1;
+
+ fields[n] = subfields[j];
+ fields[n++].offset += i * tree_to_uhwi (elt_size);
+ }
+
+ return n;
+ }
+
+ case COMPLEX_TYPE:
+ {
+ /* Complex type need consume 2 field, so n must be 0. */
+ if (n != 0)
+ return -1;
+
+ HOST_WIDE_INT elt_size = GET_MODE_SIZE (TYPE_MODE (TREE_TYPE (type)));
+
+ if (elt_size <= UNITS_PER_FP_ARG)
+ {
+ fields[0].type = TREE_TYPE (type);
+ fields[0].offset = offset;
+ fields[1].type = TREE_TYPE (type);
+ fields[1].offset = offset + elt_size;
+
+ return 2;
+ }
+
+ return -1;
+ }
+
+ default:
+ if (n < 2
+ && ((SCALAR_FLOAT_TYPE_P (type)
+ && GET_MODE_SIZE (TYPE_MODE (type)) <= UNITS_PER_FP_ARG)
+ || (INTEGRAL_TYPE_P (type)
+ && GET_MODE_SIZE (TYPE_MODE (type)) <= UNITS_PER_WORD)))
+ {
+ fields[n].type = type;
+ fields[n].offset = offset;
+ return n + 1;
+ }
+ else
+ return -1;
+ }
+}
+
+/* Identify candidate aggregates for passing in floating-point registers.
+ Candidates have at most two fields after flattening. */
+
+static int
+loongarch_flatten_aggregate_argument (const_tree type,
+ loongarch_aggregate_field fields[2])
+{
+ if (!type || TREE_CODE (type) != RECORD_TYPE)
+ return -1;
+
+ return loongarch_flatten_aggregate_field (type, fields, 0, 0);
+}
+
+/* See whether TYPE is a record whose fields should be returned in one or
+ two floating-point registers. If so, populate FIELDS accordingly. */
+
+static unsigned
+loongarch_pass_aggregate_in_fpr_pair_p (const_tree type,
+ loongarch_aggregate_field fields[2])
+{
+ int n = loongarch_flatten_aggregate_argument (type, fields);
+
+ for (int i = 0; i < n; i++)
+ if (!SCALAR_FLOAT_TYPE_P (fields[i].type))
+ return 0;
+
+ return n > 0 ? n : 0;
+}
+
+/* See whether TYPE is a record whose fields should be returned in one or
+ floating-point register and one integer register. If so, populate
+ FIELDS accordingly. */
+
+static bool
+loongarch_pass_aggregate_in_fpr_and_gpr_p (const_tree type,
+ loongarch_aggregate_field fields[2])
+{
+ unsigned num_int = 0, num_float = 0;
+ int n = loongarch_flatten_aggregate_argument (type, fields);
+
+ for (int i = 0; i < n; i++)
+ {
+ num_float += SCALAR_FLOAT_TYPE_P (fields[i].type);
+ num_int += INTEGRAL_TYPE_P (fields[i].type);
+ }
+
+ return num_int == 1 && num_float == 1;
+}
+
+/* Return the representation of an argument passed or returned in an FPR
+ when the value has mode VALUE_MODE and the type has TYPE_MODE. The
+ two modes may be different for structures like:
+
+ struct __attribute__((packed)) foo { float f; }
+
+ where the SFmode value "f" is passed in REGNO but the struct itself
+ has mode BLKmode. */
+
+static rtx
+loongarch_pass_fpr_single (machine_mode type_mode, unsigned regno,
+ machine_mode value_mode)
+{
+ rtx x = gen_rtx_REG (value_mode, regno);
+
+ if (type_mode != value_mode)
+ {
+ x = gen_rtx_EXPR_LIST (VOIDmode, x, const0_rtx);
+ x = gen_rtx_PARALLEL (type_mode, gen_rtvec (1, x));
+ }
+ return x;
+}
+
+/* Pass or return a composite value in the FPR pair REGNO and REGNO + 1.
+ MODE is the mode of the composite. MODE1 and OFFSET1 are the mode and
+ byte offset for the first value, likewise MODE2 and OFFSET2 for the
+ second value. */
+
+static rtx
+loongarch_pass_fpr_pair (machine_mode mode, unsigned regno1,
+ machine_mode mode1, HOST_WIDE_INT offset1,
+ unsigned regno2, machine_mode mode2,
+ HOST_WIDE_INT offset2)
+{
+ return gen_rtx_PARALLEL (
+ mode, gen_rtvec (2,
+ gen_rtx_EXPR_LIST (VOIDmode, gen_rtx_REG (mode1, regno1),
+ GEN_INT (offset1)),
+ gen_rtx_EXPR_LIST (VOIDmode, gen_rtx_REG (mode2, regno2),
+ GEN_INT (offset2))));
+}
+
+/* Fill INFO with information about a single argument, and return an
+ RTL pattern to pass or return the argument. CUM is the cumulative
+ state for earlier arguments. MODE is the mode of this argument and
+ TYPE is its type (if known). NAMED is true if this is a named
+ (fixed) argument rather than a variable one. RETURN_P is true if
+ returning the argument, or false if passing the argument. */
+
+static rtx
+loongarch_get_arg_info (struct loongarch_arg_info *info,
+ const CUMULATIVE_ARGS *cum, machine_mode mode,
+ const_tree type, bool named, bool return_p)
+{
+ unsigned num_bytes, num_words;
+ unsigned fpr_base = return_p ? FP_RETURN : FP_ARG_FIRST;
+ unsigned gpr_base = return_p ? GP_RETURN : GP_ARG_FIRST;
+ unsigned alignment = loongarch_function_arg_boundary (mode, type);
+
+ memset (info, 0, sizeof (*info));
+ info->gpr_offset = cum->num_gprs;
+ info->fpr_offset = cum->num_fprs;
+
+ if (named)
+ {
+ loongarch_aggregate_field fields[2];
+ unsigned fregno = fpr_base + info->fpr_offset;
+ unsigned gregno = gpr_base + info->gpr_offset;
+
+ /* Pass one- or two-element floating-point aggregates in FPRs. */
+ if ((info->num_fprs
+ = loongarch_pass_aggregate_in_fpr_pair_p (type, fields))
+ && info->fpr_offset + info->num_fprs <= MAX_ARGS_IN_REGISTERS)
+ switch (info->num_fprs)
+ {
+ case 1:
+ return loongarch_pass_fpr_single (mode, fregno,
+ TYPE_MODE (fields[0].type));
+
+ case 2:
+ return loongarch_pass_fpr_pair (mode, fregno,
+ TYPE_MODE (fields[0].type),
+ fields[0].offset,
+ fregno + 1,
+ TYPE_MODE (fields[1].type),
+ fields[1].offset);
+
+ default:
+ gcc_unreachable ();
+ }
+
+ /* Pass real and complex floating-point numbers in FPRs. */
+ if ((info->num_fprs = loongarch_pass_mode_in_fpr_p (mode))
+ && info->fpr_offset + info->num_fprs <= MAX_ARGS_IN_REGISTERS)
+ switch (GET_MODE_CLASS (mode))
+ {
+ case MODE_FLOAT:
+ return gen_rtx_REG (mode, fregno);
+
+ case MODE_COMPLEX_FLOAT:
+ return loongarch_pass_fpr_pair (mode, fregno,
+ GET_MODE_INNER (mode), 0,
+ fregno + 1, GET_MODE_INNER (mode),
+ GET_MODE_UNIT_SIZE (mode));
+
+ default:
+ gcc_unreachable ();
+ }
+
+ /* Pass structs with one float and one integer in an FPR and a GPR. */
+ if (loongarch_pass_aggregate_in_fpr_and_gpr_p (type, fields)
+ && info->gpr_offset < MAX_ARGS_IN_REGISTERS
+ && info->fpr_offset < MAX_ARGS_IN_REGISTERS)
+ {
+ info->num_gprs = 1;
+ info->num_fprs = 1;
+
+ if (!SCALAR_FLOAT_TYPE_P (fields[0].type))
+ std::swap (fregno, gregno);
+
+ return loongarch_pass_fpr_pair (mode, fregno,
+ TYPE_MODE (fields[0].type),
+ fields[0].offset, gregno,
+ TYPE_MODE (fields[1].type),
+ fields[1].offset);
+ }
+ }
+
+ /* Work out the size of the argument. */
+ num_bytes = type ? int_size_in_bytes (type) : GET_MODE_SIZE (mode);
+ num_words = (num_bytes + UNITS_PER_WORD - 1) / UNITS_PER_WORD;
+
+ /* Doubleword-aligned varargs start on an even register boundary. */
+ if (!named && num_bytes != 0 && alignment > BITS_PER_WORD)
+ info->gpr_offset += info->gpr_offset & 1;
+
+ /* Partition the argument between registers and stack. */
+ info->num_fprs = 0;
+ info->num_gprs = MIN (num_words, MAX_ARGS_IN_REGISTERS - info->gpr_offset);
+ info->stack_p = (num_words - info->num_gprs) != 0;
+
+ if (info->num_gprs || return_p)
+ return gen_rtx_REG (mode, gpr_base + info->gpr_offset);
+
+ return NULL_RTX;
+}
+
+/* Implement TARGET_FUNCTION_ARG. */
+
+static rtx
+loongarch_function_arg (cumulative_args_t cum_v, const function_arg_info &arg)
+{
+ CUMULATIVE_ARGS *cum = get_cumulative_args (cum_v);
+ struct loongarch_arg_info info;
+
+ if (arg.end_marker_p ())
+ return NULL;
+
+ return loongarch_get_arg_info (&info, cum, arg.mode, arg.type, arg.named,
+ false);
+}
+
+/* Implement TARGET_FUNCTION_ARG_ADVANCE. */
+
+static void
+loongarch_function_arg_advance (cumulative_args_t cum_v,
+ const function_arg_info &arg)
+{
+ CUMULATIVE_ARGS *cum = get_cumulative_args (cum_v);
+ struct loongarch_arg_info info;
+
+ loongarch_get_arg_info (&info, cum, arg.mode, arg.type, arg.named, false);
+
+ /* Advance the register count. This has the effect of setting
+ num_gprs to MAX_ARGS_IN_REGISTERS if a doubleword-aligned
+ argument required us to skip the final GPR and pass the whole
+ argument on the stack. */
+ cum->num_fprs = info.fpr_offset + info.num_fprs;
+ cum->num_gprs = info.gpr_offset + info.num_gprs;
+}
+
+/* Implement TARGET_ARG_PARTIAL_BYTES. */
+
+static int
+loongarch_arg_partial_bytes (cumulative_args_t cum,
+ const function_arg_info &generic_arg)
+{
+ struct loongarch_arg_info arg;
+
+ loongarch_get_arg_info (&arg, get_cumulative_args (cum), generic_arg.mode,
+ generic_arg.type, generic_arg.named, false);
+ return arg.stack_p ? arg.num_gprs * UNITS_PER_WORD : 0;
+}
+
+/* Implement FUNCTION_VALUE and LIBCALL_VALUE. For normal calls,
+ VALTYPE is the return type and MODE is VOIDmode. For libcalls,
+ VALTYPE is null and MODE is the mode of the return value. */
+
+static rtx
+loongarch_function_value_1 (const_tree type, const_tree func,
+ machine_mode mode)
+{
+ struct loongarch_arg_info info;
+ CUMULATIVE_ARGS args;
+
+ if (type)
+ {
+ int unsigned_p = TYPE_UNSIGNED (type);
+
+ mode = TYPE_MODE (type);
+
+ /* Since TARGET_PROMOTE_FUNCTION_MODE unconditionally promotes,
+ return values, promote the mode here too. */
+ mode = promote_function_mode (type, mode, &unsigned_p, func, 1);
+ }
+
+ memset (&args, 0, sizeof (args));
+ return loongarch_get_arg_info (&info, &args, mode, type, true, true);
+}
+
+
+/* Implement TARGET_FUNCTION_VALUE. */
+
+static rtx
+loongarch_function_value (const_tree valtype, const_tree fn_decl_or_type,
+ bool outgoing ATTRIBUTE_UNUSED)
+{
+ return loongarch_function_value_1 (valtype, fn_decl_or_type, VOIDmode);
+}
+
+/* Implement TARGET_LIBCALL_VALUE. */
+
+static rtx
+loongarch_libcall_value (machine_mode mode, const_rtx fun ATTRIBUTE_UNUSED)
+{
+ return loongarch_function_value_1 (NULL_TREE, NULL_TREE, mode);
+}
+
+
+/* Implement TARGET_PASS_BY_REFERENCE. */
+
+static bool
+loongarch_pass_by_reference (cumulative_args_t cum_v,
+ const function_arg_info &arg)
+{
+ HOST_WIDE_INT size = arg.type_size_in_bytes ();
+ struct loongarch_arg_info info;
+ CUMULATIVE_ARGS *cum = get_cumulative_args (cum_v);
+
+ /* ??? std_gimplify_va_arg_expr passes NULL for cum. Fortunately, we
+ never pass variadic arguments in floating-point registers, so we can
+ avoid the call to loongarch_get_arg_info in this case. */
+ if (cum != NULL)
+ {
+ /* Don't pass by reference if we can use a floating-point register. */
+ loongarch_get_arg_info (&info, cum, arg.mode, arg.type, arg.named,
+ false);
+ if (info.num_fprs)
+ return false;
+ }
+
+ /* Pass by reference if the data do not fit in two integer registers. */
+ return !IN_RANGE (size, 0, 2 * UNITS_PER_WORD);
+}
+
+/* Implement TARGET_RETURN_IN_MEMORY. */
+
+static bool
+loongarch_return_in_memory (const_tree type,
+ const_tree fndecl ATTRIBUTE_UNUSED)
+{
+ CUMULATIVE_ARGS args;
+ cumulative_args_t cum = pack_cumulative_args (&args);
+
+ /* The rules for returning in memory are the same as for passing the
+ first named argument by reference. */
+ memset (&args, 0, sizeof (args));
+ function_arg_info arg (const_cast<tree> (type), /*named=*/true);
+ return loongarch_pass_by_reference (cum, arg);
+}
+
+/* Implement TARGET_SETUP_INCOMING_VARARGS. */
+
+static void
+loongarch_setup_incoming_varargs (cumulative_args_t cum,
+ const function_arg_info &arg,
+ int *pretend_size ATTRIBUTE_UNUSED,
+ int no_rtl)
+{
+ CUMULATIVE_ARGS local_cum;
+ int gp_saved;
+
+ /* The caller has advanced CUM up to, but not beyond, the last named
+ argument. Advance a local copy of CUM past the last "real" named
+ argument, to find out how many registers are left over. */
+ local_cum = *get_cumulative_args (cum);
+ loongarch_function_arg_advance (pack_cumulative_args (&local_cum), arg);
+
+ /* Found out how many registers we need to save. */
+ gp_saved = MAX_ARGS_IN_REGISTERS - local_cum.num_gprs;
+
+ if (!no_rtl && gp_saved > 0)
+ {
+ rtx ptr = plus_constant (Pmode, virtual_incoming_args_rtx,
+ REG_PARM_STACK_SPACE (cfun->decl)
+ - gp_saved * UNITS_PER_WORD);
+ rtx mem = gen_frame_mem (BLKmode, ptr);
+ set_mem_alias_set (mem, get_varargs_alias_set ());
+
+ move_block_from_reg (local_cum.num_gprs + GP_ARG_FIRST, mem, gp_saved);
+ }
+ if (REG_PARM_STACK_SPACE (cfun->decl) == 0)
+ cfun->machine->varargs_size = gp_saved * UNITS_PER_WORD;
+}
+
+/* Make the last instruction frame-related and note that it performs
+ the operation described by FRAME_PATTERN. */
+
+static void
+loongarch_set_frame_expr (rtx frame_pattern)
+{
+ rtx insn;
+
+ insn = get_last_insn ();
+ RTX_FRAME_RELATED_P (insn) = 1;
+ REG_NOTES (insn) = alloc_EXPR_LIST (REG_FRAME_RELATED_EXPR, frame_pattern,
+ REG_NOTES (insn));
+}
+
+/* Return a frame-related rtx that stores REG at MEM.
+ REG must be a single register. */
+
+static rtx
+loongarch_frame_set (rtx mem, rtx reg)
+{
+ rtx set = gen_rtx_SET (mem, reg);
+ RTX_FRAME_RELATED_P (set) = 1;
+ return set;
+}
+
+/* Return true if the current function must save register REGNO. */
+
+static bool
+loongarch_save_reg_p (unsigned int regno)
+{
+ bool call_saved = !global_regs[regno] && !call_used_regs[regno];
+ bool might_clobber
+ = crtl->saves_all_registers || df_regs_ever_live_p (regno);
+
+ if (call_saved && might_clobber)
+ return true;
+
+ if (regno == HARD_FRAME_POINTER_REGNUM && frame_pointer_needed)
+ return true;
+
+ if (regno == RETURN_ADDR_REGNUM && crtl->calls_eh_return)
+ return true;
+
+ return false;
+}
+
+/* Determine which GPR save/restore routine to call. */
+
+static unsigned
+loongarch_save_libcall_count (unsigned mask)
+{
+ for (unsigned n = GP_REG_LAST; n > GP_REG_FIRST; n--)
+ if (BITSET_P (mask, n))
+ return CALLEE_SAVED_REG_NUMBER (n) + 1;
+ abort ();
+}
+
+/* Populate the current function's loongarch_frame_info structure.
+
+ LoongArch stack frames grown downward. High addresses are at the top.
+
+ +-------------------------------+
+ | |
+ | incoming stack arguments |
+ | |
+ +-------------------------------+ <-- incoming stack pointer
+ | |
+ | callee-allocated save area |
+ | for arguments that are |
+ | split between registers and |
+ | the stack |
+ | |
+ +-------------------------------+ <-- arg_pointer_rtx (virtual)
+ | |
+ | callee-allocated save area |
+ | for register varargs |
+ | |
+ +-------------------------------+ <-- hard_frame_pointer_rtx;
+ | | stack_pointer_rtx + gp_sp_offset
+ | GPR save area | + UNITS_PER_WORD
+ | |
+ +-------------------------------+ <-- stack_pointer_rtx + fp_sp_offset
+ | | + UNITS_PER_HWVALUE
+ | FPR save area |
+ | |
+ +-------------------------------+ <-- frame_pointer_rtx (virtual)
+ | |
+ | local variables |
+ | |
+ P +-------------------------------+
+ | |
+ | outgoing stack arguments |
+ | |
+ +-------------------------------+ <-- stack_pointer_rtx
+
+ Dynamic stack allocations such as alloca insert data at point P.
+ They decrease stack_pointer_rtx but leave frame_pointer_rtx and
+ hard_frame_pointer_rtx unchanged. */
+
+static void
+loongarch_compute_frame_info (void)
+{
+ struct loongarch_frame_info *frame;
+ HOST_WIDE_INT offset;
+ unsigned int regno, i, num_x_saved = 0, num_f_saved = 0;
+
+ frame = &cfun->machine->frame;
+ memset (frame, 0, sizeof (*frame));
+
+ /* Find out which GPRs we need to save. */
+ for (regno = GP_REG_FIRST; regno <= GP_REG_LAST; regno++)
+ if (loongarch_save_reg_p (regno))
+ frame->mask |= 1 << (regno - GP_REG_FIRST), num_x_saved++;
+
+ /* If this function calls eh_return, we must also save and restore the
+ EH data registers. */
+ if (crtl->calls_eh_return)
+ for (i = 0; (regno = EH_RETURN_DATA_REGNO (i)) != INVALID_REGNUM; i++)
+ frame->mask |= 1 << (regno - GP_REG_FIRST), num_x_saved++;
+
+ /* Find out which FPRs we need to save. This loop must iterate over
+ the same space as its companion in loongarch_for_each_saved_reg. */
+ if (TARGET_HARD_FLOAT)
+ for (regno = FP_REG_FIRST; regno <= FP_REG_LAST; regno++)
+ if (loongarch_save_reg_p (regno))
+ frame->fmask |= 1 << (regno - FP_REG_FIRST), num_f_saved++;
+
+ /* At the bottom of the frame are any outgoing stack arguments. */
+ offset = LARCH_STACK_ALIGN (crtl->outgoing_args_size);
+ /* Next are local stack variables. */
+ offset += LARCH_STACK_ALIGN (get_frame_size ());
+ /* The virtual frame pointer points above the local variables. */
+ frame->frame_pointer_offset = offset;
+ /* Next are the callee-saved FPRs. */
+ if (frame->fmask)
+ offset += LARCH_STACK_ALIGN (num_f_saved * UNITS_PER_FP_REG);
+ frame->fp_sp_offset = offset - UNITS_PER_FP_REG;
+ /* Next are the callee-saved GPRs. */
+ if (frame->mask)
+ {
+ unsigned x_save_size = LARCH_STACK_ALIGN (num_x_saved * UNITS_PER_WORD);
+ unsigned num_save_restore
+ = 1 + loongarch_save_libcall_count (frame->mask);
+
+ /* Only use save/restore routines if they don't alter the stack size. */
+ if (LARCH_STACK_ALIGN (num_save_restore * UNITS_PER_WORD) == x_save_size)
+ frame->save_libcall_adjustment = x_save_size;
+
+ offset += x_save_size;
+ }
+ frame->gp_sp_offset = offset - UNITS_PER_WORD;
+ /* The hard frame pointer points above the callee-saved GPRs. */
+ frame->hard_frame_pointer_offset = offset;
+ /* Above the hard frame pointer is the callee-allocated varags save area. */
+ offset += LARCH_STACK_ALIGN (cfun->machine->varargs_size);
+ /* Next is the callee-allocated area for pretend stack arguments. */
+ offset += LARCH_STACK_ALIGN (crtl->args.pretend_args_size);
+ /* Arg pointer must be below pretend args, but must be above alignment
+ padding. */
+ frame->arg_pointer_offset = offset - crtl->args.pretend_args_size;
+ frame->total_size = offset;
+ /* Next points the incoming stack pointer and any incoming arguments. */
+
+ /* Only use save/restore routines when the GPRs are atop the frame. */
+ if (frame->hard_frame_pointer_offset != frame->total_size)
+ frame->save_libcall_adjustment = 0;
+}
+
+/* Implement INITIAL_ELIMINATION_OFFSET. FROM is either the frame pointer
+ or argument pointer. TO is either the stack pointer or hard frame
+ pointer. */
+
+HOST_WIDE_INT
+loongarch_initial_elimination_offset (int from, int to)
+{
+ HOST_WIDE_INT src, dest;
+
+ loongarch_compute_frame_info ();
+
+ if (to == HARD_FRAME_POINTER_REGNUM)
+ dest = cfun->machine->frame.hard_frame_pointer_offset;
+ else if (to == STACK_POINTER_REGNUM)
+ dest = 0; /* The stack pointer is the base of all offsets, hence 0. */
+ else
+ gcc_unreachable ();
+
+ if (from == FRAME_POINTER_REGNUM)
+ src = cfun->machine->frame.frame_pointer_offset;
+ else if (from == ARG_POINTER_REGNUM)
+ src = cfun->machine->frame.arg_pointer_offset;
+ else
+ gcc_unreachable ();
+
+ return src - dest;
+}
+
+/* A function to save or store a register. The first argument is the
+ register and the second is the stack slot. */
+typedef void (*loongarch_save_restore_fn) (rtx, rtx);
+
+/* Use FN to save or restore register REGNO. MODE is the register's
+ mode and OFFSET is the offset of its save slot from the current
+ stack pointer. */
+
+static void
+loongarch_save_restore_reg (machine_mode mode, int regno, HOST_WIDE_INT offset,
+ loongarch_save_restore_fn fn)
+{
+ rtx mem;
+
+ mem = gen_frame_mem (mode, plus_constant (Pmode, stack_pointer_rtx, offset));
+ fn (gen_rtx_REG (mode, regno), mem);
+}
+
+/* Call FN for each register that is saved by the current function.
+ SP_OFFSET is the offset of the current stack pointer from the start
+ of the frame. */
+
+static void
+loongarch_for_each_saved_reg (HOST_WIDE_INT sp_offset,
+ loongarch_save_restore_fn fn)
+{
+ HOST_WIDE_INT offset;
+
+ /* Save the link register and s-registers. */
+ offset = cfun->machine->frame.gp_sp_offset - sp_offset;
+ for (int regno = GP_REG_FIRST; regno <= GP_REG_LAST; regno++)
+ if (BITSET_P (cfun->machine->frame.mask, regno - GP_REG_FIRST))
+ {
+ loongarch_save_restore_reg (word_mode, regno, offset, fn);
+ offset -= UNITS_PER_WORD;
+ }
+
+ /* This loop must iterate over the same space as its companion in
+ loongarch_compute_frame_info. */
+ offset = cfun->machine->frame.fp_sp_offset - sp_offset;
+ for (int regno = FP_REG_FIRST; regno <= FP_REG_LAST; regno++)
+ if (BITSET_P (cfun->machine->frame.fmask, regno - FP_REG_FIRST))
+ {
+ machine_mode mode = TARGET_DOUBLE_FLOAT ? DFmode : SFmode;
+
+ loongarch_save_restore_reg (mode, regno, offset, fn);
+ offset -= GET_MODE_SIZE (mode);
+ }
+}
+
+/* Emit a move from SRC to DEST. Assume that the move expanders can
+ handle all moves if !can_create_pseudo_p (). The distinction is
+ important because, unlike emit_move_insn, the move expanders know
+ how to force Pmode objects into the constant pool even when the
+ constant pool address is not itself legitimate. */
+
+rtx
+loongarch_emit_move (rtx dest, rtx src)
+{
+ return (can_create_pseudo_p () ? emit_move_insn (dest, src)
+ : emit_move_insn_1 (dest, src));
+}
+
+/* Save register REG to MEM. Make the instruction frame-related. */
+
+static void
+loongarch_save_reg (rtx reg, rtx mem)
+{
+ loongarch_emit_move (mem, reg);
+ loongarch_set_frame_expr (loongarch_frame_set (mem, reg));
+}
+
+/* Restore register REG from MEM. */
+
+static void
+loongarch_restore_reg (rtx reg, rtx mem)
+{
+ rtx insn = loongarch_emit_move (reg, mem);
+ rtx dwarf = NULL_RTX;
+ dwarf = alloc_reg_note (REG_CFA_RESTORE, reg, dwarf);
+ REG_NOTES (insn) = dwarf;
+
+ RTX_FRAME_RELATED_P (insn) = 1;
+}
+
+/* For stack frames that can't be allocated with a single ADDI instruction,
+ compute the best value to initially allocate. It must at a minimum
+ allocate enough space to spill the callee-saved registers. */
+
+static HOST_WIDE_INT
+loongarch_first_stack_step (struct loongarch_frame_info *frame)
+{
+ if (IMM12_OPERAND (frame->total_size))
+ return frame->total_size;
+
+ HOST_WIDE_INT min_first_step
+ = LARCH_STACK_ALIGN (frame->total_size - frame->fp_sp_offset);
+ HOST_WIDE_INT max_first_step = IMM_REACH / 2 - PREFERRED_STACK_BOUNDARY / 8;
+ HOST_WIDE_INT min_second_step = frame->total_size - max_first_step;
+ gcc_assert (min_first_step <= max_first_step);
+
+ /* As an optimization, use the least-significant bits of the total frame
+ size, so that the second adjustment step is just LU12I + ADD. */
+ if (!IMM12_OPERAND (min_second_step)
+ && frame->total_size % IMM_REACH < IMM_REACH / 2
+ && frame->total_size % IMM_REACH >= min_first_step)
+ return frame->total_size % IMM_REACH;
+
+ return max_first_step;
+}
+
+static void
+loongarch_emit_stack_tie (void)
+{
+ if (Pmode == SImode)
+ emit_insn (gen_stack_tiesi (stack_pointer_rtx, hard_frame_pointer_rtx));
+ else
+ emit_insn (gen_stack_tiedi (stack_pointer_rtx, hard_frame_pointer_rtx));
+}
+
+#define PROBE_INTERVAL (1 << STACK_CHECK_PROBE_INTERVAL_EXP)
+
+#if PROBE_INTERVAL > 16384
+#error Cannot use indexed addressing mode for stack probing
+#endif
+
+/* Emit code to probe a range of stack addresses from FIRST to FIRST+SIZE,
+ inclusive. These are offsets from the current stack pointer. */
+
+static void
+loongarch_emit_probe_stack_range (HOST_WIDE_INT first, HOST_WIDE_INT size)
+{
+ /* See if we have a constant small number of probes to generate. If so,
+ that's the easy case. */
+ if ((TARGET_64BIT && (first + size <= 32768))
+ || (!TARGET_64BIT && (first + size <= 2048)))
+ {
+ HOST_WIDE_INT i;
+
+ /* Probe at FIRST + N * PROBE_INTERVAL for values of N from 1 until
+ it exceeds SIZE. If only one probe is needed, this will not
+ generate any code. Then probe at FIRST + SIZE. */
+ for (i = PROBE_INTERVAL; i < size; i += PROBE_INTERVAL)
+ emit_stack_probe (plus_constant (Pmode, stack_pointer_rtx,
+ -(first + i)));
+
+ emit_stack_probe (plus_constant (Pmode, stack_pointer_rtx,
+ -(first + size)));
+ }
+
+ /* Otherwise, do the same as above, but in a loop. Note that we must be
+ extra careful with variables wrapping around because we might be at
+ the very top (or the very bottom) of the address space and we have
+ to be able to handle this case properly; in particular, we use an
+ equality test for the loop condition. */
+ else
+ {
+ HOST_WIDE_INT rounded_size;
+ rtx r13 = LARCH_PROLOGUE_TEMP (Pmode);
+ rtx r12 = LARCH_PROLOGUE_TEMP2 (Pmode);
+ rtx r14 = LARCH_PROLOGUE_TEMP3 (Pmode);
+
+ /* Sanity check for the addressing mode we're going to use. */
+ gcc_assert (first <= 16384);
+
+
+ /* Step 1: round SIZE to the previous multiple of the interval. */
+
+ rounded_size = ROUND_DOWN (size, PROBE_INTERVAL);
+
+ /* TEST_ADDR = SP + FIRST */
+ if (first != 0)
+ {
+ emit_move_insn (r14, GEN_INT (first));
+ emit_insn (gen_rtx_SET (r13, gen_rtx_MINUS (Pmode,
+ stack_pointer_rtx,
+ r14)));
+ }
+ else
+ emit_move_insn (r13, stack_pointer_rtx);
+
+ /* Step 2: compute initial and final value of the loop counter. */
+
+ emit_move_insn (r14, GEN_INT (PROBE_INTERVAL));
+ /* LAST_ADDR = SP + FIRST + ROUNDED_SIZE. */
+ if (rounded_size == 0)
+ emit_move_insn (r12, r13);
+ else
+ {
+ emit_move_insn (r12, GEN_INT (rounded_size));
+ emit_insn (gen_rtx_SET (r12, gen_rtx_MINUS (Pmode, r13, r12)));
+ /* Step 3: the loop
+
+ do
+ {
+ TEST_ADDR = TEST_ADDR + PROBE_INTERVAL
+ probe at TEST_ADDR
+ }
+ while (TEST_ADDR != LAST_ADDR)
+
+ probes at FIRST + N * PROBE_INTERVAL for values of N from 1
+ until it is equal to ROUNDED_SIZE. */
+
+ emit_insn (PMODE_INSN (gen_probe_stack_range, (r13, r13, r12, r14)));
+ }
+
+ /* Step 4: probe at FIRST + SIZE if we cannot assert at compile-time
+ that SIZE is equal to ROUNDED_SIZE. */
+
+ if (size != rounded_size)
+ {
+ if (TARGET_64BIT)
+ emit_stack_probe (plus_constant (Pmode, r12, rounded_size - size));
+ else
+ {
+ HOST_WIDE_INT i;
+ for (i = 2048; i < (size - rounded_size); i += 2048)
+ {
+ emit_stack_probe (plus_constant (Pmode, r12, -i));
+ emit_insn (gen_rtx_SET (r12,
+ plus_constant (Pmode, r12, -2048)));
+ }
+ rtx r1 = plus_constant (Pmode, r12,
+ -(size - rounded_size - i + 2048));
+ emit_stack_probe (r1);
+ }
+ }
+ }
+
+ /* Make sure nothing is scheduled before we are done. */
+ emit_insn (gen_blockage ());
+}
+
+/* Probe a range of stack addresses from REG1 to REG2 inclusive. These are
+ absolute addresses. */
+const char *
+loongarch_output_probe_stack_range (rtx reg1, rtx reg2, rtx reg3)
+{
+ static int labelno = 0;
+ char loop_lab[32], tmp[64];
+ rtx xops[3];
+
+ ASM_GENERATE_INTERNAL_LABEL (loop_lab, "LPSRL", labelno++);
+
+ /* Loop. */
+ ASM_OUTPUT_INTERNAL_LABEL (asm_out_file, loop_lab);
+
+ /* TEST_ADDR = TEST_ADDR + PROBE_INTERVAL. */
+ xops[0] = reg1;
+ xops[1] = GEN_INT (-PROBE_INTERVAL);
+ xops[2] = reg3;
+ if (TARGET_64BIT)
+ output_asm_insn ("sub.d\t%0,%0,%2", xops);
+ else
+ output_asm_insn ("sub.w\t%0,%0,%2", xops);
+
+ /* Probe at TEST_ADDR, test if TEST_ADDR == LAST_ADDR and branch. */
+ xops[1] = reg2;
+ strcpy (tmp, "bne\t%0,%1,");
+ if (TARGET_64BIT)
+ output_asm_insn ("st.d\t$r0,%0,0", xops);
+ else
+ output_asm_insn ("st.w\t$r0,%0,0", xops);
+ output_asm_insn (strcat (tmp, &loop_lab[1]), xops);
+
+ return "";
+}
+
+/* Expand the "prologue" pattern. */
+
+void
+loongarch_expand_prologue (void)
+{
+ struct loongarch_frame_info *frame = &cfun->machine->frame;
+ HOST_WIDE_INT size = frame->total_size;
+ HOST_WIDE_INT tmp;
+ unsigned mask = frame->mask;
+ rtx insn;
+
+ if (flag_stack_usage_info)
+ current_function_static_stack_size = size;
+
+ if (flag_stack_check == STATIC_BUILTIN_STACK_CHECK
+ || flag_stack_clash_protection)
+ {
+ if (crtl->is_leaf && !cfun->calls_alloca)
+ {
+ if (size > PROBE_INTERVAL && size > get_stack_check_protect ())
+ {
+ tmp = size - get_stack_check_protect ();
+ loongarch_emit_probe_stack_range (get_stack_check_protect (),
+ tmp);
+ }
+ }
+ else if (size > 0)
+ loongarch_emit_probe_stack_range (get_stack_check_protect (), size);
+ }
+
+ /* Save the registers. */
+ if ((frame->mask | frame->fmask) != 0)
+ {
+ HOST_WIDE_INT step1 = MIN (size, loongarch_first_stack_step (frame));
+
+ insn = gen_add3_insn (stack_pointer_rtx, stack_pointer_rtx,
+ GEN_INT (-step1));
+ RTX_FRAME_RELATED_P (emit_insn (insn)) = 1;
+ size -= step1;
+ loongarch_for_each_saved_reg (size, loongarch_save_reg);
+ }
+
+ frame->mask = mask; /* Undo the above fib. */
+
+ /* Set up the frame pointer, if we're using one. */
+ if (frame_pointer_needed)
+ {
+ insn = gen_add3_insn (hard_frame_pointer_rtx, stack_pointer_rtx,
+ GEN_INT (frame->hard_frame_pointer_offset - size));
+ RTX_FRAME_RELATED_P (emit_insn (insn)) = 1;
+
+ loongarch_emit_stack_tie ();
+ }
+
+ /* Allocate the rest of the frame. */
+ if (size > 0)
+ {
+ if (IMM12_OPERAND (-size))
+ {
+ insn = gen_add3_insn (stack_pointer_rtx, stack_pointer_rtx,
+ GEN_INT (-size));
+ RTX_FRAME_RELATED_P (emit_insn (insn)) = 1;
+ }
+ else
+ {
+ loongarch_emit_move (LARCH_PROLOGUE_TEMP (Pmode), GEN_INT (-size));
+ emit_insn (gen_add3_insn (stack_pointer_rtx, stack_pointer_rtx,
+ LARCH_PROLOGUE_TEMP (Pmode)));
+
+ /* Describe the effect of the previous instructions. */
+ insn = plus_constant (Pmode, stack_pointer_rtx, -size);
+ insn = gen_rtx_SET (stack_pointer_rtx, insn);
+ loongarch_set_frame_expr (insn);
+ }
+ }
+}
+
+/* Return nonzero if this function is known to have a null epilogue.
+ This allows the optimizer to omit jumps to jumps if no stack
+ was created. */
+
+bool
+loongarch_can_use_return_insn (void)
+{
+ return reload_completed && cfun->machine->frame.total_size == 0;
+}
+
+/* Expand an "epilogue" or "sibcall_epilogue" pattern; SIBCALL_P
+ says which. */
+
+void
+loongarch_expand_epilogue (bool sibcall_p)
+{
+ /* Split the frame into two. STEP1 is the amount of stack we should
+ deallocate before restoring the registers. STEP2 is the amount we
+ should deallocate afterwards.
+
+ Start off by assuming that no registers need to be restored. */
+ struct loongarch_frame_info *frame = &cfun->machine->frame;
+ HOST_WIDE_INT step1 = frame->total_size;
+ HOST_WIDE_INT step2 = 0;
+ rtx ra = gen_rtx_REG (Pmode, RETURN_ADDR_REGNUM);
+ rtx insn;
+
+ /* We need to add memory barrier to prevent read from deallocated stack. */
+ bool need_barrier_p
+ = (get_frame_size () + cfun->machine->frame.arg_pointer_offset) != 0;
+
+ if (!sibcall_p && loongarch_can_use_return_insn ())
+ {
+ emit_jump_insn (gen_return ());
+ return;
+ }
+
+ /* Move past any dynamic stack allocations. */
+ if (cfun->calls_alloca)
+ {
+ /* Emit a barrier to prevent loads from a deallocated stack. */
+ loongarch_emit_stack_tie ();
+ need_barrier_p = false;
+
+ rtx adjust = GEN_INT (-frame->hard_frame_pointer_offset);
+ if (!IMM12_OPERAND (INTVAL (adjust)))
+ {
+ loongarch_emit_move (LARCH_PROLOGUE_TEMP (Pmode), adjust);
+ adjust = LARCH_PROLOGUE_TEMP (Pmode);
+ }
+
+ insn = emit_insn (gen_add3_insn (stack_pointer_rtx,
+ hard_frame_pointer_rtx,
+ adjust));
+
+ rtx dwarf = NULL_RTX;
+ rtx minus_offset = NULL_RTX;
+ minus_offset = GEN_INT (-frame->hard_frame_pointer_offset);
+ rtx cfa_adjust_value = gen_rtx_PLUS (Pmode,
+ hard_frame_pointer_rtx,
+ minus_offset);
+
+ rtx cfa_adjust_rtx = gen_rtx_SET (stack_pointer_rtx, cfa_adjust_value);
+ dwarf = alloc_reg_note (REG_CFA_ADJUST_CFA, cfa_adjust_rtx, dwarf);
+ RTX_FRAME_RELATED_P (insn) = 1;
+
+ REG_NOTES (insn) = dwarf;
+ }
+
+ /* If we need to restore registers, deallocate as much stack as
+ possible in the second step without going out of range. */
+ if ((frame->mask | frame->fmask) != 0)
+ {
+ step2 = loongarch_first_stack_step (frame);
+ step1 -= step2;
+ }
+
+ /* Set TARGET to BASE + STEP1. */
+ if (step1 > 0)
+ {
+ /* Emit a barrier to prevent loads from a deallocated stack. */
+ loongarch_emit_stack_tie ();
+ need_barrier_p = false;
+
+ /* Get an rtx for STEP1 that we can add to BASE. */
+ rtx adjust = GEN_INT (step1);
+ if (!IMM12_OPERAND (step1))
+ {
+ loongarch_emit_move (LARCH_PROLOGUE_TEMP (Pmode), adjust);
+ adjust = LARCH_PROLOGUE_TEMP (Pmode);
+ }
+
+ insn = emit_insn (gen_add3_insn (stack_pointer_rtx,
+ stack_pointer_rtx,
+ adjust));
+
+ rtx dwarf = NULL_RTX;
+ rtx cfa_adjust_rtx = gen_rtx_PLUS (Pmode, stack_pointer_rtx,
+ GEN_INT (step2));
+
+ dwarf = alloc_reg_note (REG_CFA_DEF_CFA, cfa_adjust_rtx, dwarf);
+ RTX_FRAME_RELATED_P (insn) = 1;
+
+ REG_NOTES (insn) = dwarf;
+ }
+
+ /* Restore the registers. */
+ loongarch_for_each_saved_reg (frame->total_size - step2,
+ loongarch_restore_reg);
+
+ if (need_barrier_p)
+ loongarch_emit_stack_tie ();
+
+ /* Deallocate the final bit of the frame. */
+ if (step2 > 0)
+ {
+ insn = emit_insn (gen_add3_insn (stack_pointer_rtx,
+ stack_pointer_rtx,
+ GEN_INT (step2)));
+
+ rtx dwarf = NULL_RTX;
+ rtx cfa_adjust_rtx = gen_rtx_PLUS (Pmode, stack_pointer_rtx, const0_rtx);
+ dwarf = alloc_reg_note (REG_CFA_DEF_CFA, cfa_adjust_rtx, dwarf);
+ RTX_FRAME_RELATED_P (insn) = 1;
+
+ REG_NOTES (insn) = dwarf;
+ }
+
+ /* Add in the __builtin_eh_return stack adjustment. */
+ if (crtl->calls_eh_return)
+ emit_insn (gen_add3_insn (stack_pointer_rtx, stack_pointer_rtx,
+ EH_RETURN_STACKADJ_RTX));
+
+ if (!sibcall_p)
+ emit_jump_insn (gen_simple_return_internal (ra));
+}
+
+#define LU32I_B (0xfffffUL << 32)
+#define LU52I_B (0xfffUL << 52)
+
+static unsigned int
+loongarch_build_integer (struct loongarch_integer_op *codes,
+ HOST_WIDE_INT value)
+
+{
+ unsigned int cost = 0;
+
+ HOST_WIDE_INT low_part = TARGET_64BIT ? value << 32 >> 32 : value;
+
+ if (IMM12_OPERAND (low_part) || IMM12_OPERAND_UNSIGNED (low_part))
+ {
+ codes[0].code = UNKNOWN;
+ codes[0].method = METHOD_NORMAL;
+ codes[0].value = low_part;
+ cost++;
+ }
+ else
+ {
+ codes[0].code = UNKNOWN;
+ codes[0].method = METHOD_NORMAL;
+ codes[0].value = low_part & ~(IMM_REACH - 1);
+ cost++;
+ HOST_WIDE_INT iorv = low_part & (IMM_REACH - 1);
+ if (iorv != 0)
+ {
+ codes[1].code = IOR;
+ codes[1].method = METHOD_NORMAL;
+ codes[1].value = iorv;
+ cost++;
+ }
+ }
+
+ if (TARGET_64BIT)
+ {
+ bool lu32i[2] = {(value & LU32I_B) == 0, (value & LU32I_B) == LU32I_B};
+ bool lu52i[2] = {(value & LU52I_B) == 0, (value & LU52I_B) == LU52I_B};
+
+ int sign31 = (value & (1UL << 31)) >> 31;
+ if (lu32i[sign31] && lu52i[sign31])
+ return cost;
+ else if (lu32i[sign31])
+ {
+ codes[cost].method = METHOD_LU52I;
+ codes[cost].value = (value >> 52) << 52;
+ return cost + 1;
+ }
+
+ codes[cost].method = METHOD_LU32I;
+ codes[cost].value = ((value << 12) >> 44) << 32;
+ cost++;
+
+ if (!lu52i[(value & (1UL << 51)) >> 51])
+ {
+ codes[cost].method = METHOD_LU52I;
+ codes[cost].value = (value >> 52) << 52;
+ cost++;
+ }
+ }
+
+ gcc_assert (cost <= LARCH_MAX_INTEGER_OPS);
+
+ return cost;
+}
+
+/* Fill CODES with a sequence of rtl operations to load VALUE.
+ Return the number of operations needed.
+ Split interger in loongarch_output_move. */
+
+static unsigned int
+loongarch_integer_cost (HOST_WIDE_INT value)
+{
+ struct loongarch_integer_op codes[LARCH_MAX_INTEGER_OPS];
+ return loongarch_build_integer (codes, value);
+}
+
+/* Implement TARGET_LEGITIMATE_CONSTANT_P. */
+
+static bool
+loongarch_legitimate_constant_p (machine_mode mode ATTRIBUTE_UNUSED, rtx x)
+{
+ return loongarch_const_insns (x) > 0;
+}
+
+/* Return true if X is a thread-local symbol. */
+
+static bool
+loongarch_tls_symbol_p (rtx x)
+{
+ return GET_CODE (x) == SYMBOL_REF && SYMBOL_REF_TLS_MODEL (x) != 0;
+}
+
+/* Return true if SYMBOL_REF X is associated with a global symbol
+ (in the STB_GLOBAL sense). */
+
+bool
+loongarch_global_symbol_p (const_rtx x)
+{
+ if (GET_CODE (x) == LABEL_REF)
+ return false;
+
+ const_tree decl = SYMBOL_REF_DECL (x);
+
+ if (!decl)
+ return !SYMBOL_REF_LOCAL_P (x) || SYMBOL_REF_EXTERNAL_P (x);
+
+ /* Weakref symbols are not TREE_PUBLIC, but their targets are global
+ or weak symbols. Relocations in the object file will be against
+ the target symbol, so it's that symbol's binding that matters here. */
+ return DECL_P (decl) && (TREE_PUBLIC (decl) || DECL_WEAK (decl));
+}
+
+bool
+loongarch_global_symbol_noweak_p (const_rtx x)
+{
+ if (GET_CODE (x) == LABEL_REF)
+ return false;
+
+ const_tree decl = SYMBOL_REF_DECL (x);
+
+ if (!decl)
+ return !SYMBOL_REF_LOCAL_P (x) || SYMBOL_REF_EXTERNAL_P (x);
+
+ /* Weakref symbols are not TREE_PUBLIC, but their targets are global
+ or weak symbols. Relocations in the object file will be against
+ the target symbol, so it's that symbol's binding that matters here. */
+ return DECL_P (decl) && TREE_PUBLIC (decl);
+}
+
+bool
+loongarch_weak_symbol_p (const_rtx x)
+{
+ const_tree decl;
+ if (GET_CODE (x) == LABEL_REF || !(decl = SYMBOL_REF_DECL (x)))
+ return false;
+ return DECL_P (decl) && DECL_WEAK (decl);
+}
+
+/* Return true if SYMBOL_REF X binds locally. */
+
+bool
+loongarch_symbol_binds_local_p (const_rtx x)
+{
+ if (GET_CODE (x) == LABEL_REF)
+ return false;
+
+ return (SYMBOL_REF_DECL (x) ? targetm.binds_local_p (SYMBOL_REF_DECL (x))
+ : SYMBOL_REF_LOCAL_P (x));
+}
+
+/* Return true if rtx constants of mode MODE should be put into a small
+ data section. */
+
+static bool
+loongarch_rtx_constant_in_small_data_p (machine_mode mode)
+{
+ return (GET_MODE_SIZE (mode) <= g_switch_value);
+}
+
+/* Return the method that should be used to access SYMBOL_REF or
+ LABEL_REF X in context CONTEXT. */
+
+static enum loongarch_symbol_type
+loongarch_classify_symbol (const_rtx x,
+ enum loongarch_symbol_context context \
+ ATTRIBUTE_UNUSED)
+{
+ if (GET_CODE (x) == LABEL_REF)
+ {
+ return SYMBOL_GOT_DISP;
+ }
+
+ gcc_assert (GET_CODE (x) == SYMBOL_REF);
+
+ if (SYMBOL_REF_TLS_MODEL (x))
+ return SYMBOL_TLS;
+
+ if (GET_CODE (x) == SYMBOL_REF)
+ return SYMBOL_GOT_DISP;
+
+ return SYMBOL_GOT_DISP;
+}
+
+/* Return true if X is a symbolic constant that can be used in context
+ CONTEXT. If it is, store the type of the symbol in *SYMBOL_TYPE. */
+
+bool
+loongarch_symbolic_constant_p (rtx x, enum loongarch_symbol_context context,
+ enum loongarch_symbol_type *symbol_type)
+{
+ rtx offset;
+
+ split_const (x, &x, &offset);
+ if (UNSPEC_ADDRESS_P (x))
+ {
+ *symbol_type = UNSPEC_ADDRESS_TYPE (x);
+ x = UNSPEC_ADDRESS (x);
+ }
+ else if (GET_CODE (x) == SYMBOL_REF || GET_CODE (x) == LABEL_REF)
+ {
+ *symbol_type = loongarch_classify_symbol (x, context);
+ if (*symbol_type == SYMBOL_TLS)
+ return true;
+ }
+ else
+ return false;
+
+ if (offset == const0_rtx)
+ return true;
+
+ /* Check whether a nonzero offset is valid for the underlying
+ relocations. */
+ switch (*symbol_type)
+ {
+ /* Fall through. */
+
+ case SYMBOL_GOT_DISP:
+ case SYMBOL_TLSGD:
+ case SYMBOL_TLSLDM:
+ case SYMBOL_TLS:
+ return false;
+ }
+ gcc_unreachable ();
+}
+
+/* Like loongarch_symbol_insns We rely on the fact that, in the worst case. */
+
+static int
+loongarch_symbol_insns_1 (enum loongarch_symbol_type type, machine_mode mode)
+{
+ switch (type)
+ {
+ case SYMBOL_GOT_DISP:
+ /* The constant will have to be loaded from the GOT before it
+ is used in an address. */
+ if (mode != MAX_MACHINE_MODE)
+ return 0;
+
+ /* Fall through. */
+
+ return 3;
+
+ case SYMBOL_TLSGD:
+ case SYMBOL_TLSLDM:
+ return 1;
+
+ case SYMBOL_TLS:
+ /* We don't treat a bare TLS symbol as a constant. */
+ return 0;
+ }
+ gcc_unreachable ();
+}
+
+/* If MODE is MAX_MACHINE_MODE, return the number of instructions needed
+ to load symbols of type TYPE into a register. Return 0 if the given
+ type of symbol cannot be used as an immediate operand.
