[PATCH v2 00/14] LoongArch: Add loongarch32 and ilp32 abi

[PATCH v2 00/14] LoongArch: Add loongarch32 and ilp32 abi

[Jiajie Chen]

The patch series add loongarch32 and ilp32 abi support to gcc. One can
build libgcc, libatomic and glibc etc and generate a complete
loongarch32-unknown-linux-gnu-toolchain with minimal patches at:

- binutils: https://github.com/jiegec/binutils-gdb/tree/loongarch32
- glibc: https://github.com/jiegec/glibc/tree/loongarch32
- crosstool-ng: https://github.com/jiegec/crosstool-ng/tree/loongarch32

We will wait for folks at Loongson to complete abi documentation of
ilp32d abi.

Changes since v1:

- Can build a complete toolchain with glibc
- Fix sizeof(long double) in ilp32 abi
- Fix ftintrz generation for loongarch32

Full changes:

Jiajie Chen (14):
  LoongArch: Introduce loongarch32 target
  LoongArch: Fix default ISA setting
  LoongArch: Fix SI division for loongarch32 target
  LoongArch: Fix movgr2frh.w operand order
  LoongArch: Fix 64-bit move for loongarch32 target
  LoongArch: Fix 64-bit immediate move for loongarch32 target
  LoongArch: Fix signed 32-bit overflow for loongarch32 target
  LoongArch: Disable SF/DF -> unsigned DI expand in loongarch32
  LoongArch: Add -march=loongarch64 to tests with -mabi=lp64d
  LoongArch: Forbid ADDRESS_REG_REG in loongarch32
  LoongArch: Mark am* instructions as LA64-only
  LoongArch: Set long double width to 128 in la32
  LoongArch: Fix ilp32 detection
  LoongArch: Allow ftintrz for DF->DI in loongarch32

 contrib/config-list.mk                        |  1 +
 gcc/config.gcc                                | 61 ++++++++++++++++---
 .../loongarch/genopts/loongarch-strings       |  5 ++
 gcc/config/loongarch/genopts/loongarch.opt.in | 12 ++++
 gcc/config/loongarch/gnu-user.h               |  3 +
 gcc/config/loongarch/linux.h                  |  8 ++-
 gcc/config/loongarch/loongarch-c.cc           | 12 ++++
 gcc/config/loongarch/loongarch-def.c          | 33 ++++++++++
 gcc/config/loongarch/loongarch-def.h          | 25 +++++---
 gcc/config/loongarch/loongarch-driver.h       |  4 ++
 gcc/config/loongarch/loongarch-opts.cc        | 27 ++++++--
 gcc/config/loongarch/loongarch-opts.h         | 20 ++++--
 gcc/config/loongarch/loongarch-str.h          |  5 ++
 gcc/config/loongarch/loongarch.cc             |  7 ++-
 gcc/config/loongarch/loongarch.h              |  2 +-
 gcc/config/loongarch/loongarch.md             | 39 ++++++------
 gcc/config/loongarch/loongarch.opt            | 12 ++++
 gcc/config/loongarch/sync.md                  | 10 +--
 gcc/config/loongarch/t-linux                  | 16 ++++-
 gcc/testsuite/g++.target/loongarch/bytepick.C |  2 +-
 gcc/testsuite/g++.target/loongarch/pr106828.C |  2 +-
 .../gcc.target/loongarch/add-const.c          |  2 +-
 gcc/testsuite/gcc.target/loongarch/arch-1.c   |  5 ++
 gcc/testsuite/gcc.target/loongarch/arch-2.c   |  5 ++
 .../gcc.target/loongarch/array-ldx.c          |  6 ++
 .../gcc.target/loongarch/attr-model-1.c       |  2 +-
 .../gcc.target/loongarch/attr-model-2.c       |  2 +-
 .../gcc.target/loongarch/flt-abi-isa-1.c      |  2 +-
 gcc/testsuite/gcc.target/loongarch/fscaleb.c  |  2 +-
 .../gcc.target/loongarch/ftint-no-inexact.c   |  2 +-
 gcc/testsuite/gcc.target/loongarch/ftint.c    |  2 +-
 .../gcc.target/loongarch/func-call-1.c        |  2 +-
 .../gcc.target/loongarch/func-call-2.c        |  2 +-
 .../gcc.target/loongarch/func-call-3.c        |  2 +-
 .../gcc.target/loongarch/func-call-4.c        |  2 +-
 .../gcc.target/loongarch/func-call-5.c        |  2 +-
 .../gcc.target/loongarch/func-call-6.c        |  2 +-
 .../gcc.target/loongarch/func-call-7.c        |  2 +-
 .../gcc.target/loongarch/func-call-8.c        |  2 +-
 .../loongarch/func-call-extreme-1.c           |  2 +-
 .../loongarch/func-call-extreme-2.c           |  2 +-
 .../gcc.target/loongarch/func-call-medium-1.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-2.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-3.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-4.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-5.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-6.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-7.c |  2 +-
 .../gcc.target/loongarch/func-call-medium-8.c |  2 +-
 gcc/testsuite/gcc.target/loongarch/imm-load.c |  2 +-
 .../gcc.target/loongarch/imm-load1.c          |  2 +-
 gcc/testsuite/gcc.target/loongarch/mulw_d_w.c |  2 +-
 .../gcc.target/loongarch/pr109465-1.c         |  2 +-
 .../gcc.target/loongarch/pr109465-2.c         |  2 +-
 .../gcc.target/loongarch/pr109465-3.c         |  2 +-
 .../gcc.target/loongarch/prolog-opt.c         |  2 +-
 .../loongarch/relocs-symbol-noaddend.c        |  2 +-
 .../loongarch/zero-size-field-pass.c          |  2 +-
 .../loongarch/zero-size-field-ret.c           |  2 +-
 libitm/config/loongarch/asm.h                 |  2 +-
 60 files changed, 298 insertions(+), 96 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/arch-1.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/arch-2.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/array-ldx.c

-- 
2.41.0

[PATCH v2 01/14] LoongArch: Introduce loongarch32 target

[Jiajie Chen]

Introduce loongarch32 target and ilp32 abi variants. The ilp32d abi
variant is selected as the default abi of loongarch32 target.

Currently, ilp32 abi can only be used for loongarch32, but in the
future, it might be possible to use ilp32 abi for loongarch64.

contrib/ChangeLog:

	* config-list.mk: Add loongarch32-linux-gnu*.

gcc/ChangeLog:

	* config.gcc: Add target triple loongarch32-*-*-* and
	corresponding abi ilp32f, ilp32d and ilp32s.
	* config/loongarch/genopts/loongarch-strings: Add strings for
	loongarch32 and ilp32 abi variants.
	* config/loongarch/genopts/loongarch.opt.in: Add
	-march=loongarch32 and -mabi=ilp32d/ilp32f/ilp32s.
	* config/loongarch/gnu-user.h: Add ilp32 abi variants to spec.
	* config/loongarch/linux.h: Add ABI_LIBDIR for ilp32 abi
	variants.
	* config/loongarch/loongarch-c.cc (loongarch_cpu_cpp_builtins):
	Add builtin definitions for loongarch32 target.
	* config/loongarch/loongarch-def.c: Add loongarch32 and ilp32
	definitions.
	* config/loongarch/loongarch-def.h: Add loongarch32 and ilp32
	definitions.
	* config/loongarch/loongarch-driver.h: Add ilp32 abi variants to
	spec.
	* config/loongarch/loongarch-opts.cc: Handle ilp32 abi variants.
	* config/loongarch/loongarch-opts.h: Add loongarch32 case to
	macros.
	* config/loongarch/loongarch-str.h: Add loongarch32 and ilp32
	strings.
	* config/loongarch/loongarch.cc: Disable -fpcc-struct-return for
	ilp32.
	* config/loongarch/loongarch.opt: Add -march=loongarch32 and
	-mabi=ilp32d/ilp32f/ilp32s.
	* config/loongarch/t-linux: Add ilp32 abi variants to multilib.
---
 contrib/config-list.mk                        |  1 +
 gcc/config.gcc                                | 61 ++++++++++++++++---
 .../loongarch/genopts/loongarch-strings       |  5 ++
 gcc/config/loongarch/genopts/loongarch.opt.in | 12 ++++
 gcc/config/loongarch/gnu-user.h               |  3 +
 gcc/config/loongarch/linux.h                  |  8 ++-
 gcc/config/loongarch/loongarch-c.cc           | 12 ++++
 gcc/config/loongarch/loongarch-def.c          | 33 ++++++++++
 gcc/config/loongarch/loongarch-def.h          | 25 +++++---
 gcc/config/loongarch/loongarch-driver.h       |  4 ++
 gcc/config/loongarch/loongarch-opts.cc        | 22 ++++++-
 gcc/config/loongarch/loongarch-opts.h         | 20 ++++--
 gcc/config/loongarch/loongarch-str.h          |  5 ++
 gcc/config/loongarch/loongarch.cc             |  2 +-
 gcc/config/loongarch/loongarch.opt            | 12 ++++
 gcc/config/loongarch/t-linux                  | 16 ++++-
 16 files changed, 210 insertions(+), 31 deletions(-)

diff --git a/contrib/config-list.mk b/contrib/config-list.mk
index e570b13c71b..3c00ce5410a 100644
--- a/contrib/config-list.mk
+++ b/contrib/config-list.mk
@@ -57,6 +57,7 @@ LIST = aarch64-elf aarch64-freebsd13 aarch64-linux-gnu aarch64-rtems \
   i686-cygwinOPT-enable-threads=yes i686-mingw32crt ia64-elf \
   ia64-linux ia64-hpux ia64-hp-vms iq2000-elf lm32-elf \
   lm32-rtems lm32-uclinux \
+  loongarch32-linux-gnuf64 loongarch32-linux-gnuf32 loongarch32-linux-gnusf \
   loongarch64-linux-gnuf64 loongarch64-linux-gnuf32 loongarch64-linux-gnusf \
   m32c-elf m32r-elf m32rle-elf \
   m68k-elf m68k-netbsdelf \
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 415e0e1ebc5..45e69b24b44 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -4901,10 +4901,24 @@ case "${target}" in
 			arch_pattern     arch_default   \
 			fpu_pattern      fpu_default    \
 			tune_pattern     tune_default   \
-			triplet_os       triplet_abi
+			triplet_os       triplet_abi    \
+			triplet_cpu
 
 		# Infer ABI from the triplet.
 		case ${target} in
+		loongarch32-*-*-*f64)
+			abi_pattern="ilp32d"
+			;;
+		loongarch32-*-*-*f32)
+			abi_pattern="ilp32f"
+			;;
+		loongarch32-*-*-*sf)
+			abi_pattern="ilp32s"
+			;;
+		loongarch32-*-*-*)
+			abi_pattern="ilp32[dfs]"
+			abi_default="ilp32d"
+			;;
 		loongarch64-*-*-*f64)
 			abi_pattern="lp64d"
 			;;
@@ -4939,7 +4953,7 @@ case "${target}" in
 
 		# Perform initial sanity checks on --with-* options.
 		case ${with_arch} in
-		"" | loongarch64 | la464) ;; # OK, append here.
+		"" | loongarch32 | loongarch64 | la464) ;; # OK, append here.
 		native)
 			if test x${host} != x${target}; then
 				echo "--with-arch=native is illegal for cross-compiler." 1>&2
@@ -4958,7 +4972,7 @@ case "${target}" in
 		esac
 
 		case ${with_abi} in
-		"" | lp64d | lp64f | lp64s) ;; # OK, append here.
+		"" | ilp32d | ilp32f | ilp32s | lp64d | lp64f | lp64s) ;; # OK, append here.
 		*)
 			echo "Unsupported ABI given in --with-abi=$with_abi" 1>&2
 			exit 1
@@ -5007,12 +5021,21 @@ case "${target}" in
 			;;
 		esac
 
+		case ${target} in
+		loongarch32-*-*-*)
+			triplet_cpu="loongarch32"
+			;;
+		loongarch64-*-*-*)
+			triplet_cpu="loongarch64"
+			;;
+		esac
+
 		case ${with_abi} in
-		  "lp64d") triplet_abi="";;
-		  "lp64f") triplet_abi="f32";;
-		  "lp64s") triplet_abi="sf";;
+		  "ilp32d" | "lp64d") triplet_abi="";;
+		  "ilp32f" | "lp64f") triplet_abi="f32";;
+		  "ilp32s" | "lp64s") triplet_abi="sf";;
 		esac
-		la_canonical_triplet="loongarch64-${triplet_os}${triplet_abi}"
+		la_canonical_triplet="${triplet_cpu}-${triplet_os}${triplet_abi}"
 
 		# Set default value for with_abiext (internal)
 		case ${with_abiext} in
@@ -5038,6 +5061,12 @@ case "${target}" in
 
 		# Infer ISA-related default options from the ABI: pass 1
 		case ${with_abi}/${with_abiext} in
+		ilp32*/base)
+			# architectures that support ilp32* ABI
+			arch_pattern="native|loongarch32"
+			# default architecture for ilp32* ABI
+			arch_default="loongarch32"
+			;;
 		lp64*/base)
 			# architectures that support lp64* ABI
 			arch_pattern="native|loongarch64|la464"
@@ -5052,14 +5081,14 @@ case "${target}" in
 