+
+ Otherwise, return the number of instructions needed to load or store
+ values of mode MODE to or from addresses of type TYPE. Return 0 if
+ the given type of symbol is not valid in addresses.
+
+ In both cases, instruction counts are based off BASE_INSN_LENGTH. */
+
+static int
+loongarch_symbol_insns (enum loongarch_symbol_type type, machine_mode mode)
+{
+ return loongarch_symbol_insns_1 (type, mode) * (1);
+}
+
+/* Implement TARGET_CANNOT_FORCE_CONST_MEM. */
+
+static bool
+loongarch_cannot_force_const_mem (machine_mode mode, rtx x)
+{
+ enum loongarch_symbol_type type;
+ rtx base, offset;
+
+ /* As an optimization, reject constants that loongarch_legitimize_move
+ can expand inline.
+
+ Suppose we have a multi-instruction sequence that loads constant C
+ into register R. If R does not get allocated a hard register, and
+ R is used in an operand that allows both registers and memory
+ references, reload will consider forcing C into memory and using
+ one of the instruction's memory alternatives. Returning false
+ here will force it to use an input reload instead. */
+ if (CONST_INT_P (x) && loongarch_legitimate_constant_p (mode, x))
+ return true;
+
+ split_const (x, &base, &offset);
+ if (loongarch_symbolic_constant_p (base, SYMBOL_CONTEXT_LEA, &type))
+ {
+ /* The same optimization as for CONST_INT. */
+ if (IMM12_INT (offset)
+ && loongarch_symbol_insns (type, MAX_MACHINE_MODE) > 0)
+ return true;
+ }
+
+ /* TLS symbols must be computed by loongarch_legitimize_move. */
+ if (tls_referenced_p (x))
+ return true;
+
+ return false;
+}
+
+/* Return true if register REGNO is a valid base register for mode MODE.
+ STRICT_P is true if REG_OK_STRICT is in effect. */
+
+int
+loongarch_regno_mode_ok_for_base_p (int regno,
+ machine_mode mode ATTRIBUTE_UNUSED,
+ bool strict_p)
+{
+ if (!HARD_REGISTER_NUM_P (regno))
+ {
+ if (!strict_p)
+ return true;
+ regno = reg_renumber[regno];
+ }
+
+ /* These fake registers will be eliminated to either the stack or
+ hard frame pointer, both of which are usually valid base registers.
+ Reload deals with the cases where the eliminated form isn't valid. */
+ if (regno == ARG_POINTER_REGNUM || regno == FRAME_POINTER_REGNUM)
+ return true;
+
+ return GP_REG_P (regno);
+}
+
+/* Return true if X is a valid base register for mode MODE.
+ STRICT_P is true if REG_OK_STRICT is in effect. */
+
+static bool
+loongarch_valid_base_register_p (rtx x, machine_mode mode, bool strict_p)
+{
+ if (!strict_p && GET_CODE (x) == SUBREG)
+ x = SUBREG_REG (x);
+
+ return (REG_P (x)
+ && loongarch_regno_mode_ok_for_base_p (REGNO (x), mode, strict_p));
+}
+
+/* Return true if, for every base register BASE_REG, (plus BASE_REG X)
+ can address a value of mode MODE. */
+
+static bool
+loongarch_valid_offset_p (rtx x, machine_mode mode)
+{
+ /* Check that X is a signed 12-bit number,
+ * or check that X is a signed 16-bit number
+ * and offset 4 byte aligned. */
+ if (!(const_arith_operand (x, Pmode)
+ || ((mode == E_SImode || mode == E_DImode)
+ && const_imm16_operand (x, Pmode)
+ && (loongarch_signed_immediate_p (INTVAL (x), 14, 2)))))
+ return false;
+
+ /* We may need to split multiword moves, so make sure that every word
+ is accessible. */
+ if (GET_MODE_SIZE (mode) > UNITS_PER_WORD
+ && !IMM12_OPERAND (INTVAL (x) + GET_MODE_SIZE (mode) - UNITS_PER_WORD))
+ return false;
+
+ return true;
+}
+
+/* Return true if X is a valid address for machine mode MODE. If it is,
+ fill in INFO appropriately. STRICT_P is true if REG_OK_STRICT is in
+ effect. */
+
+static bool
+loongarch_classify_address (struct loongarch_address_info *info, rtx x,
+ machine_mode mode, bool strict_p)
+{
+ switch (GET_CODE (x))
+ {
+ case REG:
+ case SUBREG:
+ info->type = ADDRESS_REG;
+ info->reg = x;
+ info->offset = const0_rtx;
+ return loongarch_valid_base_register_p (info->reg, mode, strict_p);
+
+ case PLUS:
+ info->type = ADDRESS_REG;
+ info->reg = XEXP (x, 0);
+ info->offset = XEXP (x, 1);
+ return (loongarch_valid_base_register_p (info->reg, mode, strict_p)
+ && loongarch_valid_offset_p (info->offset, mode));
+
+ default:
+ return false;
+ }
+}
+
+/* Implement TARGET_LEGITIMATE_ADDRESS_P. */
+
+static bool
+loongarch_legitimate_address_p (machine_mode mode, rtx x, bool strict_p)
+{
+ struct loongarch_address_info addr;
+
+ return loongarch_classify_address (&addr, x, mode, strict_p);
+}
+
+/* Return true if X is a legitimate $sp-based address for mode MODE. */
+
+bool
+loongarch_stack_address_p (rtx x, machine_mode mode)
+{
+ struct loongarch_address_info addr;
+
+ return (loongarch_classify_address (&addr, x, mode, false)
+ && addr.type == ADDRESS_REG
+ && addr.reg == stack_pointer_rtx);
+}
+
+/* Return true if ADDR matches the pattern for the L{B,H,W,D}{,U}X load
+ indexed address instruction. Note that such addresses are
+ not considered legitimate in the TARGET_LEGITIMATE_ADDRESS_P
+ sense, because their use is so restricted. */
+
+static bool
+loongarch_lx_address_p (rtx addr, machine_mode mode ATTRIBUTE_UNUSED)
+{
+ if (GET_CODE (addr) != PLUS
+ || !REG_P (XEXP (addr, 0))
+ || !REG_P (XEXP (addr, 1)))
+ return false;
+ return false;
+}
+
+/* Return the number of instructions needed to load or store a value
+ of mode MODE at address X, assuming that BASE_INSN_LENGTH is the
+ length of one instruction. Return 0 if X isn't valid for MODE.
+ Assume that multiword moves may need to be split into word moves
+ if MIGHT_SPLIT_P, otherwise assume that a single load or store is
+ enough. */
+
+int
+loongarch_address_insns (rtx x, machine_mode mode, bool might_split_p)
+{
+ struct loongarch_address_info addr;
+ int factor;
+
+ if (!loongarch_classify_address (&addr, x, mode, false))
+ return 0;
+
+ /* BLKmode is used for single unaligned loads and stores and should
+ not count as a multiword mode. (GET_MODE_SIZE (BLKmode) is pretty
+ meaningless, so we have to single it out as a special case one way
+ or the other.) */
+ if (mode != BLKmode && might_split_p)
+ factor = (GET_MODE_SIZE (mode) + UNITS_PER_WORD - 1) / UNITS_PER_WORD;
+ else
+ factor = 1;
+
+ if (loongarch_classify_address (&addr, x, mode, false))
+ switch (addr.type)
+ {
+ case ADDRESS_REG:
+ return factor;
+
+ case ADDRESS_CONST_INT:
+ return factor;
+
+ case ADDRESS_SYMBOLIC:
+ return factor * loongarch_symbol_insns (addr.symbol_type, mode);
+ }
+ return 0;
+}
+
+/* Return true if X fits within an unsigned field of BITS bits that is
+ shifted left SHIFT bits before being used. */
+
+bool
+loongarch_unsigned_immediate_p (unsigned HOST_WIDE_INT x, int bits,
+ int shift = 0)
+{
+ return (x & ((1 << shift) - 1)) == 0 && x < ((unsigned) 1 << (shift + bits));
+}
+
+/* Return true if X fits within a signed field of BITS bits that is
+ shifted left SHIFT bits before being used. */
+
+bool
+loongarch_signed_immediate_p (unsigned HOST_WIDE_INT x, int bits,
+ int shift = 0)
+{
+ x += 1 << (bits + shift - 1);
+ return loongarch_unsigned_immediate_p (x, bits, shift);
+}
+
+/* Return true if X is a legitimate address with a 12-bit offset.
+ MODE is the mode of the value being accessed. */
+
+bool
+loongarch_12bit_offset_address_p (rtx x, machine_mode mode)
+{
+ struct loongarch_address_info addr;
+
+ return (loongarch_classify_address (&addr, x, mode, false)
+ && addr.type == ADDRESS_REG
+ && CONST_INT_P (addr.offset)
+ && LARCH_U12BIT_OFFSET_P (INTVAL (addr.offset)));
+}
+
+/* Return true if X is a legitimate address with a 14-bit offset shifted 2.
+ MODE is the mode of the value being accessed. */
+
+bool
+loongarch_14bit_shifted_offset_address_p (rtx x, machine_mode mode)
+{
+ struct loongarch_address_info addr;
+
+ return (loongarch_classify_address (&addr, x, mode, false)
+ && addr.type == ADDRESS_REG
+ && CONST_INT_P (addr.offset)
+ && LARCH_16BIT_OFFSET_P (INTVAL (addr.offset))
+ && LARCH_SHIFT_2_OFFSET_P (INTVAL (addr.offset)));
+}
+
+/* Return the number of instructions needed to load constant X,
+ assuming that BASE_INSN_LENGTH is the length of one instruction.
+ Return 0 if X isn't a valid constant. */
+
+int
+loongarch_const_insns (rtx x)
+{
+ enum loongarch_symbol_type symbol_type;
+ rtx offset;
+
+ switch (GET_CODE (x))
+ {
+ case CONST_INT:
+ return loongarch_integer_cost (INTVAL (x));
+
+ case CONST_VECTOR:
+ /* Fall through. */
+ case CONST_DOUBLE:
+ /* Allow zeros for normal mode, where we can use $0. */
+ return x == CONST0_RTX (GET_MODE (x)) ? 1 : 0;
+
+ case CONST:
+ /* See if we can refer to X directly. */
+ if (loongarch_symbolic_constant_p (x, SYMBOL_CONTEXT_LEA, &symbol_type))
+ return loongarch_symbol_insns (symbol_type, MAX_MACHINE_MODE);
+
+ /* Otherwise try splitting the constant into a base and offset.
+ If the offset is a 12-bit value, we can load the base address
+ into a register and then use ADDI.{W/D} to add in the offset.
+ If the offset is larger, we can load the base and offset
+ into separate registers and add them together with ADD.{W/D}.
+ However, the latter is only possible before reload; during
+ and after reload, we must have the option of forcing the
+ constant into the pool instead. */
+ split_const (x, &x, &offset);
+ if (offset != 0)
+ {
+ int n = loongarch_const_insns (x);
+ if (n != 0)
+ {
+ if (IMM12_INT (offset))
+ return n + 1;
+ else if (!targetm.cannot_force_const_mem (GET_MODE (x), x))
+ return n + 1 + loongarch_integer_cost (INTVAL (offset));
+ }
+ }
+ return 0;
+
+ case SYMBOL_REF:
+ case LABEL_REF:
+ return loongarch_symbol_insns (
+ loongarch_classify_symbol (x, SYMBOL_CONTEXT_LEA), MAX_MACHINE_MODE);
+
+ default:
+ return 0;
+ }
+}
+
+/* X is a doubleword constant that can be handled by splitting it into
+ two words and loading each word separately. Return the number of
+ instructions required to do this, assuming that BASE_INSN_LENGTH
+ is the length of one instruction. */
+
+int
+loongarch_split_const_insns (rtx x)
+{
+ unsigned int low, high;
+
+ low = loongarch_const_insns (loongarch_subword (x, false));
+ high = loongarch_const_insns (loongarch_subword (x, true));
+ gcc_assert (low > 0 && high > 0);
+ return low + high;
+}
+
+/* Return the number of instructions needed to implement INSN,
+ given that it loads from or stores to MEM. Assume that
+ BASE_INSN_LENGTH is the length of one instruction. */
+
+int
+loongarch_load_store_insns (rtx mem, rtx_insn *insn)
+{
+ machine_mode mode;
+ bool might_split_p;
+ rtx set;
+
+ gcc_assert (MEM_P (mem));
+ mode = GET_MODE (mem);
+
+ /* Try to prove that INSN does not need to be split. */
+ might_split_p = GET_MODE_SIZE (mode) > UNITS_PER_WORD;
+ if (might_split_p)
+ {
+ set = single_set (insn);
+ if (set
+ && !loongarch_split_move_insn_p (SET_DEST (set), SET_SRC (set),
+ insn))
+ might_split_p = false;
+ }
+
+ return loongarch_address_insns (XEXP (mem, 0), mode, might_split_p);
+}
+
+/* Return the number of instructions needed for an integer division,
+ assuming that BASE_INSN_LENGTH is the length of one instruction. */
+
+int
+loongarch_idiv_insns (machine_mode mode ATTRIBUTE_UNUSED)
+{
+ int count;
+
+ count = 1;
+ if (TARGET_CHECK_ZERO_DIV)
+ count += 2;
+
+ return count;
+}
+
+/* Emit an instruction of the form (set TARGET (CODE OP0 OP1)). */
+
+void
+loongarch_emit_binary (enum rtx_code code, rtx target, rtx op0, rtx op1)
+{
+ emit_insn (gen_rtx_SET (target, gen_rtx_fmt_ee (code, GET_MODE (target),
+ op0, op1)));
+}
+
+/* Compute (CODE OP0 OP1) and store the result in a new register
+ of mode MODE. Return that new register. */
+
+static rtx
+loongarch_force_binary (machine_mode mode, enum rtx_code code, rtx op0,
+ rtx op1)
+{
+ rtx reg;
+
+ reg = gen_reg_rtx (mode);
+ loongarch_emit_binary (code, reg, op0, op1);
+ return reg;
+}
+
+/* Copy VALUE to a register and return that register. If new pseudos
+ are allowed, copy it into a new register, otherwise use DEST. */
+
+static rtx
+loongarch_force_temporary (rtx dest, rtx value)
+{
+ if (can_create_pseudo_p ())
+ return force_reg (Pmode, value);
+ else
+ {
+ loongarch_emit_move (dest, value);
+ return dest;
+ }
+}
+
+/* Wrap symbol or label BASE in an UNSPEC address of type SYMBOL_TYPE,
+ then add CONST_INT OFFSET to the result. */
+
+static rtx
+loongarch_unspec_address_offset (rtx base, rtx offset,
+ enum loongarch_symbol_type symbol_type)
+{
+ base = gen_rtx_UNSPEC (Pmode, gen_rtvec (1, base),
+ UNSPEC_ADDRESS_FIRST + symbol_type);
+ if (offset != const0_rtx)
+ base = gen_rtx_PLUS (Pmode, base, offset);
+ return gen_rtx_CONST (Pmode, base);
+}
+
+/* Return an UNSPEC address with underlying address ADDRESS and symbol
+ type SYMBOL_TYPE. */
+
+rtx
+loongarch_unspec_address (rtx address, enum loongarch_symbol_type symbol_type)
+{
+ rtx base, offset;
+
+ split_const (address, &base, &offset);
+ return loongarch_unspec_address_offset (base, offset, symbol_type);
+}
+
+/* If OP is an UNSPEC address, return the address to which it refers,
+ otherwise return OP itself. */
+
+rtx
+loongarch_strip_unspec_address (rtx op)
+{
+ rtx base, offset;
+
+ split_const (op, &base, &offset);
+ if (UNSPEC_ADDRESS_P (base))
+ op = plus_constant (Pmode, UNSPEC_ADDRESS (base), INTVAL (offset));
+ return op;
+}
+
+/* Return a legitimate address for REG + OFFSET. TEMP is as for
+ loongarch_force_temporary; it is only needed when OFFSET is not a
+ IMM12_OPERAND. */
+
+static rtx
+loongarch_add_offset (rtx temp, rtx reg, HOST_WIDE_INT offset)
+{
+ if (!IMM12_OPERAND (offset))
+ {
+ rtx high;
+
+ /* Leave OFFSET as a 12-bit offset and put the excess in HIGH.
+ The addition inside the macro CONST_HIGH_PART may cause an
+ overflow, so we need to force a sign-extension check. */
+ high = gen_int_mode (CONST_HIGH_PART (offset), Pmode);
+ offset = CONST_LOW_PART (offset);
+ high = loongarch_force_temporary (temp, high);
+ reg = loongarch_force_temporary (temp, gen_rtx_PLUS (Pmode, high, reg));
+ }
+ return plus_constant (Pmode, reg, offset);
+}
+
+/* The __tls_get_attr symbol. */
+static GTY (()) rtx loongarch_tls_symbol;
+
+/* Load an entry from the GOT for a TLS GD access. */
+
+static rtx
+loongarch_got_load_tls_gd (rtx dest, rtx sym)
+{
+ if (Pmode == DImode)
+ return gen_got_load_tls_gddi (dest, sym);
+ else
+ return gen_got_load_tls_gdsi (dest, sym);
+}
+
+/* Load an entry from the GOT for a TLS LD access. */
+
+static rtx
+loongarch_got_load_tls_ld (rtx dest, rtx sym)
+{
+ if (Pmode == DImode)
+ return gen_got_load_tls_lddi (dest, sym);
+ else
+ return gen_got_load_tls_ldsi (dest, sym);
+}
+
+/* Load an entry from the GOT for a TLS IE access. */
+
+static rtx
+loongarch_got_load_tls_ie (rtx dest, rtx sym)
+{
+ if (Pmode == DImode)
+ return gen_got_load_tls_iedi (dest, sym);
+ else
+ return gen_got_load_tls_iesi (dest, sym);
+}
+
+/* Add in the thread pointer for a TLS LE access. */
+
+static rtx
+loongarch_got_load_tls_le (rtx dest, rtx sym)
+{
+ if (Pmode == DImode)
+ return gen_got_load_tls_ledi (dest, sym);
+ else
+ return gen_got_load_tls_lesi (dest, sym);
+}
+
+/* Return an instruction sequence that calls __tls_get_addr. SYM is
+ the TLS symbol we are referencing and TYPE is the symbol type to use
+ (either global dynamic or local dynamic). V0 is an RTX for the
+ return value location. */
+
+static rtx_insn *
+loongarch_call_tls_get_addr (rtx sym, enum loongarch_symbol_type type, rtx v0)
+{
+ rtx loc, a0;
+ rtx_insn *insn;
+
+ a0 = gen_rtx_REG (Pmode, GP_ARG_FIRST);
+
+ if (!loongarch_tls_symbol)
+ loongarch_tls_symbol = init_one_libfunc ("__tls_get_addr");
+
+ loc = loongarch_unspec_address (sym, type);
+
+ start_sequence ();
+
+ if (type == SYMBOL_TLSLDM)
+ emit_insn (loongarch_got_load_tls_ld (a0, loc));
+ else if (type == SYMBOL_TLSGD)
+ emit_insn (loongarch_got_load_tls_gd (a0, loc));
+ else
+ gcc_unreachable ();
+
+ insn = emit_call_insn (gen_call_value_internal (v0, loongarch_tls_symbol,
+ const0_rtx));
+ RTL_CONST_CALL_P (insn) = 1;
+ use_reg (&CALL_INSN_FUNCTION_USAGE (insn), a0);
+ insn = get_insns ();
+
+ end_sequence ();
+
+ return insn;
+}
+
+/* Generate the code to access LOC, a thread-local SYMBOL_REF, and return
+ its address. The return value will be both a valid address and a valid
+ SET_SRC (either a REG or a LO_SUM). */
+
+static rtx
+loongarch_legitimize_tls_address (rtx loc)
+{
+ rtx dest, tp, tmp;
+ enum tls_model model = SYMBOL_REF_TLS_MODEL (loc);
+ rtx_insn *insn;
+
+ switch (model)
+ {
+ case TLS_MODEL_LOCAL_DYNAMIC:
+ tmp = gen_rtx_REG (Pmode, GP_RETURN);
+ dest = gen_reg_rtx (Pmode);
+ insn = loongarch_call_tls_get_addr (loc, SYMBOL_TLSLDM, tmp);
+ emit_libcall_block (insn, dest, tmp, loc);
+ break;
+
+ case TLS_MODEL_GLOBAL_DYNAMIC:
+ tmp = gen_rtx_REG (Pmode, GP_RETURN);
+ dest = gen_reg_rtx (Pmode);
+ insn = loongarch_call_tls_get_addr (loc, SYMBOL_TLSGD, tmp);
+ emit_libcall_block (insn, dest, tmp, loc);
+ break;
+
+ case TLS_MODEL_INITIAL_EXEC:
+ /* la.tls.ie; tp-relative add */
+ tp = gen_rtx_REG (Pmode, THREAD_POINTER_REGNUM);
+ tmp = gen_reg_rtx (Pmode);
+ emit_insn (loongarch_got_load_tls_ie (tmp, loc));
+ dest = gen_reg_rtx (Pmode);
+ emit_insn (gen_add3_insn (dest, tmp, tp));
+ break;
+
+ case TLS_MODEL_LOCAL_EXEC:
+ /* la.tls.le; tp-relative add */
+ tp = gen_rtx_REG (Pmode, THREAD_POINTER_REGNUM);
+ tmp = gen_reg_rtx (Pmode);
+ emit_insn (loongarch_got_load_tls_le (tmp, loc));
+ dest = gen_reg_rtx (Pmode);
+ emit_insn (gen_add3_insn (dest, tmp, tp));
+ break;
+
+ default:
+ gcc_unreachable ();
+ }
+ return dest;
+}
+
+rtx
+loongarch_legitimize_call_address (rtx addr)
+{
+ if (!call_insn_operand (addr, VOIDmode))
+ {
+ rtx reg = gen_reg_rtx (Pmode);
+ loongarch_emit_move (reg, addr);
+ return reg;
+ }
+ return addr;
+}
+
+/* If X is a PLUS of a CONST_INT, return the two terms in *BASE_PTR
+ and *OFFSET_PTR. Return X in *BASE_PTR and 0 in *OFFSET_PTR otherwise. */
+
+static void
+loongarch_split_plus (rtx x, rtx *base_ptr, HOST_WIDE_INT *offset_ptr)
+{
+ if (GET_CODE (x) == PLUS && CONST_INT_P (XEXP (x, 1)))
+ {
+ *base_ptr = XEXP (x, 0);
+ *offset_ptr = INTVAL (XEXP (x, 1));
+ }
+ else
+ {
+ *base_ptr = x;
+ *offset_ptr = 0;
+ }
+}
+
+/* If X is not a valid address for mode MODE, force it into a register. */
+
+static rtx
+loongarch_force_address (rtx x, machine_mode mode)
+{
+ if (!loongarch_legitimate_address_p (mode, x, false))
+ x = force_reg (Pmode, x);
+ return x;
+}
+
+/* This function is used to implement LEGITIMIZE_ADDRESS. If X can
+ be legitimized in a way that the generic machinery might not expect,
+ return a new address, otherwise return NULL. MODE is the mode of
+ the memory being accessed. */
+
+static rtx
+loongarch_legitimize_address (rtx x, rtx oldx ATTRIBUTE_UNUSED,
+ machine_mode mode)
+{
+ rtx base, addr;
+ HOST_WIDE_INT offset;
+
+ if (loongarch_tls_symbol_p (x))
+ return loongarch_legitimize_tls_address (x);
+
+ /* Handle BASE + OFFSET using loongarch_add_offset. */
+ loongarch_split_plus (x, &base, &offset);
+ if (offset != 0)
+ {
+ if (!loongarch_valid_base_register_p (base, mode, false))
+ base = copy_to_mode_reg (Pmode, base);
+ addr = loongarch_add_offset (NULL, base, offset);
+ return loongarch_force_address (addr, mode);
+ }
+
+ return x;
+}
+
+/* Load VALUE into DEST. TEMP is as for loongarch_force_temporary. */
+
+void
+loongarch_move_integer (rtx temp, rtx dest, unsigned HOST_WIDE_INT value)
+{
+ struct loongarch_integer_op codes[LARCH_MAX_INTEGER_OPS];
+ machine_mode mode;
+ unsigned int i, num_ops;
+ rtx x;
+
+ mode = GET_MODE (dest);
+ num_ops = loongarch_build_integer (codes, value);
+
+ /* Apply each binary operation to X. Invariant: X is a legitimate
+ source operand for a SET pattern. */
+ x = GEN_INT (codes[0].value);
+ for (i = 1; i < num_ops; i++)
+ {
+ if (!can_create_pseudo_p ())
+ {
+ emit_insn (gen_rtx_SET (temp, x));
+ x = temp;
+ }
+ else
+ x = force_reg (mode, x);
+
+ switch (codes[i].method)
+ {
+ case METHOD_NORMAL:
+ x = gen_rtx_fmt_ee (codes[i].code, mode, x,
+ GEN_INT (codes[i].value));
+ break;
+ case METHOD_LU32I:
+ emit_insn (
+ gen_rtx_SET (x,
+ gen_rtx_IOR (DImode,
+ gen_rtx_ZERO_EXTEND (
+ DImode, gen_rtx_SUBREG (SImode, x, 0)),
+ GEN_INT (codes[i].value))));
+ break;
+ case METHOD_LU52I:
+ emit_insn (gen_lu52i_d (x, x, GEN_INT (0xfffffffffffff),
+ GEN_INT (codes[i].value)));
+ break;
+ case METHOD_INSV:
+ emit_insn (
+ gen_rtx_SET (gen_rtx_ZERO_EXTRACT (DImode, x, GEN_INT (20),
+ GEN_INT (32)),
+ gen_rtx_REG (DImode, 0)));
+ break;
+ default:
+ gcc_unreachable ();
+ }
+ }
+
+ emit_insn (gen_rtx_SET (dest, x));
+}
+
+/* Subroutine of loongarch_legitimize_move. Move constant SRC into register
+ DEST given that SRC satisfies immediate_operand but doesn't satisfy
+ move_operand. */
+
+static void
+loongarch_legitimize_const_move (machine_mode mode, rtx dest, rtx src)
+{
+ rtx base, offset;
+
+ /* Split moves of big integers into smaller pieces. */
+ if (splittable_const_int_operand (src, mode))
+ {
+ loongarch_move_integer (dest, dest, INTVAL (src));
+ return;
+ }
+
+ /* Generate the appropriate access sequences for TLS symbols. */
+ if (loongarch_tls_symbol_p (src))
+ {
+ loongarch_emit_move (dest, loongarch_legitimize_tls_address (src));
+ return;
+ }
+
+ /* If we have (const (plus symbol offset)), and that expression cannot
+ be forced into memory, load the symbol first and add in the offset.
+ prefer to do this even if the constant _can_ be forced into memory,
+ as it usually produces better code. */
+ split_const (src, &base, &offset);
+ if (offset != const0_rtx
+ && (targetm.cannot_force_const_mem (mode, src)
+ || (can_create_pseudo_p ())))
+ {
+ base = loongarch_force_temporary (dest, base);
+ loongarch_emit_move (dest,
+ loongarch_add_offset (NULL, base, INTVAL (offset)));
+ return;
+ }
+
+ src = force_const_mem (mode, src);
+
+ loongarch_emit_move (dest, src);
+}
+
+/* If (set DEST SRC) is not a valid move instruction, emit an equivalent
+ sequence that is valid. */
+
+bool
+loongarch_legitimize_move (machine_mode mode, rtx dest, rtx src)
+{
+ if (!register_operand (dest, mode) && !reg_or_0_operand (src, mode))
+ {
+ loongarch_emit_move (dest, force_reg (mode, src));
+ return true;
+ }
+
+ /* Both src and dest are non-registers; one special case is supported where
+ the source is (const_int 0) and the store can source the zero register.
+ */
+ if (!register_operand (dest, mode) && !register_operand (src, mode)
+ && !const_0_operand (src, mode))
+ {
+ loongarch_emit_move (dest, force_reg (mode, src));
+ return true;
+ }
+
+ /* We need to deal with constants that would be legitimate
+ immediate_operands but aren't legitimate move_operands. */
+ if (CONSTANT_P (src) && !move_operand (src, mode))
+ {
+ loongarch_legitimize_const_move (mode, dest, src);
+ set_unique_reg_note (get_last_insn (), REG_EQUAL, copy_rtx (src));
+ return true;
+ }
+ return false;
+}
+
+/* Return true if OP refers to small data symbols directly, not through
+ a LO_SUM. CONTEXT is the context in which X appears. */
+
+static int
+loongarch_small_data_pattern_1 (rtx x,
+ enum loongarch_symbol_context context \
+ ATTRIBUTE_UNUSED)
+{
+ subrtx_var_iterator::array_type array;
+ FOR_EACH_SUBRTX_VAR (iter, array, x, ALL)
+ {
+ rtx x = *iter;
+
+ /* We make no particular guarantee about which symbolic constants are
+ acceptable as asm operands versus which must be forced into a GPR. */
+ if (GET_CODE (x) == ASM_OPERANDS)
+ iter.skip_subrtxes ();
+ else if (MEM_P (x))
+ {
+ if (loongarch_small_data_pattern_1 (XEXP (x, 0), SYMBOL_CONTEXT_MEM))
+ return true;
+ iter.skip_subrtxes ();
+ }
+ }
+ return false;
+}
+
+/* Return true if OP refers to small data symbols directly, not through
+ a LO_SUM. */
+
+bool
+loongarch_small_data_pattern_p (rtx op)
+{
+ return loongarch_small_data_pattern_1 (op, SYMBOL_CONTEXT_LEA);
+}
+
+/* Rewrite *LOC so that it refers to small data using explicit
+ relocations. CONTEXT is the context in which *LOC appears. */
+
+static void
+loongarch_rewrite_small_data_1 (rtx *loc,
+ enum loongarch_symbol_context context \
+ ATTRIBUTE_UNUSED)
+{
+ subrtx_ptr_iterator::array_type array;
+ FOR_EACH_SUBRTX_PTR (iter, array, loc, ALL)
+ {
+ rtx *loc = *iter;
+ if (MEM_P (*loc))
+ {
+ loongarch_rewrite_small_data_1 (&XEXP (*loc, 0), SYMBOL_CONTEXT_MEM);
+ iter.skip_subrtxes ();
+ }
+ }
+}
+
+/* Rewrite instruction pattern PATTERN so that it refers to small data
+ using explicit relocations. */
+
+rtx
+loongarch_rewrite_small_data (rtx pattern)
+{
+ pattern = copy_insn (pattern);
+ loongarch_rewrite_small_data_1 (&pattern, SYMBOL_CONTEXT_LEA);
+ return pattern;
+}
+
+/* The cost of loading values from the constant pool. It should be
+ larger than the cost of any constant we want to synthesize inline. */
+#define CONSTANT_POOL_COST COSTS_N_INSNS (8)
+
+/* Return true if there is a instruction that implements CODE
+ and if that instruction accepts X as an immediate operand. */
+
+static int
+loongarch_immediate_operand_p (int code, HOST_WIDE_INT x)
+{
+ switch (code)
+ {
+ case ASHIFT:
+ case ASHIFTRT:
+ case LSHIFTRT:
+ /* All shift counts are truncated to a valid constant. */
+ return true;
+
+ case ROTATE:
+ case ROTATERT:
+ /* Likewise rotates, if the target supports rotates at all. */
+ return true;
+
+ case AND:
+ case IOR:
+ case XOR:
+ /* These instructions take 12-bit unsigned immediates. */
+ return IMM12_OPERAND_UNSIGNED (x);
+
+ case PLUS:
+ case LT:
+ case LTU:
+ /* These instructions take 12-bit signed immediates. */
+ return IMM12_OPERAND (x);
+
+ case EQ:
+ case NE:
+ case GT:
+ case GTU:
+ /* The "immediate" forms of these instructions are really
+ implemented as comparisons with register 0. */
+ return x == 0;
+
+ case GE:
+ case GEU:
+ /* Likewise, meaning that the only valid immediate operand is 1. */
+ return x == 1;
+
+ case LE:
+ /* We add 1 to the immediate and use SLT. */
+ return IMM12_OPERAND (x + 1);
+
+ case LEU:
+ /* Likewise SLTU, but reject the always-true case. */
+ return IMM12_OPERAND (x + 1) && x + 1 != 0;
+
+ case SIGN_EXTRACT:
+ case ZERO_EXTRACT:
+ /* The bit position and size are immediate operands. */
+ return 1;
+
+ default:
+ /* By default assume that $0 can be used for 0. */
+ return x == 0;
+ }
+}
+
+/* Return the cost of binary operation X, given that the instruction
+ sequence for a word-sized or smaller operation has cost SINGLE_COST
+ and that the sequence of a double-word operation has cost DOUBLE_COST.
+ If SPEED is true, optimize for speed otherwise optimize for size. */
+
+static int
+loongarch_binary_cost (rtx x, int single_cost, int double_cost, bool speed)
+{
+ int cost;
+
+ if (GET_MODE_SIZE (GET_MODE (x)) == UNITS_PER_WORD * 2)
+ cost = double_cost;
+ else
+ cost = single_cost;
+ return (cost
+ + set_src_cost (XEXP (x, 0), GET_MODE (x), speed)
+ + rtx_cost (XEXP (x, 1), GET_MODE (x), GET_CODE (x), 1, speed));
+}
+
+/* Return the cost of floating-point multiplications of mode MODE. */
+
+static int
+loongarch_fp_mult_cost (machine_mode mode)
+{
+ return mode == DFmode ? loongarch_cost->fp_mult_df
+ : loongarch_cost->fp_mult_sf;
+}
+
+/* Return the cost of floating-point divisions of mode MODE. */
+
+static int
+loongarch_fp_div_cost (machine_mode mode)
+{
+ return mode == DFmode ? loongarch_cost->fp_div_df
+ : loongarch_cost->fp_div_sf;
+}
+
+/* Return the cost of sign-extending OP to mode MODE, not including the
+ cost of OP itself. */
+
+static int
+loongarch_sign_extend_cost (machine_mode mode, rtx op)
+{
+ if (MEM_P (op))
+ /* Extended loads are as cheap as unextended ones. */
+ return 0;
+
+ if (TARGET_64BIT && mode == DImode && GET_MODE (op) == SImode)
+ /* A sign extension from SImode to DImode in 64-bit mode is free. */
+ return 0;
+
+ return COSTS_N_INSNS (1);
+}
+
+/* Return the cost of zero-extending OP to mode MODE, not including the
+ cost of OP itself. */
+
+static int
+loongarch_zero_extend_cost (machine_mode mode, rtx op)
+{
+ if (MEM_P (op))
+ /* Extended loads are as cheap as unextended ones. */
+ return 0;
+
+ if (TARGET_64BIT && mode == DImode && GET_MODE (op) == SImode)
+ /* We need a shift left by 32 bits and a shift right by 32 bits. */
+ return COSTS_N_INSNS (2);
+
+ /* We can use ANDI. */
+ return COSTS_N_INSNS (1);
+}
+
+/* Return the cost of moving between two registers of mode MODE,
+ assuming that the move will be in pieces of at most UNITS bytes. */
+
+static int
+loongarch_set_reg_reg_piece_cost (machine_mode mode, unsigned int units)
+{
+ return COSTS_N_INSNS ((GET_MODE_SIZE (mode) + units - 1) / units);
+}
+
+/* Return the cost of moving between two registers of mode MODE. */
+
+static int
+loongarch_set_reg_reg_cost (machine_mode mode)
+{
+ switch (GET_MODE_CLASS (mode))
+ {
+ case MODE_CC:
+ return loongarch_set_reg_reg_piece_cost (mode, GET_MODE_SIZE (CCmode));
+
+ case MODE_FLOAT:
+ case MODE_COMPLEX_FLOAT:
+ case MODE_VECTOR_FLOAT:
+ if (TARGET_HARD_FLOAT)
+ return loongarch_set_reg_reg_piece_cost (mode, UNITS_PER_HWFPVALUE);
+ /* Fall through. */
+
+ default:
+ return loongarch_set_reg_reg_piece_cost (mode, UNITS_PER_WORD);
+ }
+}
+
+/* Implement TARGET_RTX_COSTS. */
+
+static bool
+loongarch_rtx_costs (rtx x, machine_mode mode, int outer_code,
+ int opno ATTRIBUTE_UNUSED, int *total, bool speed)
+{
+ int code = GET_CODE (x);
+ bool float_mode_p = FLOAT_MODE_P (mode);
+ int cost;
+ rtx addr;
+
+ if (outer_code == COMPARE)
+ {
+ gcc_assert (CONSTANT_P (x));
+ *total = 0;
+ return true;
+ }
+
+ switch (code)
+ {
+ case CONST_INT:
+ if (TARGET_64BIT && outer_code == AND && UINTVAL (x) == 0xffffffff)
+ {
+ *total = 0;
+ return true;
+ }
+
+ /* When not optimizing for size, we care more about the cost
+ of hot code, and hot code is often in a loop. If a constant
+ operand needs to be forced into a register, we will often be
+ able to hoist the constant load out of the loop, so the load
+ should not contribute to the cost. */
+ if (speed || loongarch_immediate_operand_p (outer_code, INTVAL (x)))
+ {
+ *total = 0;
+ return true;
+ }
+ /* Fall through. */
+
+ case CONST:
+ case SYMBOL_REF:
+ case LABEL_REF:
+ case CONST_DOUBLE:
+ cost = loongarch_const_insns (x);
+ if (cost > 0)
+ {
+ if (cost == 1 && outer_code == SET
+ && !(float_mode_p && TARGET_HARD_FLOAT))
+ cost = 0;
+ else if ((outer_code == SET || GET_MODE (x) == VOIDmode))
+ cost = 1;
+ *total = COSTS_N_INSNS (cost);
+ return true;
+ }
+ /* The value will need to be fetched from the constant pool. */
+ *total = CONSTANT_POOL_COST;
+ return true;
+
+ case MEM:
+ /* If the address is legitimate, return the number of
+ instructions it needs. */
+ addr = XEXP (x, 0);
+ cost = loongarch_address_insns (addr, mode, true);
+ if (cost > 0)
+ {
+ *total = COSTS_N_INSNS (cost + 1);
+ return true;
+ }
+ /* Check for a scaled indexed address. */
+ if (loongarch_lx_address_p (addr, mode))
+ {
+ *total = COSTS_N_INSNS (2);
+ return true;
+ }
+ /* Otherwise use the default handling. */
+ return false;
+
+ case FFS:
+ *total = COSTS_N_INSNS (6);
+ return false;
+
+ case NOT:
+ *total = COSTS_N_INSNS (GET_MODE_SIZE (mode) > UNITS_PER_WORD ? 2 : 1);
+ return false;
+
+ case AND:
+ /* Check for a *clear_upper32 pattern and treat it like a zero
+ extension. See the pattern's comment for details. */
+ if (TARGET_64BIT && mode == DImode && CONST_INT_P (XEXP (x, 1))
+ && UINTVAL (XEXP (x, 1)) == 0xffffffff)
+ {
+ *total = (loongarch_zero_extend_cost (mode, XEXP (x, 0))
+ + set_src_cost (XEXP (x, 0), mode, speed));
+ return true;
+ }
+ /* (AND (NOT op0) (NOT op1) is a nor operation that can be done in
+ a single instruction. */
+ if (GET_CODE (XEXP (x, 0)) == NOT && GET_CODE (XEXP (x, 1)) == NOT)
+ {
+ cost = GET_MODE_SIZE (mode) > UNITS_PER_WORD ? 2 : 1;
+ *total = (COSTS_N_INSNS (cost)
+ + set_src_cost (XEXP (XEXP (x, 0), 0), mode, speed)
+ + set_src_cost (XEXP (XEXP (x, 1), 0), mode, speed));
+ return true;
+ }
+
+ /* Fall through. */
+
+ case IOR:
+ case XOR:
+ /* Double-word operations use two single-word operations. */
+ *total = loongarch_binary_cost (x, COSTS_N_INSNS (1), COSTS_N_INSNS (2),
+ speed);
+ return true;
+
+ case ASHIFT:
+ case ASHIFTRT:
+ case LSHIFTRT:
+ case ROTATE:
+ case ROTATERT:
+ if (CONSTANT_P (XEXP (x, 1)))
+ *total = loongarch_binary_cost (x, COSTS_N_INSNS (1),
+ COSTS_N_INSNS (4), speed);
+ else
+ *total = loongarch_binary_cost (x, COSTS_N_INSNS (1),
+ COSTS_N_INSNS (12), speed);
+ return true;
+
+ case ABS:
+ if (float_mode_p)
+ *total = loongarch_cost->fp_add;
+ else
+ *total = COSTS_N_INSNS (4);
+ return false;
+
+ case LT:
+ case LTU:
+ case LE:
+ case LEU:
+ case GT:
+ case GTU:
+ case GE:
+ case GEU:
+ case EQ:
+ case NE:
+ case UNORDERED:
+ case LTGT:
+ case UNGE:
+ case UNGT:
+ case UNLE:
+ case UNLT:
+ /* Branch comparisons have VOIDmode, so use the first operand's
+ mode instead. */
+ mode = GET_MODE (XEXP (x, 0));
+ if (FLOAT_MODE_P (mode))
+ {
+ *total = loongarch_cost->fp_add;
+ return false;
+ }
+ *total = loongarch_binary_cost (x, COSTS_N_INSNS (1), COSTS_N_INSNS (4),
+ speed);
+ return true;
+
+ case MINUS:
+ case PLUS:
+ if (float_mode_p)
+ {
+ *total = loongarch_cost->fp_add;
+ return false;
+ }
+
+ /* If it's an add + mult (which is equivalent to shift left) and
+ it's immediate operand satisfies const_immlsa_operand predicate. */
+ if ((mode == SImode || (TARGET_64BIT && mode == DImode))
+ && GET_CODE (XEXP (x, 0)) == MULT)
+ {
+ rtx op2 = XEXP (XEXP (x, 0), 1);
+ if (const_immlsa_operand (op2, mode))
+ {
+ *total = (COSTS_N_INSNS (1)
+ + set_src_cost (XEXP (XEXP (x, 0), 0), mode, speed)
+ + set_src_cost (XEXP (x, 1), mode, speed));
+ return true;
+ }
+ }
+
+ /* Double-word operations require three single-word operations and
+ an SLTU. */
+ *total = loongarch_binary_cost (x, COSTS_N_INSNS (1), COSTS_N_INSNS (4),
+ speed);
+ return true;
+
+ case NEG:
+ if (float_mode_p)
+ *total = loongarch_cost->fp_add;
+ else
+ *total = COSTS_N_INSNS (GET_MODE_SIZE (mode) > UNITS_PER_WORD ? 4 : 1);
+ return false;
+
+ case FMA:
+ *total = loongarch_fp_mult_cost (mode);
+ return false;
+
+ case MULT:
+ if (float_mode_p)
+ *total = loongarch_fp_mult_cost (mode);
+ else if (mode == DImode && !TARGET_64BIT)
+ *total = (speed
+ ? loongarch_cost->int_mult_si * 3 + 6
+ : COSTS_N_INSNS (7));
+ else if (!speed)
+ *total = COSTS_N_INSNS (1) + 1;
+ else if (mode == DImode)
+ *total = loongarch_cost->int_mult_di;
+ else
+ *total = loongarch_cost->int_mult_si;
+ return false;
+
+ case DIV:
+ /* Check for a reciprocal. */
+ if (float_mode_p
+ && flag_unsafe_math_optimizations
+ && XEXP (x, 0) == CONST1_RTX (mode))
+ {
+ if (outer_code == SQRT || GET_CODE (XEXP (x, 1)) == SQRT)
+ /* An rsqrt<mode>a or rsqrt<mode>b pattern. Count the
+ division as being free. */
+ *total = set_src_cost (XEXP (x, 1), mode, speed);
+ else
+ *total = (loongarch_fp_div_cost (mode)
+ + set_src_cost (XEXP (x, 1), mode, speed));
+ return true;
+ }
+ /* Fall through. */
+
+ case SQRT:
+ case MOD:
+ if (float_mode_p)
+ {
+ *total = loongarch_fp_div_cost (mode);
+ return false;
+ }
+ /* Fall through. */
+
+ case UDIV:
+ case UMOD:
+ if (!speed)
+ {
+ *total = COSTS_N_INSNS (loongarch_idiv_insns (mode));
+ }
+ else if (mode == DImode)
+ *total = loongarch_cost->int_div_di;
+ else
+ *total = loongarch_cost->int_div_si;
+ return false;
+
+ case SIGN_EXTEND:
+ *total = loongarch_sign_extend_cost (mode, XEXP (x, 0));
+ return false;
+
+ case ZERO_EXTEND:
+ *total = loongarch_zero_extend_cost (mode, XEXP (x, 0));
+ return false;
+ case TRUNCATE:
+ /* Costings for highpart multiplies. Matching patterns of the form:
+
+ (lshiftrt:DI (mult:DI (sign_extend:DI (...)
+ (sign_extend:DI (...))
+ (const_int 32)
+ */
+ if ((GET_CODE (XEXP (x, 0)) == ASHIFTRT
+ || GET_CODE (XEXP (x, 0)) == LSHIFTRT)
+ && CONST_INT_P (XEXP (XEXP (x, 0), 1))
+ && ((INTVAL (XEXP (XEXP (x, 0), 1)) == 32
+ && GET_MODE (XEXP (x, 0)) == DImode)
+ || (TARGET_64BIT
+ && INTVAL (XEXP (XEXP (x, 0), 1)) == 64
+ && GET_MODE (XEXP (x, 0)) == TImode))
+ && GET_CODE (XEXP (XEXP (x, 0), 0)) == MULT
+ && ((GET_CODE (XEXP (XEXP (XEXP (x, 0), 0), 0)) == SIGN_EXTEND
+ && GET_CODE (XEXP (XEXP (XEXP (x, 0), 0), 1)) == SIGN_EXTEND)
+ || (GET_CODE (XEXP (XEXP (XEXP (x, 0), 0), 0)) == ZERO_EXTEND
+ && (GET_CODE (XEXP (XEXP (XEXP (x, 0), 0), 1))
+ == ZERO_EXTEND))))
+ {
+ if (!speed)
+ *total = COSTS_N_INSNS (1) + 1;
+ else if (mode == DImode)
+ *total = loongarch_cost->int_mult_di;
+ else
+ *total = loongarch_cost->int_mult_si;
+
+ /* Sign extension is free, zero extension costs for DImode when
+ on a 64bit core / when DMUL is present. */
+ for (int i = 0; i < 2; ++i)
+ {
+ rtx op = XEXP (XEXP (XEXP (x, 0), 0), i);
+ if (TARGET_64BIT
+ && GET_CODE (op) == ZERO_EXTEND
+ && GET_MODE (op) == DImode)
+ *total += rtx_cost (op, DImode, MULT, i, speed);
+ else
+ *total += rtx_cost (XEXP (op, 0), VOIDmode, GET_CODE (op), 0,
+ speed);
+ }
+
+ return true;
+ }
+ return false;
+
+ case FLOAT:
+ case UNSIGNED_FLOAT:
+ case FIX:
+ case FLOAT_EXTEND:
+ case FLOAT_TRUNCATE:
+ *total = loongarch_cost->fp_add;
+ return false;
+
+ case SET:
+ if (register_operand (SET_DEST (x), VOIDmode)
+ && reg_or_0_operand (SET_SRC (x), VOIDmode))
+ {
+ *total = loongarch_set_reg_reg_cost (GET_MODE (SET_DEST (x)));
+ return true;
+ }
+ return false;
+
+ default:
+ return false;
+ }
+}
+
+/* Implement TARGET_ADDRESS_COST. */
+
+static int
+loongarch_address_cost (rtx addr, machine_mode mode,
+ addr_space_t as ATTRIBUTE_UNUSED,
+ bool speed ATTRIBUTE_UNUSED)
+{
+ return loongarch_address_insns (addr, mode, false);
+}
+
+/* Return one word of double-word value OP, taking into account the fixed
+ endianness of certain registers. HIGH_P is true to select the high part,
+ false to select the low part. */
+
+rtx
+loongarch_subword (rtx op, bool high_p)
+{
+ unsigned int byte, offset;
+ machine_mode mode;
+
+ mode = GET_MODE (op);
+ if (mode == VOIDmode)
+ mode = TARGET_64BIT ? TImode : DImode;
+
+ if (high_p)
+ byte = UNITS_PER_WORD;
+ else
+ byte = 0;
+
+ if (FP_REG_RTX_P (op))
+ {
+ /* Paired FPRs are always ordered little-endian. */
+ offset = (UNITS_PER_WORD < UNITS_PER_HWFPVALUE ? high_p : byte != 0);
+ return gen_rtx_REG (word_mode, REGNO (op) + offset);
+ }
+
+ if (MEM_P (op))
+ return loongarch_rewrite_small_data (adjust_address (op, word_mode, byte));
+
+ return simplify_gen_subreg (word_mode, op, mode, byte);
+}
+
+/* Return true if a move from SRC to DEST should be split into two.