 		# Infer ISA-related default options from the ABI: pass 2
 		case ${with_abi}/${with_abiext} in
-		lp64d/base)
+		ilp32d/base | lp64d/base)
 			fpu_pattern="64"
 			;;
-		lp64f/base)
+		ilp32f/base | lp64f/base)
 			fpu_pattern="32|64"
 			fpu_default="32"
 			;;
-		lp64s/base)
+		ilp32s/base | lp64s/base)
 			fpu_pattern="none|32|64"
 			fpu_default="none"
 			;;
@@ -5118,6 +5147,10 @@ case "${target}" in
 			tune_pattern="*"
 			tune_default="native"
 			;;
+		loongarch32)
+			tune_pattern="loongarch32"
+			tune_default="loongarch32"
+			;;
 		loongarch64)
 			tune_pattern="loongarch64|la464"
 			tune_default="la464"
@@ -5187,6 +5220,9 @@ case "${target}" in
 				1)
 					# Component 1: Base ABI type
 					case ${component} in
+					ilp32d) elem_tmp="ABI_BASE_ILP32D,";;
+					ilp32f) elem_tmp="ABI_BASE_ILP32F,";;
+					ilp32s) elem_tmp="ABI_BASE_ILP32S,";;
 					lp64d) elem_tmp="ABI_BASE_LP64D,";;
 					lp64f) elem_tmp="ABI_BASE_LP64F,";;
 					lp64s) elem_tmp="ABI_BASE_LP64S,";;
@@ -5741,17 +5777,22 @@ case ${target} in
 		# See macro definitions from loongarch-opts.h and loongarch-cpu.h.
 		case ${with_arch} in
 		native)		tm_defines="${tm_defines} DEFAULT_CPU_ARCH=CPU_NATIVE" ;;
+		loongarch32)	tm_defines="${tm_defines} DEFAULT_CPU_ARCH=CPU_LOONGARCH32" ;;
 		la464)		tm_defines="${tm_defines} DEFAULT_CPU_ARCH=CPU_LA464" ;;
 		loongarch64)	tm_defines="${tm_defines} DEFAULT_CPU_ARCH=CPU_LOONGARCH64" ;;
 		esac
 
 		case ${with_tune} in
 		native)		tm_defines="${tm_defines} DEFAULT_CPU_TUNE=CPU_NATIVE" ;;
+		loongarch32)	tm_defines="${tm_defines} DEFAULT_CPU_TUNE=CPU_LOONGARCH32" ;;
 		la464)		tm_defines="${tm_defines} DEFAULT_CPU_TUNE=CPU_LA464" ;;
 		loongarch64)	tm_defines="${tm_defines} DEFAULT_CPU_TUNE=CPU_LOONGARCH64" ;;
 		esac
 
 		case ${with_abi} in
+		ilp32d)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_ILP32D" ;;
+		ilp32f)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_ILP32F" ;;
+		ilp32s)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_ILP32S" ;;
 		lp64d)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_LP64D" ;;
 		lp64f)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_LP64F" ;;
 		lp64s)     tm_defines="${tm_defines} DEFAULT_ABI_BASE=ABI_BASE_LP64S" ;;
diff --git a/gcc/config/loongarch/genopts/loongarch-strings b/gcc/config/loongarch/genopts/loongarch-strings
index a40998ead97..1e20b70dde2 100644
--- a/gcc/config/loongarch/genopts/loongarch-strings
+++ b/gcc/config/loongarch/genopts/loongarch-strings
@@ -23,10 +23,12 @@ OPTSTR_ARCH	      arch
 OPTSTR_TUNE	      tune
 
 STR_CPU_NATIVE	      native
+STR_CPU_LOONGARCH32   loongarch32
 STR_CPU_LOONGARCH64   loongarch64
 STR_CPU_LA464	      la464
 
 # Base architecture
+STR_ISA_BASE_LA32V100 la32
 STR_ISA_BASE_LA64V100 la64
 
 # -mfpu
@@ -42,6 +44,9 @@ OPTSTR_DOUBLE_FLOAT   double-float
 
 # -mabi=
 OPTSTR_ABI_BASE	      abi
+STR_ABI_BASE_ILP32D   ilp32d
+STR_ABI_BASE_ILP32F   ilp32f
+STR_ABI_BASE_ILP32S   ilp32s
 STR_ABI_BASE_LP64D    lp64d
 STR_ABI_BASE_LP64F    lp64f
 STR_ABI_BASE_LP64S    lp64s
diff --git a/gcc/config/loongarch/genopts/loongarch.opt.in b/gcc/config/loongarch/genopts/loongarch.opt.in
index 4b9b4ac273e..47f77345890 100644
--- a/gcc/config/loongarch/genopts/loongarch.opt.in
+++ b/gcc/config/loongarch/genopts/loongarch.opt.in
@@ -85,6 +85,9 @@ LoongArch CPU types:
 EnumValue
 Enum(cpu_type) String(@@STR_CPU_NATIVE@@) Value(CPU_NATIVE)
 
+EnumValue
+Enum(cpu_type) String(@@STR_CPU_LOONGARCH32@@) Value(CPU_LOONGARCH32)
+
 EnumValue
 Enum(cpu_type) String(@@STR_CPU_LOONGARCH64@@) Value(CPU_LOONGARCH64)
 
@@ -108,6 +111,15 @@ Enum
 Name(abi_base) Type(int)
 Base ABI types for LoongArch:
 
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_ILP32D@@) Value(ABI_BASE_ILP32D)
+
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_ILP32F@@) Value(ABI_BASE_ILP32F)
+
+EnumValue
+Enum(abi_base) String(@@STR_ABI_BASE_ILP32S@@) Value(ABI_BASE_ILP32S)
+
 EnumValue
 Enum(abi_base) String(@@STR_ABI_BASE_LP64D@@) Value(ABI_BASE_LP64D)
 
diff --git a/gcc/config/loongarch/gnu-user.h b/gcc/config/loongarch/gnu-user.h
index fa1a5211419..bf6d3a4fe98 100644
--- a/gcc/config/loongarch/gnu-user.h
+++ b/gcc/config/loongarch/gnu-user.h
@@ -34,6 +34,9 @@ along with GCC; see the file COPYING3.  If not see
   "/lib" ABI_GRLEN_SPEC "/ld-linux-loongarch-" ABI_SPEC ".so.1"
 
 #define MUSL_ABI_SPEC \
+  "%{mabi=ilp32d:-ilp32d}" \
+  "%{mabi=ilp32f:-ilp32f}" \
+  "%{mabi=ilp32s:-ilp32s}" \
   "%{mabi=lp64d:-lp64d}" \
   "%{mabi=lp64f:-lp64f}" \
   "%{mabi=lp64s:-lp64s}"
diff --git a/gcc/config/loongarch/linux.h b/gcc/config/loongarch/linux.h
index 9059d244190..45aa3531cd3 100644
--- a/gcc/config/loongarch/linux.h
+++ b/gcc/config/loongarch/linux.h
@@ -23,7 +23,13 @@ along with GCC; see the file COPYING3.  If not see
 
 #if defined(LA_DISABLE_MULTILIB) && defined(LA_DISABLE_MULTIARCH)
 