+ SPLIT_TYPE describes the split condition. */
+
+bool
+loongarch_split_move_p (rtx dest, rtx src,
+ enum loongarch_split_type split_type ATTRIBUTE_UNUSED)
+{
+ /* FPR-to-FPR moves can be done in a single instruction, if they're
+ allowed at all. */
+ unsigned int size = GET_MODE_SIZE (GET_MODE (dest));
+ if (size == 8 && FP_REG_RTX_P (src) && FP_REG_RTX_P (dest))
+ return false;
+
+ /* Check for floating-point loads and stores. */
+ if (size == 8)
+ {
+ if (FP_REG_RTX_P (dest) && MEM_P (src))
+ return false;
+ if (FP_REG_RTX_P (src) && MEM_P (dest))
+ return false;
+ }
+ /* Otherwise split all multiword moves. */
+ return size > UNITS_PER_WORD;
+}
+
+/* Split a move from SRC to DEST, given that loongarch_split_move_p holds.
+ SPLIT_TYPE describes the split condition. */
+
+void
+loongarch_split_move (rtx dest, rtx src, enum loongarch_split_type split_type,
+ rtx insn_)
+{
+ rtx low_dest;
+
+ gcc_checking_assert (loongarch_split_move_p (dest, src, split_type));
+ if (FP_REG_RTX_P (dest) || FP_REG_RTX_P (src))
+ {
+ if (!TARGET_64BIT && GET_MODE (dest) == DImode)
+ emit_insn (gen_move_doubleword_fprdi (dest, src));
+ else if (!TARGET_64BIT && GET_MODE (dest) == DFmode)
+ emit_insn (gen_move_doubleword_fprdf (dest, src));
+ else if (TARGET_64BIT && GET_MODE (dest) == TFmode)
+ emit_insn (gen_move_doubleword_fprtf (dest, src));
+ else
+ gcc_unreachable ();
+ }
+ else
+ {
+ /* The operation can be split into two normal moves. Decide in
+ which order to do them. */
+ low_dest = loongarch_subword (dest, false);
+ if (REG_P (low_dest) && reg_overlap_mentioned_p (low_dest, src))
+ {
+ loongarch_emit_move (loongarch_subword (dest, true),
+ loongarch_subword (src, true));
+ loongarch_emit_move (low_dest, loongarch_subword (src, false));
+ }
+ else
+ {
+ loongarch_emit_move (low_dest, loongarch_subword (src, false));
+ loongarch_emit_move (loongarch_subword (dest, true),
+ loongarch_subword (src, true));
+ }
+ }
+
+ /* This is a hack. See if the next insn uses DEST and if so, see if we
+ can forward SRC for DEST. This is most useful if the next insn is a
+ simple store. */
+ rtx_insn *insn = (rtx_insn *) insn_;
+ struct loongarch_address_info addr = {};
+ if (insn)
+ {
+ rtx_insn *next = next_nonnote_nondebug_insn_bb (insn);
+ if (next)
+ {
+ rtx set = single_set (next);
+ if (set && SET_SRC (set) == dest)
+ {
+ if (MEM_P (src))
+ {
+ rtx tmp = XEXP (src, 0);
+ loongarch_classify_address (&addr, tmp, GET_MODE (tmp),
+ true);
+ if (addr.reg && !reg_overlap_mentioned_p (dest, addr.reg))
+ validate_change (next, &SET_SRC (set), src, false);
+ }
+ else
+ validate_change (next, &SET_SRC (set), src, false);
+ }
+ }
+ }
+}
+
+/* Return the split type for instruction INSN. */
+
+static enum loongarch_split_type
+loongarch_insn_split_type (rtx insn)
+{
+ basic_block bb = BLOCK_FOR_INSN (insn);
+ if (bb)
+ {
+ if (optimize_bb_for_speed_p (bb))
+ return SPLIT_FOR_SPEED;
+ else
+ return SPLIT_FOR_SIZE;
+ }
+ /* Once CFG information has been removed, we should trust the optimization
+ decisions made by previous passes and only split where necessary. */
+ return SPLIT_IF_NECESSARY;
+}
+
+/* Return true if a move from SRC to DEST in INSN should be split. */
+
+bool
+loongarch_split_move_insn_p (rtx dest, rtx src, rtx insn)
+{
+ return loongarch_split_move_p (dest, src, loongarch_insn_split_type (insn));
+}
+
+/* Split a move from SRC to DEST in INSN, given that
+ loongarch_split_move_insn_p holds. */
+
+void
+loongarch_split_move_insn (rtx dest, rtx src, rtx insn)
+{
+ loongarch_split_move (dest, src, loongarch_insn_split_type (insn), insn);
+}
+
+/* Forward declaration. Used below. */
+static HOST_WIDE_INT
+loongarch_constant_alignment (const_tree exp, HOST_WIDE_INT align);
+
+/* Return the appropriate instructions to move SRC into DEST. Assume
+ that SRC is operand 1 and DEST is operand 0. */
+
+const char *
+loongarch_output_move (rtx dest, rtx src)
+{
+ enum rtx_code dest_code = GET_CODE (dest);
+ enum rtx_code src_code = GET_CODE (src);
+ machine_mode mode = GET_MODE (dest);
+ bool dbl_p = (GET_MODE_SIZE (mode) == 8);
+
+ if (loongarch_split_move_p (dest, src, SPLIT_IF_NECESSARY))
+ return "#";
+
+ if ((src_code == REG && GP_REG_P (REGNO (src)))
+ || (src == CONST0_RTX (mode)))
+ {
+ if (dest_code == REG)
+ {
+ if (GP_REG_P (REGNO (dest)))
+ return "or\t%0,%z1,$r0";
+
+ if (FP_REG_P (REGNO (dest)))
+ return dbl_p ? "movgr2fr.d\t%0,%z1" : "movgr2fr.w\t%0,%z1";
+ }
+ if (dest_code == MEM)
+ {
+ rtx offset = XEXP (dest, 0);
+ if (GET_CODE (offset) == PLUS)
+ offset = XEXP (offset, 1);
+ switch (GET_MODE_SIZE (mode))
+ {
+ case 1:
+ return "st.b\t%z1,%0";
+ case 2:
+ return "st.h\t%z1,%0";
+ case 4:
+ if (const_arith_operand (offset, Pmode))
+ return "st.w\t%z1,%0";
+ else
+ return "stptr.w\t%z1,%0";
+ case 8:
+ if (const_arith_operand (offset, Pmode))
+ return "st.d\t%z1,%0";
+ else
+ return "stptr.d\t%z1,%0";
+ default:
+ gcc_unreachable ();
+ }
+ }
+ }
+ if (dest_code == REG && GP_REG_P (REGNO (dest)))
+ {
+ if (src_code == REG)
+ if (FP_REG_P (REGNO (src)))
+ return dbl_p ? "movfr2gr.d\t%0,%1" : "movfr2gr.s\t%0,%1";
+
+ if (src_code == MEM)
+ {
+ rtx offset = XEXP (src, 0);
+ if (GET_CODE (offset) == PLUS)
+ offset = XEXP (offset, 1);
+ switch (GET_MODE_SIZE (mode))
+ {
+ case 1:
+ return "ld.bu\t%0,%1";
+ case 2:
+ return "ld.hu\t%0,%1";
+ case 4:
+ if (const_arith_operand (offset, Pmode))
+ return "ld.w\t%0,%1";
+ else
+ return "ldptr.w\t%0,%1";
+ case 8:
+ if (const_arith_operand (offset, Pmode))
+ return "ld.d\t%0,%1";
+ else
+ return "ldptr.d\t%0,%1";
+ default:
+ gcc_unreachable ();
+ }
+ }
+
+ if (src_code == CONST_INT)
+ {
+ if (LU12I_INT (src))
+ return "lu12i.w\t%0,%1>>12\t\t\t# %X1";
+ else if (IMM12_INT (src))
+ return "addi.w\t%0,$r0,%1\t\t\t# %X1";
+ else if (IMM12_INT_UNSIGNED (src))
+ return "ori\t%0,$r0,%1\t\t\t# %X1";
+ else if (LU52I_INT (src))
+ return "lu52i.d\t%0,$r0,%X1>>52\t\t\t# %1";
+ else
+ gcc_unreachable ();
+ }
+
+ if (symbolic_operand (src, VOIDmode))
+ {
+ if ((TARGET_CMODEL_TINY && (!loongarch_global_symbol_p (src)
+ || loongarch_symbol_binds_local_p (src)))
+ || (TARGET_CMODEL_TINY_STATIC && !loongarch_weak_symbol_p (src)))
+ {
+ /* The symbol must be aligned to 4 byte. */
+ unsigned int align;
+
+ if (GET_CODE (src) == LABEL_REF)
+ align = 128 /* Whatever. */;
+ else if (CONSTANT_POOL_ADDRESS_P (src))
+ align = GET_MODE_ALIGNMENT (get_pool_mode (src));
+ else if (TREE_CONSTANT_POOL_ADDRESS_P (src))
+ {
+ tree exp = SYMBOL_REF_DECL (src);
+ align = TYPE_ALIGN (TREE_TYPE (exp));
+ align = loongarch_constant_alignment (exp, align);
+ }
+ else if (SYMBOL_REF_DECL (src))
+ align = DECL_ALIGN (SYMBOL_REF_DECL (src));
+ else if (SYMBOL_REF_HAS_BLOCK_INFO_P (src)
+ && SYMBOL_REF_BLOCK (src) != NULL)
+ align = SYMBOL_REF_BLOCK (src)->alignment;
+ else
+ align = BITS_PER_UNIT;
+
+ if (align % (4 * 8) == 0)
+ return "pcaddi\t%0,%%pcrel(%1)>>2";
+ }
+ if (TARGET_CMODEL_TINY
+ || TARGET_CMODEL_TINY_STATIC
+ || TARGET_CMODEL_NORMAL
+ || TARGET_CMODEL_LARGE)
+ {
+ if (!loongarch_global_symbol_p (src)
+ || loongarch_symbol_binds_local_p (src))
+ return "la.local\t%0,%1";
+ else
+ return "la.global\t%0,%1";
+ }
+ if (TARGET_CMODEL_EXTREME)
+ {
+ sorry ("Normal symbol loading not implemented in extreme mode.");
+ /* GCC complains. */
+ /* return ""; */
+ }
+ }
+ }
+ if (src_code == REG && FP_REG_P (REGNO (src)))
+ {
+ if (dest_code == REG && FP_REG_P (REGNO (dest)))
+ return dbl_p ? "fmov.d\t%0,%1" : "fmov.s\t%0,%1";
+
+ if (dest_code == MEM)
+ return dbl_p ? "fst.d\t%1,%0" : "fst.s\t%1,%0";
+ }
+ if (dest_code == REG && FP_REG_P (REGNO (dest)))
+ {
+ if (src_code == MEM)
+ return dbl_p ? "fld.d\t%0,%1" : "fld.s\t%0,%1";
+ }
+ gcc_unreachable ();
+}
+
+/* Return true if CMP1 is a suitable second operand for integer ordering
+ test CODE. See also the *sCC patterns in loongarch.md. */
+
+static bool
+loongarch_int_order_operand_ok_p (enum rtx_code code, rtx cmp1)
+{
+ switch (code)
+ {
+ case GT:
+ case GTU:
+ return reg_or_0_operand (cmp1, VOIDmode);
+
+ case GE:
+ case GEU:
+ return cmp1 == const1_rtx;
+
+ case LT:
+ case LTU:
+ return arith_operand (cmp1, VOIDmode);
+
+ case LE:
+ return sle_operand (cmp1, VOIDmode);
+
+ case LEU:
+ return sleu_operand (cmp1, VOIDmode);
+
+ default:
+ gcc_unreachable ();
+ }
+}
+
+/* Return true if *CMP1 (of mode MODE) is a valid second operand for
+ integer ordering test *CODE, or if an equivalent combination can
+ be formed by adjusting *CODE and *CMP1. When returning true, update
+ *CODE and *CMP1 with the chosen code and operand, otherwise leave
+ them alone. */
+
+static bool
+loongarch_canonicalize_int_order_test (enum rtx_code *code, rtx *cmp1,
+ machine_mode mode)
+{
+ HOST_WIDE_INT plus_one;
+
+ if (loongarch_int_order_operand_ok_p (*code, *cmp1))
+ return true;
+
+ if (CONST_INT_P (*cmp1))
+ switch (*code)
+ {
+ case LE:
+ plus_one = trunc_int_for_mode (UINTVAL (*cmp1) + 1, mode);
+ if (INTVAL (*cmp1) < plus_one)
+ {
+ *code = LT;
+ *cmp1 = force_reg (mode, GEN_INT (plus_one));
+ return true;
+ }
+ break;
+
+ case LEU:
+ plus_one = trunc_int_for_mode (UINTVAL (*cmp1) + 1, mode);
+ if (plus_one != 0)
+ {
+ *code = LTU;
+ *cmp1 = force_reg (mode, GEN_INT (plus_one));
+ return true;
+ }
+ break;
+
+ default:
+ break;
+ }
+ return false;
+}
+
+/* Compare CMP0 and CMP1 using ordering test CODE and store the result
+ in TARGET. CMP0 and TARGET are register_operands. If INVERT_PTR
+ is nonnull, it's OK to set TARGET to the inverse of the result and
+ flip *INVERT_PTR instead. */
+
+static void
+loongarch_emit_int_order_test (enum rtx_code code, bool *invert_ptr,
+ rtx target, rtx cmp0, rtx cmp1)
+{
+ machine_mode mode;
+
+ /* First see if there is a LoongArch instruction that can do this operation.
+ If not, try doing the same for the inverse operation. If that also
+ fails, force CMP1 into a register and try again. */
+ mode = GET_MODE (cmp0);
+ if (loongarch_canonicalize_int_order_test (&code, &cmp1, mode))
+ loongarch_emit_binary (code, target, cmp0, cmp1);
+ else
+ {
+ enum rtx_code inv_code = reverse_condition (code);
+ if (!loongarch_canonicalize_int_order_test (&inv_code, &cmp1, mode))
+ {
+ cmp1 = force_reg (mode, cmp1);
+ loongarch_emit_int_order_test (code, invert_ptr, target, cmp0, cmp1);
+ }
+ else if (invert_ptr == 0)
+ {
+ rtx inv_target;
+
+ inv_target = loongarch_force_binary (GET_MODE (target),
+ inv_code, cmp0, cmp1);
+ loongarch_emit_binary (XOR, target, inv_target, const1_rtx);
+ }
+ else
+ {
+ *invert_ptr = !*invert_ptr;
+ loongarch_emit_binary (inv_code, target, cmp0, cmp1);
+ }
+ }
+}
+
+/* Return a register that is zero if CMP0 and CMP1 are equal.
+ The register will have the same mode as CMP0. */
+
+static rtx
+loongarch_zero_if_equal (rtx cmp0, rtx cmp1)
+{
+ if (cmp1 == const0_rtx)
+ return cmp0;
+
+ if (uns_arith_operand (cmp1, VOIDmode))
+ return expand_binop (GET_MODE (cmp0), xor_optab, cmp0, cmp1, 0, 0,
+ OPTAB_DIRECT);
+
+ return expand_binop (GET_MODE (cmp0), sub_optab, cmp0, cmp1, 0, 0,
+ OPTAB_DIRECT);
+}
+
+/* Allocate a floating-point condition-code register of mode MODE. */
+
+static rtx
+loongarch_allocate_fcc (machine_mode mode)
+{
+ unsigned int regno, count;
+
+ gcc_assert (TARGET_HARD_FLOAT);
+
+ if (mode == FCCmode)
+ count = 1;
+ else
+ gcc_unreachable ();
+
+ cfun->machine->next_fcc += -cfun->machine->next_fcc & (count - 1);
+ if (cfun->machine->next_fcc > FCC_REG_LAST - FCC_REG_FIRST)
+ cfun->machine->next_fcc = 0;
+
+ regno = FCC_REG_FIRST + cfun->machine->next_fcc;
+ cfun->machine->next_fcc += count;
+ return gen_rtx_REG (mode, regno);
+}
+
+/* Sign- or zero-extend OP0 and OP1 for integer comparisons. */
+
+static void
+loongarch_extend_comparands (rtx_code code, rtx *op0, rtx *op1)
+{
+ /* Comparisons consider all XLEN bits, so extend sub-XLEN values. */
+ if (GET_MODE_SIZE (word_mode) > GET_MODE_SIZE (GET_MODE (*op0)))
+ {
+ /* TODO: checkout It is more profitable to zero-extend QImode values. */
+ if (unsigned_condition (code) == code && GET_MODE (*op0) == QImode)
+ {
+ *op0 = gen_rtx_ZERO_EXTEND (word_mode, *op0);
+ if (CONST_INT_P (*op1))
+ *op1 = GEN_INT ((uint8_t) INTVAL (*op1));
+ else
+ *op1 = gen_rtx_ZERO_EXTEND (word_mode, *op1);
+ }
+ else
+ {
+ *op0 = gen_rtx_SIGN_EXTEND (word_mode, *op0);
+ if (*op1 != const0_rtx)
+ *op1 = gen_rtx_SIGN_EXTEND (word_mode, *op1);
+ }
+ }
+}
+
+/* Convert a comparison into something that can be used in a branch. On
+ entry, *OP0 and *OP1 are the values being compared and *CODE is the code
+ used to compare them. Update them to describe the final comparison. */
+
+static void
+loongarch_emit_int_compare (enum rtx_code *code, rtx *op0, rtx *op1)
+{
+ static const enum rtx_code
+ mag_comparisons[][2] = {{LEU, LTU}, {GTU, GEU}, {LE, LT}, {GT, GE}};
+
+ if (splittable_const_int_operand (*op1, VOIDmode))
+ {
+ HOST_WIDE_INT rhs = INTVAL (*op1);
+
+ if (*code == EQ || *code == NE)
+ {
+ /* Convert e.g. OP0 == 2048 into OP0 - 2048 == 0. */
+ if (IMM12_OPERAND (-rhs))
+ {
+ *op0 = loongarch_force_binary (GET_MODE (*op0), PLUS, *op0,
+ GEN_INT (-rhs));
+ *op1 = const0_rtx;
+ }
+ }
+ else
+ {
+ /* Convert e.g. (OP0 <= 0xFFF) into (OP0 < 0x1000). */
+ for (size_t i = 0; i < ARRAY_SIZE (mag_comparisons); i++)
+ {
+ HOST_WIDE_INT new_rhs;
+ bool increment = *code == mag_comparisons[i][0];
+ bool decrement = *code == mag_comparisons[i][1];
+ if (!increment && !decrement)
+ continue;
+
+ new_rhs = rhs + (increment ? 1 : -1);
+ if (loongarch_integer_cost (new_rhs)
+ < loongarch_integer_cost (rhs)
+ && (rhs < 0) == (new_rhs < 0))
+ {
+ *op1 = GEN_INT (new_rhs);
+ *code = mag_comparisons[i][increment];
+ }
+ break;
+ }
+ }
+ }
+
+ loongarch_extend_comparands (*code, op0, op1);
+
+ *op0 = force_reg (word_mode, *op0);
+ if (*op1 != const0_rtx)
+ *op1 = force_reg (word_mode, *op1);
+}
+
+/* Like riscv_emit_int_compare, but for floating-point comparisons. */
+
+static void
+loongarch_emit_float_compare (enum rtx_code *code, rtx *op0, rtx *op1)
+{
+ rtx cmp_op0 = *op0;
+ rtx cmp_op1 = *op1;
+
+ /* Floating-point tests use a separate FCMP.cond.fmt
+ comparison to set a register. The branch or conditional move will
+ then compare that register against zero.
+
+ Set CMP_CODE to the code of the comparison instruction and
+ *CODE to the code that the branch or move should use. */
+ enum rtx_code cmp_code = *code;
+ /* Three FP conditions cannot be implemented by reversing the
+ operands for FCMP.cond.fmt, instead a reversed condition code is
+ required and a test for false. */
+ *code = NE;
+ *op0 = loongarch_allocate_fcc (FCCmode);
+
+ *op1 = const0_rtx;
+ loongarch_emit_binary (cmp_code, *op0, cmp_op0, cmp_op1);
+}
+
+/* Try performing the comparison in OPERANDS[1], whose arms are OPERANDS[2]
+ and OPERAND[3]. Store the result in OPERANDS[0].
+
+ On 64-bit targets, the mode of the comparison and target will always be
+ SImode, thus possibly narrower than that of the comparison's operands. */
+
+void
+loongarch_expand_scc (rtx operands[])
+{
+ rtx target = operands[0];
+ enum rtx_code code = GET_CODE (operands[1]);
+ rtx op0 = operands[2];
+ rtx op1 = operands[3];
+
+ gcc_assert (GET_MODE_CLASS (GET_MODE (op0)) == MODE_INT);
+
+ if (code == EQ || code == NE)
+ {
+ rtx zie = loongarch_zero_if_equal (op0, op1);
+ loongarch_emit_binary (code, target, zie, const0_rtx);
+ }
+ else
+ loongarch_emit_int_order_test (code, 0, target, op0, op1);
+}
+
+/* Compare OPERANDS[1] with OPERANDS[2] using comparison code
+ CODE and jump to OPERANDS[3] if the condition holds. */
+
+void
+loongarch_expand_conditional_branch (rtx *operands)
+{
+ enum rtx_code code = GET_CODE (operands[0]);
+ rtx op0 = operands[1];
+ rtx op1 = operands[2];
+ rtx condition;
+
+ if (FLOAT_MODE_P (GET_MODE (op1)))
+ loongarch_emit_float_compare (&code, &op0, &op1);
+ else
+ loongarch_emit_int_compare (&code, &op0, &op1);
+
+ condition = gen_rtx_fmt_ee (code, VOIDmode, op0, op1);
+ emit_jump_insn (gen_condjump (condition, operands[3]));
+}
+
+/* Perform the comparison in OPERANDS[1]. Move OPERANDS[2] into OPERANDS[0]
+ if the condition holds, otherwise move OPERANDS[3] into OPERANDS[0]. */
+
+void
+loongarch_expand_conditional_move (rtx *operands)
+{
+ enum rtx_code code = GET_CODE (operands[1]);
+ rtx op0 = XEXP (operands[1], 0);
+ rtx op1 = XEXP (operands[1], 1);
+
+ if (FLOAT_MODE_P (GET_MODE (op1)))
+ loongarch_emit_float_compare (&code, &op0, &op1);
+ else
+ {
+ /* See test-cbranchqi4.c and pr53645.c. */
+ loongarch_extend_comparands (code, &op0, &op1);
+
+ op0 = force_reg (word_mode, op0);
+
+ if (code == EQ || code == NE) /* See test-mask-1.c && test-mask-5.c. */
+ {
+ op0 = loongarch_zero_if_equal (op0, op1);
+ op1 = const0_rtx;
+ }
+ else /* See test-mask-2.c. */
+ {
+ /* The comparison needs a separate scc instruction. Store the
+ result of the scc in *OP0 and compare it against zero. */
+ bool invert = false;
+ rtx target = gen_reg_rtx (GET_MODE (op0));
+ loongarch_emit_int_order_test (code, &invert, target, op0, op1);
+ code = invert ? EQ : NE;
+ op0 = target;
+ op1 = const0_rtx;
+ }
+ }
+
+ rtx cond = gen_rtx_fmt_ee (code, GET_MODE (op0), op0, op1);
+ /* There is no direct support for general conditional GP move involving
+ two registers using SEL. See test-mask-3.c. */
+ if (INTEGRAL_MODE_P (GET_MODE (operands[2]))
+ && register_operand (operands[2], VOIDmode)
+ && register_operand (operands[3], VOIDmode))
+ {
+ machine_mode mode = GET_MODE (operands[0]);
+ rtx temp = gen_reg_rtx (mode);
+ rtx temp2 = gen_reg_rtx (mode);
+
+ emit_insn (gen_rtx_SET (temp,
+ gen_rtx_IF_THEN_ELSE (mode, cond,
+ operands[2], const0_rtx)));
+
+ /* Flip the test for the second operand. */
+ cond = gen_rtx_fmt_ee ((code == EQ) ? NE : EQ, GET_MODE (op0), op0, op1);
+
+ emit_insn (gen_rtx_SET (temp2,
+ gen_rtx_IF_THEN_ELSE (mode, cond,
+ operands[3], const0_rtx)));
+
+ /* Merge the two results, at least one is guaranteed to be zero. */
+ emit_insn (gen_rtx_SET (operands[0], gen_rtx_IOR (mode, temp, temp2)));
+ }
+ else
+ emit_insn (gen_rtx_SET (operands[0],
+ gen_rtx_IF_THEN_ELSE (GET_MODE (operands[0]), cond,
+ operands[2], operands[3])));
+}
+
+/* Implement TARGET_EXPAND_BUILTIN_VA_START. */
+
+static void
+loongarch_va_start (tree valist, rtx nextarg)
+{
+ nextarg = plus_constant (Pmode, nextarg, -cfun->machine->varargs_size);
+ std_expand_builtin_va_start (valist, nextarg);
+}
+
+/* Start a definition of function NAME. */
+
+static void
+loongarch_start_function_definition (const char *name)
+{
+ ASM_OUTPUT_TYPE_DIRECTIVE (asm_out_file, name, "function");
+
+ /* Start the definition proper. */
+ assemble_name (asm_out_file, name);
+ fputs (":\n", asm_out_file);
+}
+
+/* Implement TARGET_FUNCTION_OK_FOR_SIBCALL. */
+
+static bool
+loongarch_function_ok_for_sibcall (tree decl ATTRIBUTE_UNUSED,
+ tree exp ATTRIBUTE_UNUSED)
+{
+ /* Always OK. */
+ return true;
+}
+
+/* Implement TARGET_USE_BY_PIECES_INFRASTRUCTURE_P. */
+
+bool
+loongarch_use_by_pieces_infrastructure_p (unsigned HOST_WIDE_INT size,
+ unsigned int align,
+ enum by_pieces_operation op,
+ bool speed_p)
+{
+ if (op == STORE_BY_PIECES)
+ return loongarch_store_by_pieces_p (size, align);
+ if (op == MOVE_BY_PIECES && HAVE_cpymemsi)
+ {
+ /* cpymemsi is meant to generate code that is at least as good as
+ move_by_pieces. However, cpymemsi effectively uses a by-pieces
+ implementation both for moves smaller than a word and for
+ word-aligned moves of no more than LARCH_MAX_MOVE_BYTES_STRAIGHT
+ bytes. We should allow the tree-level optimisers to do such
+ moves by pieces, as it often exposes other optimization
+ opportunities. We might as well continue to use cpymemsi at
+ the rtl level though, as it produces better code when
+ scheduling is disabled (such as at -O). */
+ if (currently_expanding_to_rtl)
+ return false;
+ if (align < BITS_PER_WORD)
+ return size < UNITS_PER_WORD;
+ return size <= LARCH_MAX_MOVE_BYTES_STRAIGHT;
+ }
+
+ return default_use_by_pieces_infrastructure_p (size, align, op, speed_p);
+}
+
+/* Implement a handler for STORE_BY_PIECES operations
+ for TARGET_USE_BY_PIECES_INFRASTRUCTURE_P. */
+
+bool
+loongarch_store_by_pieces_p (unsigned HOST_WIDE_INT size, unsigned int align)
+{
+ /* Storing by pieces involves moving constants into registers
+ of size MIN (ALIGN, BITS_PER_WORD), then storing them.
+ We need to decide whether it is cheaper to load the address of
+ constant data into a register and use a block move instead. */
+
+ /* If the data is only byte aligned, then:
+
+ (a1) A block move of less than 4 bytes would involve three 3 LD.Bs and
+ 3 ST.Bs. We might as well use 3 single-instruction LIs and 3 SD.Bs
+ instead.
+
+ (a2) A block move of 4 bytes from aligned source data can use an
+ LD.W/ST.W sequence. This is often better than the 4 LIs and
+ 4 SD.Bs that we would generate when storing by pieces. */
+ if (align <= BITS_PER_UNIT)
+ return size < 4;
+
+ /* If the data is 2-byte aligned, then:
+
+ (b1) A block move of less than 4 bytes would use a combination of LD.Bs,
+ LD.Hs, SD.Bs and SD.Hs. We get better code by using
+ single-instruction LIs, SD.Bs and SD.Hs instead.
+
+ (b2) A block move of 4 bytes from aligned source data would again use
+ an LD.W/ST.W sequence. In most cases, loading the address of
+ the source data would require at least one extra instruction.
+ It is often more efficient to use 2 single-instruction LIs and
+ 2 SHs instead.
+
+ (b3) A block move of up to 3 additional bytes would be like (b1).
+
+ (b4) A block move of 8 bytes from aligned source data can use two
+ LD.W/ST.W sequences. Both sequences are better than the 4 LIs
+ and 4 ST.Hs that we'd generate when storing by pieces.
+
+ The reasoning for higher alignments is similar:
+
+ (c1) A block move of less than 4 bytes would be the same as (b1).
+
+ (c2) A block move of 4 bytes would use an LD.W/ST.W sequence. Again,
+ loading the address of the source data would typically require
+ at least one extra instruction. It is generally better to use
+ LU12I/ORI/SW instead.
+
+ (c3) A block move of up to 3 additional bytes would be like (b1).
+
+ (c4) A block move of 8 bytes can use two LD.W/ST.W sequences or a single
+ LD.D/ST.D sequence, and in these cases we've traditionally preferred
+ the memory copy over the more bulky constant moves. */
+ return size < 8;
+}
+
+/* Emit straight-line code to move LENGTH bytes from SRC to DEST.
+ Assume that the areas do not overlap. */
+
+static void
+loongarch_block_move_straight (rtx dest, rtx src, HOST_WIDE_INT length)
+{
+ HOST_WIDE_INT offset, delta;
+ unsigned HOST_WIDE_INT bits;
+ int i;
+ machine_mode mode;
+ rtx *regs;
+
+ bits = MIN (BITS_PER_WORD, MIN (MEM_ALIGN (src), MEM_ALIGN (dest)));
+
+ mode = int_mode_for_size (bits, 0).require ();
+ delta = bits / BITS_PER_UNIT;
+
+ /* Allocate a buffer for the temporary registers. */
+ regs = XALLOCAVEC (rtx, length / delta);
+
+ /* Load as many BITS-sized chunks as possible. Use a normal load if
+ the source has enough alignment, otherwise use left/right pairs. */
+ for (offset = 0, i = 0; offset + delta <= length; offset += delta, i++)
+ {
+ regs[i] = gen_reg_rtx (mode);
+ loongarch_emit_move (regs[i], adjust_address (src, mode, offset));
+ }
+
+ for (offset = 0, i = 0; offset + delta <= length; offset += delta, i++)
+ loongarch_emit_move (adjust_address (dest, mode, offset), regs[i]);
+
+ /* Mop up any left-over bytes. */
+ if (offset < length)
+ {
+ src = adjust_address (src, BLKmode, offset);
+ dest = adjust_address (dest, BLKmode, offset);
+ move_by_pieces (dest, src, length - offset,
+ MIN (MEM_ALIGN (src), MEM_ALIGN (dest)),
+ (enum memop_ret) 0);
+ }
+}
+
+/* Helper function for doing a loop-based block operation on memory
+ reference MEM. Each iteration of the loop will operate on LENGTH
+ bytes of MEM.
+
+ Create a new base register for use within the loop and point it to
+ the start of MEM. Create a new memory reference that uses this
+ register. Store them in *LOOP_REG and *LOOP_MEM respectively. */
+
+static void
+loongarch_adjust_block_mem (rtx mem, HOST_WIDE_INT length, rtx *loop_reg,
+ rtx *loop_mem)
+{
+ *loop_reg = copy_addr_to_reg (XEXP (mem, 0));
+
+ /* Although the new mem does not refer to a known location,
+ it does keep up to LENGTH bytes of alignment. */
+ *loop_mem = change_address (mem, BLKmode, *loop_reg);
+ set_mem_align (*loop_mem, MIN (MEM_ALIGN (mem), length * BITS_PER_UNIT));
+}
+
+/* Move LENGTH bytes from SRC to DEST using a loop that moves BYTES_PER_ITER
+ bytes at a time. LENGTH must be at least BYTES_PER_ITER. Assume that
+ the memory regions do not overlap. */
+
+static void
+loongarch_block_move_loop (rtx dest, rtx src, HOST_WIDE_INT length,
+ HOST_WIDE_INT bytes_per_iter)
+{
+ rtx_code_label *label;
+ rtx src_reg, dest_reg, final_src, test;
+ HOST_WIDE_INT leftover;
+
+ leftover = length % bytes_per_iter;
+ length -= leftover;
+
+ /* Create registers and memory references for use within the loop. */
+ loongarch_adjust_block_mem (src, bytes_per_iter, &src_reg, &src);
+ loongarch_adjust_block_mem (dest, bytes_per_iter, &dest_reg, &dest);
+
+ /* Calculate the value that SRC_REG should have after the last iteration
+ of the loop. */
+ final_src = expand_simple_binop (Pmode, PLUS, src_reg, GEN_INT (length), 0,
+ 0, OPTAB_WIDEN);
+
+ /* Emit the start of the loop. */
+ label = gen_label_rtx ();
+ emit_label (label);
+
+ /* Emit the loop body. */
+ loongarch_block_move_straight (dest, src, bytes_per_iter);
+
+ /* Move on to the next block. */
+ loongarch_emit_move (src_reg,
+ plus_constant (Pmode, src_reg, bytes_per_iter));
+ loongarch_emit_move (dest_reg,
+ plus_constant (Pmode, dest_reg, bytes_per_iter));
+
+ /* Emit the loop condition. */
+ test = gen_rtx_NE (VOIDmode, src_reg, final_src);
+ if (Pmode == DImode)
+ emit_jump_insn (gen_cbranchdi4 (test, src_reg, final_src, label));
+ else
+ emit_jump_insn (gen_cbranchsi4 (test, src_reg, final_src, label));
+
+ /* Mop up any left-over bytes. */
+ if (leftover)
+ loongarch_block_move_straight (dest, src, leftover);
+ else
+ /* Temporary fix for PR79150. */
+ emit_insn (gen_nop ());
+}
+
+/* Expand a cpymemsi instruction, which copies LENGTH bytes from
+ memory reference SRC to memory reference DEST. */
+
+bool
+loongarch_expand_block_move (rtx dest, rtx src, rtx length)
+{
+ int max_move_bytes = LARCH_MAX_MOVE_BYTES_STRAIGHT;
+
+ if (CONST_INT_P (length)
+ && INTVAL (length) <= loongarch_max_inline_memcpy_size)
+ {
+ if (INTVAL (length) <= max_move_bytes)
+ {
+ loongarch_block_move_straight (dest, src, INTVAL (length));
+ return true;
+ }
+ else if (optimize)
+ {
+ loongarch_block_move_loop (dest, src, INTVAL (length),
+ LARCH_MAX_MOVE_BYTES_PER_LOOP_ITER);
+ return true;
+ }
+ }
+ return false;
+}
+
+/* Expand a QI or HI mode atomic memory operation.
+
+ GENERATOR contains a pointer to the gen_* function that generates
+ the SI mode underlying atomic operation using masks that we
+ calculate.
+
+ RESULT is the return register for the operation. Its value is NULL
+ if unused.
+
+ MEM is the location of the atomic access.
+
+ OLDVAL is the first operand for the operation.
+
+ NEWVAL is the optional second operand for the operation. Its value
+ is NULL if unused. */
+
+void
+loongarch_expand_atomic_qihi (union loongarch_gen_fn_ptrs generator,
+ rtx result, rtx mem, rtx oldval, rtx newval,
+ rtx model)
+{
+ rtx orig_addr, memsi_addr, memsi, shift, shiftsi, unshifted_mask;
+ rtx unshifted_mask_reg, mask, inverted_mask, si_op;
+ rtx res = NULL;
+ machine_mode mode;
+
+ mode = GET_MODE (mem);
+
+ /* Compute the address of the containing SImode value. */
+ orig_addr = force_reg (Pmode, XEXP (mem, 0));
+ memsi_addr = loongarch_force_binary (Pmode, AND, orig_addr,
+ force_reg (Pmode, GEN_INT (-4)));
+
+ /* Create a memory reference for it. */
+ memsi = gen_rtx_MEM (SImode, memsi_addr);
+ set_mem_alias_set (memsi, ALIAS_SET_MEMORY_BARRIER);
+ MEM_VOLATILE_P (memsi) = MEM_VOLATILE_P (mem);
+
+ /* Work out the byte offset of the QImode or HImode value,
+ counting from the least significant byte. */
+ shift = loongarch_force_binary (Pmode, AND, orig_addr, GEN_INT (3));
+ /* Multiply by eight to convert the shift value from bytes to bits. */
+ loongarch_emit_binary (ASHIFT, shift, shift, GEN_INT (3));
+
+ /* Make the final shift an SImode value, so that it can be used in
+ SImode operations. */
+ shiftsi = force_reg (SImode, gen_lowpart (SImode, shift));
+
+ /* Set MASK to an inclusive mask of the QImode or HImode value. */
+ unshifted_mask = GEN_INT (GET_MODE_MASK (mode));
+ unshifted_mask_reg = force_reg (SImode, unshifted_mask);
+ mask = loongarch_force_binary (SImode, ASHIFT, unshifted_mask_reg, shiftsi);
+
+ /* Compute the equivalent exclusive mask. */
+ inverted_mask = gen_reg_rtx (SImode);
+ emit_insn (gen_rtx_SET (inverted_mask, gen_rtx_NOT (SImode, mask)));
+
+ /* Shift the old value into place. */
+ if (oldval != const0_rtx)
+ {
+ oldval = convert_modes (SImode, mode, oldval, true);
+ oldval = force_reg (SImode, oldval);
+ oldval = loongarch_force_binary (SImode, ASHIFT, oldval, shiftsi);
+ }
+
+ /* Do the same for the new value. */
+ if (newval && newval != const0_rtx)
+ {
+ newval = convert_modes (SImode, mode, newval, true);
+ newval = force_reg (SImode, newval);
+ newval = loongarch_force_binary (SImode, ASHIFT, newval, shiftsi);
+ }
+
+ /* Do the SImode atomic access. */
+ if (result)
+ res = gen_reg_rtx (SImode);
+
+ if (newval)
+ si_op = generator.fn_7 (res, memsi, mask, inverted_mask, oldval, newval,
+ model);
+ else if (result)
+ si_op = generator.fn_6 (res, memsi, mask, inverted_mask, oldval, model);
+ else
+ si_op = generator.fn_5 (memsi, mask, inverted_mask, oldval, model);
+
+ emit_insn (si_op);
+
+ if (result)
+ {
+ /* Shift and convert the result. */
+ loongarch_emit_binary (AND, res, res, mask);
+ loongarch_emit_binary (LSHIFTRT, res, res, shiftsi);
+ loongarch_emit_move (result, gen_lowpart (GET_MODE (result), res));
+ }
+}
+
+/* Return true if (zero_extract OP WIDTH BITPOS) can be used as the
+ source of an "ext" instruction or the destination of an "ins"
+ instruction. OP must be a register operand and the following
+ conditions must hold:
+
+ 0 <= BITPOS < GET_MODE_BITSIZE (GET_MODE (op))
+ 0 < WIDTH <= GET_MODE_BITSIZE (GET_MODE (op))
+ 0 < BITPOS + WIDTH <= GET_MODE_BITSIZE (GET_MODE (op))
+
+ Also reject lengths equal to a word as they are better handled
+ by the move patterns. */
+
+bool
+loongarch_use_ins_ext_p (rtx op, HOST_WIDE_INT width, HOST_WIDE_INT bitpos)
+{
+ if (!register_operand (op, VOIDmode)
+ || GET_MODE_BITSIZE (GET_MODE (op)) > BITS_PER_WORD)
+ return false;
+
+ if (!IN_RANGE (width, 1, GET_MODE_BITSIZE (GET_MODE (op)) - 1))
+ return false;
+
+ if (bitpos < 0 || bitpos + width > GET_MODE_BITSIZE (GET_MODE (op)))
+ return false;
+
+ return true;
+}
+
+/* Print the text for PRINT_OPERAND punctation character CH to FILE.
+ The punctuation characters are:
+
+ '.' Print the name of the register with a hard-wired zero (zero or $r0).
+ '$' Print the name of the stack pointer register (sp or $r3).
+
+ See also loongarch_init_print_operand_punct. */
+
+static void
+loongarch_print_operand_punctuation (FILE *file, int ch)
+{
+ switch (ch)
+ {
+ case '.':
+ fputs (reg_names[GP_REG_FIRST + 0], file);
+ break;
+
+ case '$':
+ fputs (reg_names[STACK_POINTER_REGNUM], file);
+ break;
+
+ default:
+ gcc_unreachable ();
+ break;
+ }
+}
+
+/* Initialize loongarch_print_operand_punct. */
+
+static void
+loongarch_init_print_operand_punct (void)
+{
+ const char *p;
+
+ for (p = ".$"; *p; p++)
+ loongarch_print_operand_punct[(unsigned char) *p] = true;
+}
+
+/* PRINT_OPERAND prefix LETTER refers to the integer branch instruction
+ associated with condition CODE. Print the condition part of the
+ opcode to FILE. */
+
+static void
+loongarch_print_int_branch_condition (FILE *file, enum rtx_code code,
+ int letter)
+{
+ switch (code)
+ {
+ case EQ:
+ case NE:
+ case GT:
+ case GE:
+ case LT:
+ case LE:
+ case GTU:
+ case GEU:
+ case LTU:
+ case LEU:
+ /* Conveniently, the LoongArch names for these conditions are the same
+ as their RTL equivalents. */
+ fputs (GET_RTX_NAME (code), file);
+ break;
+
+ default:
+ output_operand_lossage ("'%%%c' is not a valid operand prefix", letter);
+ break;
+ }
+}
+
+/* Likewise floating-point branches. */
+
+static void
+loongarch_print_float_branch_condition (FILE *file, enum rtx_code code,
+ int letter)
+{
+ switch (code)
+ {
+ case EQ:
+ fputs ("ceqz", file);
+ break;
+
+ case NE:
+ fputs ("cnez", file);
+ break;
+
+ default:
+ output_operand_lossage ("'%%%c' is not a valid operand prefix", letter);
+ break;
+ }
+}
+
+/* Implement TARGET_PRINT_OPERAND_PUNCT_VALID_P. */
+
+static bool
+loongarch_print_operand_punct_valid_p (unsigned char code)
+{
+ return loongarch_print_operand_punct[code];
+}
+
+/* Return true if a FENCE should be emitted to before a memory access to
+ implement the release portion of memory model MODEL. */
+
+static bool
+loongarch_memmodel_needs_release_fence (enum memmodel model)
+{
+ switch (model)
+ {
+ case MEMMODEL_ACQ_REL:
+ case MEMMODEL_SEQ_CST:
+ case MEMMODEL_SYNC_SEQ_CST:
+ case MEMMODEL_RELEASE:
+ case MEMMODEL_SYNC_RELEASE:
+ return true;
+
+ case MEMMODEL_ACQUIRE:
+ case MEMMODEL_CONSUME:
+ case MEMMODEL_SYNC_ACQUIRE:
+ case MEMMODEL_RELAXED:
+ return false;
+
+ default:
+ gcc_unreachable ();
+ }
+}
+
+/* Implement TARGET_PRINT_OPERAND. The LoongArch-specific operand codes are:
+
+ 'X' Print CONST_INT OP in hexadecimal format.
+ 'x' Print the low 16 bits of CONST_INT OP in hexadecimal format.
+ 'd' Print CONST_INT OP in decimal.
+ 'm' Print one less than CONST_INT OP in decimal.
+ 'y' Print exact log2 of CONST_INT OP in decimal.
+ 'C' Print the integer branch condition for comparison OP.
+ 'N' Print the inverse of the integer branch condition for comparison OP.
+ 'F' Print the FPU branch condition for comparison OP.
+ 'W' Print the inverse of the FPU branch condition for comparison OP.
+ 'T' Print 'f' for (eq:CC ...), 't' for (ne:CC ...),
+ 'z' for (eq:?I ...), 'n' for (ne:?I ...).
+ 't' Like 'T', but with the EQ/NE cases reversed
+ 'Y' Print loongarch_fp_conditions[INTVAL (OP)]
+ 'Z' Print OP and a comma for 8CC, otherwise print nothing.
+ 'z' Print $0 if OP is zero, otherwise print OP normally.
+ 'b' Print the address of a memory operand, without offset.
+ 'V' Print exact log2 of CONST_INT OP element 0 of a replicated
+ CONST_VECTOR in decimal.
+ 'A' Print a _DB suffix if the memory model requires a release.
+ 'G' Print a DBAR insn if the memory model requires a release.
+ 'i' Print i if the operand is not a register. */
+
+static void
+loongarch_print_operand (FILE *file, rtx op, int letter)
+{
+ enum rtx_code code;
+
+ if (loongarch_print_operand_punct_valid_p (letter))
+ {
+ loongarch_print_operand_punctuation (file, letter);
+ return;
+ }
+
+ gcc_assert (op);
+ code = GET_CODE (op);
+
+ switch (letter)
+ {
+ case 'X':
+ if (CONST_INT_P (op))
+ fprintf (file, HOST_WIDE_INT_PRINT_HEX, INTVAL (op));
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'x':
+ if (CONST_INT_P (op))
+ fprintf (file, HOST_WIDE_INT_PRINT_HEX, INTVAL (op) & 0xffff);
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'd':
+ if (CONST_INT_P (op))
+ fprintf (file, HOST_WIDE_INT_PRINT_DEC, INTVAL (op));
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'm':
+ if (CONST_INT_P (op))
+ fprintf (file, HOST_WIDE_INT_PRINT_DEC, INTVAL (op) - 1);
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'y':
+ if (CONST_INT_P (op))
+ {
+ int val = exact_log2 (INTVAL (op));
+ if (val != -1)
+ fprintf (file, "%d", val);
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ }
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'V':
+ if (GET_CODE (op) == CONST_VECTOR)
+ {
+ machine_mode mode = GET_MODE_INNER (GET_MODE (op));
+ unsigned HOST_WIDE_INT val = UINTVAL (CONST_VECTOR_ELT (op, 0));
+ int vlog2 = exact_log2 (val & GET_MODE_MASK (mode));
+ if (vlog2 != -1)
+ fprintf (file, "%d", vlog2);
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ }
+ else
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ break;
+
+ case 'C':
+ loongarch_print_int_branch_condition (file, code, letter);
+ break;
+
+ case 'N':
+ loongarch_print_int_branch_condition (file, reverse_condition (code),
+ letter);
+ break;
+
+ case 'F':
+ loongarch_print_float_branch_condition (file, code, letter);
+ break;
+
+ case 'W':
+ loongarch_print_float_branch_condition (file, reverse_condition (code),
+ letter);
+ break;
+
+ case 'T':
+ case 't':
+ {
+ int truth = (code == NE) == (letter == 'T');
+ fputc ("zfnt"[truth * 2 + FCC_REG_P (REGNO (XEXP (op, 0)))], file);
+ }
+ break;
+
+ case 'Y':
+ if (code == CONST_INT
+ && UINTVAL (op) < ARRAY_SIZE (loongarch_fp_conditions))
+ fputs (loongarch_fp_conditions[UINTVAL (op)], file);
+ else
+ output_operand_lossage ("'%%%c' is not a valid operand prefix",
+ letter);
+ break;
+
+ case 'Z':
+ loongarch_print_operand (file, op, 0);
+ fputc (',', file);
+ break;
+
+ case 'A':
+ if (loongarch_memmodel_needs_release_fence ((enum memmodel) INTVAL (op)))
+ fputs ("_db", file);
+ break;
+
+ case 'G':
+ if (loongarch_memmodel_needs_release_fence ((enum memmodel) INTVAL (op)))
+ fputs ("dbar\t0", file);
+ break;
+
+ case 'i':
+ if (code != REG)
+ fputs ("i", file);
+ break;
+
+ default:
+ switch (code)
+ {
+ case REG:
+ {
+ unsigned int regno = REGNO (op);
+ if (letter && letter != 'z')
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ fprintf (file, "%s", reg_names[regno]);
+ }
+ break;
+
+ case MEM:
+ if (letter == 'D')
+ output_address (GET_MODE (op),
+ plus_constant (Pmode, XEXP (op, 0), 4));
+ else if (letter == 'b')
+ {
+ gcc_assert (REG_P (XEXP (op, 0)));
+ loongarch_print_operand (file, XEXP (op, 0), 0);
+ }
+ else if (letter && letter != 'z')
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ else
+ output_address (GET_MODE (op), XEXP (op, 0));
+ break;
+
+ default:
+ if (letter == 'z' && op == CONST0_RTX (GET_MODE (op)))
+ fputs (reg_names[GP_REG_FIRST], file);
+ else if (letter && letter != 'z')
+ output_operand_lossage ("invalid use of '%%%c'", letter);
+ else
+ output_addr_const (file, loongarch_strip_unspec_address (op));
+ break;
+ }
+ }
+}
+
+/* Implement TARGET_PRINT_OPERAND_ADDRESS. */
+
+static void
+loongarch_print_operand_address (FILE *file, machine_mode /* mode */, rtx x)
+{
+ struct loongarch_address_info addr;
+
+ if (loongarch_classify_address (&addr, x, word_mode, true))
+ switch (addr.type)
+ {
+ case ADDRESS_REG:
+ fprintf (file, "%s,", reg_names[REGNO (addr.reg)]);
+ loongarch_print_operand (file, addr.offset, 0);
+ return;
+
+ case ADDRESS_CONST_INT:
+ fprintf (file, "%s,", reg_names[GP_REG_FIRST]);
+ output_addr_const (file, x);
+ return;
+
+ case ADDRESS_SYMBOLIC:
+ output_addr_const (file, loongarch_strip_unspec_address (x));
+ return;
+ }
+ if (GET_CODE (x) == CONST_INT)
+ output_addr_const (file, x);
+ else
+ gcc_unreachable ();
+}
+
+/* Implement TARGET_ASM_SELECT_RTX_SECTION. */
+
+static section *
+loongarch_select_rtx_section (machine_mode mode, rtx x,
+ unsigned HOST_WIDE_INT align)
+{
+ /* ??? Consider using mergeable small data sections. */
+ if (loongarch_rtx_constant_in_small_data_p (mode))
+ return get_named_section (NULL, ".sdata", 0);
+
+ return default_elf_select_rtx_section (mode, x, align);
+}
+
+/* Implement TARGET_ASM_FUNCTION_RODATA_SECTION.
+
+ The complication here is that, with the combination
+ !TARGET_ABSOLUTE_ABICALLS , jump tables will use
+ absolute addresses, and should therefore not be included in the
+ read-only part of a DSO. Handle such cases by selecting a normal
+ data section instead of a read-only one. The logic apes that in
+ default_function_rodata_section. */
+
+static section *
+loongarch_function_rodata_section (tree decl, bool)
+{
+ return default_function_rodata_section (decl, false);
+}
+
+/* Implement TARGET_IN_SMALL_DATA_P. */
+
+static bool
+loongarch_in_small_data_p (const_tree decl)
+{
+ int size;
+
+ if (TREE_CODE (decl) == STRING_CST || TREE_CODE (decl) == FUNCTION_DECL)
+ return false;
+
+ if (TREE_CODE (decl) == VAR_DECL && DECL_SECTION_NAME (decl) != 0)
+ {
+ const char *name;
+
+ /* Reject anything that isn't in a known small-data section. */
+ name = DECL_SECTION_NAME (decl);
+ if (strcmp (name, ".sdata") != 0 && strcmp (name, ".sbss") != 0)
+ return false;
+
+ /* If a symbol is defined externally, the assembler will use the
+ usual -G rules when deciding how to implement macros. */
+ if (!DECL_EXTERNAL (decl))
+ return true;
+ }
+
+ /* We have traditionally not treated zero-sized objects as small data,
+ so this is now effectively part of the ABI. */
+ size = int_size_in_bytes (TREE_TYPE (decl));
+ return size > 0 && size <= g_switch_value;
+}
+
+/* Implement TARGET_USE_ANCHORS_FOR_SYMBOL_P. We don't want to use
+ anchors for small data: the GP register acts as an anchor in that
+ case. We also don't want to use them for PC-relative accesses,
+ where the PC acts as an anchor. */
+
+static bool
+loongarch_use_anchors_for_symbol_p (const_rtx symbol)
+{
+ return default_use_anchors_for_symbol_p (symbol);
+}
+
+/* The LoongArch debug format wants all automatic variables and arguments
+ to be in terms of the virtual frame pointer (stack pointer before
+ any adjustment in the function), while the LoongArch linker wants
+ the frame pointer to be the stack pointer after the initial
+ adjustment. So, we do the adjustment here. The arg pointer (which
+ is eliminated) points to the virtual frame pointer, while the frame
+ pointer (which may be eliminated) points to the stack pointer after
+ the initial adjustments. */
+
+HOST_WIDE_INT
+loongarch_debugger_offset (rtx addr, HOST_WIDE_INT offset)
+{
+ rtx offset2 = const0_rtx;
+ rtx reg = eliminate_constant_term (addr, &offset2);
+
+ if (offset == 0)
+ offset = INTVAL (offset2);
+
+ if (reg == stack_pointer_rtx
+ || reg == frame_pointer_rtx
+ || reg == hard_frame_pointer_rtx)
+ {
+ offset -= cfun->machine->frame.total_size;
+ if (reg == hard_frame_pointer_rtx)
+ offset += cfun->machine->frame.hard_frame_pointer_offset;
+ }
+
+ return offset;
+}
+
+/* Implement ASM_OUTPUT_EXTERNAL. */
+
+void
+loongarch_output_external (FILE *file, tree decl, const char *name)
+{
+ default_elf_asm_output_external (file, decl, name);
+
+ /* We output the name if and only if TREE_SYMBOL_REFERENCED is
+ set in order to avoid putting out names that are never really
+ used. */
+ if (TREE_SYMBOL_REFERENCED (DECL_ASSEMBLER_NAME (decl)))
+ {
+ if (loongarch_in_small_data_p (decl))
+ {
+ /* When using assembler macros, emit .extern directives for
+ all small-data externs so that the assembler knows how
+ big they are.
+
+ In most cases it would be safe (though pointless) to emit
+ .externs for other symbols too. One exception is when an
+ object is within the -G limit but declared by the user to
+ be in a section other than .sbss or .sdata. */
+ fputs ("\t.extern\t", file);
+ assemble_name (file, name);
+ fprintf (file, ", " HOST_WIDE_INT_PRINT_DEC "\n",
+ int_size_in_bytes (TREE_TYPE (decl)));
+ }
+ }
+}
+
+/* Implement TARGET_ASM_OUTPUT_DWARF_DTPREL. */
+
+static void ATTRIBUTE_UNUSED
+loongarch_output_dwarf_dtprel (FILE *file, int size, rtx x)
+{
+ switch (size)
+ {
+ case 4:
+ fputs ("\t.dtprelword\t", file);
+ break;
+
+ case 8:
+ fputs ("\t.dtpreldword\t", file);
+ break;
+
+ default:
+ gcc_unreachable ();
+ }
+ output_addr_const (file, x);
+ fputs ("+0x8000", file);
+}
+
+/* Implement TARGET_DWARF_FRAME_REG_MODE. */
+
+static machine_mode
+loongarch_dwarf_frame_reg_mode (int regno)
+{
+ machine_mode mode = default_dwarf_frame_reg_mode (regno);
+
+ return mode;
+}
+
+/* Implement ASM_OUTPUT_ASCII. */
+
+void
+loongarch_output_ascii (FILE *stream, const char *string, size_t len)
+{
+ size_t i;
+ int cur_pos;
+
+ cur_pos = 17;
+ fprintf (stream, "\t.ascii\t\"");
+ for (i = 0; i < len; i++)
+ {
+ int c;
+
+ c = (unsigned char) string[i];
+ if (ISPRINT (c))
+ {
+ if (c == '\\' || c == '\"')
+ {
+ putc ('\\', stream);
+ cur_pos++;
+ }
+ putc (c, stream);
+ cur_pos++;
+ }
+ else
+ {
+ fprintf (stream, "\\%03o", c);
+ cur_pos += 4;
+ }
+
+ if (cur_pos > 72 && i + 1 < len)
+ {
+ cur_pos = 17;
+ fprintf (stream, "\"\n\t.ascii\t\"");
+ }
+ }
+ fprintf (stream, "\"\n");
+}
+
+/* Emit either a label, .comm, or .lcomm directive. When using assembler
+ macros, mark the symbol as written so that loongarch_asm_output_external
+ won't emit an .extern for it. STREAM is the output file, NAME is the
+ name of the symbol, INIT_STRING is the string that should be written
+ before the symbol and FINAL_STRING is the string that should be
+ written after it. FINAL_STRING is a printf format that consumes the
+ remaining arguments. */
+
+void
+loongarch_declare_object (FILE *stream, const char *name,
+ const char *init_string, const char *final_string,
+ ...)
+{
+ va_list ap;
+
+ fputs (init_string, stream);
+ assemble_name (stream, name);
+ va_start (ap, final_string);
+ vfprintf (stream, final_string, ap);
+ va_end (ap);
+
+ tree name_tree = get_identifier (name);
+ TREE_ASM_WRITTEN (name_tree) = 1;
+}
+
+/* Declare a common object of SIZE bytes using asm directive INIT_STRING.
+ NAME is the name of the object and ALIGN is the required alignment
+ in bytes. TAKES_ALIGNMENT_P is true if the directive takes a third
+ alignment argument. */
+
+void
+loongarch_declare_common_object (FILE *stream, const char *name,
+ const char *init_string,
+ unsigned HOST_WIDE_INT size,
+ unsigned int align, bool takes_alignment_p)
+{
+ if (!takes_alignment_p)
+ {
+ size += (align / BITS_PER_UNIT) - 1;
+ size -= size % (align / BITS_PER_UNIT);
+ loongarch_declare_object (stream, name, init_string,
+ "," HOST_WIDE_INT_PRINT_UNSIGNED "\n", size);
+ }
+ else
+ loongarch_declare_object (stream, name, init_string,
+ "," HOST_WIDE_INT_PRINT_UNSIGNED ",%u\n", size,
+ align / BITS_PER_UNIT);
+}
+
+/* Implement ASM_OUTPUT_ALIGNED_DECL_COMMON. This is usually the same as the
+ elfos.h version. */
+
+void
+loongarch_output_aligned_decl_common (FILE *stream,
+ tree decl ATTRIBUTE_UNUSED,
+ const char *name,
+ unsigned HOST_WIDE_INT size,
+ unsigned int align)
+{
+ loongarch_declare_common_object (stream, name, "\n\t.comm\t", size, align,
+ true);
+}
+
+#ifdef ASM_OUTPUT_SIZE_DIRECTIVE
+extern int size_directive_output;
+
+/* Implement ASM_DECLARE_OBJECT_NAME. This is like most of the standard ELF
+ definitions except that it uses loongarch_declare_object to emit the label.
+*/
+
+void
+loongarch_declare_object_name (FILE *stream, const char *name,
+ tree decl ATTRIBUTE_UNUSED)
+{
+#ifdef ASM_OUTPUT_TYPE_DIRECTIVE
+#ifdef USE_GNU_UNIQUE_OBJECT
+ /* As in elfos.h. */
+ if (USE_GNU_UNIQUE_OBJECT && DECL_ONE_ONLY (decl)
+ && (!DECL_ARTIFICIAL (decl) || !TREE_READONLY (decl)))
+ ASM_OUTPUT_TYPE_DIRECTIVE (stream, name, "gnu_unique_object");
+ else
+#endif
+ ASM_OUTPUT_TYPE_DIRECTIVE (stream, name, "object");
+#endif
+
+ size_directive_output = 0;
+ if (!flag_inhibit_size_directive && DECL_SIZE (decl))
+ {
+ HOST_WIDE_INT size;
+
+ size_directive_output = 1;
+ size = int_size_in_bytes (TREE_TYPE (decl));
+ ASM_OUTPUT_SIZE_DIRECTIVE (stream, name, size);
+ }
+
+ loongarch_declare_object (stream, name, "", ":\n");
+}
+
+/* Implement ASM_FINISH_DECLARE_OBJECT. This is generic ELF stuff. */
+
+void
+loongarch_finish_declare_object (FILE *stream, tree decl, int top_level,
+ int at_end)
+{
+ const char *name;
+
+ name = XSTR (XEXP (DECL_RTL (decl), 0), 0);
+ if (!flag_inhibit_size_directive
+ && DECL_SIZE (decl) != 0
+ && !at_end
+ && top_level
+ && DECL_INITIAL (decl) == error_mark_node
+ && !size_directive_output)
+ {
+ HOST_WIDE_INT size;
+
+ size_directive_output = 1;
+ size = int_size_in_bytes (TREE_TYPE (decl));
+ ASM_OUTPUT_SIZE_DIRECTIVE (stream, name, size);
+ }
+}
+#endif
+
+/* Mark text contents as code or data, mainly for the purpose of correct
+ disassembly. Emit a local symbol and set its type appropriately for
+ that purpose. */
+
+void
+loongarch_set_text_contents_type (FILE *file ATTRIBUTE_UNUSED,
+ const char *prefix ATTRIBUTE_UNUSED,
+ unsigned long num ATTRIBUTE_UNUSED,
+ bool function_p ATTRIBUTE_UNUSED)
+{
+#ifdef ASM_OUTPUT_TYPE_DIRECTIVE
+ char buf[(sizeof (num) * 10) / 4 + 2];
+ const char *fnname;
+ char *sname;
+ rtx symbol;
+
+ sprintf (buf, "%lu", num);
+ symbol = XEXP (DECL_RTL (current_function_decl), 0);
+ fnname = targetm.strip_name_encoding (XSTR (symbol, 0));
+ sname = ACONCAT ((prefix, fnname, "_", buf, NULL));
+
+ ASM_OUTPUT_TYPE_DIRECTIVE (file, sname, function_p ? "function" : "object");
+ assemble_name (file, sname);
+ fputs (":\n", file);
+#endif
+}
+
+/* Implement TARGET_ASM_FILE_START. */
+
+static void
+loongarch_file_start (void)
+{
+ default_file_start ();
+}
+
+
+/* Implement TARGET_FRAME_POINTER_REQUIRED. */
+
+static bool
+loongarch_frame_pointer_required (void)
+{
+ /* If the function contains dynamic stack allocations, we need to
+ use the frame pointer to access the static parts of the frame. */
+ if (cfun->calls_alloca)
+ return true;
+
+ return false;
+}
+
+/* Implement TARGET_CAN_ELIMINATE. Make sure that we're not trying
+ to eliminate to the wrong hard frame pointer. */
+
+static bool
+loongarch_can_eliminate (const int from ATTRIBUTE_UNUSED, const int to)
+{
+ return (to == HARD_FRAME_POINTER_REGNUM || to == STACK_POINTER_REGNUM);
+}
+
+/* Implement RETURN_ADDR_RTX. We do not support moving back to a
+ previous frame. */
+
+rtx
+loongarch_return_addr (int count, rtx frame ATTRIBUTE_UNUSED)
+{
+ if (count != 0)
+ return const0_rtx;
+
+ return get_hard_reg_initial_val (Pmode, RETURN_ADDR_REGNUM);
+}
+
+/* Emit code to change the current function's return address to
+ ADDRESS. SCRATCH is available as a scratch register, if needed.
+ ADDRESS and SCRATCH are both word-mode GPRs. */
+
+void
+loongarch_set_return_address (rtx address, rtx scratch)
+{
+ rtx slot_address;
+
+ gcc_assert (BITSET_P (cfun->machine->frame.mask, RETURN_ADDR_REGNUM));
+
+ if (frame_pointer_needed)
+ slot_address = loongarch_add_offset (scratch, hard_frame_pointer_rtx,
+ -UNITS_PER_WORD);
+ else
+ slot_address = loongarch_add_offset (scratch, stack_pointer_rtx,
+ cfun->machine->frame.gp_sp_offset);
+
+ loongarch_emit_move (gen_frame_mem (GET_MODE (address), slot_address),
+ address);
+}
+
+/* Implement ASM_DECLARE_FUNCTION_NAME. */
+
+void
+loongarch_declare_function_name (FILE *stream ATTRIBUTE_UNUSED,
+ const char *name,
+ tree fndecl ATTRIBUTE_UNUSED)
+{
+ loongarch_start_function_definition (name);
+}
+
+/* Return true if register REGNO can store a value of mode MODE.
+ The result of this function is cached in loongarch_hard_regno_mode_ok. */
+
+static bool
+loongarch_hard_regno_mode_ok_uncached (unsigned int regno, machine_mode mode)
+{
+ unsigned int size;
+ enum mode_class mclass;
+
+ if (mode == FCCmode)
+ return FCC_REG_P (regno);
+
+ size = GET_MODE_SIZE (mode);
+ mclass = GET_MODE_CLASS (mode);
+
+ if (GP_REG_P (regno))
+ return ((regno - GP_REG_FIRST) & 1) == 0 || size <= UNITS_PER_WORD;
+
+ if (FP_REG_P (regno)
+ && (((regno - FP_REG_FIRST) % MAX_FPRS_PER_FMT) == 0
+ || (MIN_FPRS_PER_FMT == 1 && size <= UNITS_PER_FPREG)))
+ {
+ if (mclass == MODE_FLOAT
+ || mclass == MODE_COMPLEX_FLOAT
+ || mclass == MODE_VECTOR_FLOAT)
+ return size <= UNITS_PER_FPVALUE;
+
+ /* Allow integer modes that fit into a single register. We need
+ to put integers into FPRs when using instructions like CVT
+ and TRUNC. There's no point allowing sizes smaller than a word,
+ because the FPU has no appropriate load/store instructions. */
+ if (mclass == MODE_INT)
+ return size >= MIN_UNITS_PER_WORD && size <= UNITS_PER_FPREG;
+ }
+
+ return false;
+}
+
+/* Implement TARGET_HARD_REGNO_MODE_OK. */
+
+static bool
+loongarch_hard_regno_mode_ok (unsigned int regno, machine_mode mode)
+{
+ return loongarch_hard_regno_mode_ok_p[mode][regno];
+}
+
+/* Return nonzero if register OLD_REG can be renamed to register NEW_REG. */
+
+bool
+loongarch_hard_regno_rename_ok (unsigned int old_reg ATTRIBUTE_UNUSED,
+ unsigned int new_reg ATTRIBUTE_UNUSED)
+{
+ return true;
+}
+
+/* Implement TARGET_HARD_REGNO_NREGS. */
+
+static unsigned int
+loongarch_hard_regno_nregs (unsigned int regno, machine_mode mode)
+{
+ if (FCC_REG_P (regno))
+ /* The size of FP status registers is always 4, because they only hold
+ FCCmode values, and FCCmode is always considered to be 4 bytes wide. */
+ return (GET_MODE_SIZE (mode) + 3) / 4;
+
+ if (FP_REG_P (regno))
+ return (GET_MODE_SIZE (mode) + UNITS_PER_FPREG - 1) / UNITS_PER_FPREG;
+
+ /* All other registers are word-sized. */
+ return (GET_MODE_SIZE (mode) + UNITS_PER_WORD - 1) / UNITS_PER_WORD;
+}
+
+/* Implement CLASS_MAX_NREGS, taking the maximum of the cases
+ in loongarch_hard_regno_nregs. */
+
+int
+loongarch_class_max_nregs (enum reg_class rclass, machine_mode mode)
+{
+ int size;
+ HARD_REG_SET left;
+
+ size = 0x8000;
+ left = reg_class_contents[rclass];
+ if (hard_reg_set_intersect_p (left, reg_class_contents[(int) FCC_REGS]))
+ {
+ if (loongarch_hard_regno_mode_ok (FCC_REG_FIRST, mode))
+ size = MIN (size, 4);
+
+ left &= ~reg_class_contents[FCC_REGS];
+ }
+ if (hard_reg_set_intersect_p (left, reg_class_contents[(int) FP_REGS]))
+ {
+ if (loongarch_hard_regno_mode_ok (FP_REG_FIRST, mode))
+ size = MIN (size, UNITS_PER_FPREG);
+
+ left &= ~reg_class_contents[FP_REGS];
+ }
+ if (!hard_reg_set_empty_p (left))
+ size = MIN (size, UNITS_PER_WORD);
+ return (GET_MODE_SIZE (mode) + size - 1) / size;
+}
+
+/* Implement TARGET_CAN_CHANGE_MODE_CLASS. */
+
+static bool
+loongarch_can_change_mode_class (machine_mode from, machine_mode to,
+ reg_class_t rclass)
+{
+ /* Allow conversions between different Loongson integer vectors,
+ and between those vectors and DImode. */
+ if (GET_MODE_SIZE (from) == 8 && GET_MODE_SIZE (to) == 8
+ && INTEGRAL_MODE_P (from) && INTEGRAL_MODE_P (to))
+ return true;
+
+ return !reg_classes_intersect_p (FP_REGS, rclass);
+}
+
+/* Return true if moves in mode MODE can use the FPU's fmov.fmt instruction,
+*/
+
+static bool
+loongarch_mode_ok_for_mov_fmt_p (machine_mode mode)
+{
+ switch (mode)
+ {
+ case E_FCCmode:
+ case E_SFmode:
+ return TARGET_HARD_FLOAT;
+
+ case E_DFmode:
+ return TARGET_HARD_FLOAT && TARGET_DOUBLE_FLOAT;
+
+ case E_V2SFmode:
+ return 0;
+
+ default:
+ return 0;
+ }
+}
+
+/* Implement TARGET_MODES_TIEABLE_P. */
+
+static bool
+loongarch_modes_tieable_p (machine_mode mode1, machine_mode mode2)
+{
+ /* FPRs allow no mode punning, so it's not worth tying modes if we'd
+ prefer to put one of them in FPRs. */
+ return (mode1 == mode2
+ || (!loongarch_mode_ok_for_mov_fmt_p (mode1)
+ && !loongarch_mode_ok_for_mov_fmt_p (mode2)));
+}
+
+/* Implement TARGET_PREFERRED_RELOAD_CLASS. */
+
+static reg_class_t
+loongarch_preferred_reload_class (rtx x, reg_class_t rclass)
+{
+ if (reg_class_subset_p (FP_REGS, rclass)
+ && loongarch_mode_ok_for_mov_fmt_p (GET_MODE (x)))
+ return FP_REGS;
+
+ if (reg_class_subset_p (GR_REGS, rclass))
+ rclass = GR_REGS;
+
+ return rclass;
+}
+
+/* RCLASS is a class involved in a REGISTER_MOVE_COST calculation.
+ Return a "canonical" class to represent it in later calculations. */
+
+static reg_class_t
+loongarch_canonicalize_move_class (reg_class_t rclass)
+{
+ if (reg_class_subset_p (rclass, GENERAL_REGS))
+ rclass = GENERAL_REGS;
+
+ return rclass;
+}
+
+/* Return the cost of moving a value from a register of class FROM to a GPR.
+ Return 0 for classes that are unions of other classes handled by this
+ function. */
+
+static int
+loongarch_move_to_gpr_cost (reg_class_t from)
+{
+ switch (from)
+ {
+ case GENERAL_REGS:
+ /* MOVE macro. */
+ return 2;
+
+ case FP_REGS:
+ /* MOVFR2GR, etc. */
+ return 4;
+
+ default:
+ return 0;
+ }
+}
+
+/* Return the cost of moving a value from a GPR to a register of class TO.
+ Return 0 for classes that are unions of other classes handled by this
+ function. */
+
+static int
+loongarch_move_from_gpr_cost (reg_class_t to)
+{
+ switch (to)
+ {
+ case GENERAL_REGS:
+ /*MOVE macro. */
+ return 2;
+
+ case FP_REGS:
+ /* MOVGR2FR, etc. */
+ return 4;
+
+ default:
+ return 0;
+ }
+}
+
+/* Implement TARGET_REGISTER_MOVE_COST. Return 0 for classes that are the
+ maximum of the move costs for subclasses; regclass will work out
+ the maximum for us. */
+
+static int
+loongarch_register_move_cost (machine_mode mode, reg_class_t from,
+ reg_class_t to)
+{
+ reg_class_t dregs;
+ int cost1, cost2;
+
+ from = loongarch_canonicalize_move_class (from);
+ to = loongarch_canonicalize_move_class (to);
+
+ /* Handle moves that can be done without using general-purpose registers. */
+ if (from == FP_REGS)
+ {
+ if (to == FP_REGS && loongarch_mode_ok_for_mov_fmt_p (mode))
+ /* FMOV.FMT. */
+ return 4;
+ }
+
+ /* Handle cases in which only one class deviates from the ideal. */
+ dregs = GENERAL_REGS;
+ if (from == dregs)
+ return loongarch_move_from_gpr_cost (to);
+ if (to == dregs)
+ return loongarch_move_to_gpr_cost (from);
+
+ /* Handles cases that require a GPR temporary. */
+ cost1 = loongarch_move_to_gpr_cost (from);
+ if (cost1 != 0)
+ {
+ cost2 = loongarch_move_from_gpr_cost (to);
+ if (cost2 != 0)
+ return cost1 + cost2;
+ }
+
+ return 0;
+}
+
+/* Implement TARGET_MEMORY_MOVE_COST. */
+
+static int
+loongarch_memory_move_cost (machine_mode mode, reg_class_t rclass, bool in)
+{
+ return (loongarch_cost->memory_latency
+ + memory_move_secondary_cost (mode, rclass, in));
+}
+
+/* Implement TARGET_SECONDARY_MEMORY_NEEDED.
+
+ This can be achieved using MOVFRH2GR.S/MOVGR2FRH.W when these instructions
+ are available but otherwise moves must go via memory. Using
+ MOVGR2FR/MOVFR2GR to access the lower-half of these registers would require
+ a forbidden single-precision access. We require all double-word moves to
+ use memory because adding even and odd floating-point registers classes
+ would have a significant impact on the backend. */
+
+static bool
+loongarch_secondary_memory_needed (machine_mode mode ATTRIBUTE_UNUSED,
+ reg_class_t class1, reg_class_t class2)
+{
+ /* Ignore spilled pseudos. */
+ if (lra_in_progress && (class1 == NO_REGS || class2 == NO_REGS))
+ return false;
+
+ return false;
+}
+
+/* Return the register class required for a secondary register when
+ copying between one of the registers in RCLASS and value X, which
+ has mode MODE. X is the source of the move if IN_P, otherwise it
+ is the destination. Return NO_REGS if no secondary register is
+ needed. */
+
+static reg_class_t
+loongarch_secondary_reload (bool in_p ATTRIBUTE_UNUSED, rtx x,
+ reg_class_t rclass, machine_mode mode,
+ secondary_reload_info *sri ATTRIBUTE_UNUSED)
+{
+ int regno;
+
+ regno = true_regnum (x);
+
+ /* Copying from accumulator registers to anywhere other than a general
+ register requires a temporary general register. */
+ if (reg_class_subset_p (rclass, FP_REGS))
+ {
+ if (regno < 0
+ || (MEM_P (x)
+ && (GET_MODE_SIZE (mode) == 4 || GET_MODE_SIZE (mode) == 8)))
+ /* In this case we can use fld.s, fst.s, fld.d or fst.d. */
+ return NO_REGS;
+
+ if (GP_REG_P (regno) || x == CONST0_RTX (mode))
+ /* In this case we can use movgr2fr.s, movfr2gr.s, movgr2fr.d or
+ * movfr2gr.d. */
+ return NO_REGS;
+
+ if (CONSTANT_P (x) && !targetm.cannot_force_const_mem (mode, x))
+ /* We can force the constant to memory and use fld.s
+ and fld.d. As above, we will use pairs of lwc1s if
+ ldc1 is not supported. */
+ return NO_REGS;
+
+ if (FP_REG_P (regno) && loongarch_mode_ok_for_mov_fmt_p (mode))
+ /* In this case we can use fmov.{s/d}. */
+ return NO_REGS;
+
+ /* Otherwise, we need to reload through an integer register. */
+ return GR_REGS;
+ }
+ if (FP_REG_P (regno))
+ return reg_class_subset_p (rclass, GR_REGS) ? NO_REGS : GR_REGS;
+
+ return NO_REGS;
+}
+
+/* Implement TARGET_VALID_POINTER_MODE. */
+
+static bool
+loongarch_valid_pointer_mode (scalar_int_mode mode)
+{
+ return mode == SImode || (TARGET_64BIT && mode == DImode);
+}
+
+/* Implement TARGET_SCALAR_MODE_SUPPORTED_P. */
+
+static bool
+loongarch_scalar_mode_supported_p (scalar_mode mode)
+{
+ if (ALL_FIXED_POINT_MODE_P (mode)
+ && GET_MODE_PRECISION (mode) <= 2 * BITS_PER_WORD)
+ return true;
+
+ return default_scalar_mode_supported_p (mode);
+}
+
+/* Implement TARGET_VECTORIZE_PREFERRED_SIMD_MODE. */
+
+static machine_mode
+loongarch_preferred_simd_mode (scalar_mode mode ATTRIBUTE_UNUSED)
+{
+ return word_mode;
+}
+
+/* Return the length of INSN. LENGTH is the initial length computed by
+ attributes in the machine-description file. */
+
+int
+loongarch_adjust_insn_length (rtx_insn *insn, int length)
+{
+ /* loongarch.md uses MAX_PIC_BRANCH_LENGTH as a placeholder for the length
+ of a PIC long-branch sequence. Substitute the correct value. */
+ if (length == MAX_PIC_BRANCH_LENGTH
+ && JUMP_P (insn)
+ && INSN_CODE (insn) >= 0
+ && get_attr_type (insn) == TYPE_BRANCH)
+ {
+ /* Add the branch-over instruction and its delay slot, if this
+ is a conditional branch. */
+ length = simplejump_p (insn) ? 0 : 8;
+
+ /* Add the length of an indirect jump, ignoring the delay slot. */
+ length += 4;
+ }
+
+ /* A unconditional jump has an unfilled delay slot if it is not part
+ of a sequence. A conditional jump normally has a delay slot. */
+ if (CALL_P (insn) || (JUMP_P (insn)))
+ length += 4;
+
+ /* See how many nops might be needed to avoid hardware hazards. */
+ if (!cfun->machine->ignore_hazard_length_p
+ && INSN_P (insn)
+ && INSN_CODE (insn) >= 0)
+ switch (get_attr_hazard (insn))
+ {
+ case HAZARD_NONE:
+ break;
+
+ case HAZARD_DELAY:
+ case HAZARD_FORBIDDEN_SLOT:
+ /* nop insn length is 4 bytes. */
+ length += 4;
+ break;
+ }
+
+ return length;
+}
+
+/* Return the assembly code for INSN, which has the operands given by
+ OPERANDS, and which branches to OPERANDS[0] if some condition is true.
+ BRANCH_IF_TRUE is the asm template that should be used if OPERANDS[0]
+ is in range of a direct branch. BRANCH_IF_FALSE is an inverted
+ version of BRANCH_IF_TRUE. */
+
+const char *
+loongarch_output_conditional_branch (rtx_insn *insn, rtx *operands,
+ const char *branch_if_true,
+ const char *branch_if_false)
+{
+ unsigned int length;
+ rtx taken;
+
+ gcc_assert (LABEL_P (operands[0]));
+
+ length = get_attr_length (insn);
+ if (length <= 12)
+ {
+ return branch_if_true;
+ }
+
+ /* Generate a reversed branch around a direct jump. */
+ rtx_code_label *not_taken = gen_label_rtx ();
+ taken = operands[0];
+
+ /* Generate the reversed branch to NOT_TAKEN. */
+ operands[0] = not_taken;
+ output_asm_insn (branch_if_false, operands);
+
+ /* If INSN has a delay slot, we must provide delay slots for both the
+ branch to NOT_TAKEN and the conditional jump. We must also ensure
+ that INSN's delay slot is executed in the appropriate cases. */
+ if (final_sequence)
+ {
+ /* This first delay slot will always be executed, so use INSN's
+ delay slot if is not annulled. */
+ if (!INSN_ANNULLED_BRANCH_P (insn))
+ {
+ final_scan_insn (final_sequence->insn (1), asm_out_file, optimize, 1,
+ NULL);
+ final_sequence->insn (1)->set_deleted ();
+ }
+ fprintf (asm_out_file, "\n");
+ }
+
+ output_asm_insn ("b\t%0", &taken);
+
+ /* Now deal with its delay slot; see above. */
+ if (final_sequence)
+ {
+ /* This delay slot will only be executed if the branch is taken.
+ Use INSN's delay slot if is annulled. */
+ if (INSN_ANNULLED_BRANCH_P (insn))
+ {
+ final_scan_insn (final_sequence->insn (1), asm_out_file, optimize, 1,
+ NULL);
+ final_sequence->insn (1)->set_deleted ();
+ }
+ fprintf (asm_out_file, "\n");
+ }
+
+ /* Output NOT_TAKEN. */
+ targetm.asm_out.internal_label (asm_out_file, "L",
+ CODE_LABEL_NUMBER (not_taken));
+ return "";
+}
+
+/* Return the assembly code for INSN, which branches to OPERANDS[0]
+ if some equality condition is true. The condition is given by
+ OPERANDS[1] if !INVERTED_P, otherwise it is the inverse of
+ OPERANDS[1]. OPERANDS[2] is the comparison's first operand;
+ OPERANDS[3] is the second operand and may be zero or a register. */
+
+const char *
+loongarch_output_equal_conditional_branch (rtx_insn *insn, rtx *operands,
+ bool inverted_p)
+{
+ const char *branch[2];
+ if (operands[3] == const0_rtx)
+ {
+ branch[!inverted_p] = LARCH_BRANCH ("b%C1z", "%2,%0");
+ branch[inverted_p] = LARCH_BRANCH ("b%N1z", "%2,%0");
+ }
+ else
+ {
+ branch[!inverted_p] = LARCH_BRANCH ("b%C1", "%2,%z3,%0");
+ branch[inverted_p] = LARCH_BRANCH ("b%N1", "%2,%z3,%0");
+ }
+
+ return loongarch_output_conditional_branch (insn, operands, branch[1],
+ branch[0]);
+}
+
+/* Return the assembly code for INSN, which branches to OPERANDS[0]
+ if some ordering condition is true. The condition is given by
+ OPERANDS[1] if !INVERTED_P, otherwise it is the inverse of
+ OPERANDS[1]. OPERANDS[2] is the comparison's first operand;
+ OPERANDS[3] is the second operand and may be zero or a register. */
+
+const char *
+loongarch_output_order_conditional_branch (rtx_insn *insn, rtx *operands,
+ bool inverted_p)
+{
+ const char *branch[2];
+
+ /* Make BRANCH[1] branch to OPERANDS[0] when the condition is true.
+ Make BRANCH[0] branch on the inverse condition. */
+ if (operands[3] != const0_rtx)
+ {
+ /* Handle degenerate cases that should not, but do, occur. */
+ if (REGNO (operands[2]) == REGNO (operands[3]))
+ {
+ switch (GET_CODE (operands[1]))
+ {
+ case LT:
+ case LTU:
+ inverted_p = !inverted_p;
+ /* Fall through. */
+ case GE:
+ case GEU:
+ branch[!inverted_p] = LARCH_BRANCH ("b", "%0");
+ branch[inverted_p] = "\t# branch never";
+ break;
+ default:
+ gcc_unreachable ();
+ }
+ }
+ else
+ {
+ switch (GET_CODE (operands[1]))
+ {
+ case LE:
+ case LEU:
+ case GT:
+ case GTU:
+ case LT:
+ case LTU:
+ case GE:
+ case GEU:
+ branch[!inverted_p] = LARCH_BRANCH ("b%C1", "%2,%3,%0");
+ branch[inverted_p] = LARCH_BRANCH ("b%N1", "%2,%3,%0");
+ break;
+ default:
+ gcc_unreachable ();
+ }
+ }
+ }
+ else
+ {
+ switch (GET_CODE (operands[1]))
+ {
+ /* These cases are equivalent to comparisons against zero. */
+ case LEU:
+ case GTU:
+ case LTU:
+ case GEU:
+ case LE:
+ case GT:
+ case LT:
+ case GE:
+ branch[!inverted_p] = LARCH_BRANCH ("b%C1", "%2,$r0,%0");
+ branch[inverted_p] = LARCH_BRANCH ("b%N1", "%2,$r0,%0");
+ break;
+ default:
+ gcc_unreachable ();
+ }
+ }
+ return loongarch_output_conditional_branch (insn, operands, branch[1],
+ branch[0]);
+}
+
+/* Return the assembly code for DIV.{W/D} instruction DIVISION, which has
+ the operands given by OPERANDS. Add in a divide-by-zero check if needed.
+ */
+
+const char *
+loongarch_output_division (const char *division, rtx *operands)
+{
+ const char *s;
+
+ s = division;
+ if (TARGET_CHECK_ZERO_DIV)
+ {
+ output_asm_insn (s, operands);
+ s = "bne\t%2,%.,1f\n\tbreak\t7\n1:";
+ }
+ return s;
+}
+
+/* Implement TARGET_SCHED_ADJUST_COST. We assume that anti and output
+ dependencies have no cost. */
+
+static int
+loongarch_adjust_cost (rtx_insn *, int dep_type, rtx_insn *, int cost,
+ unsigned int)
+{
+ if (dep_type != 0 && (dep_type != REG_DEP_OUTPUT))
+ return 0;
+ return cost;
+}
+
+/* Return the number of instructions that can be issued per cycle. */
+
+static int
+loongarch_issue_rate (void)
+{
+ if ((unsigned long) __ACTUAL_TUNE < N_TUNE_TYPES)
+ return loongarch_cpu_issue_rate[__ACTUAL_TUNE];
+ else
+ return 1;
+}
+
+/* Implement TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD. This should
+ be as wide as the scheduling freedom in the DFA. */
+
+static int
+loongarch_multipass_dfa_lookahead (void)
+{
+ if ((unsigned long) __ACTUAL_TUNE < N_ARCH_TYPES)
+ return loongarch_cpu_multipass_dfa_lookahead[__ACTUAL_TUNE];
+ else
+ return 0;
+}
+
+/* Implement TARGET_SCHED_REORDER. */
+
+static int
+loongarch_sched_reorder (FILE *file ATTRIBUTE_UNUSED,
+ int verbose ATTRIBUTE_UNUSED,
+ rtx_insn **ready ATTRIBUTE_UNUSED,
+ int *nreadyp ATTRIBUTE_UNUSED,
+ int cycle ATTRIBUTE_UNUSED)
+{
+ return loongarch_issue_rate ();
+}
+
+/* Implement TARGET_SCHED_REORDER2. */
+
+static int
+loongarch_sched_reorder2 (FILE *file ATTRIBUTE_UNUSED,
+ int verbose ATTRIBUTE_UNUSED,
+ rtx_insn **ready ATTRIBUTE_UNUSED,
+ int *nreadyp ATTRIBUTE_UNUSED,
+ int cycle ATTRIBUTE_UNUSED)
+{
+ return cached_can_issue_more;
+}
+
+/* Implement TARGET_SCHED_INIT. */
+
+static void
+loongarch_sched_init (FILE *file ATTRIBUTE_UNUSED,
+ int verbose ATTRIBUTE_UNUSED,
+ int max_ready ATTRIBUTE_UNUSED)
+{}
+
+/* Implement TARGET_SCHED_VARIABLE_ISSUE. */
+
+static int
+loongarch_variable_issue (FILE *file ATTRIBUTE_UNUSED,
+ int verbose ATTRIBUTE_UNUSED, rtx_insn *insn,
+ int more)
+{
+ /* Ignore USEs and CLOBBERs; don't count them against the issue rate. */
+ if (USEFUL_INSN_P (insn))
+ {
+ if (get_attr_type (insn) != TYPE_GHOST)
+ more--;
+ }
+
+ /* Instructions of type 'multi' should all be split before
+ the second scheduling pass. */
+ gcc_assert (!reload_completed
+ || recog_memoized (insn) < 0
+ || get_attr_type (insn) != TYPE_MULTI);
+
+ cached_can_issue_more = more;
+ return more;
+}
+
+/* A structure representing the state of the processor pipeline.
+ Used by the loongarch_sim_* family of functions. */
+
+struct loongarch_sim
+{
+ /* The maximum number of instructions that can be issued in a cycle.
+ (Caches loongarch_issue_rate.) */
+ unsigned int issue_rate;
+
+ /* The current simulation time. */
+ unsigned int time;
+
+ /* How many more instructions can be issued in the current cycle. */
+ unsigned int insns_left;
+
+ /* LAST_SET[X].INSN is the last instruction to set register X.
+ LAST_SET[X].TIME is the time at which that instruction was issued.
+ INSN is null if no instruction has yet set register X. */
+ struct
+ {
+ rtx_insn *insn;
+ unsigned int time;
+ } last_set[FIRST_PSEUDO_REGISTER];
+
+ /* The pipeline's current DFA state. */
+ state_t dfa_state;
+};
+
+/* Reset STATE to the initial simulation state. */
+
+static void
+loongarch_sim_reset (struct loongarch_sim *state)
+{
+ curr_state = state->dfa_state;
+
+ state->time = 0;
+ state->insns_left = state->issue_rate;
+ memset (&state->last_set, 0, sizeof (state->last_set));
+ state_reset (curr_state);
+
+ targetm.sched.init (0, false, 0);
+ advance_state (curr_state);
+}
+
+/* Initialize STATE before its first use. DFA_STATE points to an
+ allocated but uninitialized DFA state. */
+
+static void
+loongarch_sim_init (struct loongarch_sim *state, state_t dfa_state)
+{
+ if (targetm.sched.init_dfa_pre_cycle_insn)
+ targetm.sched.init_dfa_pre_cycle_insn ();
+
+ if (targetm.sched.init_dfa_post_cycle_insn)
+ targetm.sched.init_dfa_post_cycle_insn ();
+
+ state->issue_rate = loongarch_issue_rate ();
+ state->dfa_state = dfa_state;
+ loongarch_sim_reset (state);
+}
+
+/* Set up costs based on the current architecture and tuning settings. */
+
+static void
+loongarch_set_tuning_info (void)
+{
+ dfa_start ();
+
+ struct loongarch_sim state;
+ loongarch_sim_init (&state, alloca (state_size ()));
+
+ dfa_finish ();
+}
+
+/* Implement TARGET_EXPAND_TO_RTL_HOOK. */
+
+static void
+loongarch_expand_to_rtl_hook (void)
+{
+ /* We need to call this at a point where we can safely create sequences
+ of instructions, so TARGET_OVERRIDE_OPTIONS is too early. We also
+ need to call it at a point where the DFA infrastructure is not
+ already in use, so we can't just call it lazily on demand.
+
+ At present, loongarch_tuning_info is only needed during post-expand
+ RTL passes such as split_insns, so this hook should be early enough.
+ We may need to move the call elsewhere if loongarch_tuning_info starts
+ to be used for other things (such as rtx_costs, or expanders that
+ could be called during gimple optimization). */
+
+ loongarch_set_tuning_info ();
+}
+
+
+/* Given that we have an rtx of the form (prefetch ... WRITE LOCALITY),
+ return the first operand of the associated PREF or PREFX insn. */
+
+rtx
+loongarch_prefetch_cookie (rtx write, rtx locality)
+{
+ /* store_streamed / load_streamed. */
+ if (INTVAL (locality) <= 0)
+ return GEN_INT (INTVAL (write) + 4);
+
+ /* store / load. */
+ if (INTVAL (locality) <= 2)
+ return write;
+
+ /* store_retained / load_retained. */
+ return GEN_INT (INTVAL (write) + 6);
+}
+
+/* Implement TARGET_ASM_OUTPUT_MI_THUNK. Generate rtl rather than asm text
+ in order to avoid duplicating too much logic from elsewhere. */
+
+static void
+loongarch_output_mi_thunk (FILE *file, tree thunk_fndecl ATTRIBUTE_UNUSED,
+ HOST_WIDE_INT delta, HOST_WIDE_INT vcall_offset,
+ tree function)
+{
+ const char *fnname = IDENTIFIER_POINTER (DECL_ASSEMBLER_NAME (thunk_fndecl));
+ rtx this_rtx, temp1, temp2, fnaddr;
+ rtx_insn *insn;
+ bool use_sibcall_p;
+
+ /* Pretend to be a post-reload pass while generating rtl. */
+ reload_completed = 1;
+
+ /* Mark the end of the (empty) prologue. */
+ emit_note (NOTE_INSN_PROLOGUE_END);
+
+ /* Determine if we can use a sibcall to call FUNCTION directly. */
+ fnaddr = XEXP (DECL_RTL (function), 0);
+ use_sibcall_p = const_call_insn_operand (fnaddr, Pmode);
+
+ /* We need two temporary registers in some cases. */
+ temp1 = gen_rtx_REG (Pmode, 12);
+ temp2 = gen_rtx_REG (Pmode, 13);
+
+ /* Find out which register contains the "this" pointer. */
+ if (aggregate_value_p (TREE_TYPE (TREE_TYPE (function)), function))
+ this_rtx = gen_rtx_REG (Pmode, GP_ARG_FIRST + 1);
+ else
+ this_rtx = gen_rtx_REG (Pmode, GP_ARG_FIRST);
+
+ /* Add DELTA to THIS_RTX. */
+ if (delta != 0)
+ {
+ rtx offset = GEN_INT (delta);
+ if (!IMM12_OPERAND (delta))
+ {
+ loongarch_emit_move (temp1, offset);
+ offset = temp1;
+ }
+ emit_insn (gen_add3_insn (this_rtx, this_rtx, offset));
+ }
+
+ /* If needed, add *(*THIS_RTX + VCALL_OFFSET) to THIS_RTX. */
+ if (vcall_offset != 0)
+ {
+ rtx addr;
+
+ /* Set TEMP1 to *THIS_RTX. */
+ loongarch_emit_move (temp1, gen_rtx_MEM (Pmode, this_rtx));
+
+ /* Set ADDR to a legitimate address for *THIS_RTX + VCALL_OFFSET. */
+ addr = loongarch_add_offset (temp2, temp1, vcall_offset);
+
+ /* Load the offset and add it to THIS_RTX. */
+ loongarch_emit_move (temp1, gen_rtx_MEM (Pmode, addr));
+ emit_insn (gen_add3_insn (this_rtx, this_rtx, temp1));
+ }
+
+ /* Jump to the target function. Use a sibcall if direct jumps are
+ allowed, otherwise load the address into a register first. */
+ if (use_sibcall_p)
+ {
+ insn = emit_call_insn (gen_sibcall_internal (fnaddr, const0_rtx));
+ SIBLING_CALL_P (insn) = 1;
+ }
+ else
+ {
+ loongarch_emit_move (temp1, fnaddr);
+ emit_jump_insn (gen_indirect_jump (temp1));
+ }
+
+ /* Run just enough of rest_of_compilation. This sequence was
+ "borrowed" from alpha.c. */
+ insn = get_insns ();
+ split_all_insns_noflow ();
+ shorten_branches (insn);
+ assemble_start_function (thunk_fndecl, fnname);
+ final_start_function (insn, file, 1);
+ final (insn, file, 1);
+ final_end_function ();
+ assemble_end_function (thunk_fndecl, fnname);
+
+ /* Clean up the vars set above. Note that final_end_function resets
+ the global pointer for us. */
+ reload_completed = 0;
+}
+
+/* Allocate a chunk of memory for per-function machine-dependent data. */
+
+static struct machine_function *
+loongarch_init_machine_status (void)
+{
+ return ggc_cleared_alloc<machine_function> ();
+}
+
+static void
+loongarch_option_override_internal (struct gcc_options *opts,
+ struct gcc_options *opts_set);
+
+/* Implement TARGET_OPTION_OVERRIDE. */
+
+static void
+loongarch_option_override (void)
+{
+ loongarch_option_override_internal (&global_options, &global_options_set);
+}
+
+static void
+loongarch_option_override_internal (struct gcc_options *opts,
+ struct gcc_options *opts_set)
+{
+ int i, regno, mode;
+
+ (void) opts_set;
+
+ if (flag_pic)
+ g_switch_value = 0;
+
+ /* Handle target-specific options: compute defaults/conflicts etc. */
+ loongarch_config_target (&la_target, la_opt_switches,
+ la_opt_cpu_arch, la_opt_cpu_tune, la_opt_fpu,
+ la_opt_abi_base, la_opt_abi_ext, la_opt_cmodel, 0);
+
+ /* End of code shared with GAS. */
+ if (TARGET_ABI_LP64)
+ flag_pcc_struct_return = 0;
+
+ /* Decide which rtx_costs structure to use. */
+ if (optimize_size)
+ loongarch_cost = &loongarch_rtx_cost_optimize_size;
+ else
+ loongarch_cost = &loongarch_cpu_rtx_cost_data[__ACTUAL_TUNE];
+
+ /* If the user hasn't specified a branch cost, use the processor's
+ default. */
+ if (loongarch_branch_cost == 0)
+ loongarch_branch_cost = loongarch_cost->branch_cost;
+
+
+ switch (la_target.cmodel)
+ {
+ case CMODEL_TINY_STATIC:
+ case CMODEL_EXTREME:
+ if (opts->x_flag_plt)
+ error ("code model %qs and %qs not support %s mode",
+ "tiny-static", "extreme", "plt");
+ break;
+
+ case CMODEL_NORMAL:
+ case CMODEL_TINY:
+ case CMODEL_LARGE:
+ break;
+
+ default:
+ gcc_unreachable ();
+ }
+
+ /* .cfi_* directives generate a read-only section, so fall back on
+ manual .eh_frame creation if we need the section to be writable. */
+ if (flag_pic)
+ flag_dwarf2_cfi_asm = 0;
+
+ loongarch_init_print_operand_punct ();
+
+ /* Set up array to map GCC register number to debug register number.