-  #if DEFAULT_ABI_BASE == ABI_BASE_LP64D
+  #if DEFAULT_ABI_BASE == ABI_BASE_ILP32D
+    #define ABI_LIBDIR "lib32"
+  #elif DEFAULT_ABI_BASE == ABI_BASE_ILP32F
+    #define ABI_LIBDIR "lib32/f32"
+  #elif DEFAULT_ABI_BASE == ABI_BASE_ILP32S
+    #define ABI_LIBDIR "lib32/sf"
+  #elif DEFAULT_ABI_BASE == ABI_BASE_LP64D
     #define ABI_LIBDIR "lib64"
   #elif DEFAULT_ABI_BASE == ABI_BASE_LP64F
     #define ABI_LIBDIR "lib64/f32"
diff --git a/gcc/config/loongarch/loongarch-c.cc b/gcc/config/loongarch/loongarch-c.cc
index 67911b78f28..e3783a325d7 100644
--- a/gcc/config/loongarch/loongarch-c.cc
+++ b/gcc/config/loongarch/loongarch-c.cc
@@ -65,12 +65,24 @@ loongarch_cpu_cpp_builtins (cpp_reader *pfile)
   LARCH_CPP_SET_PROCESSOR ("_LOONGARCH_TUNE", LARCH_ACTUAL_TUNE);
 
   /* Base architecture / ABI.  */
+  if (TARGET_32BIT)
+    {
+      builtin_define ("__loongarch_grlen=32");
+      builtin_define ("__loongarch32");
+    }
   if (TARGET_64BIT)
     {
       builtin_define ("__loongarch_grlen=64");
       builtin_define ("__loongarch64");
     }
 
+  if (TARGET_ABI_ILP32)
+    {
+      builtin_define ("_ABILP32=3");
+      builtin_define ("_LOONGARCH_SIM=_ABILP32");
+      builtin_define ("__loongarch_ilp32");
+    }
+
   if (TARGET_ABI_LP64)
     {
       builtin_define ("_ABILP64=3");
diff --git a/gcc/config/loongarch/loongarch-def.c b/gcc/config/loongarch/loongarch-def.c
index 6729c857f7c..5ae8bd24c9f 100644
--- a/gcc/config/loongarch/loongarch-def.c
+++ b/gcc/config/loongarch/loongarch-def.c
@@ -40,12 +40,17 @@ along with GCC; see the file COPYING3.  If not see
 const char*
 loongarch_cpu_strings[N_TUNE_TYPES] = {
   [CPU_NATIVE]		  = STR_CPU_NATIVE,
+  [CPU_LOONGARCH32]	  = STR_CPU_LOONGARCH32,
   [CPU_LOONGARCH64]	  = STR_CPU_LOONGARCH64,
   [CPU_LA464]		  = STR_CPU_LA464,
 };
 
 struct loongarch_isa
 loongarch_cpu_default_isa[N_ARCH_TYPES] = {
+  [CPU_LOONGARCH32] = {
+      .base = ISA_BASE_LA32V100,
+      .fpu = ISA_EXT_FPU32,
+  },
   [CPU_LOONGARCH64] = {
       .base = ISA_BASE_LA64V100,
       .fpu = ISA_EXT_FPU64,
@@ -58,6 +63,12 @@ loongarch_cpu_default_isa[N_ARCH_TYPES] = {
 
 struct loongarch_cache
 loongarch_cpu_cache[N_TUNE_TYPES] = {
+  [CPU_LOONGARCH32] = {
+      .l1d_line_size = 64,
+      .l1d_size = 64,
+      .l2d_size = 256,
+      .simultaneous_prefetches = 4,
+  },
   [CPU_LOONGARCH64] = {
       .l1d_line_size = 64,
       .l1d_size = 64,
@@ -74,6 +85,10 @@ loongarch_cpu_cache[N_TUNE_TYPES] = {
 
 struct loongarch_align
 loongarch_cpu_align[N_TUNE_TYPES] = {
+  [CPU_LOONGARCH32] = {
+    .function = "32",
+    .label = "16",
+  },
   [CPU_LOONGARCH64] = {
     .function = "32",
     .label = "16",
@@ -94,6 +109,9 @@ loongarch_cpu_rtx_cost_data[N_TUNE_TYPES] = {
   [CPU_NATIVE] = {
       DEFAULT_COSTS
   },
+  [CPU_LOONGARCH32] = {
+      DEFAULT_COSTS
+  },
   [CPU_LOONGARCH64] = {
       DEFAULT_COSTS
   },
@@ -121,6 +139,7 @@ loongarch_rtx_cost_optimize_size = {
 int
 loongarch_cpu_issue_rate[N_TUNE_TYPES] = {
   [CPU_NATIVE]	      = 4,
+  [CPU_LOONGARCH32]   = 4,
   [CPU_LOONGARCH64]   = 4,
   [CPU_LA464]	      = 4,
 };
@@ -128,6 +147,7 @@ loongarch_cpu_issue_rate[N_TUNE_TYPES] = {
 int
 loongarch_cpu_multipass_dfa_lookahead[N_TUNE_TYPES] = {
   [CPU_NATIVE]	      = 4,
+  [CPU_LOONGARCH32]   = 4,
   [CPU_LOONGARCH64]   = 4,
   [CPU_LA464]	      = 4,
 };
@@ -139,6 +159,7 @@ loongarch_cpu_multipass_dfa_lookahead[N_TUNE_TYPES] = {
 
 const char*
 loongarch_isa_base_strings[N_ISA_BASE_TYPES] = {
+  [ISA_BASE_LA32V100] = STR_ISA_BASE_LA32V100,
   [ISA_BASE_LA64V100] = STR_ISA_BASE_LA64V100,
 };
 
@@ -151,6 +172,9 @@ loongarch_isa_ext_strings[N_ISA_EXT_TYPES] = {
 
 const char*
 loongarch_abi_base_strings[N_ABI_BASE_TYPES] = {
+  [ABI_BASE_ILP32D] = STR_ABI_BASE_ILP32D,
+  [ABI_BASE_ILP32F] = STR_ABI_BASE_ILP32F,
+  [ABI_BASE_ILP32S] = STR_ABI_BASE_ILP32S,
   [ABI_BASE_LP64D] = STR_ABI_BASE_LP64D,
   [ABI_BASE_LP64F] = STR_ABI_BASE_LP64F,
   [ABI_BASE_LP64S] = STR_ABI_BASE_LP64S,
@@ -182,6 +206,15 @@ loongarch_switch_strings[] = {
 /* ABI-related definitions.  */
 const struct loongarch_isa
 abi_minimal_isa[N_ABI_BASE_TYPES][N_ABI_EXT_TYPES] = {
+  [ABI_BASE_ILP32D] = {
+      [ABI_EXT_BASE] = {.base = ISA_BASE_LA32V100, .fpu = ISA_EXT_FPU64},
+  },
+  [ABI_BASE_ILP32F] = {
+      [ABI_EXT_BASE] = {.base = ISA_BASE_LA32V100, .fpu = ISA_EXT_FPU32},
+  },
+  [ABI_BASE_ILP32S] = {
+      [ABI_EXT_BASE] = {.base = ISA_BASE_LA32V100, .fpu = ISA_EXT_NOFPU},
+  },
   [ABI_BASE_LP64D] = {
       [ABI_EXT_BASE] = {.base = ISA_BASE_LA64V100, .fpu = ISA_EXT_FPU64},
   },
diff --git a/gcc/config/loongarch/loongarch-def.h b/gcc/config/loongarch/loongarch-def.h
index fb8bb88eb52..6c4b9fa3385 100644
--- a/gcc/config/loongarch/loongarch-def.h
+++ b/gcc/config/loongarch/loongarch-def.h
@@ -54,8 +54,9 @@ extern "C" {
 