+ Ignore the special purpose register numbers. */
+
+ for (i = 0; i < FIRST_PSEUDO_REGISTER; i++)
+ {
+ if (GP_REG_P (i) || FP_REG_P (i))
+ loongarch_dwarf_regno[i] = i;
+ else
+ loongarch_dwarf_regno[i] = INVALID_REGNUM;
+ }
+
+ /* Set up loongarch_hard_regno_mode_ok. */
+ for (mode = 0; mode < MAX_MACHINE_MODE; mode++)
+ for (regno = 0; regno < FIRST_PSEUDO_REGISTER; regno++)
+ loongarch_hard_regno_mode_ok_p[mode][regno]
+ = loongarch_hard_regno_mode_ok_uncached (regno, (machine_mode) mode);
+
+ /* Function to allocate machine-dependent function status. */
+ init_machine_status = &loongarch_init_machine_status;
+}
+
+/* Implement TARGET_CONDITIONAL_REGISTER_USAGE. */
+
+static void
+loongarch_conditional_register_usage (void)
+{
+ if (!TARGET_HARD_FLOAT)
+ accessible_reg_set &= ~(reg_class_contents[FP_REGS]
+ | reg_class_contents[FCC_REGS]);
+}
+
+/* Implement EH_USES. */
+
+bool
+loongarch_eh_uses (unsigned int regno ATTRIBUTE_UNUSED)
+{
+ return false;
+}
+
+/* Implement EPILOGUE_USES. */
+
+bool
+loongarch_epilogue_uses (unsigned int regno)
+{
+ /* Say that the epilogue uses the return address register. Note that
+ in the case of sibcalls, the values "used by the epilogue" are
+ considered live at the start of the called function. */
+ if (regno == RETURN_ADDR_REGNUM)
+ return true;
+
+ return false;
+}
+
+bool
+loongarch_load_store_bonding_p (rtx *operands, machine_mode mode, bool load_p)
+{
+ rtx reg1, reg2, mem1, mem2, base1, base2;
+ enum reg_class rc1, rc2;
+ HOST_WIDE_INT offset1, offset2;
+
+ if (load_p)
+ {
+ reg1 = operands[0];
+ reg2 = operands[2];
+ mem1 = operands[1];
+ mem2 = operands[3];
+ }
+ else
+ {
+ reg1 = operands[1];
+ reg2 = operands[3];
+ mem1 = operands[0];
+ mem2 = operands[2];
+ }
+
+ if (loongarch_address_insns (XEXP (mem1, 0), mode, false) == 0
+ || loongarch_address_insns (XEXP (mem2, 0), mode, false) == 0)
+ return false;
+
+ loongarch_split_plus (XEXP (mem1, 0), &base1, &offset1);
+ loongarch_split_plus (XEXP (mem2, 0), &base2, &offset2);
+
+ /* Base regs do not match. */
+ if (!REG_P (base1) || !rtx_equal_p (base1, base2))
+ return false;
+
+ /* Either of the loads is clobbering base register. It is legitimate to bond
+ loads if second load clobbers base register. However, hardware does not
+ support such bonding. */
+ if (load_p
+ && (REGNO (reg1) == REGNO (base1) || (REGNO (reg2) == REGNO (base1))))
+ return false;
+
+ /* Loading in same registers. */
+ if (load_p && REGNO (reg1) == REGNO (reg2))
+ return false;
+
+ /* The loads/stores are not of same type. */
+ rc1 = REGNO_REG_CLASS (REGNO (reg1));
+ rc2 = REGNO_REG_CLASS (REGNO (reg2));
+ if (rc1 != rc2 && !reg_class_subset_p (rc1, rc2)
+ && !reg_class_subset_p (rc2, rc1))
+ return false;
+
+ if (abs (offset1 - offset2) != GET_MODE_SIZE (mode))
+ return false;
+
+ return true;
+}
+
+/* Implement TARGET_TRAMPOLINE_INIT. */
+
+static void
+loongarch_trampoline_init (rtx m_tramp, tree fndecl, rtx chain_value)
+{
+ rtx addr, end_addr, mem;
+ rtx trampoline[8];
+ unsigned int i, j;
+ HOST_WIDE_INT end_addr_offset, static_chain_offset, target_function_offset;
+
+ /* Work out the offsets of the pointers from the start of the
+ trampoline code. */
+ end_addr_offset = TRAMPOLINE_CODE_SIZE;
+ static_chain_offset = end_addr_offset;
+ target_function_offset = static_chain_offset + GET_MODE_SIZE (ptr_mode);
+
+ /* Get pointers to the beginning and end of the code block. */
+ addr = force_reg (Pmode, XEXP (m_tramp, 0));
+ end_addr
+ = loongarch_force_binary (Pmode, PLUS, addr, GEN_INT (end_addr_offset));
+
+#define OP(X) gen_int_mode (X, SImode)
+
+ /* Build up the code in TRAMPOLINE. */
+ i = 0;
+ /*pcaddi $static_chain,0
+ ld.[dw] $tmp,$static_chain,target_function_offset
+ ld.[dw] $static_chain,$static_chain,static_chain_offset
+ jirl $r0,$tmp,0 */
+ trampoline[i++] = OP (0x18000000 | (STATIC_CHAIN_REGNUM - GP_REG_FIRST));
+ trampoline[i++] = OP ((ptr_mode == DImode ? 0x28c00000 : 0x28800000)
+ | 19 /* $t7 */
+ | ((STATIC_CHAIN_REGNUM - GP_REG_FIRST) << 5)
+ | ((target_function_offset & 0xfff) << 10));
+ trampoline[i++] = OP ((ptr_mode == DImode ? 0x28c00000 : 0x28800000)
+ | (STATIC_CHAIN_REGNUM - GP_REG_FIRST)
+ | ((STATIC_CHAIN_REGNUM - GP_REG_FIRST) << 5)
+ | ((static_chain_offset & 0xfff) << 10));
+ trampoline[i++] = OP (0x4c000000 | (19 << 5));
+#undef OP
+
+ for (j = 0; j < i; j++)
+ {
+ mem = adjust_address (m_tramp, SImode, j * GET_MODE_SIZE (SImode));
+ loongarch_emit_move (mem, trampoline[j]);
+ }
+
+ /* Set up the static chain pointer field. */
+ mem = adjust_address (m_tramp, ptr_mode, static_chain_offset);
+ loongarch_emit_move (mem, chain_value);
+
+ /* Set up the target function field. */
+ mem = adjust_address (m_tramp, ptr_mode, target_function_offset);
+ loongarch_emit_move (mem, XEXP (DECL_RTL (fndecl), 0));
+
+ /* Flush the code part of the trampoline. */
+ emit_insn (gen_add3_insn (end_addr, addr, GEN_INT (TRAMPOLINE_SIZE)));
+ emit_insn (gen_clear_cache (addr, end_addr));
+}
+
+/* Implement TARGET_SHIFT_TRUNCATION_MASK. We want to keep the default
+ behavior of TARGET_SHIFT_TRUNCATION_MASK for non-vector modes even
+ when TARGET_LOONGSON_MMI is true. */
+
+static unsigned HOST_WIDE_INT
+loongarch_shift_truncation_mask (machine_mode mode)
+{
+ return GET_MODE_BITSIZE (mode) - 1;
+}
+
+/* Implement TARGET_SCHED_REASSOCIATION_WIDTH. */
+
+static int
+loongarch_sched_reassociation_width (unsigned int opc ATTRIBUTE_UNUSED,
+ machine_mode mode ATTRIBUTE_UNUSED)
+{
+ return 1;
+}
+
+/* Implement HARD_REGNO_CALLER_SAVE_MODE. */
+
+machine_mode
+loongarch_hard_regno_caller_save_mode (unsigned int regno, unsigned int nregs,
+ machine_mode mode)
+{
+ /* For performance, avoid saving/restoring upper parts of a register
+ by returning MODE as save mode when the mode is known. */
+ if (mode == VOIDmode)
+ return choose_hard_reg_mode (regno, nregs, NULL);
+ else
+ return mode;
+}
+
+/* Implement TARGET_CASE_VALUES_THRESHOLD. */
+
+unsigned int
+loongarch_case_values_threshold (void)
+{
+ return default_case_values_threshold ();
+}
+
+/* Implement TARGET_SPILL_CLASS. */
+
+static reg_class_t
+loongarch_spill_class (reg_class_t rclass ATTRIBUTE_UNUSED,
+ machine_mode mode ATTRIBUTE_UNUSED)
+{
+ return NO_REGS;
+}
+
+/* Implement TARGET_LRA_P. */
+
+static bool
+loongarch_lra_p (void)
+{
+ return loongarch_lra_flag;
+}
+
+/* Implement TARGET_IRA_CHANGE_PSEUDO_ALLOCNO_CLASS. */
+
+static reg_class_t
+loongarch_ira_change_pseudo_allocno_class (int regno,
+ reg_class_t allocno_class,
+ reg_class_t best_class \
+ ATTRIBUTE_UNUSED)
+{
+ /* LRA will allocate an FPR for an integer mode pseudo instead of spilling
+ to memory if an FPR is present in the allocno class. It is rare that
+ we actually need to place an integer mode value in an FPR so where
+ possible limit the allocation to GR_REGS. This will slightly pessimize
+ code that involves integer to/from float conversions as these will have
+ to reload into FPRs in LRA. Such reloads are sometimes eliminated and
+ sometimes only partially eliminated. We choose to take this penalty
+ in order to eliminate usage of FPRs in code that does not use floating
+ point data.
+
+ This change has a similar effect to increasing the cost of FPR->GPR
+ register moves for integer modes so that they are higher than the cost
+ of memory but changing the allocno class is more reliable.
+
+ This is also similar to forbidding integer mode values in FPRs entirely
+ but this would lead to an inconsistency in the integer to/from float
+ instructions that say integer mode values must be placed in FPRs. */
+ if (INTEGRAL_MODE_P (PSEUDO_REGNO_MODE (regno)) && allocno_class == ALL_REGS)
+ return GR_REGS;
+ return allocno_class;
+}
+
+/* Implement TARGET_PROMOTE_FUNCTION_MODE. */
+
+/* This function is equivalent to default_promote_function_mode_always_promote
+ except that it returns a promoted mode even if type is NULL_TREE. This is
+ needed by libcalls which have no type (only a mode) such as fixed conversion
+ routines that take a signed or unsigned char/short argument and convert it
+ to a fixed type. */
+
+static machine_mode
+loongarch_promote_function_mode (const_tree type ATTRIBUTE_UNUSED,
+ machine_mode mode,
+ int *punsignedp ATTRIBUTE_UNUSED,
+ const_tree fntype ATTRIBUTE_UNUSED,
+ int for_return ATTRIBUTE_UNUSED)
+{
+ int unsignedp;
+
+ if (type != NULL_TREE)
+ return promote_mode (type, mode, punsignedp);
+
+ unsignedp = *punsignedp;
+ PROMOTE_MODE (mode, unsignedp, type);
+ *punsignedp = unsignedp;
+ return mode;
+}
+
+/* Implement TARGET_TRULY_NOOP_TRUNCATION. */
+
+static bool
+loongarch_truly_noop_truncation (poly_uint64 outprec, poly_uint64 inprec)
+{
+ return !TARGET_64BIT || inprec <= 32 || outprec > 32;
+}
+
+/* Implement TARGET_CONSTANT_ALIGNMENT. */
+
+static HOST_WIDE_INT
+loongarch_constant_alignment (const_tree exp, HOST_WIDE_INT align)
+{
+ if (TREE_CODE (exp) == STRING_CST || TREE_CODE (exp) == CONSTRUCTOR)
+ return MAX (align, BITS_PER_WORD);
+ return align;
+}
+
+/* Implement TARGET_STARTING_FRAME_OFFSET. See loongarch_compute_frame_info
+ for details about the frame layout. */
+
+static HOST_WIDE_INT
+loongarch_starting_frame_offset (void)
+{
+ if (FRAME_GROWS_DOWNWARD)
+ return 0;
+ return crtl->outgoing_args_size;
+}
+
+/* Initialize the GCC target structure. */
+#undef TARGET_ASM_ALIGNED_HI_OP
+#define TARGET_ASM_ALIGNED_HI_OP "\t.half\t"
+#undef TARGET_ASM_ALIGNED_SI_OP
+#define TARGET_ASM_ALIGNED_SI_OP "\t.word\t"
+#undef TARGET_ASM_ALIGNED_DI_OP
+#define TARGET_ASM_ALIGNED_DI_OP "\t.dword\t"
+
+#undef TARGET_OPTION_OVERRIDE
+#define TARGET_OPTION_OVERRIDE loongarch_option_override
+
+#undef TARGET_LEGITIMIZE_ADDRESS
+#define TARGET_LEGITIMIZE_ADDRESS loongarch_legitimize_address
+
+#undef TARGET_ASM_SELECT_RTX_SECTION
+#define TARGET_ASM_SELECT_RTX_SECTION loongarch_select_rtx_section
+#undef TARGET_ASM_FUNCTION_RODATA_SECTION
+#define TARGET_ASM_FUNCTION_RODATA_SECTION loongarch_function_rodata_section
+
+#undef TARGET_SCHED_INIT
+#define TARGET_SCHED_INIT loongarch_sched_init
+#undef TARGET_SCHED_REORDER
+#define TARGET_SCHED_REORDER loongarch_sched_reorder
+#undef TARGET_SCHED_REORDER2
+#define TARGET_SCHED_REORDER2 loongarch_sched_reorder2
+#undef TARGET_SCHED_VARIABLE_ISSUE
+#define TARGET_SCHED_VARIABLE_ISSUE loongarch_variable_issue
+#undef TARGET_SCHED_ADJUST_COST
+#define TARGET_SCHED_ADJUST_COST loongarch_adjust_cost
+#undef TARGET_SCHED_ISSUE_RATE
+#define TARGET_SCHED_ISSUE_RATE loongarch_issue_rate
+#undef TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD
+#define TARGET_SCHED_FIRST_CYCLE_MULTIPASS_DFA_LOOKAHEAD \
+ loongarch_multipass_dfa_lookahead
+
+#undef TARGET_FUNCTION_OK_FOR_SIBCALL
+#define TARGET_FUNCTION_OK_FOR_SIBCALL loongarch_function_ok_for_sibcall
+
+#undef TARGET_VALID_POINTER_MODE
+#define TARGET_VALID_POINTER_MODE loongarch_valid_pointer_mode
+#undef TARGET_REGISTER_MOVE_COST
+#define TARGET_REGISTER_MOVE_COST loongarch_register_move_cost
+#undef TARGET_MEMORY_MOVE_COST
+#define TARGET_MEMORY_MOVE_COST loongarch_memory_move_cost
+#undef TARGET_RTX_COSTS
+#define TARGET_RTX_COSTS loongarch_rtx_costs
+#undef TARGET_ADDRESS_COST
+#define TARGET_ADDRESS_COST loongarch_address_cost
+
+#undef TARGET_IN_SMALL_DATA_P
+#define TARGET_IN_SMALL_DATA_P loongarch_in_small_data_p
+
+#undef TARGET_PREFERRED_RELOAD_CLASS
+#define TARGET_PREFERRED_RELOAD_CLASS loongarch_preferred_reload_class
+
+#undef TARGET_EXPAND_TO_RTL_HOOK
+#define TARGET_EXPAND_TO_RTL_HOOK loongarch_expand_to_rtl_hook
+#undef TARGET_ASM_FILE_START
+#define TARGET_ASM_FILE_START loongarch_file_start
+#undef TARGET_ASM_FILE_START_FILE_DIRECTIVE
+#define TARGET_ASM_FILE_START_FILE_DIRECTIVE true
+
+#undef TARGET_EXPAND_BUILTIN_VA_START
+#define TARGET_EXPAND_BUILTIN_VA_START loongarch_va_start
+
+#undef TARGET_PROMOTE_FUNCTION_MODE
+#define TARGET_PROMOTE_FUNCTION_MODE loongarch_promote_function_mode
+#undef TARGET_RETURN_IN_MEMORY
+#define TARGET_RETURN_IN_MEMORY loongarch_return_in_memory
+
+#undef TARGET_FUNCTION_VALUE
+#define TARGET_FUNCTION_VALUE loongarch_function_value
+#undef TARGET_LIBCALL_VALUE
+#define TARGET_LIBCALL_VALUE loongarch_libcall_value
+
+#undef TARGET_ASM_OUTPUT_MI_THUNK
+#define TARGET_ASM_OUTPUT_MI_THUNK loongarch_output_mi_thunk
+#undef TARGET_ASM_CAN_OUTPUT_MI_THUNK
+#define TARGET_ASM_CAN_OUTPUT_MI_THUNK \
+ hook_bool_const_tree_hwi_hwi_const_tree_true
+
+#undef TARGET_PRINT_OPERAND
+#define TARGET_PRINT_OPERAND loongarch_print_operand
+#undef TARGET_PRINT_OPERAND_ADDRESS
+#define TARGET_PRINT_OPERAND_ADDRESS loongarch_print_operand_address
+#undef TARGET_PRINT_OPERAND_PUNCT_VALID_P
+#define TARGET_PRINT_OPERAND_PUNCT_VALID_P \
+ loongarch_print_operand_punct_valid_p
+
+#undef TARGET_SETUP_INCOMING_VARARGS
+#define TARGET_SETUP_INCOMING_VARARGS loongarch_setup_incoming_varargs
+#undef TARGET_STRICT_ARGUMENT_NAMING
+#define TARGET_STRICT_ARGUMENT_NAMING hook_bool_CUMULATIVE_ARGS_true
+#undef TARGET_MUST_PASS_IN_STACK
+#define TARGET_MUST_PASS_IN_STACK must_pass_in_stack_var_size
+#undef TARGET_PASS_BY_REFERENCE
+#define TARGET_PASS_BY_REFERENCE loongarch_pass_by_reference
+#undef TARGET_ARG_PARTIAL_BYTES
+#define TARGET_ARG_PARTIAL_BYTES loongarch_arg_partial_bytes
+#undef TARGET_FUNCTION_ARG
+#define TARGET_FUNCTION_ARG loongarch_function_arg
+#undef TARGET_FUNCTION_ARG_ADVANCE
+#define TARGET_FUNCTION_ARG_ADVANCE loongarch_function_arg_advance
+#undef TARGET_FUNCTION_ARG_BOUNDARY
+#define TARGET_FUNCTION_ARG_BOUNDARY loongarch_function_arg_boundary
+
+#undef TARGET_SCALAR_MODE_SUPPORTED_P
+#define TARGET_SCALAR_MODE_SUPPORTED_P loongarch_scalar_mode_supported_p
+
+#undef TARGET_VECTORIZE_PREFERRED_SIMD_MODE
+#define TARGET_VECTORIZE_PREFERRED_SIMD_MODE loongarch_preferred_simd_mode
+
+#undef TARGET_INIT_BUILTINS
+#define TARGET_INIT_BUILTINS loongarch_init_builtins
+#undef TARGET_BUILTIN_DECL
+#define TARGET_BUILTIN_DECL loongarch_builtin_decl
+#undef TARGET_EXPAND_BUILTIN
+#define TARGET_EXPAND_BUILTIN loongarch_expand_builtin
+
+/* The generic ELF target does not always have TLS support. */
+#ifdef HAVE_AS_TLS
+#undef TARGET_HAVE_TLS
+#define TARGET_HAVE_TLS HAVE_AS_TLS
+#endif
+
+#undef TARGET_CANNOT_FORCE_CONST_MEM
+#define TARGET_CANNOT_FORCE_CONST_MEM loongarch_cannot_force_const_mem
+
+#undef TARGET_LEGITIMATE_CONSTANT_P
+#define TARGET_LEGITIMATE_CONSTANT_P loongarch_legitimate_constant_p
+
+#undef TARGET_ATTRIBUTE_TABLE
+#define TARGET_ATTRIBUTE_TABLE loongarch_attribute_table
+/* All our function attributes are related to how out-of-line copies should
+ be compiled or called. They don't in themselves prevent inlining. */
+#undef TARGET_FUNCTION_ATTRIBUTE_INLINABLE_P
+#define TARGET_FUNCTION_ATTRIBUTE_INLINABLE_P hook_bool_const_tree_true
+
+#undef TARGET_USE_BLOCKS_FOR_CONSTANT_P
+#define TARGET_USE_BLOCKS_FOR_CONSTANT_P hook_bool_mode_const_rtx_true
+#undef TARGET_USE_ANCHORS_FOR_SYMBOL_P
+#define TARGET_USE_ANCHORS_FOR_SYMBOL_P loongarch_use_anchors_for_symbol_p
+
+#ifdef HAVE_AS_DTPRELWORD
+#undef TARGET_ASM_OUTPUT_DWARF_DTPREL
+#define TARGET_ASM_OUTPUT_DWARF_DTPREL loongarch_output_dwarf_dtprel
+#endif
+#undef TARGET_DWARF_FRAME_REG_MODE
+#define TARGET_DWARF_FRAME_REG_MODE loongarch_dwarf_frame_reg_mode
+
+#undef TARGET_LEGITIMATE_ADDRESS_P
+#define TARGET_LEGITIMATE_ADDRESS_P loongarch_legitimate_address_p
+
+#undef TARGET_FRAME_POINTER_REQUIRED
+#define TARGET_FRAME_POINTER_REQUIRED loongarch_frame_pointer_required
+
+#undef TARGET_CAN_ELIMINATE
+#define TARGET_CAN_ELIMINATE loongarch_can_eliminate
+
+#undef TARGET_CONDITIONAL_REGISTER_USAGE
+#define TARGET_CONDITIONAL_REGISTER_USAGE loongarch_conditional_register_usage
+
+#undef TARGET_TRAMPOLINE_INIT
+#define TARGET_TRAMPOLINE_INIT loongarch_trampoline_init
+
+#undef TARGET_SHIFT_TRUNCATION_MASK
+#define TARGET_SHIFT_TRUNCATION_MASK loongarch_shift_truncation_mask
+
+#undef TARGET_SCHED_REASSOCIATION_WIDTH
+#define TARGET_SCHED_REASSOCIATION_WIDTH loongarch_sched_reassociation_width
+
+#undef TARGET_CASE_VALUES_THRESHOLD
+#define TARGET_CASE_VALUES_THRESHOLD loongarch_case_values_threshold
+
+#undef TARGET_ATOMIC_ASSIGN_EXPAND_FENV
+#define TARGET_ATOMIC_ASSIGN_EXPAND_FENV loongarch_atomic_assign_expand_fenv
+
+#undef TARGET_CALL_FUSAGE_CONTAINS_NON_CALLEE_CLOBBERS
+#define TARGET_CALL_FUSAGE_CONTAINS_NON_CALLEE_CLOBBERS true
+
+#undef TARGET_USE_BY_PIECES_INFRASTRUCTURE_P
+#define TARGET_USE_BY_PIECES_INFRASTRUCTURE_P \
+ loongarch_use_by_pieces_infrastructure_p
+
+#undef TARGET_SPILL_CLASS
+#define TARGET_SPILL_CLASS loongarch_spill_class
+#undef TARGET_LRA_P
+#define TARGET_LRA_P loongarch_lra_p
+#undef TARGET_IRA_CHANGE_PSEUDO_ALLOCNO_CLASS
+#define TARGET_IRA_CHANGE_PSEUDO_ALLOCNO_CLASS \
+ loongarch_ira_change_pseudo_allocno_class
+
+#undef TARGET_HARD_REGNO_NREGS
+#define TARGET_HARD_REGNO_NREGS loongarch_hard_regno_nregs
+#undef TARGET_HARD_REGNO_MODE_OK
+#define TARGET_HARD_REGNO_MODE_OK loongarch_hard_regno_mode_ok
+
+#undef TARGET_MODES_TIEABLE_P
+#define TARGET_MODES_TIEABLE_P loongarch_modes_tieable_p
+
+#undef TARGET_CUSTOM_FUNCTION_DESCRIPTORS
+#define TARGET_CUSTOM_FUNCTION_DESCRIPTORS 2
+
+#undef TARGET_SECONDARY_MEMORY_NEEDED
+#define TARGET_SECONDARY_MEMORY_NEEDED loongarch_secondary_memory_needed
+
+#undef TARGET_CAN_CHANGE_MODE_CLASS
+#define TARGET_CAN_CHANGE_MODE_CLASS loongarch_can_change_mode_class
+
+#undef TARGET_TRULY_NOOP_TRUNCATION
+#define TARGET_TRULY_NOOP_TRUNCATION loongarch_truly_noop_truncation
+
+#undef TARGET_CONSTANT_ALIGNMENT
+#define TARGET_CONSTANT_ALIGNMENT loongarch_constant_alignment
+
+#undef TARGET_STARTING_FRAME_OFFSET
+#define TARGET_STARTING_FRAME_OFFSET loongarch_starting_frame_offset
+
+#undef TARGET_SECONDARY_RELOAD
+#define TARGET_SECONDARY_RELOAD loongarch_secondary_reload
+
+struct gcc_target targetm = TARGET_INITIALIZER;
+
+#include "gt-loongarch.h"
diff --git a/gcc/config/loongarch/loongarch.h b/gcc/config/loongarch/loongarch.h
new file mode 100644
index 00000000000..f829ae53f21
--- /dev/null
+++ b/gcc/config/loongarch/loongarch.h
@@ -0,0 +1,1291 @@
+/* Definitions of target machine for GNU compiler. LoongArch version.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS and RISC-V target for GNU compiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+/* LoongArch external variables defined in loongarch.c. */
+
+#include "config/loongarch/loongarch-opts.h"
+
+/* Macros to silence warnings about numbers being signed in traditional
+ C and unsigned in ISO C when compiled on 32-bit hosts. */
+
+#define BITMASK_HIGH (((unsigned long) 1) << 31) /* 0x80000000 */
+
+/* Run-time compilation parameters selecting different hardware subsets. */
+
+/* Target CPU builtins. */
+#define TARGET_CPU_CPP_BUILTINS() loongarch_cpu_cpp_builtins (pfile)
+
+/* Default target_flags if no switches are specified. */
+
+#ifdef IN_LIBGCC2
+#undef TARGET_64BIT
+/* Make this compile time constant for libgcc2. */
+#ifdef __loongarch64
+#define TARGET_64BIT 1
+#else
+#define TARGET_64BIT 0
+#endif
+#endif /* IN_LIBGCC2 */
+
+#define TARGET_LIBGCC_SDATA_SECTION ".sdata"
+
+/* Support for a compile-time default CPU, et cetera. The rules are:
+ --with-divide is ignored if -mdivide-traps or -mdivide-breaks are
+ specified. */
+#define OPTION_DEFAULT_SPECS \
+ {"divide", "%{!mdivide-traps:%{!mdivide-breaks:-mdivide-%(VALUE)}}"},
+
+/* Driver native functions for SPEC processing in the GCC driver. */
+#include "loongarch-driver.h"
+
+/* This definition replaces the formerly used 'm' constraint with a
+ different constraint letter in order to avoid changing semantics of
+ the 'm' constraint when accepting new address formats in
+ TARGET_LEGITIMATE_ADDRESS_P. The constraint letter defined here
+ must not be used in insn definitions or inline assemblies. */
+#define TARGET_MEM_CONSTRAINT 'w'
+
+/* Tell collect what flags to pass to nm. */
+#ifndef NM_FLAGS
+#define NM_FLAGS "-Bn"
+#endif
+
+/* SUBTARGET_ASM_DEBUGGING_SPEC handles passing debugging options to
+ the assembler. It may be overridden by subtargets. */
+
+#ifndef SUBTARGET_ASM_DEBUGGING_SPEC
+#define SUBTARGET_ASM_DEBUGGING_SPEC "\
+%{g} %{g0} %{g1} %{g2} %{g3} \
+%{ggdb:-g} %{ggdb0:-g0} %{ggdb1:-g1} %{ggdb2:-g2} %{ggdb3:-g3} \
+%{gstabs:-g} %{gstabs0:-g0} %{gstabs1:-g1} %{gstabs2:-g2} %{gstabs3:-g3} \
+%{gstabs+:-g} %{gstabs+0:-g0} %{gstabs+1:-g1} %{gstabs+2:-g2} %{gstabs+3:-g3}"
+#endif
+
+/* SUBTARGET_ASM_SPEC is always passed to the assembler. It may be
+ overridden by subtargets. */
+
+#ifndef SUBTARGET_ASM_SPEC
+#define SUBTARGET_ASM_SPEC ""
+#endif
+
+#undef ASM_SPEC
+#define ASM_SPEC "%{mabi=*} %{subtarget_asm_spec}"
+
+/* Extra switches sometimes passed to the linker. */
+
+#ifndef LINK_SPEC
+#define LINK_SPEC ""
+#endif /* LINK_SPEC defined */
+
+/* Specs for the compiler proper. */
+
+/* CC1_SPEC is the set of arguments to pass to the compiler proper. */
+
+#undef CC1_SPEC
+#define CC1_SPEC "\
+%{G*} \
+%(subtarget_cc1_spec)"
+
+/* Preprocessor specs. */
+
+/* SUBTARGET_CPP_SPEC is passed to the preprocessor. It may be
+ overridden by subtargets. */
+#ifndef SUBTARGET_CPP_SPEC
+#define SUBTARGET_CPP_SPEC ""
+#endif
+
+#define CPP_SPEC "%(subtarget_cpp_spec)"
+
+/* This macro defines names of additional specifications to put in the specs
+ that can be used in various specifications like CC1_SPEC. Its definition
+ is an initializer with a subgrouping for each command option.
+
+ Each subgrouping contains a string constant, that defines the
+ specification name, and a string constant that used by the GCC driver
+ program.
+
+ Do not define this macro if it does not need to do anything. */
+
+#define EXTRA_SPECS \
+ {"subtarget_cc1_spec", SUBTARGET_CC1_SPEC}, \
+ {"subtarget_cpp_spec", SUBTARGET_CPP_SPEC}, \
+ {"subtarget_asm_debugging_spec", SUBTARGET_ASM_DEBUGGING_SPEC}, \
+ {"subtarget_asm_spec", SUBTARGET_ASM_SPEC},
+
+/* Registers may have a prefix which can be ignored when matching
+ user asm and register definitions. */
+#ifndef REGISTER_PREFIX
+#define REGISTER_PREFIX "$"
+#endif
+
+/* Local compiler-generated symbols must have a prefix that the assembler
+ understands. */
+
+#define LOCAL_LABEL_PREFIX "."
+
+/* By default on the loongarch, external symbols do not have an underscore
+ prepended. */
+
+#define USER_LABEL_PREFIX ""
+
+#ifndef PREFERRED_DEBUGGING_TYPE
+#define PREFERRED_DEBUGGING_TYPE DWARF2_DEBUG
+#endif
+
+/* The size of DWARF addresses should be the same as the size of symbols
+ in the target file format. */
+#define DWARF2_ADDR_SIZE (TARGET_64BIT ? 8 : 4)
+
+/* By default, turn on GDB extensions. */
+#define DEFAULT_GDB_EXTENSIONS 1
+
+/* By default, produce dwarf version 2 format debugging output in response
+ to the ‘-g’ option. */
+#define DWARF2_DEBUGGING_INFO 1
+
+/* The mapping from gcc register number to DWARF 2 CFA column number. */
+#define DWARF_FRAME_REGNUM(REGNO) loongarch_dwarf_regno[REGNO]
+
+/* The DWARF 2 CFA column which tracks the return address. */
+#define DWARF_FRAME_RETURN_COLUMN RETURN_ADDR_REGNUM
+
+/* Before the prologue, RA lives in r1. */
+#define INCOMING_RETURN_ADDR_RTX gen_rtx_REG (Pmode, RETURN_ADDR_REGNUM)
+
+/* Describe how we implement __builtin_eh_return. */
+#define EH_RETURN_DATA_REGNO(N) \
+ ((N) < (4) ? (N) + GP_ARG_FIRST : INVALID_REGNUM)
+
+#define EH_RETURN_STACKADJ_RTX gen_rtx_REG (Pmode, GP_ARG_FIRST + 4)
+
+#define EH_USES(N) loongarch_eh_uses (N)
+
+/* Offsets recorded in opcodes are a multiple of this alignment factor.
+ The default for this in 64-bit mode is 8, which causes problems with
+ SFmode register saves. */
+#define DWARF_CIE_DATA_ALIGNMENT -4
+
+/* Target machine storage layout. */
+
+#define BITS_BIG_ENDIAN 0
+#define BYTES_BIG_ENDIAN 0
+#define WORDS_BIG_ENDIAN 0
+
+#define MAX_BITS_PER_WORD 64
+
+/* Width of a word, in units (bytes). */
+#define UNITS_PER_WORD (TARGET_64BIT ? 8 : 4)
+#ifndef IN_LIBGCC2
+#define MIN_UNITS_PER_WORD 4
+#endif
+
+/* For LARCH, width of a floating point register. */
+#define UNITS_PER_FPREG (TARGET_DOUBLE_FLOAT ? 8 : 4)
+
+/* The number of consecutive floating-point registers needed to store the
+ largest format supported by the FPU. */
+#define MAX_FPRS_PER_FMT 1
+
+/* The number of consecutive floating-point registers needed to store the
+ smallest format supported by the FPU. */
+#define MIN_FPRS_PER_FMT 1
+
+/* The largest size of value that can be held in floating-point
+ registers and moved with a single instruction. */
+#define UNITS_PER_HWFPVALUE \
+ (TARGET_SOFT_FLOAT ? 0 : MAX_FPRS_PER_FMT * UNITS_PER_FPREG)
+
+/* The largest size of value that can be held in floating-point
+ registers. */
+#define UNITS_PER_FPVALUE \
+ (TARGET_SOFT_FLOAT ? 0 \
+ : TARGET_SINGLE_FLOAT ? UNITS_PER_FPREG \
+ : LONG_DOUBLE_TYPE_SIZE / BITS_PER_UNIT)
+
+/* The number of bytes in a double. */
+#define UNITS_PER_DOUBLE (TYPE_PRECISION (double_type_node) / BITS_PER_UNIT)
+
+/* Set the sizes of the core types. */
+#define SHORT_TYPE_SIZE 16
+#define INT_TYPE_SIZE 32
+#define LONG_TYPE_SIZE (TARGET_64BIT ? 64 : 32)
+#define LONG_LONG_TYPE_SIZE 64
+
+#define FLOAT_TYPE_SIZE 32
+#define DOUBLE_TYPE_SIZE 64
+#define LONG_DOUBLE_TYPE_SIZE (TARGET_64BIT ? 128 : 64)
+
+/* Define the sizes of fixed-point types. */
+#define SHORT_FRACT_TYPE_SIZE 8
+#define FRACT_TYPE_SIZE 16
+#define LONG_FRACT_TYPE_SIZE 32
+#define LONG_LONG_FRACT_TYPE_SIZE 64
+
+#define SHORT_ACCUM_TYPE_SIZE 16
+#define ACCUM_TYPE_SIZE 32
+#define LONG_ACCUM_TYPE_SIZE 64
+/* FIXME. LONG_LONG_ACCUM_TYPE_SIZE should be 128 bits, but GCC
+ doesn't support 128-bit integers for loongarch32 currently. */
+#define LONG_LONG_ACCUM_TYPE_SIZE (TARGET_64BIT ? 128 : 64)
+
+/* long double is not a fixed mode, but the idea is that, if we
+ support long double, we also want a 128-bit integer type. */
+#define MAX_FIXED_MODE_SIZE LONG_DOUBLE_TYPE_SIZE
+
+/* Width in bits of a pointer. */
+#ifndef POINTER_SIZE
+#define POINTER_SIZE (TARGET_64BIT ? 64 : 32)
+#endif
+
+/* Allocation boundary (in *bits*) for storing arguments in argument list. */
+#define PARM_BOUNDARY BITS_PER_WORD
+
+/* Allocation boundary (in *bits*) for the code of a function. */
+#define FUNCTION_BOUNDARY 32
+
+/* Alignment of field after `int : 0' in a structure. */
+#define EMPTY_FIELD_BOUNDARY 32
+
+/* Number of bits which any structure or union's size must be a multiple of.
+ Each structure or union's size is rounded up to a multiple of this. */
+#define STRUCTURE_SIZE_BOUNDARY 8
+
+/* There is no point aligning anything to a rounder boundary than
+ LONG_DOUBLE_TYPE_SIZE. */
+#define BIGGEST_ALIGNMENT (LONG_DOUBLE_TYPE_SIZE)
+
+/* All accesses must be aligned. */
+#define STRICT_ALIGNMENT (TARGET_STRICT_ALIGN)
+
+/* Define this if you wish to imitate the way many other C compilers
+ handle alignment of bitfields and the structures that contain
+ them.
+
+ The behavior is that the type written for a bit-field (`int',
+ `short', or other integer type) imposes an alignment for the
+ entire structure, as if the structure really did contain an
+ ordinary field of that type. In addition, the bit-field is placed
+ within the structure so that it would fit within such a field,
+ not crossing a boundary for it.
+
+ Thus, on most machines, a bit-field whose type is written as `int'
+ would not cross a four-byte boundary, and would force four-byte
+ alignment for the whole structure. (The alignment used may not
+ be four bytes; it is controlled by the other alignment
+ parameters.)
+
+ If the macro is defined, its definition should be a C expression;
+ a nonzero value for the expression enables this behavior. */
+
+#define PCC_BITFIELD_TYPE_MATTERS 1
+
+/* If defined, a C expression to compute the alignment for a static
+ variable. TYPE is the data type, and ALIGN is the alignment that
+ the object would ordinarily have. The value of this macro is used
+ instead of that alignment to align the object.
+
+ If this macro is not defined, then ALIGN is used.
+
+ One use of this macro is to increase alignment of medium-size
+ data to make it all fit in fewer cache lines. Another is to
+ cause character arrays to be word-aligned so that `strcpy' calls
+ that copy constants to character arrays can be done inline. */
+
+#undef DATA_ALIGNMENT
+#define DATA_ALIGNMENT(TYPE, ALIGN) \
+ ((((ALIGN) < BITS_PER_WORD) \
+ && (TREE_CODE (TYPE) == ARRAY_TYPE \
+ || TREE_CODE (TYPE) == UNION_TYPE \
+ || TREE_CODE (TYPE) == RECORD_TYPE)) ? BITS_PER_WORD : (ALIGN))
+
+/* We need this for the same reason as DATA_ALIGNMENT, namely to cause
+ character arrays to be word-aligned so that `strcpy' calls that copy
+ constants to character arrays can be done inline, and 'strcmp' can be
+ optimised to use word loads. */
+#define LOCAL_ALIGNMENT(TYPE, ALIGN) DATA_ALIGNMENT (TYPE, ALIGN)
+
+/* Define if operations between registers always perform the operation
+ on the full register even if a narrower mode is specified. */
+#define WORD_REGISTER_OPERATIONS 1
+
+/* When in 64-bit mode, move insns will sign extend SImode and FCCmode
+ moves. All other references are zero extended. */
+#define LOAD_EXTEND_OP(MODE) \
+ (TARGET_64BIT && ((MODE) == SImode || (MODE) == FCCmode) ? SIGN_EXTEND \
+ : ZERO_EXTEND)
+
+/* Define this macro if it is advisable to hold scalars in registers
+ in a wider mode than that declared by the program. In such cases,
+ the value is constrained to be within the bounds of the declared
+ type, but kept valid in the wider mode. The signedness of the
+ extension may differ from that of the type. */
+
+#define PROMOTE_MODE(MODE, UNSIGNEDP, TYPE) \
+ if (GET_MODE_CLASS (MODE) == MODE_INT \
+ && GET_MODE_SIZE (MODE) < UNITS_PER_WORD) \
+ { \
+ if ((MODE) == SImode) \
+ (UNSIGNEDP) = 0; \
+ (MODE) = Pmode; \
+ }
+
+/* Pmode is always the same as ptr_mode, but not always the same as word_mode.
+ Extensions of pointers to word_mode must be signed. */
+#define POINTERS_EXTEND_UNSIGNED false
+
+/* Define if loading short immediate values into registers sign extends. */
+#define SHORT_IMMEDIATES_SIGN_EXTEND 1
+
+/* The clz.{w/d} instructions have the natural values at 0. */
+
+#define CLZ_DEFINED_VALUE_AT_ZERO(MODE, VALUE) \
+ ((VALUE) = GET_MODE_UNIT_BITSIZE (MODE), 2)
+
+/* Standard register usage. */
+
+/* Number of hardware registers. We have:
+
+ - 32 integer registers
+ - 32 floating point registers
+ - 8 condition code registers
+ - 2 fake registers:
+ - ARG_POINTER_REGNUM
+ - FRAME_POINTER_REGNUM
+*/
+
+#define FIRST_PSEUDO_REGISTER 74
+
+/* zero, tp, sp and x are fixed. */
+#define FIXED_REGISTERS \
+{ /* General registers. */ \
+ 1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ /* Floating-point registers. */ \
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ /* Others. */ \
+ 0, 0, 0, 0, 0, 0, 0, 0, 1, 1}
+
+/* The call RTLs themselves clobber ra. */
+#define CALL_USED_REGISTERS \
+{ /* General registers. */ \
+ 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, \
+ /* Floating-point registers. */ \
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, \
+ 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, \
+ /* Others. */ \
+ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1}
+
+/* Internal macros to classify a register number as to whether it's a
+ general purpose register, a floating point register, or a status
+ register. */
+
+#define GP_REG_FIRST 0
+#define GP_REG_LAST 31
+#define GP_REG_NUM (GP_REG_LAST - GP_REG_FIRST + 1)
+
+#define FP_REG_FIRST 32
+#define FP_REG_LAST 63
+#define FP_REG_NUM (FP_REG_LAST - FP_REG_FIRST + 1)
+
+/* The DWARF 2 CFA column which tracks the return address from a
+ signal handler context. This means that to maintain backwards
+ compatibility, no hard register can be assigned this column if it
+ would need to be handled by the DWARF unwinder. */
+#define DWARF_ALT_FRAME_RETURN_COLUMN 72
+
+#define FCC_REG_FIRST 64
+#define FCC_REG_LAST 71
+#define FCC_REG_NUM (FCC_REG_LAST - FCC_REG_FIRST + 1)
+
+#define GP_REG_P(REGNO) \
+ ((unsigned int) ((int) (REGNO) - GP_REG_FIRST) < GP_REG_NUM)
+#define FP_REG_P(REGNO) \
+ ((unsigned int) ((int) (REGNO) - FP_REG_FIRST) < FP_REG_NUM)
+#define FCC_REG_P(REGNO) \
+ ((unsigned int) ((int) (REGNO) - FCC_REG_FIRST) < FCC_REG_NUM)
+
+#define FP_REG_RTX_P(X) (REG_P (X) && FP_REG_P (REGNO (X)))
+
+#define HARD_REGNO_RENAME_OK(OLD_REG, NEW_REG) \
+ loongarch_hard_regno_rename_ok (OLD_REG, NEW_REG)
+
+/* Select a register mode required for caller save of hard regno REGNO. */
+#define HARD_REGNO_CALLER_SAVE_MODE(REGNO, NREGS, MODE) \
+ loongarch_hard_regno_caller_save_mode (REGNO, NREGS, MODE)
+
+/* Register to use for pushing function arguments. */
+#define STACK_POINTER_REGNUM (GP_REG_FIRST + 3)
+
+/* These two registers don't really exist: they get eliminated to either
+ the stack or hard frame pointer. */
+#define ARG_POINTER_REGNUM 72
+#define FRAME_POINTER_REGNUM 73
+
+#define HARD_FRAME_POINTER_REGNUM (GP_REG_FIRST + 22)
+
+#define HARD_FRAME_POINTER_IS_FRAME_POINTER 0
+#define HARD_FRAME_POINTER_IS_ARG_POINTER 0
+
+/* Register in which static-chain is passed to a function. */
+#define STATIC_CHAIN_REGNUM (GP_REG_FIRST + 20) /* $t8 */
+
+#define GP_TEMP_FIRST (GP_REG_FIRST + 12)
+#define LARCH_PROLOGUE_TEMP_REGNUM (GP_TEMP_FIRST + 1)
+#define LARCH_PROLOGUE_TEMP2_REGNUM (GP_TEMP_FIRST)
+#define LARCH_PROLOGUE_TEMP3_REGNUM (GP_TEMP_FIRST + 2)
+#define LARCH_EPILOGUE_TEMP_REGNUM (GP_TEMP_FIRST)
+
+#define CALLEE_SAVED_REG_NUMBER(REGNO) \
+ ((REGNO) >= 22 && (REGNO) <= 31 ? (REGNO) -22 : -1)
+
+#define LARCH_PROLOGUE_TEMP(MODE) \
+ gen_rtx_REG (MODE, LARCH_PROLOGUE_TEMP_REGNUM)
+#define LARCH_PROLOGUE_TEMP2(MODE) \
+ gen_rtx_REG (MODE, LARCH_PROLOGUE_TEMP2_REGNUM)
+#define LARCH_PROLOGUE_TEMP3(MODE) \
+ gen_rtx_REG (MODE, LARCH_PROLOGUE_TEMP3_REGNUM)
+#define LARCH_EPILOGUE_TEMP(MODE) \
+ gen_rtx_REG (MODE, LARCH_EPILOGUE_TEMP_REGNUM)
+
+/* Define this macro if it is as good or better to call a constant
+ function address than to call an address kept in a register. */
+#define NO_FUNCTION_CSE 1
+
+#define THREAD_POINTER_REGNUM (GP_REG_FIRST + 2)
+
+/* Define the classes of registers for register constraints in the
+ machine description. Also define ranges of constants.
+
+ One of the classes must always be named ALL_REGS and include all hard regs.
+ If there is more than one class, another class must be named NO_REGS
+ and contain no registers.
+
+ The name GENERAL_REGS must be the name of a class (or an alias for
+ another name such as ALL_REGS). This is the class of registers
+ that is allowed by "r" in a register constraint.
+ Also, registers outside this class are allocated only when
+ instructions express preferences for them.
+
+ The classes must be numbered in nondecreasing order; that is,
+ a larger-numbered class must never be contained completely
+ in a smaller-numbered class.
+
+ For any two classes, it is very desirable that there be another
+ class that represents their union. */
+
+enum reg_class
+{
+ NO_REGS, /* no registers in set */
+ SIBCALL_REGS, /* SIBCALL_REGS */
+ JIRL_REGS, /* JIRL_REGS */
+ GR_REGS, /* integer registers */
+ CSR_REGS, /* integer registers except for $r0 and $r1 for lcsr. */
+ FP_REGS, /* floating point registers */
+ FCC_REGS, /* status registers (fp status) */
+ FRAME_REGS, /* arg pointer and frame pointer */
+ ALL_REGS, /* all registers */
+ LIM_REG_CLASSES /* max value + 1 */
+};
+
+#define N_REG_CLASSES (int) LIM_REG_CLASSES
+
+#define GENERAL_REGS GR_REGS
+
+/* An initializer containing the names of the register classes as C
+ string constants. These names are used in writing some of the
+ debugging dumps. */
+
+#define REG_CLASS_NAMES \
+{ \
+ "NO_REGS", \
+ "SIBCALL_REGS", \
+ "JIRL_REGS", \
+ "GR_REGS", \
+ "CSR_REGS", \
+ "FP_REGS", \
+ "FCC_REGS", \
+ "FRAME_REGS", \
+ "ALL_REGS" \
+}
+
+/* An initializer containing the contents of the register classes,
+ as integers which are bit masks. The Nth integer specifies the
+ contents of class N. The way the integer MASK is interpreted is
+ that register R is in the class if `MASK & (1 << R)' is 1.
+
+ When the machine has more than 32 registers, an integer does not
+ suffice. Then the integers are replaced by sub-initializers,
+ braced groupings containing several integers. Each
+ sub-initializer must be suitable as an initializer for the type
+ `HARD_REG_SET' which is defined in `hard-reg-set.h'. */
+
+#define REG_CLASS_CONTENTS \
+{ \
+ { 0x00000000, 0x00000000, 0x00000000 }, /* NO_REGS */ \
+ { 0x001ff000, 0x00000000, 0x00000000 }, /* SIBCALL_REGS */ \
+ { 0xff9ffff0, 0x00000000, 0x00000000 }, /* JIRL_REGS */ \
+ { 0xffffffff, 0x00000000, 0x00000000 }, /* GR_REGS */ \
+ { 0xfffffffc, 0x00000000, 0x00000000 }, /* CSR_REGS */ \
+ { 0x00000000, 0xffffffff, 0x00000000 }, /* FP_REGS */ \
+ { 0x00000000, 0x00000000, 0x000000ff }, /* FCC_REGS */ \
+ { 0x00000000, 0x00000000, 0x00000300 }, /* FRAME_REGS */ \
+ { 0xffffffff, 0xffffffff, 0x000003ff } /* ALL_REGS */ \
+}
+
+/* A C expression whose value is a register class containing hard
+ register REGNO. In general there is more that one such class;
+ choose a class which is "minimal", meaning that no smaller class
+ also contains the register. */
+
+#define REGNO_REG_CLASS(REGNO) loongarch_regno_to_class[(REGNO)]
+
+/* A macro whose definition is the name of the class to which a
+ valid base register must belong. A base register is one used in
+ an address which is the register value plus a displacement. */
+
+#define BASE_REG_CLASS (GR_REGS)
+
+/* A macro whose definition is the name of the class to which a
+ valid index register must belong. An index register is one used
+ in an address where its value is either multiplied by a scale
+ factor or added to another register (as well as added to a
+ displacement). */
+
+#define INDEX_REG_CLASS NO_REGS
+
+/* We generally want to put call-clobbered registers ahead of
+ call-saved ones. (IRA expects this.) */
+
+#define REG_ALLOC_ORDER \
+{ /* Call-clobbered GPRs. */ \
+ 12, 13, 14, 15, 16, 17, 18, 19, 20, 4, 5, 6, 7, 8, 9, 10, 11, 1, \
+ /* Call-saved GPRs. */ \
+ 23, 24, 25, 26, 27, 28, 29, 30, 31, \
+ /* GPRs that can never be exposed to the register allocator. */ \
+ 0, 2, 3, 21, 22, \
+ /* Call-clobbered FPRs. */ \
+ 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, \
+ 48, 49, 50, 51,52, 53, 54, 55, \
+ 56, 57, 58, 59, 60, 61, 62, 63, \
+ /* None of the remaining classes have defined call-saved \
+ registers. */ \
+ 64, 65, 66, 67, 68, 69, 70, 71, 72, 73}
+
+#define IMM_BITS 12
+#define IMM_REACH (1LL << IMM_BITS)
+
+/* True if VALUE is an unsigned 6-bit number. */
+
+#define UIMM6_OPERAND(VALUE) (((VALUE) & ~(unsigned HOST_WIDE_INT) 0x3f) == 0)
+
+/* True if VALUE is a signed 10-bit number. */
+
+#define IMM10_OPERAND(VALUE) ((unsigned HOST_WIDE_INT) (VALUE) + 0x200 < 0x400)
+
+/* True if VALUE is a signed 12-bit number. */
+
+#define IMM12_OPERAND(VALUE) \
+ ((unsigned HOST_WIDE_INT) (VALUE) + IMM_REACH / 2 < IMM_REACH)
+
+/* True if VALUE is a signed 16-bit number. */
+
+#define IMM16_OPERAND(VALUE) \
+ ((unsigned HOST_WIDE_INT) (VALUE) + 0x8000 < 0x10000)
+
+/* True if VALUE is an unsigned 12-bit number. */
+
+#define IMM12_OPERAND_UNSIGNED(VALUE) \
+ (((VALUE) & ~(unsigned HOST_WIDE_INT) (IMM_REACH - 1)) == 0)
+
+/* True if VALUE can be loaded into a register using LU12I. */
+
+#define LU12I_OPERAND(VALUE) \
+ (((VALUE) | ((1UL << 31) - IMM_REACH)) == ((1UL << 31) - IMM_REACH) \
+ || ((VALUE) | ((1UL << 31) - IMM_REACH)) + IMM_REACH == 0)
+
+/* True if VALUE can be loaded into a register using LU32I. */
+
+#define LU32I_OPERAND(VALUE) \
+ (((VALUE) | (((1UL << 19) - 1) << 32)) == (((1UL << 19) - 1) << 32) \
+ || ((VALUE) | (((1UL << 19) - 1) << 32)) + (1UL << 32) == 0)
+
+/* True if VALUE can be loaded into a register using LU52I. */
+
+#define LU52I_OPERAND(VALUE) (((VALUE) | (0xfffUL << 52)) == (0xfffUL << 52))
+
+/* Return a value X with the low 12 bits clear, and such that
+ VALUE - X is a signed 12-bit value. */
+
+#define CONST_HIGH_PART(VALUE) (((VALUE) + (IMM_REACH / 2)) & ~(IMM_REACH - 1))
+
+#define CONST_LOW_PART(VALUE) ((VALUE) -CONST_HIGH_PART (VALUE))
+
+#define IMM12_INT(X) IMM12_OPERAND (INTVAL (X))
+#define IMM12_INT_UNSIGNED(X) IMM12_OPERAND_UNSIGNED (INTVAL (X))
+#define LU12I_INT(X) LU12I_OPERAND (INTVAL (X))
+#define LU32I_INT(X) LU32I_OPERAND (INTVAL (X))
+#define LU52I_INT(X) LU52I_OPERAND (INTVAL (X))
+#define LARCH_U12BIT_OFFSET_P(OFFSET) (IN_RANGE (OFFSET, -2048, 2047))
+#define LARCH_9BIT_OFFSET_P(OFFSET) (IN_RANGE (OFFSET, -256, 255))
+#define LARCH_16BIT_OFFSET_P(OFFSET) (IN_RANGE (OFFSET, -32768, 32767))
+#define LARCH_SHIFT_2_OFFSET_P(OFFSET) (((OFFSET) & 0x3) == 0)
+
+/* Return the maximum number of consecutive registers
+ needed to represent mode MODE in a register of class CLASS. */
+
+#define CLASS_MAX_NREGS(CLASS, MODE) loongarch_class_max_nregs (CLASS, MODE)
+
+/* Stack layout; function entry, exit and calling. */
+
+#define STACK_GROWS_DOWNWARD 1
+
+#define FRAME_GROWS_DOWNWARD 1
+
+#define RETURN_ADDR_RTX loongarch_return_addr
+
+/* Similarly, don't use the least-significant bit to tell pointers to
+ code from vtable index. */
+
+#define TARGET_PTRMEMFUNC_VBIT_LOCATION ptrmemfunc_vbit_in_delta
+
+#define ELIMINABLE_REGS \
+ { \
+ {ARG_POINTER_REGNUM, STACK_POINTER_REGNUM}, \
+ {ARG_POINTER_REGNUM, HARD_FRAME_POINTER_REGNUM}, \
+ {FRAME_POINTER_REGNUM, STACK_POINTER_REGNUM}, \
+ {FRAME_POINTER_REGNUM, HARD_FRAME_POINTER_REGNUM}, \
+ }
+
+#define INITIAL_ELIMINATION_OFFSET(FROM, TO, OFFSET) \
+ (OFFSET) = loongarch_initial_elimination_offset ((FROM), (TO))
+
+/* Allocate stack space for arguments at the beginning of each function. */
+#define ACCUMULATE_OUTGOING_ARGS 1
+
+/* The argument pointer always points to the first argument. */
+#define FIRST_PARM_OFFSET(FNDECL) 0
+
+#define REG_PARM_STACK_SPACE(FNDECL) 0
+
+/* Define this if it is the responsibility of the caller to
+ allocate the area reserved for arguments passed in registers.
+ If `ACCUMULATE_OUTGOING_ARGS' is also defined, the only effect
+ of this macro is to determine whether the space is included in
+ `crtl->outgoing_args_size'. */
+#define OUTGOING_REG_PARM_STACK_SPACE(FNTYPE) 1
+
+#define STACK_BOUNDARY (TARGET_ABI_LP64 ? 128 : 64)
+
+/* Symbolic macros for the registers used to return integer and floating
+ point values. */
+
+#define GP_RETURN (GP_REG_FIRST + 4)
+#define FP_RETURN ((TARGET_SOFT_FLOAT) ? GP_RETURN : (FP_REG_FIRST + 0))
+
+#define MAX_ARGS_IN_REGISTERS 8
+
+/* Symbolic macros for the first/last argument registers. */
+
+#define GP_ARG_FIRST (GP_REG_FIRST + 4)
+#define GP_ARG_LAST (GP_ARG_FIRST + MAX_ARGS_IN_REGISTERS - 1)
+#define FP_ARG_FIRST (FP_REG_FIRST + 0)
+#define FP_ARG_LAST (FP_ARG_FIRST + MAX_ARGS_IN_REGISTERS - 1)
+
+/* 1 if N is a possible register number for function argument passing.
+ We have no FP argument registers when soft-float. */
+
+/* Accept arguments in a0-a7, and in fa0-fa7 if permitted by the ABI. */
+#define FUNCTION_ARG_REGNO_P(N) \
+ (IN_RANGE ((N), GP_ARG_FIRST, GP_ARG_LAST) \
+ || (UNITS_PER_FP_ARG && IN_RANGE ((N), FP_ARG_FIRST, FP_ARG_LAST)))
+
+/* This structure has to cope with two different argument allocation
+ schemes. Most LoongArch ABIs view the arguments as a structure, of which
+ the first N words go in registers and the rest go on the stack. If I
+ < N, the Ith word might go in Ith integer argument register or in a
+ floating-point register. For these ABIs, we only need to remember
+ the offset of the current argument into the structure.
+
+ So for the standard ABIs, the first N words are allocated to integer
+ registers, and loongarch_function_arg decides on an argument-by-argument
+ basis whether that argument should really go in an integer register,
+ or in a floating-point one. */
+
+typedef struct loongarch_args
+{
+ /* Number of integer registers used so far, up to MAX_ARGS_IN_REGISTERS. */
+ unsigned int num_gprs;
+
+ /* Number of floating-point registers used so far, likewise. */
+ unsigned int num_fprs;
+
+} CUMULATIVE_ARGS;
+
+/* Initialize a variable CUM of type CUMULATIVE_ARGS
+ for a call to a function whose data type is FNTYPE.
+ For a library call, FNTYPE is 0. */
+
+#define INIT_CUMULATIVE_ARGS(CUM, FNTYPE, LIBNAME, INDIRECT, N_NAMED_ARGS) \
+ memset (&(CUM), 0, sizeof (CUM))
+
+#define EPILOGUE_USES(REGNO) loongarch_epilogue_uses (REGNO)
+
+/* Treat LOC as a byte offset from the stack pointer and round it up
+ to the next fully-aligned offset. */
+#define LARCH_STACK_ALIGN(LOC) \
+ (TARGET_ABI_LP64 ? ROUND_UP ((LOC), 16) : ROUND_UP ((LOC), 8))
+
+/* Output assembler code to FILE to increment profiler label # LABELNO
+ for profiling a function entry. */
+
+#define MCOUNT_NAME "_mcount"
+
+/* Emit rtl for profiling. Output assembler code to FILE
+ to call "_mcount" for profiling a function entry. */
+#define PROFILE_HOOK(LABEL) \
+ { \
+ rtx fun, ra; \
+ ra = get_hard_reg_initial_val (Pmode, RETURN_ADDR_REGNUM); \
+ fun = gen_rtx_SYMBOL_REF (Pmode, MCOUNT_NAME); \
+ emit_library_call (fun, LCT_NORMAL, VOIDmode, ra, Pmode); \
+ }
+
+/* All the work done in PROFILE_HOOK, but still required. */
+#define FUNCTION_PROFILER(STREAM, LABELNO) do { } while (0)
+
+#define NO_PROFILE_COUNTERS 1
+
+/* EXIT_IGNORE_STACK should be nonzero if, when returning from a function,
+ the stack pointer does not matter. The value is tested only in
+ functions that have frame pointers.
+ No definition is equivalent to always zero. */
+
+#define EXIT_IGNORE_STACK 1
+
+/* Trampolines are a block of code followed by two pointers. */
+
+#define TRAMPOLINE_CODE_SIZE 16
+#define TRAMPOLINE_SIZE \
+ ((Pmode == SImode) ? TRAMPOLINE_CODE_SIZE \
+ : (TRAMPOLINE_CODE_SIZE + POINTER_SIZE * 2))
+#define TRAMPOLINE_ALIGNMENT POINTER_SIZE
+
+/* loongarch_trampoline_init calls this library function to flush
+ program and data caches. */
+
+#ifndef CACHE_FLUSH_FUNC
+#define CACHE_FLUSH_FUNC "_flush_cache"
+#endif
+
+/* Addressing modes, and classification of registers for them. */
+
+#define REGNO_OK_FOR_INDEX_P(REGNO) 0
+#define REGNO_MODE_OK_FOR_BASE_P(REGNO, MODE) \
+ loongarch_regno_mode_ok_for_base_p (REGNO, MODE, 1)
+
+/* Maximum number of registers that can appear in a valid memory address. */
+
+#define MAX_REGS_PER_ADDRESS 1
+
+/* Check for constness inline but use loongarch_legitimate_address_p
+ to check whether a constant really is an address. */
+
+#define CONSTANT_ADDRESS_P(X) (CONSTANT_P (X) && memory_address_p (SImode, X))
+
+/* This handles the magic '..CURRENT_FUNCTION' symbol, which means
+ 'the start of the function that this code is output in'. */
+
+#define ASM_OUTPUT_LABELREF(FILE, NAME) \
+ do \
+ { \
+ if (strcmp (NAME, "..CURRENT_FUNCTION") == 0) \
+ asm_fprintf ((FILE), "%U%s", \
+ XSTR (XEXP (DECL_RTL (current_function_decl), 0), 0)); \
+ else \
+ asm_fprintf ((FILE), "%U%s", (NAME)); \
+ } \
+ while (0)
+
+#define CASE_VECTOR_MODE Pmode
+
+/* Only use short offsets if their range will not overflow. */
+#define CASE_VECTOR_SHORTEN_MODE(MIN, MAX, BODY) Pmode
+
+/* Define this as 1 if `char' should by default be signed; else as 0. */
+#ifndef DEFAULT_SIGNED_CHAR
+#define DEFAULT_SIGNED_CHAR 1
+#endif
+
+/* The SPARC port says:
+ The maximum number of bytes that a single instruction
+ can move quickly between memory and registers or between
+ two memory locations. */
+#define MOVE_MAX UNITS_PER_WORD
+#define MAX_MOVE_MAX 8
+
+/* The SPARC port says:
+ Nonzero if access to memory by bytes is slow and undesirable.
+ For RISC chips, it means that access to memory by bytes is no
+ better than access by words when possible, so grab a whole word
+ and maybe make use of that. */
+#define SLOW_BYTE_ACCESS 1
+
+/* Standard LoongArch integer shifts truncate the shift amount to the
+ width of the shifted operand. */
+#define SHIFT_COUNT_TRUNCATED 1
+
+/* Specify the machine mode that pointers have.
+ After generation of rtl, the compiler makes no further distinction
+ between pointers and any other objects of this machine mode. */
+
+#ifndef Pmode
+#define Pmode (TARGET_64BIT ? DImode : SImode)
+#endif
+
+/* Give call MEMs SImode since it is the "most permissive" mode
+ for both 32-bit and 64-bit targets. */
+
+#define FUNCTION_MODE SImode
+
+/* We allocate $fcc registers by hand and can't cope with moves of
+ CCmode registers to and from pseudos (or memory). */
+#define AVOID_CCMODE_COPIES
+
+/* A C expression for the cost of a branch instruction. A value of
+ 1 is the default; other values are interpreted relative to that. */
+
+#define BRANCH_COST(speed_p, predictable_p) loongarch_branch_cost
+#define LOGICAL_OP_NON_SHORT_CIRCUIT 0
+
+/* If defined, modifies the length assigned to instruction INSN as a
+ function of the context in which it is used. LENGTH is an lvalue
+ that contains the initially computed length of the insn and should
+ be updated with the correct length of the insn. */
+#define ADJUST_INSN_LENGTH(INSN, LENGTH) \
+ ((LENGTH) = loongarch_adjust_insn_length ((INSN), (LENGTH)))
+
+/* Return the asm template for a conditional branch instruction.
+ OPCODE is the opcode's mnemonic and OPERANDS is the asm template for
+ its operands. */
+#define LARCH_BRANCH(OPCODE, OPERANDS) OPCODE "\t" OPERANDS
+
+/* Control the assembler format that we output. */
+
+/* Output to assembler file text saying following lines
+ may contain character constants, extra white space, comments, etc. */
+
+#ifndef ASM_APP_ON
+#define ASM_APP_ON " #APP\n"
+#endif
+
+/* Output to assembler file text saying following lines
+ no longer contain unusual constructs. */
+
+#ifndef ASM_APP_OFF
+#define ASM_APP_OFF " #NO_APP\n"
+#endif
+
+#define REGISTER_NAMES \
+{ "$r0", "$r1", "$r2", "$r3", "$r4", "$r5", "$r6", "$r7", \
+ "$r8", "$r9", "$r10", "$r11", "$r12", "$r13", "$r14", "$r15", \
+ "$r16", "$r17", "$r18", "$r19", "$r20", "$r21", "$r22", "$r23", \
+ "$r24", "$r25", "$r26", "$r27", "$r28", "$r29", "$r30", "$r31", \
+ "$f0", "$f1", "$f2", "$f3", "$f4", "$f5", "$f6", "$f7", \
+ "$f8", "$f9", "$f10", "$f11", "$f12", "$f13", "$f14", "$f15", \
+ "$f16", "$f17", "$f18", "$f19", "$f20", "$f21", "$f22", "$f23", \
+ "$f24", "$f25", "$f26", "$f27", "$f28", "$f29", "$f30", "$f31", \
+ "$fcc0","$fcc1","$fcc2","$fcc3","$fcc4","$fcc5","$fcc6","$fcc7", \
+ "$arg", "$frame"}
+
+/* This macro defines additional names for hard registers. */
+
+#define ADDITIONAL_REGISTER_NAMES \
+{ \
+ { "zero", 0 + GP_REG_FIRST }, \
+ { "ra", 1 + GP_REG_FIRST }, \
+ { "tp", 2 + GP_REG_FIRST }, \
+ { "sp", 3 + GP_REG_FIRST }, \
+ { "a0", 4 + GP_REG_FIRST }, \
+ { "a1", 5 + GP_REG_FIRST }, \
+ { "a2", 6 + GP_REG_FIRST }, \
+ { "a3", 7 + GP_REG_FIRST }, \
+ { "a4", 8 + GP_REG_FIRST }, \
+ { "a5", 9 + GP_REG_FIRST }, \
+ { "a6", 10 + GP_REG_FIRST }, \
+ { "a7", 11 + GP_REG_FIRST }, \
+ { "t0", 12 + GP_REG_FIRST }, \
+ { "t1", 13 + GP_REG_FIRST }, \
+ { "t2", 14 + GP_REG_FIRST }, \
+ { "t3", 15 + GP_REG_FIRST }, \
+ { "t4", 16 + GP_REG_FIRST }, \
+ { "t5", 17 + GP_REG_FIRST }, \
+ { "t6", 18 + GP_REG_FIRST }, \
+ { "t7", 19 + GP_REG_FIRST }, \
+ { "t8", 20 + GP_REG_FIRST }, \
+ { "x", 21 + GP_REG_FIRST }, \
+ { "fp", 22 + GP_REG_FIRST }, \
+ { "s0", 23 + GP_REG_FIRST }, \
+ { "s1", 24 + GP_REG_FIRST }, \
+ { "s2", 25 + GP_REG_FIRST }, \
+ { "s3", 26 + GP_REG_FIRST }, \
+ { "s4", 27 + GP_REG_FIRST }, \
+ { "s5", 28 + GP_REG_FIRST }, \
+ { "s6", 29 + GP_REG_FIRST }, \
+ { "s7", 30 + GP_REG_FIRST }, \
+ { "s8", 31 + GP_REG_FIRST }, \
+ { "v0", 4 + GP_REG_FIRST }, \
+ { "v1", 5 + GP_REG_FIRST } \
+}
+
+#define DBR_OUTPUT_SEQEND(STREAM) \
+ do \
+ { \
+ /* Emit a blank line after the delay slot for emphasis. */ \
+ fputs ("\n", STREAM); \
+ } \
+ while (0)
+
+/* The LoongArch implementation uses some labels for its own purpose. The
+ following lists what labels are created, and are all formed by the
+ pattern $L[a-z].*. The machine independent portion of GCC creates
+ labels matching: $L[A-Z][0-9]+ and $L[0-9]+.
+
+ LM[0-9]+ Silicon Graphics/ECOFF stabs label before each stmt.
+ $Lb[0-9]+ Begin blocks for LoongArch debug support
+ $Lc[0-9]+ Label for use in s<xx> operation.
+ $Le[0-9]+ End blocks for LoongArch debug support. */
+
+#undef ASM_DECLARE_OBJECT_NAME
+#define ASM_DECLARE_OBJECT_NAME(STREAM, NAME, DECL) \
+ loongarch_declare_object (STREAM, NAME, "", ":\n")
+
+/* Globalizing directive for a label. */
+#define GLOBAL_ASM_OP "\t.globl\t"
+
+/* This says how to define a global common symbol. */
+
+#define ASM_OUTPUT_ALIGNED_DECL_COMMON loongarch_output_aligned_decl_common
+
+/* This says how to define a local common symbol (i.e., not visible to
+ linker). */
+
+#ifndef ASM_OUTPUT_ALIGNED_LOCAL
+#define ASM_OUTPUT_ALIGNED_LOCAL(STREAM, NAME, SIZE, ALIGN) \
+ loongarch_declare_common_object (STREAM, NAME, "\n\t.lcomm\t", SIZE, ALIGN, \
+ false)
+#endif
+
+/* This says how to output an external. It would be possible not to
+ output anything and let undefined symbol become external. However
+ the assembler uses length information on externals to allocate in
+ data/sdata bss/sbss, thereby saving exec time. */
+
+#undef ASM_OUTPUT_EXTERNAL
+#define ASM_OUTPUT_EXTERNAL(STREAM, DECL, NAME) \
+ loongarch_output_external (STREAM, DECL, NAME)
+
+/* This is how to declare a function name. The actual work of
+ emitting the label is moved to function_prologue, so that we can
+ get the line number correctly emitted before the .ent directive,
+ and after any .file directives. Define as empty so that the function
+ is not declared before the .ent directive elsewhere. */
+
+#undef ASM_DECLARE_FUNCTION_NAME
+#define ASM_DECLARE_FUNCTION_NAME(STREAM, NAME, DECL) \
+ loongarch_declare_function_name (STREAM, NAME, DECL)
+
+/* This is how to store into the string LABEL
+ the symbol_ref name of an internal numbered label where
+ PREFIX is the class of label and NUM is the number within the class.
+ This is suitable for output with `assemble_name'. */
+
+#undef ASM_GENERATE_INTERNAL_LABEL
+#define ASM_GENERATE_INTERNAL_LABEL(LABEL, PREFIX, NUM) \
+ sprintf ((LABEL), "*%s%s%ld", (LOCAL_LABEL_PREFIX), (PREFIX), (long) (NUM))
+
+/* Print debug labels as "foo = ." rather than "foo:" because they should
+ represent a byte pointer rather than an ISA-encoded address. This is
+ particularly important for code like:
+
+ $LFBxxx = .
+ .cfi_startproc
+ ...
+ .section .gcc_except_table,...
+ ...
+ .uleb128 foo-$LFBxxx
+
+ The .uleb128 requies $LFBxxx to match the FDE start address, which is
+ likewise a byte pointer rather than an ISA-encoded address.
+
+ At the time of writing, this hook is not used for the function end
+ label:
+
+ $LFExxx:
+ .end foo
+
+ */
+
+#define ASM_OUTPUT_DEBUG_LABEL(FILE, PREFIX, NUM) \
+ fprintf (FILE, "%s%s%d = .\n", LOCAL_LABEL_PREFIX, PREFIX, NUM)
+
+/* This is how to output an element of a case-vector that is absolute. */
+
+#define ASM_OUTPUT_ADDR_VEC_ELT(STREAM, VALUE) \
+ fprintf (STREAM, "\t%s\t%sL%d\n", ptr_mode == DImode ? ".dword" : ".word", \
+ LOCAL_LABEL_PREFIX, VALUE)
+
+/* This is how to output an element of a case-vector. */
+
+#define ASM_OUTPUT_ADDR_DIFF_ELT(STREAM, BODY, VALUE, REL) \
+ do \
+ { \
+ fprintf (STREAM, "\t%s\t%sL%d-%sL%d\n", \
+ ptr_mode == DImode ? ".dword" : ".word", LOCAL_LABEL_PREFIX, \
+ VALUE, LOCAL_LABEL_PREFIX, REL); \
+ } \
+ while (0)
+
+#define JUMP_TABLES_IN_TEXT_SECTION 0
+
+/* This is how to output an assembler line
+ that says to advance the location counter
+ to a multiple of 2**LOG bytes. */
+
+#define ASM_OUTPUT_ALIGN(STREAM, LOG) fprintf (STREAM, "\t.align\t%d\n", (LOG))
+
+/* "nop" instruction 54525952 (andi $r0,$r0,0) is
+ used for padding. */
+#define ASM_OUTPUT_ALIGN_WITH_NOP(STREAM, LOG) \
+ fprintf (STREAM, "\t.align\t%d,54525952,4\n", (LOG))
+
+/* This is how to output an assembler line to advance the location
+ counter by SIZE bytes. */
+
+#undef ASM_OUTPUT_SKIP
+#define ASM_OUTPUT_SKIP(STREAM, SIZE) \
+ fprintf (STREAM, "\t.space\t" HOST_WIDE_INT_PRINT_UNSIGNED "\n", (SIZE))
+
+/* This is how to output a string. */
+#undef ASM_OUTPUT_ASCII
+#define ASM_OUTPUT_ASCII loongarch_output_ascii
+
+/* Define the strings to put out for each section in the object file. */
+#define TEXT_SECTION_ASM_OP "\t.text" /* instructions */
+#define DATA_SECTION_ASM_OP "\t.data" /* large data */
+
+#undef READONLY_DATA_SECTION_ASM_OP
+#define READONLY_DATA_SECTION_ASM_OP "\t.section\t.rodata" /* read-only data */
+
+#define ASM_OUTPUT_REG_PUSH(STREAM, REGNO) \
+ do \
+ { \
+ fprintf (STREAM, "\t%s\t%s,%s,-8\n\t%s\t%s,%s,0\n", \
+ TARGET_64BIT ? "addi.d" : "addi.w", \
+ reg_names[STACK_POINTER_REGNUM], \
+ reg_names[STACK_POINTER_REGNUM], \
+ TARGET_64BIT ? "st.d" : "st.w", reg_names[REGNO], \
+ reg_names[STACK_POINTER_REGNUM]); \
+ } \
+ while (0)
+
+#define ASM_OUTPUT_REG_POP(STREAM, REGNO) \
+ do \
+ { \
+ fprintf (STREAM, "\t%s\t%s,%s,0\n\t%s\t%s,%s,8\n", \
+ TARGET_64BIT ? "ld.d" : "ld.w", reg_names[REGNO], \
+ reg_names[STACK_POINTER_REGNUM], \
+ TARGET_64BIT ? "addi.d" : "addi.w", \
+ reg_names[STACK_POINTER_REGNUM], \
+ reg_names[STACK_POINTER_REGNUM]); \
+ } \
+ while (0)
+
+/* How to start an assembler comment.
+ The leading space is important (the loongarch native assembler requires it).
+ */
+#ifndef ASM_COMMENT_START
+#define ASM_COMMENT_START " #"
+#endif
+
+#undef SIZE_TYPE
+#define SIZE_TYPE (POINTER_SIZE == 64 ? "long unsigned int" : "unsigned int")
+
+#undef PTRDIFF_TYPE
+#define PTRDIFF_TYPE (POINTER_SIZE == 64 ? "long int" : "int")
+
+/* The maximum number of bytes that can be copied by one iteration of
+ a cpymemsi loop; see loongarch_block_move_loop. */
+#define LARCH_MAX_MOVE_BYTES_PER_LOOP_ITER (UNITS_PER_WORD * 4)
+
+/* The maximum number of bytes that can be copied by a straight-line
+ implementation of cpymemsi; see loongarch_block_move_straight. We want
+ to make sure that any loop-based implementation will iterate at
+ least twice. */
+#define LARCH_MAX_MOVE_BYTES_STRAIGHT (LARCH_MAX_MOVE_BYTES_PER_LOOP_ITER * 2)
+
+/* The base cost of a memcpy call, for MOVE_RATIO and friends. These
+ values were determined experimentally by benchmarking with CSiBE.
+*/
+#define LARCH_CALL_RATIO 8
+
+/* Any loop-based implementation of cpymemsi will have at least
+ LARCH_MAX_MOVE_BYTES_STRAIGHT / UNITS_PER_WORD memory-to-memory
+ moves, so allow individual copies of fewer elements.
+
+ When cpymemsi is not available, use a value approximating
+ the length of a memcpy call sequence, so that move_by_pieces
+ will generate inline code if it is shorter than a function call.
+ Since move_by_pieces_ninsns counts memory-to-memory moves, but
+ we'll have to generate a load/store pair for each, halve the
+ value of LARCH_CALL_RATIO to take that into account. */
+
+#define MOVE_RATIO(speed) \
+ (HAVE_cpymemsi ? LARCH_MAX_MOVE_BYTES_STRAIGHT / MOVE_MAX \
+ : LARCH_CALL_RATIO / 2)
+
+/* For CLEAR_RATIO, when optimizing for size, give a better estimate
+ of the length of a memset call, but use the default otherwise. */
+
+#define CLEAR_RATIO(speed) ((speed) ? 15 : LARCH_CALL_RATIO)
+
+/* This is similar to CLEAR_RATIO, but for a non-zero constant, so when
+ optimizing for size adjust the ratio to account for the overhead of
+ loading the constant and replicating it across the word. */
+
+#define SET_RATIO(speed) ((speed) ? 15 : LARCH_CALL_RATIO - 2)
+
+#ifndef USED_FOR_TARGET
+extern const enum reg_class loongarch_regno_to_class[];
+extern int loongarch_dwarf_regno[];
+
+/* Information about a function's frame layout. */
+struct GTY (()) loongarch_frame_info
+{
+ /* The size of the frame in bytes. */
+ HOST_WIDE_INT total_size;
+
+ /* Bit X is set if the function saves or restores GPR X. */
+ unsigned int mask;
+
+ /* Likewise FPR X. */
+ unsigned int fmask;
+
+ /* How much the GPR save/restore routines adjust sp (or 0 if unused). */
+ unsigned save_libcall_adjustment;
+
+ /* Offsets of fixed-point and floating-point save areas from frame
+ bottom. */
+ HOST_WIDE_INT gp_sp_offset;
+ HOST_WIDE_INT fp_sp_offset;
+
+ /* Offset of virtual frame pointer from stack pointer/frame bottom. */
+ HOST_WIDE_INT frame_pointer_offset;
+
+ /* Offset of hard frame pointer from stack pointer/frame bottom. */
+ HOST_WIDE_INT hard_frame_pointer_offset;
+
+ /* The offset of arg_pointer_rtx from the bottom of the frame. */
+ HOST_WIDE_INT arg_pointer_offset;
+};
+
+struct GTY (()) machine_function
+{
+ /* The next floating-point condition-code register to allocate
+ for 8CC targets, relative to FCC_REG_FIRST. */
+ unsigned int next_fcc;
+
+ /* The number of extra stack bytes taken up by register varargs.
+ This area is allocated by the callee at the very top of the frame. */
+ int varargs_size;
+
+ /* The current frame information, calculated by loongarch_compute_frame_info.
+ */
+ struct loongarch_frame_info frame;
+
+ /* True if loongarch_adjust_insn_length should ignore an instruction's
+ hazard attribute. */
+ bool ignore_hazard_length_p;
+
+ /* True if at least one of the formal parameters to a function must be
+ written to the frame header (probably so its address can be taken). */
+ bool does_not_use_frame_header;
+
+ /* True if none of the functions that are called by this function need
+ stack space allocated for their arguments. */
+ bool optimize_call_stack;
+
+ /* True if one of the functions calling this function may not allocate
+ a frame header. */
+ bool callers_may_not_allocate_frame;
+};
+#endif
+
+/* As on most targets, we want the .eh_frame section to be read-only where
+ possible. And as on most targets, this means two things:
+
+ (a) Non-locally-binding pointers must have an indirect encoding,
+ so that the addresses in the .eh_frame section itself become
+ locally-binding.
+
+ (b) A shared library's .eh_frame section must encode locally-binding
+ pointers in a relative (relocation-free) form.
+
+ However, LoongArch has traditionally not allowed directives like:
+
+ .long x-.
+
+ in cases where "x" is in a different section, or is not defined in the
+ same assembly file. We are therefore unable to emit the PC-relative
+ form required by (b) at assembly time.
+
+ Fortunately, the linker is able to convert absolute addresses into
+ PC-relative addresses on our behalf. Unfortunately, only certain
+ versions of the linker know how to do this for indirect pointers,
+ and for personality data. We must fall back on using writable
+ .eh_frame sections for shared libraries if the linker does not
+ support this feature. */
+#define ASM_PREFERRED_EH_DATA_FORMAT(CODE, GLOBAL) \
+ (((GLOBAL) ? DW_EH_PE_indirect : 0) | DW_EH_PE_absptr)
+
+/* Several named LoongArch patterns depend on Pmode. These patterns have the
+ form <NAME>_si for Pmode == SImode and <NAME>_di for Pmode == DImode.
+ Add the appropriate suffix to generator function NAME and invoke it
+ with arguments ARGS. */
+#define PMODE_INSN(NAME, ARGS) \
+ (Pmode == SImode ? NAME##_si ARGS : NAME##_di ARGS)
+
+/* Do emit .note.GNU-stack by default. */
+#ifndef NEED_INDICATE_EXEC_STACK
+#define NEED_INDICATE_EXEC_STACK 1
+#endif
+
+/* The `Q' extension is not yet supported. */
+/* TODO: according to march. */
+#define UNITS_PER_FP_REG (TARGET_DOUBLE_FLOAT ? 8 : 4)
+
+/* The largest type that can be passed in floating-point registers. */
+/* TODO: according to mabi. */
+#define UNITS_PER_FP_ARG \
+ (TARGET_HARD_FLOAT ? (TARGET_DOUBLE_FLOAT ? 8 : 4) : 0)
+
+#define FUNCTION_VALUE_REGNO_P(N) ((N) == GP_RETURN || (N) == FP_RETURN)
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 06/12] LoongArch Port: Builtin functions.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (4 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 05/12] LoongArch Port: Machine description C files and .h files chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 07/12] LoongArch Port: Builtin macros chenglulu
` (6 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/
* config/loongarch/larchintrin.h: New file.
* config/loongarch/loongarch-builtins.c: New file.
---
gcc/config/loongarch/larchintrin.h | 413 +++++++++++++++++
gcc/config/loongarch/loongarch-builtins.c | 511 ++++++++++++++++++++++
2 files changed, 924 insertions(+)
create mode 100644 gcc/config/loongarch/larchintrin.h
create mode 100644 gcc/config/loongarch/loongarch-builtins.c
diff --git a/gcc/config/loongarch/larchintrin.h b/gcc/config/loongarch/larchintrin.h
new file mode 100644
index 00000000000..fcf5043841d
--- /dev/null
+++ b/gcc/config/loongarch/larchintrin.h
@@ -0,0 +1,413 @@
+/* Intrinsics for LoongArch BASE operations.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published
+by the Free Software Foundation; either version 3, or (at your
+option) any later version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+License for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef _GCC_LOONGARCH_BASE_INTRIN_H
+#define _GCC_LOONGARCH_BASE_INTRIN_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef struct drdtime
+{
+ unsigned long dvalue;
+ unsigned long dtimeid;
+} __drdtime_t;
+
+typedef struct rdtime
+{
+ unsigned int value;
+ unsigned int timeid;
+} __rdtime_t;
+
+#ifdef __loongarch64
+extern __inline __drdtime_t
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_rdtime_d (void)
+{
+ __drdtime_t drdtime;
+ __asm__ volatile (
+ "rdtime.d\t%[val],%[tid]\n\t"
+ : [val]"=&r"(drdtime.dvalue),[tid]"=&r"(drdtime.dtimeid)
+ :);
+ return drdtime;
+}
+#define __rdtime_d __builtin_loongarch_rdtime_d
+#endif
+
+extern __inline __rdtime_t
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_rdtimeh_w (void)
+{
+ __rdtime_t rdtime;
+ __asm__ volatile (
+ "rdtimeh.w\t%[val],%[tid]\n\t"
+ : [val]"=&r"(rdtime.value),[tid]"=&r"(rdtime.timeid)
+ :);
+ return rdtime;
+}
+#define __rdtimel_w __builtin_loongarch_rdtimel_w
+
+extern __inline __rdtime_t
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_rdtimel_w (void)
+{
+ __rdtime_t rdtime;
+ __asm__ volatile (
+ "rdtimel.w\t%[val],%[tid]\n\t"
+ : [val]"=&r"(rdtime.value),[tid]"=&r"(rdtime.timeid)
+ :);
+ return rdtime;
+}
+#define __rdtimeh_w __builtin_loongarch_rdtimeh_w
+
+/* Assembly instruction format: rj, fcsr. */
+/* Data types in instruction templates: USI, UQI. */
+#define __movfcsr2gr(/*ui5*/ _1) __builtin_loongarch_movfcsr2gr ((_1));
+
+/* Assembly instruction format: 0, fcsr, rj. */
+/* Data types in instruction templates: VOID, UQI, USI. */
+#define __movgr2fcsr(/*ui5*/ _1, _2) \
+ __builtin_loongarch_movgr2fcsr ((unsigned short) _1, (unsigned int) _2);
+
+#if defined __loongarch64
+/* Assembly instruction format: ui5, rj, si12. */
+/* Data types in instruction templates: VOID, USI, UDI, SI. */
+#define __dcacop(/*ui5*/ _1, /*unsigned long int*/ _2, /*si12*/ _3) \
+ ((void) __builtin_loongarch_dcacop ((_1), (unsigned long int) (_2), (_3)))
+#else
+#error "Don't support this ABI."
+#endif
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: USI, USI. */
+extern __inline unsigned int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__cpucfg (unsigned int _1)
+{
+ return (unsigned int) __builtin_loongarch_cpucfg ((unsigned int) _1);
+}
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: DI, DI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__asrtle_d (long int _1, long int _2)
+{
+ __builtin_loongarch_asrtle_d ((long int) _1, (long int) _2);
+}
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: DI, DI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__asrtgt_d (long int _1, long int _2)
+{
+ __builtin_loongarch_asrtgt_d ((long int) _1, (long int) _2);
+}
+#endif
+
+#if defined __loongarch64
+/* Assembly instruction format: rd, rj, ui5. */
+/* Data types in instruction templates: DI, DI, UQI. */
+#define __dlddir(/*long int*/ _1, /*ui5*/ _2) \
+ ((long int) __builtin_loongarch_dlddir ((long int) (_1), (_2)))
+#else
+#error "Don't support this ABI."
+#endif
+
+#if defined __loongarch64
+/* Assembly instruction format: rj, ui5. */
+/* Data types in instruction templates: VOID, DI, UQI. */
+#define __dldpte(/*long int*/ _1, /*ui5*/ _2) \
+ ((void) __builtin_loongarch_dldpte ((long int) (_1), (_2)))
+#else
+#error "Don't support this ABI."
+#endif
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, QI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crc_w_b_w (char _1, int _2)
+{
+ return (int) __builtin_loongarch_crc_w_b_w ((char) _1, (int) _2);
+}
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, HI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crc_w_h_w (short _1, int _2)
+{
+ return (int) __builtin_loongarch_crc_w_h_w ((short) _1, (int) _2);
+}
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, SI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crc_w_w_w (int _1, int _2)
+{
+ return (int) __builtin_loongarch_crc_w_w_w ((int) _1, (int) _2);
+}
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, DI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crc_w_d_w (long int _1, int _2)
+{
+ return (int) __builtin_loongarch_crc_w_d_w ((long int) _1, (int) _2);
+}
+#endif
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, QI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crcc_w_b_w (char _1, int _2)
+{
+ return (int) __builtin_loongarch_crcc_w_b_w ((char) _1, (int) _2);
+}
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, HI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crcc_w_h_w (short _1, int _2)
+{
+ return (int) __builtin_loongarch_crcc_w_h_w ((short) _1, (int) _2);
+}
+
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, SI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crcc_w_w_w (int _1, int _2)
+{
+ return (int) __builtin_loongarch_crcc_w_w_w ((int) _1, (int) _2);
+}
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, rj, rk. */
+/* Data types in instruction templates: SI, DI, SI. */
+extern __inline int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__crcc_w_d_w (long int _1, int _2)
+{
+ return (int) __builtin_loongarch_crcc_w_d_w ((long int) _1, (int) _2);
+}
+#endif
+
+/* Assembly instruction format: rd, ui14. */
+/* Data types in instruction templates: USI, USI. */
+#define __csrrd(/*ui14*/ _1) ((unsigned int) __builtin_loongarch_csrrd ((_1)))
+
+/* Assembly instruction format: rd, ui14. */
+/* Data types in instruction templates: USI, USI, USI. */
+#define __csrwr(/*unsigned int*/ _1, /*ui14*/ _2) \
+ ((unsigned int) __builtin_loongarch_csrwr ((unsigned int) (_1), (_2)))
+
+/* Assembly instruction format: rd, rj, ui14. */
+/* Data types in instruction templates: USI, USI, USI, USI. */
+#define __csrxchg(/*unsigned int*/ _1, /*unsigned int*/ _2, /*ui14*/ _3) \
+ ((unsigned int) __builtin_loongarch_csrxchg ((unsigned int) (_1), \
+ (unsigned int) (_2), (_3)))
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, ui14. */
+/* Data types in instruction templates: UDI, USI. */
+#define __dcsrrd(/*ui14*/ _1) \
+ ((unsigned long int) __builtin_loongarch_dcsrrd ((_1)))
+
+/* Assembly instruction format: rd, ui14. */
+/* Data types in instruction templates: UDI, UDI, USI. */
+#define __dcsrwr(/*unsigned long int*/ _1, /*ui14*/ _2) \
+ ((unsigned long int) __builtin_loongarch_dcsrwr ((unsigned long int) (_1), \
+ (_2)))
+
+/* Assembly instruction format: rd, rj, ui14. */
+/* Data types in instruction templates: UDI, UDI, UDI, USI. */
+#define __dcsrxchg(/*unsigned long int*/ _1, /*unsigned long int*/ _2, \
+ /*ui14*/ _3) \
+ ((unsigned long int) __builtin_loongarch_dcsrxchg ( \
+ (unsigned long int) (_1), (unsigned long int) (_2), (_3)))
+#endif
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: UQI, USI. */
+extern __inline unsigned char
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrrd_b (unsigned int _1)
+{
+ return (unsigned char) __builtin_loongarch_iocsrrd_b ((unsigned int) _1);
+}
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: UHI, USI. */
+extern __inline unsigned char
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrrd_h (unsigned int _1)
+{
+ return (unsigned short) __builtin_loongarch_iocsrrd_h ((unsigned int) _1);
+}
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: USI, USI. */
+extern __inline unsigned int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrrd_w (unsigned int _1)
+{
+ return (unsigned int) __builtin_loongarch_iocsrrd_w ((unsigned int) _1);
+}
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: UDI, USI. */
+extern __inline unsigned long int
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrrd_d (unsigned int _1)
+{
+ return (unsigned long int) __builtin_loongarch_iocsrrd_d ((unsigned int) _1);
+}
+#endif
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: VOID, UQI, USI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrwr_b (unsigned char _1, unsigned int _2)
+{
+ return (void) __builtin_loongarch_iocsrwr_b ((unsigned char) _1,
+ (unsigned int) _2);
+}
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: VOID, UHI, USI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrwr_h (unsigned short _1, unsigned int _2)
+{
+ return (void) __builtin_loongarch_iocsrwr_h ((unsigned short) _1,
+ (unsigned int) _2);
+}
+
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: VOID, USI, USI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrwr_w (unsigned int _1, unsigned int _2)
+{
+ return (void) __builtin_loongarch_iocsrwr_w ((unsigned int) _1,
+ (unsigned int) _2);
+}
+
+#ifdef __loongarch64
+/* Assembly instruction format: rd, rj. */
+/* Data types in instruction templates: VOID, UDI, USI. */
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__iocsrwr_d (unsigned long int _1, unsigned int _2)
+{
+ return (void) __builtin_loongarch_iocsrwr_d ((unsigned long int) _1,
+ (unsigned int) _2);
+}
+#endif
+
+/* Assembly instruction format: ui15. */
+/* Data types in instruction templates: UQI. */
+#define __dbar(/*ui15*/ _1) __builtin_loongarch_dbar ((_1))
+
+/* Assembly instruction format: ui15. */
+/* Data types in instruction templates: UQI. */
+#define __ibar(/*ui15*/ _1) __builtin_loongarch_ibar ((_1))
+
+#define __builtin_loongarch_syscall(a) \
+ { \
+ __asm__ volatile ("syscall %0\n\t" ::"I"(a)); \
+ }
+#define __syscall __builtin_loongarch_syscall
+
+#define __builtin_loongarch_break(a) \
+ { \
+ __asm__ volatile ("break %0\n\t" ::"I"(a)); \
+ }
+#define __break __builtin_loongarch_break
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbsrch (void)
+{
+ __asm__ volatile ("tlbsrch\n\t");
+}
+#define __tlbsrch __builtin_loongarch_tlbsrch
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbrd (void)
+{
+ __asm__ volatile ("tlbrd\n\t");
+}
+#define __tlbrd __builtin_loongarch_tlbrd
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbwr (void)
+{
+ __asm__ volatile ("tlbwr\n\t");
+}
+#define __tlbwr __builtin_loongarch_tlbwr
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbfill (void)
+{
+ __asm__ volatile ("tlbfill\n\t");
+}
+#define __tlbfill __builtin_loongarch_tlbfill
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbclr (void)
+{
+ __asm__ volatile ("tlbclr\n\t");
+}
+#define __tlbclr __builtin_loongarch_tlbclr
+
+extern __inline void
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+__builtin_loongarch_tlbflush (void)
+{
+ __asm__ volatile ("tlbflush\n\t");
+}
+#define __tlbflush __builtin_loongarch_tlbflush
+
+#ifdef __cplusplus
+}
+#endif
+#endif /* _GCC_LOONGARCH_BASE_INTRIN_H */
diff --git a/gcc/config/loongarch/loongarch-builtins.c b/gcc/config/loongarch/loongarch-builtins.c
new file mode 100644
index 00000000000..53112f63b8a
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-builtins.c
@@ -0,0 +1,511 @@
+/* Subroutines used for expanding LoongArch builtins.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "backend.h"
+#include "target.h"
+#include "rtl.h"
+#include "tree.h"
+#include "memmodel.h"
+#include "gimple.h"
+#include "tm_p.h"
+#include "optabs.h"
+#include "recog.h"
+#include "diagnostic.h"
+#include "fold-const.h"
+#include "expr.h"
+#include "langhooks.h"
+
+/* Macros to create an enumeration identifier for a function prototype. */
+#define LARCH_FTYPE_NAME1(A, B) LARCH_##A##_FTYPE_##B
+#define LARCH_FTYPE_NAME2(A, B, C) LARCH_##A##_FTYPE_##B##_##C
+#define LARCH_FTYPE_NAME3(A, B, C, D) LARCH_##A##_FTYPE_##B##_##C##_##D
+#define LARCH_FTYPE_NAME4(A, B, C, D, E) \
+ LARCH_##A##_FTYPE_##B##_##C##_##D##_##E
+
+/* Classifies the prototype of a built-in function. */
+enum loongarch_function_type
+{
+#define DEF_LARCH_FTYPE(NARGS, LIST) LARCH_FTYPE_NAME##NARGS LIST,
+#include "config/loongarch/loongarch-ftypes.def"
+#undef DEF_LARCH_FTYPE
+ LARCH_MAX_FTYPE_MAX
+};
+
+/* Specifies how a built-in function should be converted into rtl. */
+enum loongarch_builtin_type
+{
+ /* The function corresponds directly to an .md pattern. The return
+ value is mapped to operand 0 and the arguments are mapped to
+ operands 1 and above. */
+ LARCH_BUILTIN_DIRECT,
+
+ /* The function corresponds directly to an .md pattern. There is no return
+ value and the arguments are mapped to operands 0 and above. */
+ LARCH_BUILTIN_DIRECT_NO_TARGET,
+
+};
+
+/* Invoke MACRO (COND) for each fcmp.cond.{s/d} condition. */
+#define LARCH_FP_CONDITIONS(MACRO) \
+ MACRO (f), \
+ MACRO (un), \
+ MACRO (eq), \
+ MACRO (ueq), \
+ MACRO (olt), \
+ MACRO (ult), \
+ MACRO (ole), \
+ MACRO (ule), \
+ MACRO (sf), \
+ MACRO (ngle), \
+ MACRO (seq), \
+ MACRO (ngl), \
+ MACRO (lt), \
+ MACRO (nge), \
+ MACRO (le), \
+ MACRO (ngt)
+
+/* Enumerates the codes above as LARCH_FP_COND_<X>. */
+#define DECLARE_LARCH_COND(X) LARCH_FP_COND_##X
+enum loongarch_fp_condition
+{
+ LARCH_FP_CONDITIONS (DECLARE_LARCH_COND)
+};
+#undef DECLARE_LARCH_COND
+
+/* Index X provides the string representation of LARCH_FP_COND_<X>. */
+#define STRINGIFY(X) #X
+const char *const
+loongarch_fp_conditions[16]= {LARCH_FP_CONDITIONS (STRINGIFY)};
+#undef STRINGIFY
+
+/* Declare an availability predicate for built-in functions that require
+ * COND to be true. NAME is the main part of the predicate's name. */
+#define AVAIL_ALL(NAME, COND) \
+ static unsigned int \
+ loongarch_builtin_avail_##NAME (void) \
+ { \
+ return (COND) ? 1 : 0; \
+ }
+
+static unsigned int
+loongarch_builtin_avail_default (void)
+{
+ return 1;
+}
+/* This structure describes a single built-in function. */
+struct loongarch_builtin_description
+{
+ /* The code of the main .md file instruction. See loongarch_builtin_type
+ for more information. */
+ enum insn_code icode;
+
+ /* The floating-point comparison code to use with ICODE, if any. */
+ enum loongarch_fp_condition cond;
+
+ /* The name of the built-in function. */
+ const char *name;
+
+ /* Specifies how the function should be expanded. */
+ enum loongarch_builtin_type builtin_type;
+
+ /* The function's prototype. */
+ enum loongarch_function_type function_type;
+
+ /* Whether the function is available. */
+ unsigned int (*avail) (void);
+};
+
+AVAIL_ALL (hard_float, TARGET_HARD_FLOAT_ABI)
+
+/* Construct a loongarch_builtin_description from the given arguments.