 /* enum isa_base */
 extern const char* loongarch_isa_base_strings[];
-#define ISA_BASE_LA64V100     0
-#define N_ISA_BASE_TYPES      1
+#define ISA_BASE_LA32V100     0
+#define ISA_BASE_LA64V100     1
+#define N_ISA_BASE_TYPES      2
 
 /* enum isa_ext_* */
 extern const char* loongarch_isa_ext_strings[];
@@ -67,10 +68,13 @@ extern const char* loongarch_isa_ext_strings[];
 
 /* 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
+#define ABI_BASE_ILP32D	      0
+#define ABI_BASE_ILP32F	      1
+#define ABI_BASE_ILP32S	      2
+#define ABI_BASE_LP64D	      3
+#define ABI_BASE_LP64F	      4
+#define ABI_BASE_LP64S	      5
+#define N_ABI_BASE_TYPES      6
 
 /* enum abi_ext */
 extern const char* loongarch_abi_ext_strings[];
@@ -132,10 +136,11 @@ struct loongarch_target
 /* 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
+#define CPU_LOONGARCH32	  1
+#define CPU_LOONGARCH64	  2
+#define CPU_LA464	  3
+#define N_ARCH_TYPES	  4
+#define N_TUNE_TYPES	  4
 
 /* parallel tables.  */
 extern const char* loongarch_cpu_strings[];
diff --git a/gcc/config/loongarch/loongarch-driver.h b/gcc/config/loongarch/loongarch-driver.h
index ba8817a4621..0a22acab803 100644
--- a/gcc/config/loongarch/loongarch-driver.h
+++ b/gcc/config/loongarch/loongarch-driver.h
@@ -58,9 +58,13 @@ driver_get_normalized_m_opts (int argc, const char **argv);
 
 /* ABI spec strings.  */
 #define ABI_GRLEN_SPEC \
+  "%{mabi=ilp32*:32}"   \
   "%{mabi=lp64*:64}"   \
 
 #define ABI_SPEC \
+  "%{mabi=ilp32d:ilp32d}" \
+  "%{mabi=ilp32f:ilp32f}" \
+  "%{mabi=ilp32s:ilp32s}" \
   "%{mabi=lp64d:lp64d}" \
   "%{mabi=lp64f:lp64f}" \
   "%{mabi=lp64s:lp64s}" \
diff --git a/gcc/config/loongarch/loongarch-opts.cc b/gcc/config/loongarch/loongarch-opts.cc
index a52e25236ea..9fc0bbbcb6e 100644
--- a/gcc/config/loongarch/loongarch-opts.cc
+++ b/gcc/config/loongarch/loongarch-opts.cc
@@ -36,6 +36,9 @@ struct loongarch_target la_target;
 #define ABI_COUNT (sizeof(abi_priority_list)/sizeof(struct loongarch_abi))
 static const struct loongarch_abi
 abi_priority_list[] = {
+    {ABI_BASE_ILP32D, ABI_EXT_BASE},
+    {ABI_BASE_ILP32F, ABI_EXT_BASE},
+    {ABI_BASE_ILP32S, ABI_EXT_BASE},
     {ABI_BASE_LP64D, ABI_EXT_BASE},
     {ABI_BASE_LP64F, ABI_EXT_BASE},
     {ABI_BASE_LP64S, ABI_EXT_BASE},
@@ -410,17 +413,23 @@ isa_default_abi (const struct loongarch_isa *isa)
   switch (isa->fpu)
     {
       case ISA_EXT_FPU64:
-	if (isa->base == ISA_BASE_LA64V100)
+	if (isa->base == ISA_BASE_LA32V100)
+	  abi.base = ABI_BASE_ILP32D;
+	else if (isa->base == ISA_BASE_LA64V100)
 	  abi.base = ABI_BASE_LP64D;
 	break;
 
       case ISA_EXT_FPU32:
-	if (isa->base == ISA_BASE_LA64V100)
+	if (isa->base == ISA_BASE_LA32V100)
+	  abi.base = ABI_BASE_ILP32F;
+	else if (isa->base == ISA_BASE_LA64V100)
 	  abi.base = ABI_BASE_LP64F;
 	break;
 
       case ISA_EXT_NOFPU:
-	if (isa->base == ISA_BASE_LA64V100)
+	if (isa->base == ISA_BASE_LA32V100)
+	  abi.base = ABI_BASE_ILP32S;
+	else if (isa->base == ISA_BASE_LA64V100)
 	  abi.base = ABI_BASE_LP64S;
 	break;
 
@@ -439,6 +448,8 @@ isa_base_compat_p (const struct loongarch_isa *set1,
 {
   switch (set2->base)
     {
+      case ISA_BASE_LA32V100:
+	return (set1->base == ISA_BASE_LA32V100);
       case ISA_BASE_LA64V100:
 	return (set1->base == ISA_BASE_LA64V100);
 
@@ -487,6 +498,11 @@ abi_default_cpu_arch (struct loongarch_abi abi)
 {
   switch (abi.base)
     {
+      case ABI_BASE_ILP32D:
+      case ABI_BASE_ILP32F:
+      case ABI_BASE_ILP32S:
+	if (abi.ext == ABI_EXT_BASE)
+	  return CPU_LOONGARCH32;
       case ABI_BASE_LP64D:
       case ABI_BASE_LP64F:
       case ABI_BASE_LP64S:
diff --git a/gcc/config/loongarch/loongarch-opts.h b/gcc/config/loongarch/loongarch-opts.h
index b1ff54426e4..c8794dff0ad 100644
--- a/gcc/config/loongarch/loongarch-opts.h
+++ b/gcc/config/loongarch/loongarch-opts.h
@@ -51,15 +51,25 @@ loongarch_config_target (struct loongarch_target *target,
 #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 \
+#define TARGET_HARD_FLOAT_ABI	    (la_target.abi.base == ABI_BASE_ILP32D \
+				     || la_target.abi.base == ABI_BASE_ILP32F \
+				     || 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_SOFT_FLOAT_ABI	  (la_target.abi.base == ABI_BASE_ILP32S \
+				     || 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_SINGLE_FLOAT_ABI	  (la_target.abi.base == ABI_BASE_ILP32F \
+				     || 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_DOUBLE_FLOAT_ABI	  (la_target.abi.base == ABI_BASE_ILP32D \
+				     || la_target.abi.base == ABI_BASE_LP64D)
+
+#define TARGET_32BIT		  (la_target.isa.base == ISA_BASE_LA32V100)
+#define TARGET_ABI_ILP32	  (la_target.abi.base == ABI_BASE_ILP32D \
+				   || la_target.abi.base == ABI_BASE_ILP32F \
+				   || la_target.abi.base == ABI_BASE_ILP32S)
 