+
+ INSN is the name of the associated instruction pattern, without the
+ leading CODE_FOR_loongarch_.
+
+ CODE is the floating-point condition code associated with the
+ function. It can be 'f' if the field is not applicable.
+
+ NAME is the name of the function itself, without the leading
+ "__builtin_loongarch_".
+
+ BUILTIN_TYPE and FUNCTION_TYPE are loongarch_builtin_description fields.
+
+ AVAIL is the name of the availability predicate, without the leading
+ loongarch_builtin_avail_. */
+#define LARCH_BUILTIN(INSN, COND, NAME, BUILTIN_TYPE, FUNCTION_TYPE, AVAIL) \
+ { \
+ CODE_FOR_loongarch_##INSN, LARCH_FP_COND_##COND, \
+ "__builtin_loongarch_" NAME, BUILTIN_TYPE, FUNCTION_TYPE, \
+ loongarch_builtin_avail_##AVAIL \
+ }
+
+/* Define __builtin_loongarch_<INSN>, which is a LARCH_BUILTIN_DIRECT function
+ mapped to instruction CODE_FOR_loongarch_<INSN>, FUNCTION_TYPE and AVAIL
+ are as for LARCH_BUILTIN. */
+#define DIRECT_BUILTIN(INSN, FUNCTION_TYPE, AVAIL) \
+ LARCH_BUILTIN (INSN, f, #INSN, LARCH_BUILTIN_DIRECT, FUNCTION_TYPE, AVAIL)
+
+/* Define __builtin_loongarch_<INSN>, which is a LARCH_BUILTIN_DIRECT_NO_TARGET
+ function mapped to instruction CODE_FOR_loongarch_<INSN>, FUNCTION_TYPE
+ and AVAIL are as for LARCH_BUILTIN. */
+#define DIRECT_NO_TARGET_BUILTIN(INSN, FUNCTION_TYPE, AVAIL) \
+ LARCH_BUILTIN (INSN, f, #INSN, LARCH_BUILTIN_DIRECT_NO_TARGET, \
+ FUNCTION_TYPE, AVAIL)
+
+/* Loongson support crc. */
+#define CODE_FOR_loongarch_crc_w_b_w CODE_FOR_crc_w_b_w
+#define CODE_FOR_loongarch_crc_w_h_w CODE_FOR_crc_w_h_w
+#define CODE_FOR_loongarch_crc_w_w_w CODE_FOR_crc_w_w_w
+#define CODE_FOR_loongarch_crc_w_d_w CODE_FOR_crc_w_d_w
+#define CODE_FOR_loongarch_crcc_w_b_w CODE_FOR_crcc_w_b_w
+#define CODE_FOR_loongarch_crcc_w_h_w CODE_FOR_crcc_w_h_w
+#define CODE_FOR_loongarch_crcc_w_w_w CODE_FOR_crcc_w_w_w
+#define CODE_FOR_loongarch_crcc_w_d_w CODE_FOR_crcc_w_d_w
+
+/* Privileged state instruction. */
+#define CODE_FOR_loongarch_cpucfg CODE_FOR_cpucfg
+#define CODE_FOR_loongarch_asrtle_d CODE_FOR_asrtle_d
+#define CODE_FOR_loongarch_asrtgt_d CODE_FOR_asrtgt_d
+#define CODE_FOR_loongarch_csrrd CODE_FOR_csrrd
+#define CODE_FOR_loongarch_dcsrrd CODE_FOR_dcsrrd
+#define CODE_FOR_loongarch_csrwr CODE_FOR_csrwr
+#define CODE_FOR_loongarch_dcsrwr CODE_FOR_dcsrwr
+#define CODE_FOR_loongarch_csrxchg CODE_FOR_csrxchg
+#define CODE_FOR_loongarch_dcsrxchg CODE_FOR_dcsrxchg
+#define CODE_FOR_loongarch_iocsrrd_b CODE_FOR_iocsrrd_b
+#define CODE_FOR_loongarch_iocsrrd_h CODE_FOR_iocsrrd_h
+#define CODE_FOR_loongarch_iocsrrd_w CODE_FOR_iocsrrd_w
+#define CODE_FOR_loongarch_iocsrrd_d CODE_FOR_iocsrrd_d
+#define CODE_FOR_loongarch_iocsrwr_b CODE_FOR_iocsrwr_b
+#define CODE_FOR_loongarch_iocsrwr_h CODE_FOR_iocsrwr_h
+#define CODE_FOR_loongarch_iocsrwr_w CODE_FOR_iocsrwr_w
+#define CODE_FOR_loongarch_iocsrwr_d CODE_FOR_iocsrwr_d
+#define CODE_FOR_loongarch_lddir CODE_FOR_lddir
+#define CODE_FOR_loongarch_dlddir CODE_FOR_dlddir
+#define CODE_FOR_loongarch_ldpte CODE_FOR_ldpte
+#define CODE_FOR_loongarch_dldpte CODE_FOR_dldpte
+#define CODE_FOR_loongarch_cacop CODE_FOR_cacop
+#define CODE_FOR_loongarch_dcacop CODE_FOR_dcacop
+#define CODE_FOR_loongarch_dbar CODE_FOR_dbar
+#define CODE_FOR_loongarch_ibar CODE_FOR_ibar
+
+static const struct loongarch_builtin_description loongarch_builtins[] = {
+#define LARCH_MOVFCSR2GR 0
+ DIRECT_BUILTIN (movfcsr2gr, LARCH_USI_FTYPE_UQI, hard_float),
+#define LARCH_MOVGR2FCSR 1
+ DIRECT_NO_TARGET_BUILTIN (movgr2fcsr, LARCH_VOID_FTYPE_UQI_USI, hard_float),
+
+ DIRECT_NO_TARGET_BUILTIN (cacop, LARCH_VOID_FTYPE_USI_USI_SI, default),
+ DIRECT_NO_TARGET_BUILTIN (dcacop, LARCH_VOID_FTYPE_USI_UDI_SI, default),
+ DIRECT_NO_TARGET_BUILTIN (dbar, LARCH_VOID_FTYPE_USI, default),
+ DIRECT_NO_TARGET_BUILTIN (ibar, LARCH_VOID_FTYPE_USI, default),
+
+ DIRECT_BUILTIN (cpucfg, LARCH_USI_FTYPE_USI, default),
+ DIRECT_BUILTIN (asrtle_d, LARCH_VOID_FTYPE_DI_DI, default),
+ DIRECT_BUILTIN (asrtgt_d, LARCH_VOID_FTYPE_DI_DI, default),
+ DIRECT_BUILTIN (dlddir, LARCH_DI_FTYPE_DI_UQI, default),
+ DIRECT_BUILTIN (lddir, LARCH_SI_FTYPE_SI_UQI, default),
+ DIRECT_NO_TARGET_BUILTIN (dldpte, LARCH_VOID_FTYPE_DI_UQI, default),
+ DIRECT_NO_TARGET_BUILTIN (ldpte, LARCH_VOID_FTYPE_SI_UQI, default),
+
+ /* CRC Instrinsic */
+
+ DIRECT_BUILTIN (crc_w_b_w, LARCH_SI_FTYPE_QI_SI, default),
+ DIRECT_BUILTIN (crc_w_h_w, LARCH_SI_FTYPE_HI_SI, default),
+ DIRECT_BUILTIN (crc_w_w_w, LARCH_SI_FTYPE_SI_SI, default),
+ DIRECT_BUILTIN (crc_w_d_w, LARCH_SI_FTYPE_DI_SI, default),
+ DIRECT_BUILTIN (crcc_w_b_w, LARCH_SI_FTYPE_QI_SI, default),
+ DIRECT_BUILTIN (crcc_w_h_w, LARCH_SI_FTYPE_HI_SI, default),
+ DIRECT_BUILTIN (crcc_w_w_w, LARCH_SI_FTYPE_SI_SI, default),
+ DIRECT_BUILTIN (crcc_w_d_w, LARCH_SI_FTYPE_DI_SI, default),
+
+ DIRECT_BUILTIN (csrrd, LARCH_USI_FTYPE_USI, default),
+ DIRECT_BUILTIN (dcsrrd, LARCH_UDI_FTYPE_USI, default),
+ DIRECT_BUILTIN (csrwr, LARCH_USI_FTYPE_USI_USI, default),
+ DIRECT_BUILTIN (dcsrwr, LARCH_UDI_FTYPE_UDI_USI, default),
+ DIRECT_BUILTIN (csrxchg, LARCH_USI_FTYPE_USI_USI_USI, default),
+ DIRECT_BUILTIN (dcsrxchg, LARCH_UDI_FTYPE_UDI_UDI_USI, default),
+ DIRECT_BUILTIN (iocsrrd_b, LARCH_UQI_FTYPE_USI, default),
+ DIRECT_BUILTIN (iocsrrd_h, LARCH_UHI_FTYPE_USI, default),
+ DIRECT_BUILTIN (iocsrrd_w, LARCH_USI_FTYPE_USI, default),
+ DIRECT_BUILTIN (iocsrrd_d, LARCH_UDI_FTYPE_USI, default),
+ DIRECT_NO_TARGET_BUILTIN (iocsrwr_b, LARCH_VOID_FTYPE_UQI_USI, default),
+ DIRECT_NO_TARGET_BUILTIN (iocsrwr_h, LARCH_VOID_FTYPE_UHI_USI, default),
+ DIRECT_NO_TARGET_BUILTIN (iocsrwr_w, LARCH_VOID_FTYPE_USI_USI, default),
+ DIRECT_NO_TARGET_BUILTIN (iocsrwr_d, LARCH_VOID_FTYPE_UDI_USI, default),
+};
+
+/* Index I is the function declaration for loongarch_builtins[I], or null if
+ the function isn't defined on this target. */
+static GTY (()) tree loongarch_builtin_decls[ARRAY_SIZE (loongarch_builtins)];
+/* Get the index I of the function declaration for loongarch_builtin_decls[I]
+ using the instruction code or return null if not defined for the target. */
+static GTY (()) int loongarch_get_builtin_decl_index[NUM_INSN_CODES];
+
+/* Return a type for 'const volatile void *'. */
+
+static tree
+loongarch_build_cvpointer_type (void)
+{
+ static tree cache;
+
+ if (cache == NULL_TREE)
+ cache = build_pointer_type (build_qualified_type (void_type_node,
+ TYPE_QUAL_CONST
+ | TYPE_QUAL_VOLATILE));
+ return cache;
+}
+
+/* Source-level argument types. */
+#define LARCH_ATYPE_VOID void_type_node
+#define LARCH_ATYPE_INT integer_type_node
+#define LARCH_ATYPE_POINTER ptr_type_node
+#define LARCH_ATYPE_CVPOINTER loongarch_build_cvpointer_type ()
+
+/* Standard mode-based argument types. */
+#define LARCH_ATYPE_QI intQI_type_node
+#define LARCH_ATYPE_UQI unsigned_intQI_type_node
+#define LARCH_ATYPE_HI intHI_type_node
+#define LARCH_ATYPE_UHI unsigned_intHI_type_node
+#define LARCH_ATYPE_SI intSI_type_node
+#define LARCH_ATYPE_USI unsigned_intSI_type_node
+#define LARCH_ATYPE_DI intDI_type_node
+#define LARCH_ATYPE_UDI unsigned_intDI_type_node
+#define LARCH_ATYPE_SF float_type_node
+#define LARCH_ATYPE_DF double_type_node
+
+/* LARCH_FTYPE_ATYPESN takes N LARCH_FTYPES-like type codes and lists
+ their associated LARCH_ATYPEs. */
+#define LARCH_FTYPE_ATYPES1(A, B) LARCH_ATYPE_##A, LARCH_ATYPE_##B
+
+#define LARCH_FTYPE_ATYPES2(A, B, C) \
+ LARCH_ATYPE_##A, LARCH_ATYPE_##B, LARCH_ATYPE_##C
+
+#define LARCH_FTYPE_ATYPES3(A, B, C, D) \
+ LARCH_ATYPE_##A, LARCH_ATYPE_##B, LARCH_ATYPE_##C, LARCH_ATYPE_##D
+
+#define LARCH_FTYPE_ATYPES4(A, B, C, D, E) \
+ LARCH_ATYPE_##A, LARCH_ATYPE_##B, LARCH_ATYPE_##C, LARCH_ATYPE_##D, \
+ LARCH_ATYPE_##E
+
+/* Return the function type associated with function prototype TYPE. */
+
+static tree
+loongarch_build_function_type (enum loongarch_function_type type)
+{
+ static tree types[(int) LARCH_MAX_FTYPE_MAX];
+
+ if (types[(int) type] == NULL_TREE)
+ switch (type)
+ {
+#define DEF_LARCH_FTYPE(NUM, ARGS) \
+ case LARCH_FTYPE_NAME##NUM ARGS: \
+ types[(int) type] \
+ = build_function_type_list (LARCH_FTYPE_ATYPES##NUM ARGS, NULL_TREE); \
+ break;
+#include "config/loongarch/loongarch-ftypes.def"
+#undef DEF_LARCH_FTYPE
+ default:
+ gcc_unreachable ();
+ }
+
+ return types[(int) type];
+}
+
+/* Implement TARGET_INIT_BUILTINS. */
+
+void
+loongarch_init_builtins (void)
+{
+ const struct loongarch_builtin_description *d;
+ unsigned int i;
+ tree type;
+
+ /* Iterate through all of the bdesc arrays, initializing all of the
+ builtin functions. */
+ for (i = 0; i < ARRAY_SIZE (loongarch_builtins); i++)
+ {
+ d = &loongarch_builtins[i];
+ if (d->avail ())
+ {
+ type = loongarch_build_function_type (d->function_type);
+ loongarch_builtin_decls[i]
+ = add_builtin_function (d->name, type, i, BUILT_IN_MD, NULL,
+ NULL);
+ loongarch_get_builtin_decl_index[d->icode] = i;
+ }
+ }
+}
+
+/* Implement TARGET_BUILTIN_DECL. */
+
+tree
+loongarch_builtin_decl (unsigned int code, bool initialize_p ATTRIBUTE_UNUSED)
+{
+ if (code >= ARRAY_SIZE (loongarch_builtins))
+ return error_mark_node;
+ return loongarch_builtin_decls[code];
+}
+
+/* Take argument ARGNO from EXP's argument list and convert it into
+ an expand operand. Store the operand in *OP. */
+
+static void
+loongarch_prepare_builtin_arg (struct expand_operand *op, tree exp,
+ unsigned int argno)
+{
+ tree arg;
+ rtx value;
+
+ arg = CALL_EXPR_ARG (exp, argno);
+ value = expand_normal (arg);
+ create_input_operand (op, value, TYPE_MODE (TREE_TYPE (arg)));
+}
+
+/* Expand instruction ICODE as part of a built-in function sequence.
+ Use the first NOPS elements of OPS as the instruction's operands.
+ HAS_TARGET_P is true if operand 0 is a target; it is false if the
+ instruction has no target.
+
+ Return the target rtx if HAS_TARGET_P, otherwise return const0_rtx. */
+
+static rtx
+loongarch_expand_builtin_insn (enum insn_code icode, unsigned int nops,
+ struct expand_operand *ops, bool has_target_p)
+{
+ if (!maybe_expand_insn (icode, nops, ops))
+ {
+ error ("invalid argument to built-in function");
+ return has_target_p ? gen_reg_rtx (ops[0].mode) : const0_rtx;
+ }
+ return has_target_p ? ops[0].value : const0_rtx;
+}
+
+/* Expand a LARCH_BUILTIN_DIRECT or LARCH_BUILTIN_DIRECT_NO_TARGET function;
+ HAS_TARGET_P says which. EXP is the CALL_EXPR that calls the function
+ and ICODE is the code of the associated .md pattern. TARGET, if nonnull,
+ suggests a good place to put the result. */
+
+static rtx
+loongarch_expand_builtin_direct (enum insn_code icode, rtx target, tree exp,
+ bool has_target_p)
+{
+ struct expand_operand ops[MAX_RECOG_OPERANDS];
+ int opno, argno;
+
+ /* Map any target to operand 0. */
+ opno = 0;
+ if (has_target_p)
+ create_output_operand (&ops[opno++], target, TYPE_MODE (TREE_TYPE (exp)));
+
+ /* Map the arguments to the other operands. */
+ gcc_assert (opno + call_expr_nargs (exp)
+ == insn_data[icode].n_generator_args);
+ for (argno = 0; argno < call_expr_nargs (exp); argno++)
+ loongarch_prepare_builtin_arg (&ops[opno++], exp, argno);
+
+ return loongarch_expand_builtin_insn (icode, opno, ops, has_target_p);
+}
+
+/* Implement TARGET_EXPAND_BUILTIN. */
+
+rtx
+loongarch_expand_builtin (tree exp, rtx target, rtx subtarget ATTRIBUTE_UNUSED,
+ machine_mode mode ATTRIBUTE_UNUSED,
+ int ignore ATTRIBUTE_UNUSED)
+{
+ tree fndecl;
+ unsigned int fcode, avail;
+ const struct loongarch_builtin_description *d;
+
+ fndecl = TREE_OPERAND (CALL_EXPR_FN (exp), 0);
+ fcode = DECL_MD_FUNCTION_CODE (fndecl);
+ gcc_assert (fcode < ARRAY_SIZE (loongarch_builtins));
+ d = &loongarch_builtins[fcode];
+ avail = d->avail ();
+ gcc_assert (avail != 0);
+ switch (d->builtin_type)
+ {
+ case LARCH_BUILTIN_DIRECT:
+ return loongarch_expand_builtin_direct (d->icode, target, exp, true);
+
+ case LARCH_BUILTIN_DIRECT_NO_TARGET:
+ return loongarch_expand_builtin_direct (d->icode, target, exp, false);
+ }
+ gcc_unreachable ();
+}
+
+/* Implement TARGET_ATOMIC_ASSIGN_EXPAND_FENV. */
+
+void
+loongarch_atomic_assign_expand_fenv (tree *hold, tree *clear, tree *update)
+{
+ if (!TARGET_HARD_FLOAT_ABI)
+ return;
+ tree exceptions_var = create_tmp_var_raw (LARCH_ATYPE_USI);
+ tree fcsr_orig_var = create_tmp_var_raw (LARCH_ATYPE_USI);
+ tree fcsr_mod_var = create_tmp_var_raw (LARCH_ATYPE_USI);
+ tree const0 = build_int_cst (LARCH_ATYPE_UQI, 0);
+ tree get_fcsr = loongarch_builtin_decls[LARCH_MOVFCSR2GR];
+ tree set_fcsr = loongarch_builtin_decls[LARCH_MOVGR2FCSR];
+ tree get_fcsr_hold_call = build_call_expr (get_fcsr, 1, const0);
+ tree hold_assign_orig = build4 (TARGET_EXPR, LARCH_ATYPE_USI,
+ fcsr_orig_var, get_fcsr_hold_call,
+ NULL, NULL);
+ tree hold_mod_val = build2 (BIT_AND_EXPR, LARCH_ATYPE_USI, fcsr_orig_var,
+ build_int_cst (LARCH_ATYPE_USI, 0xffe0ffe0));
+ tree hold_assign_mod = build4 (TARGET_EXPR, LARCH_ATYPE_USI,
+ fcsr_mod_var, hold_mod_val, NULL, NULL);
+ tree set_fcsr_hold_call = build_call_expr (set_fcsr, 2, const0,
+ fcsr_mod_var);
+ tree hold_all = build2 (COMPOUND_EXPR, LARCH_ATYPE_USI, hold_assign_orig,
+ hold_assign_mod);
+ *hold = build2 (COMPOUND_EXPR, void_type_node, hold_all, set_fcsr_hold_call);
+
+ *clear = build_call_expr (set_fcsr, 2, const0, fcsr_mod_var);
+
+ tree get_fcsr_update_call = build_call_expr (get_fcsr, 1, const0);
+ *update = build4 (TARGET_EXPR, LARCH_ATYPE_USI, exceptions_var,
+ get_fcsr_update_call, NULL, NULL);
+ tree set_fcsr_update_call = build_call_expr (set_fcsr, 2, const0,
+ fcsr_orig_var);
+ *update = build2 (COMPOUND_EXPR, void_type_node, *update,
+ set_fcsr_update_call);
+ tree atomic_feraiseexcept
+ = builtin_decl_implicit (BUILT_IN_ATOMIC_FERAISEEXCEPT);
+ tree int_exceptions_var = fold_convert (integer_type_node, exceptions_var);
+ tree atomic_feraiseexcept_call = build_call_expr (atomic_feraiseexcept, 1,
+ int_exceptions_var);
+ *update = build2 (COMPOUND_EXPR, void_type_node, *update,
+ atomic_feraiseexcept_call);
+}
+
+/* Implement TARGET_BUILTIN_VA_LIST. */
+
+tree
+loongarch_build_builtin_va_list (void)
+{
+ return ptr_type_node;
+}
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 07/12] LoongArch Port: Builtin macros.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (5 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 06/12] LoongArch Port: Builtin functions chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 08/12] LoongArch Port: libgcc chenglulu
` (5 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/
*config/loongarch/loongarch-c.c
---
gcc/config/loongarch/loongarch-c.c | 111 +++++++++++++++++++++++++++++
1 file changed, 111 insertions(+)
create mode 100644 gcc/config/loongarch/loongarch-c.c
diff --git a/gcc/config/loongarch/loongarch-c.c b/gcc/config/loongarch/loongarch-c.c
new file mode 100644
index 00000000000..b577b2f952c
--- /dev/null
+++ b/gcc/config/loongarch/loongarch-c.c
@@ -0,0 +1,111 @@
+/* LoongArch-specific code for C family languages.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License
+along with GCC; see the file COPYING3. If not see
+<http://www.gnu.org/licenses/>. */
+
+#define IN_TARGET_CODE 1
+
+#include "config.h"
+#include "system.h"
+#include "coretypes.h"
+#include "tm.h"
+#include "c-family/c-common.h"
+#include "cpplib.h"
+
+#define preprocessing_asm_p() (cpp_get_options (pfile)->lang == CLK_ASM)
+#define builtin_define(TXT) cpp_define (pfile, TXT)
+#define builtin_assert(TXT) cpp_assert (pfile, TXT)
+
+/* Define preprocessor macros for the -march and -mtune options.
+ PREFIX is either _LOONGARCH_ARCH or _LOONGARCH_TUNE, INFO is
+ the selected processor. If INFO's canonical name is "foo",
+ define PREFIX to be "foo", and define an additional macro
+ PREFIX_FOO. */
+#define LARCH_CPP_SET_PROCESSOR(PREFIX, CPU_TYPE) \
+ do \
+ { \
+ char *macro, *p; \
+ int cpu_type = (CPU_TYPE); \
+ \
+ macro = concat ((PREFIX), "_", \
+ loongarch_cpu_strings[cpu_type], NULL); \
+ for (p = macro; *p != 0; p++) \
+ *p = TOUPPER (*p); \
+ \
+ builtin_define (macro); \
+ builtin_define_with_value ((PREFIX), \
+ loongarch_cpu_strings[cpu_type], 1); \
+ free (macro); \
+ } \
+ while (0)
+
+/* TODO: what is the pfile technique ??? !!! */
+
+void
+loongarch_cpu_cpp_builtins (cpp_reader *pfile)
+{
+ builtin_assert ("machine=loongarch");
+ builtin_assert ("cpu=loongarch");
+ builtin_define ("__loongarch__");
+
+ LARCH_CPP_SET_PROCESSOR ("_LOONGARCH_ARCH", __ACTUAL_ARCH);
+ LARCH_CPP_SET_PROCESSOR ("_LOONGARCH_TUNE", __ACTUAL_TUNE);
+
+ /* Base architecture / ABI. */
+ if (TARGET_64BIT)
+ {
+ builtin_define ("__loongarch_grlen=64");
+ builtin_define ("__loongarch64");
+ }
+
+ if (TARGET_ABI_LP64)
+ {
+ builtin_define ("_ABILP64=3");
+ builtin_define ("_LOONGARCH_SIM=_ABILP64");
+ builtin_define ("__loongarch_lp64");
+ }
+
+ /* These defines reflect the ABI in use, not whether the
+ FPU is directly accessible. */
+ if (TARGET_DOUBLE_FLOAT_ABI)
+ builtin_define ("__loongarch_double_float=1");
+ else if (TARGET_SINGLE_FLOAT_ABI)
+ builtin_define ("__loongarch_single_float=1");
+
+ if (TARGET_DOUBLE_FLOAT_ABI || TARGET_SINGLE_FLOAT_ABI)
+ builtin_define ("__loongarch_hard_float=1");
+ else
+ builtin_define ("__loongarch_soft_float=1");
+
+
+ /* ISA Extensions. */
+ if (TARGET_DOUBLE_FLOAT)
+ builtin_define ("__loongarch_frlen=64");
+ else if (TARGET_SINGLE_FLOAT)
+ builtin_define ("__loongarch_frlen=32");
+ else
+ builtin_define ("__loongarch_frlen=0");
+
+ /* Native Data Sizes. */
+ builtin_define_with_int_value ("_LOONGARCH_SZINT", INT_TYPE_SIZE);
+ builtin_define_with_int_value ("_LOONGARCH_SZLONG", LONG_TYPE_SIZE);
+ builtin_define_with_int_value ("_LOONGARCH_SZPTR", POINTER_SIZE);
+ builtin_define_with_int_value ("_LOONGARCH_FPSET", 32 / MAX_FPRS_PER_FMT);
+ builtin_define_with_int_value ("_LOONGARCH_SPFPSET", 32);
+
+}
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 08/12] LoongArch Port: libgcc
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (6 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 07/12] LoongArch Port: Builtin macros chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 09/12] LoongArch Port: Regenerate libgcc/configure chenglulu
` (4 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
libgcc/
* config/loongarch/crtfastmath.c: New file.
* config/loongarch/crti.S: Like wise.
* config/loongarch/crtn.S: Like wise.
* config/loongarch/lib2funcs.c: Like wise.
* config/loongarch/linux-unwind.h: Like wise.
* config/loongarch/sfp-machine.h: Like wise.
* config/loongarch/t-crtstuff: Like wise.
* config/loongarch/t-elf: Like wise.
* config/loongarch/t-loongarch: Like wise.
* config/loongarch/t-loongarch64: Like wise.
* config/loongarch/t-softfp-tf: Like wise.
* config.host: Add LoongArch tuples.
* configure.ac: Add LoongArch support.
---
libgcc/config.host | 28 ++++-
libgcc/config/loongarch/crtfastmath.c | 52 +++++++++
libgcc/config/loongarch/crti.S | 43 +++++++
libgcc/config/loongarch/crtn.S | 39 +++++++
libgcc/config/loongarch/lib2funcs.c | 0
libgcc/config/loongarch/linux-unwind.h | 80 +++++++++++++
libgcc/config/loongarch/sfp-machine.h | 152 +++++++++++++++++++++++++
libgcc/config/loongarch/t-crtstuff | 2 +
libgcc/config/loongarch/t-elf | 3 +
libgcc/config/loongarch/t-loongarch | 9 ++
libgcc/config/loongarch/t-loongarch64 | 1 +
libgcc/config/loongarch/t-softfp-tf | 3 +
libgcc/configure.ac | 2 +-
13 files changed, 412 insertions(+), 2 deletions(-)
create mode 100644 libgcc/config/loongarch/crtfastmath.c
create mode 100644 libgcc/config/loongarch/crti.S
create mode 100644 libgcc/config/loongarch/crtn.S
create mode 100644 libgcc/config/loongarch/lib2funcs.c
create mode 100644 libgcc/config/loongarch/linux-unwind.h
create mode 100644 libgcc/config/loongarch/sfp-machine.h
create mode 100644 libgcc/config/loongarch/t-crtstuff
create mode 100644 libgcc/config/loongarch/t-elf
create mode 100644 libgcc/config/loongarch/t-loongarch
create mode 100644 libgcc/config/loongarch/t-loongarch64
create mode 100644 libgcc/config/loongarch/t-softfp-tf
diff --git a/libgcc/config.host b/libgcc/config.host
index ad0cdb284c7..7e9da09cb80 100644
--- a/libgcc/config.host
+++ b/libgcc/config.host
@@ -138,6 +138,22 @@ hppa*-*-*)
lm32*-*-*)
cpu_type=lm32
;;
+loongarch*-*-*)
+ cpu_type=loongarch
+ tmake_file="loongarch/t-loongarch"
+ if test "${libgcc_cv_loongarch_hard_float}" = yes; then
+ tmake_file="${tmake_file} t-hardfp-sfdf t-hardfp"
+ else
+ tmake_file="${tmake_file} t-softfp-sfdf"
+ fi
+ if test "${ac_cv_sizeof_long_double}" = 16; then
+ tmake_file="${tmake_file} loongarch/t-softfp-tf"
+ fi
+ if test "${host_address}" = 64; then
+ tmake_file="${tmake_file} loongarch/t-loongarch64"
+ fi
+ tmake_file="${tmake_file} t-softfp"
+ ;;
m32r*-*-*)
cpu_type=m32r
;;
@@ -925,7 +941,17 @@ lm32-*-rtems*)
lm32-*-uclinux*)
extra_parts="$extra_parts crtbegin.o crtendS.o crtbeginT.o"
tmake_file="lm32/t-lm32 lm32/t-uclinux t-libgcc-pic t-softfp-sfdf t-softfp"
- ;;
+ ;;
+loongarch*-*-linux*)
+ extra_parts="$extra_parts crtfastmath.o"
+ tmake_file="${tmake_file} t-crtfm"
+ case ${host} in
+ *)
+ tmake_file="${tmake_file} t-slibgcc-libgcc"
+ ;;
+ esac
+ md_unwind_header=loongarch/linux-unwind.h
+ ;;
m32r-*-elf*)
tmake_file="$tmake_file m32r/t-m32r t-fdpbit"
extra_parts="$extra_parts crtinit.o crtfini.o"
diff --git a/libgcc/config/loongarch/crtfastmath.c b/libgcc/config/loongarch/crtfastmath.c
new file mode 100644
index 00000000000..3523d531e2b
--- /dev/null
+++ b/libgcc/config/loongarch/crtfastmath.c
@@ -0,0 +1,52 @@
+/* Copyright (C) 2021 Free Software Foundation, Inc.
+ Contributed by Loongson Ltd.
+ Based on MIPS target for GNU compiler.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
+or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public
+License for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License
+and a copy of the GCC Runtime Library Exception along with this
+program; see the files COPYING3 and COPYING.RUNTIME respectively.
+If not, see <http://www.gnu.org/licenses/>. */
+
+#ifdef __loongarch_hard_float
+
+/* Rounding control. */
+#define _FPU_RC_NEAREST 0x000 /* RECOMMENDED. */
+#define _FPU_RC_ZERO 0x100
+#define _FPU_RC_UP 0x200
+#define _FPU_RC_DOWN 0x300
+
+/* Enable interrupts for IEEE exceptions. */
+#define _FPU_IEEE 0x0000001F
+
+/* Macros for accessing the hardware control word. */
+#define _FPU_GETCW(cw) __asm__ volatile ("movfcsr2gr %0,$r0" : "=r" (cw))
+#define _FPU_SETCW(cw) __asm__ volatile ("movgr2fcsr $r0,%0" : : "r" (cw))
+
+static void __attribute__((constructor))
+set_fast_math (void)
+{
+ unsigned int fcr;
+
+ /* Flush to zero, round to nearest, IEEE exceptions disabled. */
+ fcr = _FPU_RC_NEAREST;
+
+ _FPU_SETCW (fcr);
+}
+
+#endif /* __loongarch_hard_float */
diff --git a/libgcc/config/loongarch/crti.S b/libgcc/config/loongarch/crti.S
new file mode 100644
index 00000000000..bac029874b5
--- /dev/null
+++ b/libgcc/config/loongarch/crti.S
@@ -0,0 +1,43 @@
+/* Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
+<http://www.gnu.org/licenses/>. */
+
+/* 4 slots for argument spill area. 1 for cpreturn, 1 for stack.
+ Return spill offset of 8. Aligned to 16 bytes for lp64. */
+
+ .section .init,"ax",@progbits
+ .globl _init
+ .type _init,@function
+_init:
+ addi.d $r3,$r3,-16
+ st.d $r1,$r3,8
+ addi.d $r3,$r3,16
+ jirl $r0,$r1,0
+
+ .section .fini,"ax",@progbits
+ .globl _fini
+ .type _fini,@function
+_fini:
+ addi.d $r3,$r3,-16
+ st.d $r1,$r3,8
+ addi.d $r3,$r3,16
+ jirl $r0,$r1,0
diff --git a/libgcc/config/loongarch/crtn.S b/libgcc/config/loongarch/crtn.S
new file mode 100644
index 00000000000..7af6728690f
--- /dev/null
+++ b/libgcc/config/loongarch/crtn.S
@@ -0,0 +1,39 @@
+/* Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
+<http://www.gnu.org/licenses/>. */
+
+/* 4 slots for argument spill area. 1 for cpreturn, 1 for stack.
+ Return spill offset of 8. Aligned to 16 bytes for lp64. */
+
+
+ .section .init,"ax",@progbits
+init:
+ ld.d $r1,$r3,8
+ addi.d $r3,$r3,16
+ jirl $r0,$r1,0
+
+ .section .fini,"ax",@progbits
+fini:
+ ld.d $r1,$r3,8
+ addi.d $r3,$r3,16
+ jirl $r0,$r1,0
+
diff --git a/libgcc/config/loongarch/lib2funcs.c b/libgcc/config/loongarch/lib2funcs.c
new file mode 100644
index 00000000000..e69de29bb2d
diff --git a/libgcc/config/loongarch/linux-unwind.h b/libgcc/config/loongarch/linux-unwind.h
new file mode 100644
index 00000000000..8c9dfa56a6d
--- /dev/null
+++ b/libgcc/config/loongarch/linux-unwind.h
@@ -0,0 +1,80 @@
+/* DWARF2 EH unwinding support for LoongArch Linux.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify
+it under the terms of the GNU General Public License as published by
+the Free Software Foundation; either version 3, or (at your option)
+any later version.
+
+GCC is distributed in the hope that it will be useful,
+but WITHOUT ANY WARRANTY; without even the implied warranty of
+MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+GNU General Public License for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
+<http://www.gnu.org/licenses/>. */
+
+#ifndef inhibit_libc
+/* Do code reading to identify a signal frame, and set the frame
+ state data appropriately. See unwind-dw2.c for the structs. */
+
+#include <signal.h>
+#include <sys/syscall.h>
+#include <sys/ucontext.h>
+
+#define MD_FALLBACK_FRAME_STATE_FOR loongarch_fallback_frame_state
+
+static _Unwind_Reason_Code
+loongarch_fallback_frame_state (struct _Unwind_Context *context,
+ _Unwind_FrameState *fs)
+{
+ u_int32_t *pc = (u_int32_t *) context->ra;
+ struct sigcontext *sc;
+ _Unwind_Ptr new_cfa;
+ int i;
+
+ /* 03822c0b dli a7, 0x8b (sigreturn) */
+ /* 002b0000 syscall 0 */
+ if (pc[1] != 0x002b0000)
+ return _URC_END_OF_STACK;
+ if (pc[0] == 0x03822c0b)
+ {
+ struct rt_sigframe
+ {
+ siginfo_t info;
+ ucontext_t uc;
+ } *rt_ = context->cfa;
+ sc = &rt_->uc.uc_mcontext;
+ }
+ else
+ return _URC_END_OF_STACK;
+
+ new_cfa = (_Unwind_Ptr) sc;
+ fs->regs.cfa_how = CFA_REG_OFFSET;
+ fs->regs.cfa_reg = __LIBGCC_STACK_POINTER_REGNUM__;
+ fs->regs.cfa_offset = new_cfa - (_Unwind_Ptr) context->cfa;
+
+ for (i = 0; i < 32; i++)
+ {
+ fs->regs.reg[i].how = REG_SAVED_OFFSET;
+ fs->regs.reg[i].loc.offset = (_Unwind_Ptr) & (sc->sc_regs[i]) - new_cfa;
+ }
+
+ fs->signal_frame = 1;
+ fs->regs.reg[__LIBGCC_DWARF_ALT_FRAME_RETURN_COLUMN__].how
+ = REG_SAVED_VAL_OFFSET;
+ fs->regs.reg[__LIBGCC_DWARF_ALT_FRAME_RETURN_COLUMN__].loc.offset
+ = (_Unwind_Ptr) (sc->sc_pc) - new_cfa;
+ fs->retaddr_column = __LIBGCC_DWARF_ALT_FRAME_RETURN_COLUMN__;
+
+ return _URC_NO_REASON;
+}
+#endif
diff --git a/libgcc/config/loongarch/sfp-machine.h b/libgcc/config/loongarch/sfp-machine.h
new file mode 100644
index 00000000000..cb5b04feede
--- /dev/null
+++ b/libgcc/config/loongarch/sfp-machine.h
@@ -0,0 +1,152 @@
+/* softfp machine description for LoongArch.
+ Copyright (C) 2021 Free Software Foundation, Inc.
+
+This file is part of GCC.
+
+GCC is free software; you can redistribute it and/or modify it under
+the terms of the GNU General Public License as published by the Free
+Software Foundation; either version 3, or (at your option) any later
+version.
+
+GCC is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or
+FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
+for more details.
+
+Under Section 7 of GPL version 3, you are granted additional
+permissions described in the GCC Runtime Library Exception, version
+3.1, as published by the Free Software Foundation.
+
+You should have received a copy of the GNU General Public License and
+a copy of the GCC Runtime Library Exception along with this program;
+see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
+<http://www.gnu.org/licenses/>. */
+
+#ifdef __loongarch64
+#define _FP_W_TYPE_SIZE 64
+#define _FP_W_TYPE unsigned long long
+#define _FP_WS_TYPE signed long long
+#define _FP_I_TYPE long long
+
+typedef int TItype __attribute__ ((mode (TI)));
+typedef unsigned int UTItype __attribute__ ((mode (TI)));
+#define TI_BITS (__CHAR_BIT__ * (int) sizeof (TItype))
+
+#define _FP_MUL_MEAT_S(R, X, Y) \
+ _FP_MUL_MEAT_1_wide (_FP_WFRACBITS_S, R, X, Y, umul_ppmm)
+#define _FP_MUL_MEAT_D(R, X, Y) \
+ _FP_MUL_MEAT_1_wide (_FP_WFRACBITS_D, R, X, Y, umul_ppmm)
+#define _FP_MUL_MEAT_Q(R, X, Y) \
+ _FP_MUL_MEAT_2_wide (_FP_WFRACBITS_Q, R, X, Y, umul_ppmm)
+
+#define _FP_DIV_MEAT_S(R, X, Y) _FP_DIV_MEAT_1_udiv_norm (S, R, X, Y)
+#define _FP_DIV_MEAT_D(R, X, Y) _FP_DIV_MEAT_1_udiv_norm (D, R, X, Y)
+#define _FP_DIV_MEAT_Q(R, X, Y) _FP_DIV_MEAT_2_udiv (Q, R, X, Y)
+
+#define _FP_NANFRAC_S ((_FP_QNANBIT_S << 1) - 1)
+#define _FP_NANFRAC_D ((_FP_QNANBIT_D << 1) - 1)
+#define _FP_NANFRAC_Q ((_FP_QNANBIT_Q << 1) - 1), -1
+#else
+#define _FP_W_TYPE_SIZE 32
+#define _FP_W_TYPE unsigned int
+#define _FP_WS_TYPE signed int
+#define _FP_I_TYPE int
+
+#define _FP_MUL_MEAT_S(R, X, Y) \
+ _FP_MUL_MEAT_1_wide (_FP_WFRACBITS_S, R, X, Y, umul_ppmm)
+#define _FP_MUL_MEAT_D(R, X, Y) \
+ _FP_MUL_MEAT_2_wide (_FP_WFRACBITS_D, R, X, Y, umul_ppmm)
+#define _FP_MUL_MEAT_Q(R, X, Y) \
+ _FP_MUL_MEAT_4_wide (_FP_WFRACBITS_Q, R, X, Y, umul_ppmm)
+
+#define _FP_DIV_MEAT_S(R, X, Y) _FP_DIV_MEAT_1_udiv_norm (S, R, X, Y)
+#define _FP_DIV_MEAT_D(R, X, Y) _FP_DIV_MEAT_2_udiv (D, R, X, Y)
+#define _FP_DIV_MEAT_Q(R, X, Y) _FP_DIV_MEAT_4_udiv (Q, R, X, Y)
+
+#define _FP_NANFRAC_S ((_FP_QNANBIT_S << 1) - 1)
+#define _FP_NANFRAC_D ((_FP_QNANBIT_D << 1) - 1), -1
+#define _FP_NANFRAC_Q ((_FP_QNANBIT_Q << 1) - 1), -1, -1, -1
+#endif
+
+/* The type of the result of a floating point comparison. This must
+ match __libgcc_cmp_return__ in GCC for the target. */
+typedef int __gcc_CMPtype __attribute__ ((mode (__libgcc_cmp_return__)));
+#define CMPtype __gcc_CMPtype
+
+#define _FP_NANSIGN_S 0
+#define _FP_NANSIGN_D 0
+#define _FP_NANSIGN_Q 0
+
+#define _FP_KEEPNANFRACP 1
+#define _FP_QNANNEGATEDP 0
+
+/* NaN payloads should be preserved for NAN2008. */
+#define _FP_CHOOSENAN(fs, wc, R, X, Y, OP) \
+ do \
+ { \
+ R##_s = X##_s; \
+ _FP_FRAC_COPY_##wc (R, X); \
+ R##_c = FP_CLS_NAN; \
+ } \
+ while (0)
+
+#ifdef __loongarch_hard_float
+#define FP_EX_INVALID 0x100000
+#define FP_EX_DIVZERO 0x080000
+#define FP_EX_OVERFLOW 0x040000
+#define FP_EX_UNDERFLOW 0x020000
+#define FP_EX_INEXACT 0x010000
+#define FP_EX_ALL \
+ (FP_EX_INVALID | FP_EX_DIVZERO | FP_EX_OVERFLOW | FP_EX_UNDERFLOW \
+ | FP_EX_INEXACT)
+
+#define FP_EX_ENABLE_SHIFT 16
+#define FP_EX_CAUSE_SHIFT 8
+
+#define FP_RND_NEAREST 0x000
+#define FP_RND_ZERO 0x100
+#define FP_RND_PINF 0x200
+#define FP_RND_MINF 0x300
+#define FP_RND_MASK 0x300
+
+#define _FP_DECL_EX \
+ unsigned long int _fcsr __attribute__ ((unused)) = FP_RND_NEAREST
+
+#define FP_INIT_ROUNDMODE \
+ do \
+ { \
+ _fcsr = __builtin_loongarch_movfcsr2gr (0); \
+ } \
+ while (0)
+
+#define FP_ROUNDMODE (_fcsr & FP_RND_MASK)
+
+#define FP_TRAPPING_EXCEPTIONS ((_fcsr << FP_EX_ENABLE_SHIFT) & FP_EX_ALL)
+
+#define FP_HANDLE_EXCEPTIONS \
+ do \
+ { \
+ _fcsr &= ~(FP_EX_ALL << FP_EX_CAUSE_SHIFT); \
+ _fcsr |= _fex | (_fex << FP_EX_CAUSE_SHIFT); \
+ __builtin_loongarch_movgr2fcsr (0, _fcsr); \
+ } \
+ while (0)
+
+#else
+#define FP_EX_INVALID (1 << 4)
+#define FP_EX_DIVZERO (1 << 3)
+#define FP_EX_OVERFLOW (1 << 2)
+#define FP_EX_UNDERFLOW (1 << 1)
+#define FP_EX_INEXACT (1 << 0)
+#endif
+
+#define _FP_TININESS_AFTER_ROUNDING 1
+
+#define __LITTLE_ENDIAN 1234
+
+#define __BYTE_ORDER __LITTLE_ENDIAN
+
+/* Define ALIASNAME as a strong alias for NAME. */
+#define strong_alias(name, aliasname) _strong_alias (name, aliasname)
+#define _strong_alias(name, aliasname) \
+ extern __typeof (name) aliasname __attribute__ ((alias (#name)));
diff --git a/libgcc/config/loongarch/t-crtstuff b/libgcc/config/loongarch/t-crtstuff
new file mode 100644
index 00000000000..d8ab43be2ae
--- /dev/null
+++ b/libgcc/config/loongarch/t-crtstuff
@@ -0,0 +1,2 @@
+# Don't let CTOR_LIST end up in sdata section.
+CRTSTUFF_T_CFLAGS = -G 0
diff --git a/libgcc/config/loongarch/t-elf b/libgcc/config/loongarch/t-elf
new file mode 100644
index 00000000000..651f10a5303
--- /dev/null
+++ b/libgcc/config/loongarch/t-elf
@@ -0,0 +1,3 @@
+# We must build libgcc2.a with -G 0, in case the user wants to link
+# without the $gp register.
+HOST_LIBGCC2_CFLAGS += -G 0
diff --git a/libgcc/config/loongarch/t-loongarch b/libgcc/config/loongarch/t-loongarch
new file mode 100644
index 00000000000..9508cb2fcae
--- /dev/null
+++ b/libgcc/config/loongarch/t-loongarch
@@ -0,0 +1,9 @@
+LIB2_SIDITI_CONV_FUNCS = yes
+
+softfp_float_modes :=
+softfp_int_modes := si di
+softfp_extensions :=
+softfp_truncations :=
+softfp_exclude_libgcc2 := n
+
+LIB2ADD_ST += $(srcdir)/config/loongarch/lib2funcs.c
diff --git a/libgcc/config/loongarch/t-loongarch64 b/libgcc/config/loongarch/t-loongarch64
new file mode 100644
index 00000000000..a1e3513e288
--- /dev/null
+++ b/libgcc/config/loongarch/t-loongarch64
@@ -0,0 +1 @@
+softfp_int_modes += ti
diff --git a/libgcc/config/loongarch/t-softfp-tf b/libgcc/config/loongarch/t-softfp-tf
new file mode 100644
index 00000000000..306677b1255
--- /dev/null
+++ b/libgcc/config/loongarch/t-softfp-tf
@@ -0,0 +1,3 @@
+softfp_float_modes += tf
+softfp_extensions += sftf dftf
+softfp_truncations += tfsf tfdf
diff --git a/libgcc/configure.ac b/libgcc/configure.ac
index 13a80b2551b..b17d788dbd0 100644
--- a/libgcc/configure.ac
+++ b/libgcc/configure.ac
@@ -317,7 +317,7 @@ AC_CACHE_CHECK([whether assembler supports CFI directives], [libgcc_cv_cfi],
# word size rather than the address size.
cat > conftest.c <<EOF
#if defined(__x86_64__) || (!defined(__i386__) && defined(__LP64__)) \
- || defined(__mips64)
+ || defined(__mips64) || defined(__loongarch64)
host_address=64
#else
host_address=32
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 09/12] LoongArch Port: Regenerate libgcc/configure.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (7 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 08/12] LoongArch Port: libgcc chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 10/12] LoongArch Port: libgomp chenglulu
` (3 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
---
libgcc/configure | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/libgcc/configure b/libgcc/configure
index 4919a56f518..ce04c4f529f 100755
--- a/libgcc/configure
+++ b/libgcc/configure
@@ -2412,6 +2412,9 @@ case "${host}" in
# sets the default TLS model and affects inlining.
PICFLAG=-fPIC
;;
+ loongarch*-*-*)
+ PICFLAG=-fpic
+ ;;
mips-sgi-irix6*)
# PIC is the default.
;;
@@ -5066,7 +5069,7 @@ $as_echo "$libgcc_cv_cfi" >&6; }
# word size rather than the address size.
cat > conftest.c <<EOF
#if defined(__x86_64__) || (!defined(__i386__) && defined(__LP64__)) \
- || defined(__mips64)
+ || defined(__mips64) || defined(__loongarch64)
host_address=64
#else
host_address=32
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 10/12] LoongArch Port: libgomp
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (8 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 09/12] LoongArch Port: Regenerate libgcc/configure chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 11/12] LoongArch Port: gcc/testsuite chenglulu
` (2 subsequent siblings)
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
libgomp/
* configure.tgt: Add LoongArch triplet.
---
libgomp/configure.tgt | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/libgomp/configure.tgt b/libgomp/configure.tgt
index d4f1e741b5a..2cd7272fcd8 100644
--- a/libgomp/configure.tgt
+++ b/libgomp/configure.tgt
@@ -56,6 +56,10 @@ if test x$enable_linux_futex = xyes; then
config_path="linux/ia64 linux posix"
;;
+ loongarch*-*-linux*)
+ config_path="linux posix"
+ ;;
+
mips*-*-linux*)
config_path="linux/mips linux posix"
;;
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 11/12] LoongArch Port: gcc/testsuite
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (9 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 10/12] LoongArch Port: libgomp chenglulu
@ 2021-12-24 9:28 ` chenglulu
2021-12-24 9:28 ` [PATCH v4 12/12] LoongArch Port: Add doc chenglulu
2022-01-05 14:13 ` [PATCH v4 00/12] Add LoongArch support Paul Hua
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
gcc/testsuite/
* g++.dg/cpp0x/constexpr-rom.C: Add build options for LoongArch.
* g++.old-deja/g++.abi/ptrmem.C: Add LoongArch support.
* g++.old-deja/g++.pt/ptrmem6.C: xfail for LoongArch.
* gcc.dg/20020312-2.c: Add LoongArch support.
* gcc.dg/loop-8.c: Skip on LoongArch.
* gcc.dg/torture/stackalign/builtin-apply-2.c: Likewise.
* gcc.dg/tree-ssa/ssa-fre-3.c: Likewise.
* go.test/go-test.exp: Define the LoongArch target.
* lib/target-supports.exp: Like wise.
* gcc.target/loongarch/loongarch.exp: New file.
* gcc.target/loongarch/tst-asm-const.c: Like wise.
---
gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C | 2 +-
gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C | 2 +-
gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C | 2 +-
gcc/testsuite/gcc.dg/20020312-2.c | 2 +
gcc/testsuite/gcc.dg/loop-8.c | 2 +-
.../torture/stackalign/builtin-apply-2.c | 2 +-
gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c | 2 +-
.../gcc.target/loongarch/loongarch.exp | 40 +++++++++++++++++++
.../gcc.target/loongarch/tst-asm-const.c | 16 ++++++++
gcc/testsuite/go.test/go-test.exp | 3 ++
gcc/testsuite/lib/target-supports.exp | 14 +++++++
11 files changed, 81 insertions(+), 6 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/loongarch/loongarch.exp
create mode 100644 gcc/testsuite/gcc.target/loongarch/tst-asm-const.c
diff --git a/gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C b/gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C
index 2e0ef685f36..424979a604b 100644
--- a/gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C
+++ b/gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C
@@ -1,6 +1,6 @@
// PR c++/49673: check that test_data goes into .rodata
// { dg-do compile { target c++11 } }
-// { dg-additional-options -G0 { target { { alpha*-*-* frv*-*-* ia64-*-* lm32*-*-* m32r*-*-* microblaze*-*-* mips*-*-* nios2-*-* powerpc*-*-* rs6000*-*-* } && { ! { *-*-darwin* *-*-aix* alpha*-*-*vms* } } } } }
+// { dg-additional-options -G0 { target { { alpha*-*-* frv*-*-* ia64-*-* lm32*-*-* m32r*-*-* microblaze*-*-* mips*-*-* loongarch*-*-* nios2-*-* powerpc*-*-* rs6000*-*-* } && { ! { *-*-darwin* *-*-aix* alpha*-*-*vms* } } } } }
// { dg-final { scan-assembler "\\.rdata" { target mips*-*-* } } }
// { dg-final { scan-assembler "rodata" { target { { *-*-linux-gnu *-*-gnu* *-*-elf } && { ! { mips*-*-* riscv*-*-* } } } } } }
diff --git a/gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C b/gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C
index bda7960d8a2..f69000e9081 100644
--- a/gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C
+++ b/gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C
@@ -7,7 +7,7 @@
function. However, some platforms use all bits to encode a
function pointer. Such platforms use the lowest bit of the delta,
that is shifted left by one bit. */
-#if defined __MN10300__ || defined __SH5__ || defined __arm__ || defined __thumb__ || defined __mips__ || defined __aarch64__ || defined __PRU__
+#if defined __MN10300__ || defined __SH5__ || defined __arm__ || defined __thumb__ || defined __mips__ || defined __aarch64__ || defined __PRU__ || defined __loongarch__
#define ADJUST_PTRFN(func, virt) ((void (*)())(func))
#define ADJUST_DELTA(delta, virt) (((delta) << 1) + !!(virt))
#else
diff --git a/gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C b/gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C
index 9f4bbe43f89..8f8f7017ab7 100644
--- a/gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C
+++ b/gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C
@@ -25,7 +25,7 @@ int main() {
h<&B::j>(); // { dg-error "" }
g<(void (A::*)()) &A::f>(); // { dg-error "" "" { xfail c++11 } }
h<(int A::*) &A::i>(); // { dg-error "" "" { xfail c++11 } }
- g<(void (A::*)()) &B::f>(); // { dg-error "" "" { xfail { c++11 && { aarch64*-*-* arm*-*-* mips*-*-* } } } }
+ g<(void (A::*)()) &B::f>(); // { dg-error "" "" { xfail { c++11 && { aarch64*-*-* arm*-*-* mips*-*-* loongarch*-*-* } } } }
h<(int A::*) &B::j>(); // { dg-error "" }
g<(void (A::*)()) 0>(); // { dg-error "" "" { target { ! c++11 } } }
h<(int A::*) 0>(); // { dg-error "" "" { target { ! c++11 } } }
diff --git a/gcc/testsuite/gcc.dg/20020312-2.c b/gcc/testsuite/gcc.dg/20020312-2.c
index 52c33d09b90..92bc150df0f 100644
--- a/gcc/testsuite/gcc.dg/20020312-2.c
+++ b/gcc/testsuite/gcc.dg/20020312-2.c
@@ -37,6 +37,8 @@ extern void abort (void);
/* PIC register is r1, but is used even without -fpic. */
#elif defined(__lm32__)
/* No pic register. */
+#elif defined(__loongarch__)
+/* No pic register. */
#elif defined(__M32R__)
/* No pic register. */
#elif defined(__m68k__)
diff --git a/gcc/testsuite/gcc.dg/loop-8.c b/gcc/testsuite/gcc.dg/loop-8.c
index a685fc25056..8e5f2087831 100644
--- a/gcc/testsuite/gcc.dg/loop-8.c
+++ b/gcc/testsuite/gcc.dg/loop-8.c
@@ -1,6 +1,6 @@
/* { dg-do compile } */
/* { dg-options "-O1 -fdump-rtl-loop2_invariant" } */
-/* { dg-skip-if "unexpected IV" { "hppa*-*-* mips*-*-* visium-*-* powerpc*-*-* riscv*-*-* mmix-*-* vax-*-*" } } */
+/* { dg-skip-if "unexpected IV" { "hppa*-*-* mips*-*-* visium-*-* powerpc*-*-* riscv*-*-* mmix-*-* vax-*-* loongarch*-*-*" } } */
/* Load immediate on condition is available from z13 on and prevents moving
the load out of the loop, so always run this test with -march=zEC12 that
does not have load immediate on condition. */
diff --git a/gcc/testsuite/gcc.dg/torture/stackalign/builtin-apply-2.c b/gcc/testsuite/gcc.dg/torture/stackalign/builtin-apply-2.c
index 5ec05587dba..552ca1433f4 100644
--- a/gcc/testsuite/gcc.dg/torture/stackalign/builtin-apply-2.c
+++ b/gcc/testsuite/gcc.dg/torture/stackalign/builtin-apply-2.c
@@ -9,7 +9,7 @@
/* arm_hf_eabi: Variadic funcs use Base AAPCS. Normal funcs use VFP variant.
avr: Variadic funcs don't pass arguments in registers, while normal funcs
do. */
-/* { dg-skip-if "Variadic funcs use different argument passing from normal funcs" { arm_hf_eabi || { csky*-*-* avr-*-* riscv*-*-* or1k*-*-* msp430-*-* amdgcn-*-* pru-*-* } } } */
+/* { dg-skip-if "Variadic funcs use different argument passing from normal funcs" { arm_hf_eabi || { csky*-*-* avr-*-* riscv*-*-* or1k*-*-* msp430-*-* amdgcn-*-* pru-*-* loongarch*-*-* } } } */
/* { dg-skip-if "Variadic funcs have all args on stack. Normal funcs have args in registers." { nds32*-*-* } { v850*-*-* } } */
/* { dg-require-effective-target untyped_assembly } */
diff --git a/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c
index 6b6255b9713..224dd4f72ef 100644
--- a/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c
+++ b/gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c
@@ -5,7 +5,7 @@
When the condition is true, we distribute "(int) (a + b)" as
"(int) a + (int) b", otherwise we keep the original. */
-/* { dg-do compile { target { ! mips64 } } } */
+/* { dg-do compile { target { ! mips64 } && { ! loongarch64 } } } */
/* { dg-options "-O -fno-tree-forwprop -fno-tree-ccp -fwrapv -fdump-tree-fre1-details" } */
/* From PR14844. */
diff --git a/gcc/testsuite/gcc.target/loongarch/loongarch.exp b/gcc/testsuite/gcc.target/loongarch/loongarch.exp
new file mode 100644
index 00000000000..9f374a9bc73
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/loongarch.exp
@@ -0,0 +1,40 @@
+# Copyright (C) 2021 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with GCC; see the file COPYING3. If not see
+# <http://www.gnu.org/licenses/>.
+
+# GCC testsuite that uses the `dg.exp' driver.
+
+# Exit immediately if this isn't a LoongArch target.
+if ![istarget loongarch*-*-*] then {
+ return
+}
+
+# Load support procs.
+load_lib gcc-dg.exp
+
+# If a testcase doesn't have special options, use these.
+global DEFAULT_CFLAGS
+if ![info exists DEFAULT_CFLAGS] then {
+ set DEFAULT_CFLAGS " "
+}
+
+# Initialize `dg'.
+dg-init
+
+# Main loop.
+dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/*.\[cS\]]] \
+ "" $DEFAULT_CFLAGS
+# All done.
+dg-finish
diff --git a/gcc/testsuite/gcc.target/loongarch/tst-asm-const.c b/gcc/testsuite/gcc.target/loongarch/tst-asm-const.c
new file mode 100644
index 00000000000..2e04b99e301
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/tst-asm-const.c
@@ -0,0 +1,16 @@
+/* Test asm const. */
+/* { dg-do compile } */
+/* { dg-final { scan-assembler-times "foo:.*\\.long 1061109567.*\\.long 52" 1 } } */
+int foo ()
+{
+ __asm__ volatile (
+ "foo:"
+ "\n\t"
+ ".long %a0\n\t"
+ ".long %a1\n\t"
+ :
+ :"i"(0x3f3f3f3f), "i"(52)
+ :
+ );
+}
+
diff --git a/gcc/testsuite/go.test/go-test.exp b/gcc/testsuite/go.test/go-test.exp
index f16754513e4..63755aed106 100644
--- a/gcc/testsuite/go.test/go-test.exp
+++ b/gcc/testsuite/go.test/go-test.exp
@@ -232,6 +232,9 @@ proc go-set-goarch { } {
"riscv64-*-*" {
set goarch "riscv64"
}
+ "loongarch64-*-*" {
+ set goarch "loongarch64"
+ }
"s390*-*-*" {
if [check_effective_target_ilp32] {
set goarch "s390"
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 039125c866a..6899b317b8c 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -284,6 +284,10 @@ proc check_configured_with { pattern } {
proc check_weak_available { } {
global target_cpu
+ if { [ string first "loongarch" $target_cpu ] >= 0 } {
+ return 1
+ }
+
# All mips targets should support it
if { [ string first "mips" $target_cpu ] >= 0 } {
@@ -1295,6 +1299,14 @@ proc check_effective_target_mpaired_single { } {
# Return true if the target has access to FPU instructions.
proc check_effective_target_hard_float { } {
+ if { [istarget loongarch*-*-*] } {
+ return [check_no_compiler_messages hard_float assembly {
+ #if (defined __loongarch_soft_float)
+ #error __loongarch_soft_float
+ #endif
+ }]
+ }
+
if { [istarget mips*-*-*] } {
return [check_no_compiler_messages hard_float assembly {
#if (defined __mips_soft_float || defined __mips16)
@@ -8585,6 +8597,7 @@ proc check_effective_target_sync_char_short { } {
|| [istarget cris-*-*]
|| ([istarget sparc*-*-*] && [check_effective_target_sparc_v9])
|| ([istarget arc*-*-*] && [check_effective_target_arc_atomic])
+ || [istarget loongarch*-*-*]
|| [check_effective_target_mips_llsc] }}]
}
@@ -10677,6 +10690,7 @@ proc check_effective_target_branch_cost {} {
|| [istarget epiphany*-*-*]
|| [istarget frv*-*-*]
|| [istarget i?86-*-*] || [istarget x86_64-*-*]
+ || [istarget loongarch*-*-*]
|| [istarget mips*-*-*]
|| [istarget s390*-*-*]
|| [istarget riscv*-*-*]
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH v4 12/12] LoongArch Port: Add doc.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (10 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 11/12] LoongArch Port: gcc/testsuite chenglulu
@ 2021-12-24 9:28 ` chenglulu
2022-01-05 14:13 ` [PATCH v4 00/12] Add LoongArch support Paul Hua
12 siblings, 0 replies; 16+ messages in thread
From: chenglulu @ 2021-12-24 9:28 UTC (permalink / raw)
To: gcc-patches; +Cc: xuchenghua, yangyujie, chenglulu
* contrib/config-list.mk: Add LoongArch triplet.
* gcc/doc/install.texi: Add LoongArch options section.
* gcc/doc/invoke.texi: Add LoongArch options section.
* gcc/doc/md.texi: Add LoongArch options section.
---
contrib/config-list.mk | 5 +-
gcc/doc/install.texi | 47 +++++++++-
gcc/doc/invoke.texi | 201 +++++++++++++++++++++++++++++++++++++++++
gcc/doc/md.texi | 55 +++++++++++
4 files changed, 302 insertions(+), 6 deletions(-)
diff --git a/contrib/config-list.mk b/contrib/config-list.mk
index 3e1d1321861..ba6f12e4693 100644
--- a/contrib/config-list.mk
+++ b/contrib/config-list.mk
@@ -57,7 +57,10 @@ LIST = aarch64-elf aarch64-linux-gnu aarch64-rtems \
i686-wrs-vxworksae \
i686-cygwinOPT-enable-threads=yes i686-mingw32crt ia64-elf \
ia64-freebsd6 ia64-linux ia64-hpux ia64-hp-vms iq2000-elf lm32-elf \
- lm32-rtems lm32-uclinux m32c-rtems m32c-elf m32r-elf m32rle-elf \
+ lm32-rtems lm32-uclinux \
+ loongarch64-linux-gnu loongarch64-linux-gnuf64 \
+ loongarch64-linux-gnuf32 loongarch64-linux-gnusf \
+ m32c-rtems m32c-elf m32r-elf m32rle-elf \
m68k-elf m68k-netbsdelf \
m68k-uclinux m68k-linux m68k-rtems \
mcore-elf microblaze-linux microblaze-elf \
diff --git a/gcc/doc/install.texi b/gcc/doc/install.texi
index 549da22aa16..4e0cf20347e 100644
--- a/gcc/doc/install.texi
+++ b/gcc/doc/install.texi
@@ -752,9 +752,9 @@ Here are the possible CPU types:
@quotation
aarch64, aarch64_be, alpha, alpha64, amdgcn, arc, arceb, arm, armeb, avr, bfin,
bpf, cr16, cris, csky, epiphany, fido, fr30, frv, ft32, h8300, hppa, hppa2.0,
-hppa64, i486, i686, ia64, iq2000, lm32, m32c, m32r, m32rle, m68k, mcore,
-microblaze, microblazeel, mips, mips64, mips64el, mips64octeon, mips64orion,
-mips64vr, mipsel, mipsisa32, mipsisa32r2, mipsisa64, mipsisa64r2,
+hppa64, i486, i686, ia64, iq2000, lm32, loongarch64, m32c, m32r, m32rle, m68k,
+mcore, microblaze, microblazeel, mips, mips64, mips64el, mips64octeon,
+mips64orion, mips64vr, mipsel, mipsisa32, mipsisa32r2, mipsisa64, mipsisa64r2,
mipsisa64r2el, mipsisa64sb1, mipsisa64sr71k, mipstx39, mmix, mn10300, moxie,
msp430, nds32be, nds32le, nios2, nvptx, or1k, pdp11, powerpc, powerpc64,
powerpc64le, powerpcle, pru, riscv32, riscv32be, riscv64, riscv64be, rl78, rx,
@@ -1171,8 +1171,9 @@ sysv, aix.
@itemx --without-multilib-list
Specify what multilibs to build. @var{list} is a comma separated list of
values, possibly consisting of a single value. Currently only implemented
-for aarch64*-*-*, arm*-*-*, riscv*-*-*, sh*-*-* and x86-64-*-linux*. The
-accepted values and meaning for each target is given below.
+for aarch64*-*-*, arm*-*-*, loongarch64-*-*, riscv*-*-*, sh*-*-* and
+x86-64-*-linux*. The accepted values and meaning for each target is given
+below.
@table @code
@item aarch64*-*-*
@@ -1259,6 +1260,14 @@ profile. The union of these options is considered when specifying both
@code{-mfloat-abi=hard}
@end multitable
+@item loongarch*-*-*
+@var{list} is a comma-separated list of the following ABI identifiers:
+@code{lp64d[/base]} @code{lp64f[/base]} @code{lp64d[/base]}, where the
+@code{/base} suffix may be omitted, to enable their respective run-time
+libraries. If @var{list} is empty, @code{default}
+or @option{--with-multilib-list} is not specified, then the default ABI
+as specified by @option{--with-abi} or implied by @option{--target} is selected.
+
@item riscv*-*-*
@var{list} is a single ABI name. The target architecture must be either
@code{rv32gc} or @code{rv64gc}. This will build a single multilib for the
@@ -4430,6 +4439,34 @@ This configuration is intended for embedded systems.
Lattice Mico32 processor.
This configuration is intended for embedded systems running uClinux.
+@html
+<hr />
+@end html
+@anchor{loongarch}
+@heading LoongArch
+LoongArch processor.
+The following LoongArch targets are available:
+@table @code
+@item loongarch64-linux-gnu*
+LoongArch processor running GNU/Linux. This target triplet may be coupled
+with a small set of possible suffixes to identify their default ABI type:
+@table @code
+@item f64
+Uses @code{lp64d/base} ABI by default.
+@item f32
+Uses @code{lp64f/base} ABI by default.
+@item sf
+Uses @code{lp64s/base} ABI by default.
+@end table
+
+@item loongarch64-linux-gnu
+Same as @code{loongarch64-linux-gnuf64}, but may be used with
+@option{--with-abi=*} to configure the default ABI type.
+@end table
+
+More information about LoongArch can be found at
+@uref{https://github.com/loongson/LoongArch-Documentation}.
+
@html
<hr />
@end html
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index ca621577432..2cf226604f7 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -991,6 +991,16 @@ Objective-C and Objective-C++ Dialects}.
@gccoptlist{-mbarrel-shift-enabled -mdivide-enabled -mmultiply-enabled @gol
-msign-extend-enabled -muser-enabled}
+@emph{LoongArch Options}
+@gccoptlist{-march=@var{cpu-type} -mtune=@var{cpu-type} -mabi=@var{base-abi-type} @gol
+-mfpu=@var{fpu-type} -msoft-float -msingle-float -mdouble-float @gol
+-mbranch-cost=@var{n} -mcheck-zero-division -mno-check-zero-division @gol
+-mcond-move-int -mno-cond-move-int @gol
+-mcond-move-float -mno-cond-move-float @gol
+-memcpy -mno-memcpy -mstrict-align @gol
+-mmax-inline-memcpy-size=@var{n} @gol
+-mlra -mcmodel=@var{code-model}}
+
@emph{M32R/D Options}
@gccoptlist{-m32r2 -m32rx -m32r @gol
-mdebug @gol
@@ -18638,6 +18648,7 @@ platform.
* HPPA Options::
* IA-64 Options::
* LM32 Options::
+* LoongArch Options::
* M32C Options::
* M32R/D Options::
* M680x0 Options::
@@ -24134,6 +24145,196 @@ Enable user-defined instructions.
@end table
+@node LoongArch Options
+@subsection LoongArch Options
+@cindex LoongArch Options
+
+These command-line options are defined for LoongArch targets:
+
+@table @gcctabopt
+@item -march=@var{cpu-type}
+@opindex -march
+Generate instructions for the machine type @var{cpu-type}. In contrast to
+@option{-mtune=@var{cpu-type}}, which merely tunes the generated code
+for the specified @var{cpu-type}, @option{-march=@var{cpu-type}} allows GCC
+to generate code that may not run at all on processors other than the one
+indicated. Specifying @option{-march=@var{cpu-type}} implies
+@option{-mtune=@var{cpu-type}}, except where noted otherwise.
+
+The choices for @var{cpu-type} are:
+
+@table @samp
+@item native
+This selects the CPU to generate code for at compilation time by determining
+the processor type of the compiling machine. Using @option{-march=native}
+enables all instruction subsets supported by the local machine (hence
+the result might not run on different machines). Using @option{-mtune=native}
+produces code optimized for the local machine under the constraints
+of the selected instruction set.
+@item loongarch64
+A generic CPU with 64-bit extensions.
+@item la464
+LoongArch LA464 CPU with LBT, LSX, LASX, LVZ.
+@end table
+
+
+@item -mtune=@var{cpu-type}
+@opindex mtune
+Optimize the output for the given processor, specified by microarchitecture
+name.
+
+@item -mabi=@var{base-abi-type}
+@opindex mabi
+Generate code for the specified calling convention. @gol
+Set base ABI to one of: @gol
+@table @samp
+@item lp64d
+Uses 64-bit general purpose registers and 32/64-bit floating-point
+registers for parameter passing. Data model is LP64, where int
+is 32 bits, while long int and pointers are 64 bits.
+@item lp64f
+Uses 64-bit general purpose registers and 32-bit floating-point
+registers for parameter passing. Data model is LP64, where int
+is 32 bits, while long int and pointers are 64 bits.
+@item lp64s
+Uses 64-bit general purpose registers and no floating-point
+registers for parameter passing. Data model is LP64, where int
+is 32 bits, while long int and pointers are 64 bits.
+@end table
+
+
+@item -mfpu=@var{fpu-type}
+@opindex mfpu
+Generating code for the specified FPU type: @gol
+@table @samp
+@item 64
+Allow the use of hardware floating-point instructions for 32-bit
+and 64-bit operations.
+@item 32
+Allow the use of hardware floating-point instructions for 32-bit
+operations.
+@item none
+@item 0
+Prevent the use of hardware floating-point instructions.
+@end table
+
+
+@item -msoft-float
+@opindex msoft-float
+Force @option{-mfpu=none} and prevents the use of floating-point
+registers for parameter passing. This option may change the target
+ABI.
+
+@item -msingle-float
+@opindex -msingle-float
+Force @option{-mfpu=32} and allow the use of 32-bit floating-point
+registers for parameter passing. This option may change the target
+ABI.
+
+@item -mdouble-float
+@opindex -mdouble-float
+Force @option{-mfpu=64} and allow the use of 32/64-bit floating-point
+registers for parameter passing. This option may change the target
+ABI.
+
+
+@item -mbranch-cost=@var{n}
+@opindex -mbranch-cost
+Set the cost of branches to roughly n instructions.
+
+@item -mcheck-zero-division
+@itemx -mno-check-zero-divison
+@opindex -mcheck-zero-division
+Trap (do not trap) on integer division by zero. The default is '-mcheck-zero-
+division'.
+
+
+@item -mcond-move-int
+@itemx -mno-cond-move-int
+@opindex -mcond-move-int
+Conditional moves for floating point are enabled (disabled). The default is
+'-mcond-move-float'.
+
+@item -mmemcpy
+@itemx -mno-memcpy
+@opindex -mmemcpy
+Force (do not force) the use of memcpy for non-trivial block moves. The default
+is '-mno-memcpy', which allows GCC to inline most constant-sized copies.
+
+
+@item -mlra
+@opindex -mlra
+Use the new LRA register allocator. By default, the LRA is used.
+
+@item -mstrict-align
+@itemx -mno-strict-align
+@opindex -mstrict-align
+Avoid or allow generating memory accesses that may not be aligned on a natural
+object boundary as described in the architecture specification.
+
+@item -msmall-data-limit=@var{number}
+@opindex -msmall-data-limit
+Put global and static data smaller than @code{number} bytes into a special section (on some targets).
+Default value is 0.
+
+
+@item -mmax-inline-memcpy-size=@var{n}
+@opindex -mmax-inline-memcpy-size
+Set the max size n of memcpy to inline, default n is 1024.
+
+@item -mcmodel=@var{code-model}
+Default code model is normal.
+Set the code model to one of:
+@table @samp
+@item tiny-static
+@itemize @bullet
+@item
+local symbol and global strong symbol: The data section must be within +/-2MiB addressing space.
+The text section must be within +/-128MiB addressing space.
+@item
+global weak symbol: The got table must be within +/-2GiB addressing space.
+@end itemize
+
+@item tiny
+@itemize @bullet
+@item
+local symbol: The data section must be within +/-2MiB addressing space.
+The text section must be within +/-128MiB
+addressing space.
+@item
+global symbol: The got table must be within +/-2GiB addressing space.
+@end itemize
+
+@item normal
+@itemize @bullet
+@item
+local symbol: The data section must be within +/-2GiB addressing space.
+The text section must be within +/-128MiB addressing space.
+@item
+global symbol: The got table must be within +/-2GiB addressing space.
+@end itemize
+
+@item large
+@itemize @bullet
+@item
+local symbol: The data section must be within +/-2GiB addressing space.
+The text section must be within +/-128GiB addressing space.
+@item
+global symbol: The got table must be within +/-2GiB addressing space.
+@end itemize
+
+@item extreme(Not implemented yet)
+@itemize @bullet
+@item
+local symbol: The data and text section must be within +/-8EiB addressing space.
+@item
+global symbol: The data got table must be within +/-8EiB addressing space.
+@end itemize
+@end table
+@end table
+
+
+
@node M32C Options
@subsection M32C Options
@cindex M32C options
diff --git a/gcc/doc/md.texi b/gcc/doc/md.texi
index 9ec051e94e1..8f8ab195c5f 100644
--- a/gcc/doc/md.texi
+++ b/gcc/doc/md.texi
@@ -2747,6 +2747,61 @@ Memory addressed using the small base register ($sb).
$r1h
@end table
+@item LoongArch---@file{config/loongarch/constraints.md}
+@table @code
+@item a
+A constant call global and noplt address.
+@item c
+A constant call local address.
+@item e
+A register that is used as function call.
+@item f
+A floating-point register (if available).
+@item h
+A constant call plt address.
+@item j
+A rester that is used as sibing call.
+@item l
+A signed 16-bit constant.
+@item m
+A memory operand whose address is formed by a base register and offset
+that is suitable for use in instructions with the same addressing mode
+as @code{st.w} and @code{ld.w}.
+@item q
+A general-purpose register except for $r0 and $r1 for csr instructions.
+@item t
+A constant call weak address.
+@item u
+A signed 52bit constant and low 32-bit is zero (for logic instructions).
+@item v
+A nsigned 64-bit constant and low 44-bit is zero (for logic instructions).
+@item z
+A floating-point condition code register.
+@item G
+Floating-point zero.
+@item I
+A signed 12-bit constant (for arithmetic instructions).
+@item J
+Integer zero.
+@item K
+An unsigned 12-bit constant (for logic instructions).
+@item Q
+A 12-bit constant used for logical operations.
+@item W
+A memory address based on a member of @code{BASE_REG_CLASS}. This is
+true for allreferences.
+@item Yd
+A constant @code{move_operand} that can be safely loaded using
+@code{la}.
+@item ZB
+An address that is held in a general-purpose register.
+The offset is zero.
+@item ZC
+A memory operand whose address is formed by a base register and offset
+that is suitable for use in instructions with the same addressing mode
+as @code{ll.w} and @code{sc.w}.
+@end table
+
@item MicroBlaze---@file{config/microblaze/constraints.md}
@table @code
@item d
--
2.27.0
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH v4 00/12] Add LoongArch support.
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
` (11 preceding siblings ...)
2021-12-24 9:28 ` [PATCH v4 12/12] LoongArch Port: Add doc chenglulu
@ 2022-01-05 14:13 ` Paul Hua
12 siblings, 0 replies; 16+ messages in thread
From: Paul Hua @ 2022-01-05 14:13 UTC (permalink / raw)
To: chenglulu; +Cc: gcc-patches, Xu Chenghua, yangyujie
Hi all,
Ping?
By the way, the LoongArch machine is already connected to the Cfarm
and will be announced soon.
You can login through ssh.
ssh -l your-cfarm-user-name -p 25469 114.242.206.180
On Fri, Dec 24, 2021 at 5:28 PM chenglulu <chenglulu@loongson.cn> wrote:
>
> The LoongArch architecture (LoongArch) is an Instruction Set
> Architecture (ISA) that has a Reduced Instruction Set Computer (RISC)
> style.
> The documents are on
> https://loongson.github.io/LoongArch-Documentation/README-EN.html
>
> The ELF ABI Documents are on:
> https://loongson.github.io/LoongArch-Documentation/LoongArch-ELF-ABI-EN.html
>
> The binutils has been merged into trunk:
> https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=560b3fe208255ae909b4b1c88ba9c28b09043307
>
> Note: We split -mabi= into -mabi=lp64d/f/s, the new options not support by upstream binutils yet,
> this GCC port requires the following patch applied to binutils to build.
> https://github.com/loongson/binutils-gdb/commit/aacb0bf860f02aa5a7dcb76dd0e392bf871c7586
> (will be submitted to upstream after gcc side comfirmed)
>
>
> changelog:
>
> v1 -> v2
> 1. Split patch set.
> 2. Change some code style.
> 3. Add -mabi=lp64d/f/s options.
> 4. Change GLIBC_DYNAMIC_LINKER_LP64 name.
>
> v2 -> v3
> 1. Change some code style.
> 2. Bug fix.
>
> v3 -> v4
> 1. Change some code style.
> 2. Bug fix.
> 3. Delete some builtin macros.
>
> Add LoongArch support.
>
> chenglulu (12):
> LoongArch Port: Regenerate configure
> LoongArch Port: gcc build
> LoongArch Port: Regenerate gcc/configure.
> LoongArch Port: Machine Decsription files.
> LoongArch Port: Machine description C files and .h files.
> LoongArch Port: Builtin functions.
> LoongArch Port: Builtin macros.
> LoongArch Port: libgcc
> LoongArch Port: Regenerate libgcc/configure.
> LoongArch Port: libgomp
> LoongArch Port: gcc/testsuite
> LoongArch Port: Add doc.
>
> config/picflag.m4 | 3 +
> configure | 10 +-
> configure.ac | 10 +-
> contrib/config-list.mk | 5 +-
> .../config/loongarch/loongarch-common.c | 63 +
> gcc/config.gcc | 400 +-
> gcc/config/host-linux.c | 2 +
> gcc/config/loongarch/constraints.md | 212 +
> gcc/config/loongarch/generic.md | 132 +
> gcc/config/loongarch/genopts/genstr.sh | 91 +
> .../loongarch/genopts/loongarch-strings | 58 +
> gcc/config/loongarch/genopts/loongarch.opt.in | 189 +
> gcc/config/loongarch/gnu-user.h | 78 +
> gcc/config/loongarch/la464.md | 132 +
> gcc/config/loongarch/larchintrin.h | 413 ++
> gcc/config/loongarch/linux.h | 50 +
> gcc/config/loongarch/loongarch-builtins.c | 511 ++
> gcc/config/loongarch/loongarch-c.c | 111 +
> gcc/config/loongarch/loongarch-cpu.c | 206 +
> gcc/config/loongarch/loongarch-cpu.h | 30 +
> gcc/config/loongarch/loongarch-def.c | 164 +
> gcc/config/loongarch/loongarch-def.h | 151 +
> gcc/config/loongarch/loongarch-driver.c | 187 +
> gcc/config/loongarch/loongarch-driver.h | 69 +
> gcc/config/loongarch/loongarch-ftypes.def | 106 +
> gcc/config/loongarch/loongarch-modes.def | 29 +
> gcc/config/loongarch/loongarch-opts.c | 582 ++
> gcc/config/loongarch/loongarch-opts.h | 86 +
> gcc/config/loongarch/loongarch-protos.h | 242 +
> gcc/config/loongarch/loongarch-str.h | 57 +
> gcc/config/loongarch/loongarch-tune.h | 72 +
> gcc/config/loongarch/loongarch.c | 6461 +++++++++++++++++
> gcc/config/loongarch/loongarch.h | 1291 ++++
> gcc/config/loongarch/loongarch.md | 3829 ++++++++++
> gcc/config/loongarch/loongarch.opt | 189 +
> gcc/config/loongarch/predicates.md | 553 ++
> gcc/config/loongarch/sync.md | 574 ++
> gcc/config/loongarch/t-linux | 53 +
> gcc/config/loongarch/t-loongarch | 59 +
> gcc/configure | 85 +-
> gcc/configure.ac | 33 +-
> gcc/doc/install.texi | 47 +-
> gcc/doc/invoke.texi | 201 +
> gcc/doc/md.texi | 55 +
> gcc/testsuite/g++.dg/cpp0x/constexpr-rom.C | 2 +-
> gcc/testsuite/g++.old-deja/g++.abi/ptrmem.C | 2 +-
> gcc/testsuite/g++.old-deja/g++.pt/ptrmem6.C | 2 +-
> gcc/testsuite/gcc.dg/20020312-2.c | 2 +
> gcc/testsuite/gcc.dg/loop-8.c | 2 +-
> .../torture/stackalign/builtin-apply-2.c | 2 +-
> gcc/testsuite/gcc.dg/tree-ssa/ssa-fre-3.c | 2 +-
> .../gcc.target/loongarch/loongarch.exp | 40 +
> .../gcc.target/loongarch/tst-asm-const.c | 16 +
> gcc/testsuite/go.test/go-test.exp | 3 +
> gcc/testsuite/lib/target-supports.exp | 14 +
> libgcc/config.host | 28 +-
> libgcc/config/loongarch/crtfastmath.c | 52 +
> libgcc/config/loongarch/crti.S | 43 +
> libgcc/config/loongarch/crtn.S | 39 +
> libgcc/config/loongarch/lib2funcs.c | 0
> libgcc/config/loongarch/linux-unwind.h | 80 +
> libgcc/config/loongarch/sfp-machine.h | 152 +
> libgcc/config/loongarch/t-crtstuff | 2 +
> libgcc/config/loongarch/t-elf | 3 +
> libgcc/config/loongarch/t-loongarch | 9 +
> libgcc/config/loongarch/t-loongarch64 | 1 +
> libgcc/config/loongarch/t-softfp-tf | 3 +
> libgcc/configure | 5 +-
> libgcc/configure.ac | 2 +-
> libgomp/configure.tgt | 4 +
> 70 files changed, 18363 insertions(+), 28 deletions(-)
> create mode 100644 gcc/common/config/loongarch/loongarch-common.c
> create mode 100644 gcc/config/loongarch/constraints.md
> create mode 100644 gcc/config/loongarch/generic.md
> create mode 100755 gcc/config/loongarch/genopts/genstr.sh
> create mode 100644 gcc/config/loongarch/genopts/loongarch-strings
> create mode 100644 gcc/config/loongarch/genopts/loongarch.opt.in
> create mode 100644 gcc/config/loongarch/gnu-user.h
> create mode 100644 gcc/config/loongarch/la464.md
> create mode 100644 gcc/config/loongarch/larchintrin.h
> create mode 100644 gcc/config/loongarch/linux.h
> create mode 100644 gcc/config/loongarch/loongarch-builtins.c
> create mode 100644 gcc/config/loongarch/loongarch-c.c
> create mode 100644 gcc/config/loongarch/loongarch-cpu.c
> create mode 100644 gcc/config/loongarch/loongarch-cpu.h
> create mode 100644 gcc/config/loongarch/loongarch-def.c
> create mode 100644 gcc/config/loongarch/loongarch-def.h
> create mode 100644 gcc/config/loongarch/loongarch-driver.c
> create mode 100644 gcc/config/loongarch/loongarch-driver.h
> create mode 100644 gcc/config/loongarch/loongarch-ftypes.def
> create mode 100644 gcc/config/loongarch/loongarch-modes.def
> create mode 100644 gcc/config/loongarch/loongarch-opts.c
> create mode 100644 gcc/config/loongarch/loongarch-opts.h
> create mode 100644 gcc/config/loongarch/loongarch-protos.h
> create mode 100644 gcc/config/loongarch/loongarch-str.h
> create mode 100644 gcc/config/loongarch/loongarch-tune.h
> create mode 100644 gcc/config/loongarch/loongarch.c
> create mode 100644 gcc/config/loongarch/loongarch.h
> create mode 100644 gcc/config/loongarch/loongarch.md
> create mode 100644 gcc/config/loongarch/loongarch.opt
> create mode 100644 gcc/config/loongarch/predicates.md
> create mode 100644 gcc/config/loongarch/sync.md
> create mode 100644 gcc/config/loongarch/t-linux
> create mode 100644 gcc/config/loongarch/t-loongarch
> create mode 100644 gcc/testsuite/gcc.target/loongarch/loongarch.exp
> create mode 100644 gcc/testsuite/gcc.target/loongarch/tst-asm-const.c
> create mode 100644 libgcc/config/loongarch/crtfastmath.c
> create mode 100644 libgcc/config/loongarch/crti.S
> create mode 100644 libgcc/config/loongarch/crtn.S
> create mode 100644 libgcc/config/loongarch/lib2funcs.c
> create mode 100644 libgcc/config/loongarch/linux-unwind.h
> create mode 100644 libgcc/config/loongarch/sfp-machine.h
> create mode 100644 libgcc/config/loongarch/t-crtstuff
> create mode 100644 libgcc/config/loongarch/t-elf
> create mode 100644 libgcc/config/loongarch/t-loongarch
> create mode 100644 libgcc/config/loongarch/t-loongarch64
> create mode 100644 libgcc/config/loongarch/t-softfp-tf
>
> --
> 2.27.0
>
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH v4 04/12] LoongArch Port: Machine Decsription files.
2021-12-24 9:28 ` [PATCH v4 04/12] LoongArch Port: Machine Decsription files chenglulu
@ 2022-01-06 17:54 ` Xi Ruoyao
2022-01-07 11:36 ` 程璐璐
0 siblings, 1 reply; 16+ messages in thread
From: Xi Ruoyao @ 2022-01-06 17:54 UTC (permalink / raw)
To: chenglulu, gcc-patches; +Cc: xuchenghua, yangyujie
[-- Attachment #1: Type: text/plain, Size: 1659 bytes --]
On Fri, 2021-12-24 at 17:28 +0800, chenglulu wrote:
> +(define_insn "*zero_extendsidi2_internal"
> + [(set (match_operand:DI 0 "register_operand" "=r,r,r")
> + (subreg:DI (match_operand:SI 1 "nonimmediate_operand" "r,ZC,W") 0))]
> + "TARGET_64BIT"
> + "@
> + bstrpick.d\t%0,%1,31,0
> + ldptr.w\t%0,%1\n\tlu32i.d\t%0,0
> + ld.wu\t%0,%1"
> + [(set_attr "move_type" "arith,load,load")
> + (set_attr "mode" "DI")
> + (set_attr "insn_count" "1,2,1")])
This pattern is generating wrong code, causing
FAIL: gcc.dg/compat/scalar-by-value-3 c_compat_x_tst.o-c_compat_y_tst.o execute
This failure is a real bug, the reduced testcase is attached. In the
assembly:
# ...
bstrins.d $r5,$r13,31,0
addi.d $r12,$r22,-228
bstrpick.d $r12,$r12,31,0
bstrins.d $r5,$r12,63,32
addi.w $r4,$r0,14 # 0xe
bl testvaci
This obviously does not make any sense: it calculates the *address* of
g[0]'s imaginary part, truncates it to 32-bit, then pass it as the
*value* of the imaginary part to testvaci().
The problem is: before the reload pass, the compiler consider g[0] a
(virtual) register. It only becomes MEM after the reload pass. Adding
reload_completed as the condition of this insn seems able to fix the
issue. However I'm not sure if other insns need the change too.
> +;; See the comment before the *and<mode>3 pattern why this is generated by
> +;; combine.
A minor issue: the comment before 'and<mode>3' does not exist.
--
Xi Ruoyao <xry111@mengyan1223.wang>
School of Aerospace Science and Technology, Xidian University
[-- Attachment #2: bug.c --]
[-- Type: text/x-csrc, Size: 608 bytes --]
#include <stdarg.h>
#include <assert.h>
extern void testvaci (int n, ...);
int main() {
int i;
volatile _Complex int g[14];
for (i = 0; i < 14; i++)
g[i] = i + 3 + 3i;
testvaci (14, g[0], g[1], g[2], g[3], g[4], g[5], g[6], g[7], g[8], g[9], g[10], g[11], g[12], g[13]);
return 0;
}
void testvaci (int n, ...) {
int i;
va_list ap;
va_start(ap, n);
for (i = 0; i < n; i++) {
_Complex int t = __builtin_va_arg(ap, _Complex int);
assert(t == i + 3 + 3i);
}
va_end(ap);
}
^ permalink raw reply [flat|nested] 16+ messages in thread
* Re: [PATCH v4 04/12] LoongArch Port: Machine Decsription files.
2022-01-06 17:54 ` Xi Ruoyao
@ 2022-01-07 11:36 ` 程璐璐
0 siblings, 0 replies; 16+ messages in thread
From: 程璐璐 @ 2022-01-07 11:36 UTC (permalink / raw)
To: Xi Ruoyao, gcc-patches; +Cc: xuchenghua, yangyujie
在 2022/1/7 上午1:54, Xi Ruoyao 写道:
> On Fri, 2021-12-24 at 17:28 +0800, chenglulu wrote:
>> +(define_insn "*zero_extendsidi2_internal"
>> + [(set (match_operand:DI 0 "register_operand" "=r,r,r")
>> + (subreg:DI (match_operand:SI 1 "nonimmediate_operand" "r,ZC,W") 0))]
>> + "TARGET_64BIT"
>> + "@
>> + bstrpick.d\t%0,%1,31,0
>> + ldptr.w\t%0,%1\n\tlu32i.d\t%0,0
>> + ld.wu\t%0,%1"
>> + [(set_attr "move_type" "arith,load,load")
>> + (set_attr "mode" "DI")
>> + (set_attr "insn_count" "1,2,1")])
> This pattern is generating wrong code, causing
>
> FAIL: gcc.dg/compat/scalar-by-value-3 c_compat_x_tst.o-c_compat_y_tst.o execute
>
> This failure is a real bug, the reduced testcase is attached. In the
> assembly:
>
> # ...
> bstrins.d $r5,$r13,31,0
> addi.d $r12,$r22,-228
> bstrpick.d $r12,$r12,31,0
> bstrins.d $r5,$r12,63,32
> addi.w $r4,$r0,14 # 0xe
> bl testvaci
>
> This obviously does not make any sense: it calculates the *address* of
> g[0]'s imaginary part, truncates it to 32-bit, then pass it as the
> *value* of the imaginary part to testvaci().
>
> The problem is: before the reload pass, the compiler consider g[0] a
> (virtual) register. It only becomes MEM after the reload pass. Adding
> reload_completed as the condition of this insn seems able to fix the
> issue. However I'm not sure if other insns need the change too.
>
>> +;; See the comment before the *and<mode>3 pattern why this is generated by
>> +;; combine.
> A minor issue: the comment before 'and<mode>3' does not exist.
Thanks, we are working on this. In addition, we will check other
instruction templates.
^ permalink raw reply [flat|nested] 16+ messages in thread
end of thread, other threads:[~2022-01-07 11:36 UTC | newest]
Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-24 9:28 [PATCH v4 00/12] Add LoongArch support chenglulu
2021-12-24 9:28 ` [PATCH v4 01/12] LoongArch Port: Regenerate configure chenglulu
2021-12-24 9:28 ` [PATCH v4 02/12] LoongArch Port: gcc build chenglulu
2021-12-24 9:28 ` [PATCH v4 03/12] LoongArch Port: Regenerate gcc/configure chenglulu
2021-12-24 9:28 ` [PATCH v4 04/12] LoongArch Port: Machine Decsription files chenglulu
2022-01-06 17:54 ` Xi Ruoyao
2022-01-07 11:36 ` 程璐璐
2021-12-24 9:28 ` [PATCH v4 05/12] LoongArch Port: Machine description C files and .h files chenglulu
2021-12-24 9:28 ` [PATCH v4 06/12] LoongArch Port: Builtin functions chenglulu
2021-12-24 9:28 ` [PATCH v4 07/12] LoongArch Port: Builtin macros chenglulu
2021-12-24 9:28 ` [PATCH v4 08/12] LoongArch Port: libgcc chenglulu
2021-12-24 9:28 ` [PATCH v4 09/12] LoongArch Port: Regenerate libgcc/configure chenglulu
2021-12-24 9:28 ` [PATCH v4 10/12] LoongArch Port: libgomp chenglulu
2021-12-24 9:28 ` [PATCH v4 11/12] LoongArch Port: gcc/testsuite chenglulu
2021-12-24 9:28 ` [PATCH v4 12/12] LoongArch Port: Add doc chenglulu
2022-01-05 14:13 ` [PATCH v4 00/12] Add LoongArch support Paul Hua
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).