 #define TARGET_64BIT		  (la_target.isa.base == ISA_BASE_LA64V100)
 #define TARGET_ABI_LP64		  (la_target.abi.base == ABI_BASE_LP64D	\
@@ -78,9 +88,11 @@ loongarch_config_target (struct loongarch_target *target,
 				    ? (la_target.cpu_native) : (CPU_NATIVE)) \
 				    : (la_target.cpu_tune))
 
+#define TARGET_ARCH_LOONGARCH32	  (LARCH_ACTUAL_ARCH == CPU_LOONGARCH32)
 #define TARGET_ARCH_LOONGARCH64	  (LARCH_ACTUAL_ARCH == CPU_LOONGARCH64)
 #define TARGET_ARCH_LA464	  (LARCH_ACTUAL_ARCH == CPU_LA464)
 
+#define TARGET_TUNE_LOONGARCH32	  (LARCH_ACTUAL_TUNE == CPU_LOONGARCH32)
 #define TARGET_TUNE_LOONGARCH64	  (LARCH_ACTUAL_TUNE == CPU_LOONGARCH64)
 #define TARGET_TUNE_LA464	  (LARCH_ACTUAL_TUNE == CPU_LA464)
 
diff --git a/gcc/config/loongarch/loongarch-str.h b/gcc/config/loongarch/loongarch-str.h
index af2e82a321f..60588da45e2 100644
--- a/gcc/config/loongarch/loongarch-str.h
+++ b/gcc/config/loongarch/loongarch-str.h
@@ -27,9 +27,11 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTSTR_TUNE "tune"
 
 #define STR_CPU_NATIVE "native"
+#define STR_CPU_LOONGARCH32 "loongarch32"
 #define STR_CPU_LOONGARCH64 "loongarch64"
 #define STR_CPU_LA464 "la464"
 
+#define STR_ISA_BASE_LA32V100 "la32"
 #define STR_ISA_BASE_LA64V100 "la64"
 
 #define OPTSTR_ISA_EXT_FPU "fpu"
@@ -43,6 +45,9 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTSTR_DOUBLE_FLOAT "double-float"
 
 #define OPTSTR_ABI_BASE "abi"
+#define STR_ABI_BASE_ILP32D "ilp32d"
+#define STR_ABI_BASE_ILP32F "ilp32f"
+#define STR_ABI_BASE_ILP32S "ilp32s"
 #define STR_ABI_BASE_LP64D "lp64d"
 #define STR_ABI_BASE_LP64F "lp64f"
 #define STR_ABI_BASE_LP64S "lp64s"
diff --git a/gcc/config/loongarch/loongarch.cc b/gcc/config/loongarch/loongarch.cc
index 86d58784113..c980de98758 100644
--- a/gcc/config/loongarch/loongarch.cc
+++ b/gcc/config/loongarch/loongarch.cc
@@ -6209,7 +6209,7 @@ loongarch_option_override_internal (struct gcc_options *opts)
 			   la_opt_cpu_arch, la_opt_cpu_tune, la_opt_fpu,
 			   la_opt_abi_base, la_opt_abi_ext, la_opt_cmodel, 0);
 
-  if (TARGET_ABI_LP64)
+  if (TARGET_ABI_LP64 || TARGET_ABI_ILP32)
     flag_pcc_struct_return = 0;
 
   /* Decide which rtx_costs structure to use.  */
diff --git a/gcc/config/loongarch/loongarch.opt b/gcc/config/loongarch/loongarch.opt
index 68018ade73f..2cd36c5e8fa 100644
--- a/gcc/config/loongarch/loongarch.opt
+++ b/gcc/config/loongarch/loongarch.opt
@@ -92,6 +92,9 @@ LoongArch CPU types:
 EnumValue
 Enum(cpu_type) String(native) Value(CPU_NATIVE)
 
+EnumValue
+Enum(cpu_type) String(loongarch32) Value(CPU_LOONGARCH32)
+
 EnumValue
 Enum(cpu_type) String(loongarch64) Value(CPU_LOONGARCH64)
 
@@ -115,6 +118,15 @@ Enum
 Name(abi_base) Type(int)
 Base ABI types for LoongArch:
 
+EnumValue
+Enum(abi_base) String(ilp32d) Value(ABI_BASE_ILP32D)
+
+EnumValue
+Enum(abi_base) String(ilp32f) Value(ABI_BASE_ILP32F)
+
+EnumValue
+Enum(abi_base) String(ilp32s) Value(ABI_BASE_ILP32S)
+
 EnumValue
 Enum(abi_base) String(lp64d) Value(ABI_BASE_LP64D)
 
diff --git a/gcc/config/loongarch/t-linux b/gcc/config/loongarch/t-linux
index e40da179203..1718c3e11a2 100644
--- a/gcc/config/loongarch/t-linux
+++ b/gcc/config/loongarch/t-linux
@@ -17,8 +17,8 @@
 # <http://www.gnu.org/licenses/>.
 
 # Multilib
-MULTILIB_OPTIONS = mabi=lp64d/mabi=lp64f/mabi=lp64s
-MULTILIB_DIRNAMES = base/lp64d base/lp64f base/lp64s
+MULTILIB_OPTIONS = mabi=ilp32d/mabi=ilp32f/mabi=ilp32s/mabi=lp64d/mabi=lp64f/mabi=lp64s
+MULTILIB_DIRNAMES = base/ilp32d base/ilp32f base/ilp32s 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)
@@ -38,6 +38,18 @@ endif
 # Don't define MULTILIB_OSDIRNAMES if multilib is disabled.
 ifeq ($(filter LA_DISABLE_MULTILIB,$(tm_defines)),)
 
+    MULTILIB_OSDIRNAMES = \
+      mabi.ilp32d=../lib32$\
+      $(call if_multiarch,:loongarch32-linux-gnu)
+
+    MULTILIB_OSDIRNAMES += \
+      mabi.ilp32f=../lib32/f32$\
+      $(call if_multiarch,:loongarch32-linux-gnuf32)
+
+    MULTILIB_OSDIRNAMES += \
+      mabi.ilp32s=../lib32/sf$\
+      $(call if_multiarch,:loongarch32-linux-gnusf)
+
     MULTILIB_OSDIRNAMES = \
       mabi.lp64d=../lib64$\
       $(call if_multiarch,:loongarch64-linux-gnu)
-- 
2.41.0

[PATCH v2 02/14] LoongArch: Fix default ISA setting

[Jiajie Chen]

When loongarch_arch_target is called, la_target has not been
initialized, thus the macro LARCH_ACTUAL_ARCH always equals to zero.

This commit fixes by expanding the macro and reading the latest value.
It permits -march=loongarch64 when the default target is loongarch32 and
vice versa.

gcc/ChangeLog:

	* config/loongarch/loongarch-opts.cc (loongarch_config_target):
	Fix -march detection.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/arch-1.c: New test.
	* gcc.target/loongarch/arch-2.c: New test.
---
 gcc/config/loongarch/loongarch-opts.cc      | 5 ++++-
 gcc/testsuite/gcc.target/loongarch/arch-1.c | 5 +++++
 gcc/testsuite/gcc.target/loongarch/arch-2.c | 5 +++++
 3 files changed, 14 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/arch-1.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/arch-2.c

diff --git a/gcc/config/loongarch/loongarch-opts.cc b/gcc/config/loongarch/loongarch-opts.cc
index 9fc0bbbcb6e..29c0c4468bb 100644
--- a/gcc/config/loongarch/loongarch-opts.cc
+++ b/gcc/config/loongarch/loongarch-opts.cc
@@ -246,7 +246,10 @@ loongarch_config_target (struct loongarch_target *target,
 config_target_isa:
 
   /* Get default ISA from "-march" or its default value.  */
-  t.isa = loongarch_cpu_default_isa[LARCH_ACTUAL_ARCH];
+  if (t.cpu_arch == TARGET_ARCH_NATIVE)
+    t.isa = loongarch_cpu_default_isa[t.cpu_native];
+  else
+    t.isa = loongarch_cpu_default_isa[t.cpu_arch];
 
   /* Apply incremental changes.  */
   /* "-march=native" overrides the default FPU type.  */
diff --git a/gcc/testsuite/gcc.target/loongarch/arch-1.c b/gcc/testsuite/gcc.target/loongarch/arch-1.c
new file mode 100644
index 00000000000..379036ec76f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/arch-1.c
@@ -0,0 +1,5 @@
+/* { dg-do compile } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d" } */
+int foo()
+{
+}
diff --git a/gcc/testsuite/gcc.target/loongarch/arch-2.c b/gcc/testsuite/gcc.target/loongarch/arch-2.c
new file mode 100644
index 00000000000..55d646902a6
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/arch-2.c
@@ -0,0 +1,5 @@
+/* { dg-do compile } */
+/* { dg-options "-march=loongarch32 -mabi=ilp32d" } */
+int foo()
+{
+}
-- 
2.41.0

[PATCH v2 03/14] LoongArch: Fix SI division for loongarch32 target

[Jiajie Chen]

Add TARGET_64BIT check for loongarch64-only handling of SI division. It
shall not promote SI to DI before division in loongarch32 target.

gcc/ChangeLog:

	* config/loongarch/loongarch.md: Add TARGET_64BIT check for
	loongarch64-only handling of SI division.
---
 gcc/config/loongarch/loongarch.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index b37e070660f..95c5b25d22a 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -851,7 +851,7 @@
 		     (match_operand:GPR 2 "register_operand")))]
   ""
 {
- if (GET_MODE (operands[0]) == SImode)
+ if (GET_MODE (operands[0]) == SImode && TARGET_64BIT)
   {
     rtx reg1 = gen_reg_rtx (DImode);
     rtx reg2 = gen_reg_rtx (DImode);
-- 
2.41.0

[PATCH v2 04/14] LoongArch: Fix movgr2frh.w operand order

[Jiajie Chen]

The operand order of movgr2frh.w was wrong. The correct order should be
`movgr2frh.w fd, rj`.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (movgr2frh<mode>): Correct
	movgr2frh.w operand order.
---
 gcc/config/loongarch/loongarch.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index 95c5b25d22a..93d8bf5bcca 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -2297,7 +2297,7 @@
 			(match_operand:SPLITF 2 "register_operand" "0")]
 			UNSPEC_MOVGR2FRH))]
   "TARGET_DOUBLE_FLOAT"
-  "movgr2frh.w\t%z1,%0"
+  "movgr2frh.w\t%0,%z1"
   [(set_attr "move_type" "mgtf")
    (set_attr "mode" "<HALFMODE>")])
 
-- 
2.41.0

[PATCH v2 05/14] LoongArch: Fix 64-bit move for loongarch32 target

[Jiajie Chen]

Bring back 64-bit move splitting for loongarch32. The code was removed
in commit 16fc26d4e7a (`LoongArch: Support split symbol.`) for unknown
reason.

gcc/ChangeLog:

	* config/loongarch/loongarch.md: Handle move splitting for
	64-bit operands.
---
 gcc/config/loongarch/loongarch.md | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index 93d8bf5bcca..9eb6bb75c35 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -1965,6 +1965,16 @@
   [(set_attr "move_type" "move,load,store")
    (set_attr "mode" "DF")])
 
+(define_split
+  [(set (match_operand:MOVE64 0 "nonimmediate_operand")
+       (match_operand:MOVE64 1 "move_operand"))]
+  "reload_completed && loongarch_split_move_p (operands[0], operands[1])"
+  [(const_int 0)]
+{
+  loongarch_split_move (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;
-- 
2.41.0

[PATCH v2 06/14] LoongArch: Fix 64-bit immediate move for loongarch32 target

[Jiajie Chen]

loongarch_move_integer does not support splitting 64-bit integer into
two 32-bit ones. Thus, define_split is removed from movdi_32bit and
TARGET_64BIT is added to the split condition of movdi_64bit to avoid
using it for loongarch32.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (movdi_32bit): Remove not
	working split, use existing loongarch_split_move instead.
	(movdi_64bit): Add TARGET_64BIT to split condition.
---
 gcc/config/loongarch/loongarch.md | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index 9eb6bb75c35..c611a8a822a 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -1777,22 +1777,13 @@
     DONE;
 })
 
-(define_insn_and_split "*movdi_32bit"
+(define_insn "*movdi_32bit"
   [(set (match_operand:DI 0 "nonimmediate_operand" "=r,r,r,w,*f,*f,*r,*m")
        (match_operand:DI 1 "move_operand" "r,i,w,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]); }
-  "CONST_INT_P (operands[1]) && REG_P (operands[0]) && GP_REG_P (REGNO
-  (operands[0]))"
-  [(const_int 0)]
-  "
-{
-  loongarch_move_integer (operands[0], operands[0], INTVAL (operands[1]));
-  DONE;
-}
-  "
   [(set_attr "move_type" "move,const,load,store,mgtf,fpload,mftg,fpstore")
    (set_attr "mode" "DI")])
 
@@ -1804,7 +1795,7 @@
        || reg_or_0_operand (operands[1], DImode))"
   { return loongarch_output_move (operands[0], operands[1]); }
   "CONST_INT_P (operands[1]) && REG_P (operands[0]) && GP_REG_P (REGNO
-  (operands[0]))"
+  (operands[0])) && TARGET_64BIT"
   [(const_int 0)]
   "
 {
-- 
2.41.0

[PATCH v2 07/14] LoongArch: Fix signed 32-bit overflow for loongarch32 target

[Jiajie Chen]

When rhs equals to 0x7fffffff, adding 1 to rhs overflows SI, generating
invalid const_int.

gcc/ChangeLog:

	* config/loongarch/loongarch.cc (loongarch_emit_int_compare):
	Call trunc_int_mode to ensure valid rhs.
---
 gcc/config/loongarch/loongarch.cc | 1 +
 1 file changed, 1 insertion(+)

diff --git a/gcc/config/loongarch/loongarch.cc b/gcc/config/loongarch/loongarch.cc
index c980de98758..49df9509ba9 100644
--- a/gcc/config/loongarch/loongarch.cc
+++ b/gcc/config/loongarch/loongarch.cc
@@ -4284,6 +4284,7 @@ loongarch_emit_int_compare (enum rtx_code *code, rtx *op0, rtx *op1)
 		break;
 
 	      new_rhs = rhs + (increment ? 1 : -1);
+	      new_rhs = trunc_int_for_mode (new_rhs, GET_MODE (*op0));
 	      if (loongarch_integer_cost (new_rhs)
 		    < loongarch_integer_cost (rhs))
 		{
-- 
2.41.0

[PATCH v2 08/14] LoongArch: Disable SF/DF -> unsigned DI expand in loongarch32

[Jiajie Chen]

The current SF/DF -> unsigned DI expand rules require iordi3 insn which
is not available in loongarch32.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (fixuns_truncdfdi2): Add
	TARGET_64BIT to condition.
	(fixuns_truncsfdi2): Add TARGET_64BIT to condition.
---
 gcc/config/loongarch/loongarch.md | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index c611a8a822a..31bdf3388f6 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -1576,7 +1576,7 @@
 (define_expand "fixuns_truncdfdi2"
   [(set (match_operand:DI 0 "register_operand")
 	(unsigned_fix:DI (match_operand:DF 1 "register_operand")))]
-  "TARGET_DOUBLE_FLOAT"
+  "TARGET_DOUBLE_FLOAT && TARGET_64BIT"
 {
   rtx reg1 = gen_reg_rtx (DFmode);
   rtx reg2 = gen_reg_rtx (DFmode);
@@ -1658,7 +1658,7 @@
 (define_expand "fixuns_truncsfdi2"
   [(set (match_operand:DI 0 "register_operand")
 	(unsigned_fix:DI (match_operand:SF 1 "register_operand")))]
-  "TARGET_DOUBLE_FLOAT"
+  "TARGET_DOUBLE_FLOAT && TARGET_64BIT"
 {
   rtx reg1 = gen_reg_rtx (SFmode);
   rtx reg2 = gen_reg_rtx (SFmode);
-- 
2.41.0

[PATCH v2 09/14] LoongArch: Add -march=loongarch64 to tests with -mabi=lp64d

[Jiajie Chen]

The compiler emits a warning if the current target (-march=loongarch32)
mismatches with abi(-march=lp64d). Adding: Add -march=loongarch64
explicitly fixes the tests.

gcc/testsuite/ChangeLog:

	* g++.target/loongarch/bytepick.C: Add -march=loongarch64
	* g++.target/loongarch/pr106828.C: Add -march=loongarch64
	* gcc.target/loongarch/add-const.c: Add -march=loongarch64
	* gcc.target/loongarch/arch-1.c: Add -march=loongarch64
	* gcc.target/loongarch/attr-model-1.c: Add -march=loongarch64
	* gcc.target/loongarch/attr-model-2.c: Add -march=loongarch64
	* gcc.target/loongarch/flt-abi-isa-1.c: Add -march=loongarch64
	* gcc.target/loongarch/fscaleb.c: Add -march=loongarch64
	* gcc.target/loongarch/ftint-no-inexact.c: Add
	-march=loongarch64
	* gcc.target/loongarch/ftint.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-1.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-2.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-3.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-4.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-5.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-6.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-7.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-8.c: Add -march=loongarch64
	* gcc.target/loongarch/func-call-extreme-1.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-extreme-2.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-1.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-2.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-3.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-4.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-5.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-6.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-7.c: Add
	-march=loongarch64
	* gcc.target/loongarch/func-call-medium-8.c: Add
	-march=loongarch64
	* gcc.target/loongarch/imm-load.c: Add -march=loongarch64
	* gcc.target/loongarch/imm-load1.c: Add -march=loongarch64
	* gcc.target/loongarch/mulw_d_w.c: Add -march=loongarch64
	* gcc.target/loongarch/pr109465-1.c: Add -march=loongarch64
	* gcc.target/loongarch/pr109465-2.c: Add -march=loongarch64
	* gcc.target/loongarch/pr109465-3.c: Add -march=loongarch64
	* gcc.target/loongarch/prolog-opt.c: Add -march=loongarch64
	* gcc.target/loongarch/relocs-symbol-noaddend.c: Add
	-march=loongarch64
	* gcc.target/loongarch/zero-size-field-pass.c: Add
	-march=loongarch64
	* gcc.target/loongarch/zero-size-field-ret.c: Add
	-march=loongarch64
---
 gcc/testsuite/g++.target/loongarch/bytepick.C               | 2 +-
 gcc/testsuite/g++.target/loongarch/pr106828.C               | 2 +-
 gcc/testsuite/gcc.target/loongarch/add-const.c              | 2 +-
 gcc/testsuite/gcc.target/loongarch/attr-model-1.c           | 2 +-
 gcc/testsuite/gcc.target/loongarch/attr-model-2.c           | 2 +-
 gcc/testsuite/gcc.target/loongarch/flt-abi-isa-1.c          | 2 +-
 gcc/testsuite/gcc.target/loongarch/fscaleb.c                | 2 +-
 gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c       | 2 +-
 gcc/testsuite/gcc.target/loongarch/ftint.c                  | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-1.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-2.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-3.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-4.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-5.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-6.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-7.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-8.c            | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-extreme-1.c    | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-extreme-2.c    | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-1.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-2.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-3.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-4.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-5.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-6.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-7.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/func-call-medium-8.c     | 2 +-
 gcc/testsuite/gcc.target/loongarch/imm-load.c               | 2 +-
 gcc/testsuite/gcc.target/loongarch/imm-load1.c              | 2 +-
 gcc/testsuite/gcc.target/loongarch/mulw_d_w.c               | 2 +-
 gcc/testsuite/gcc.target/loongarch/pr109465-1.c             | 2 +-
 gcc/testsuite/gcc.target/loongarch/pr109465-2.c             | 2 +-
 gcc/testsuite/gcc.target/loongarch/pr109465-3.c             | 2 +-
 gcc/testsuite/gcc.target/loongarch/prolog-opt.c             | 2 +-
 gcc/testsuite/gcc.target/loongarch/relocs-symbol-noaddend.c | 2 +-
 gcc/testsuite/gcc.target/loongarch/zero-size-field-pass.c   | 2 +-
 gcc/testsuite/gcc.target/loongarch/zero-size-field-ret.c    | 2 +-
 37 files changed, 37 insertions(+), 37 deletions(-)

diff --git a/gcc/testsuite/g++.target/loongarch/bytepick.C b/gcc/testsuite/g++.target/loongarch/bytepick.C
index a39e2fa65b7..c2b0dd5e513 100644
--- a/gcc/testsuite/g++.target/loongarch/bytepick.C
+++ b/gcc/testsuite/g++.target/loongarch/bytepick.C
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d" } */
 /* { dg-final { scan-assembler-times "bytepick.w\t\\\$r4,\\\$r5,\\\$r4" 3 } } */
 /* { dg-final { scan-assembler-times "bytepick.d\t\\\$r4,\\\$r5,\\\$r4" 7 } } */
 /* { dg-final { scan-assembler-not "slli.w" } } */
diff --git a/gcc/testsuite/g++.target/loongarch/pr106828.C b/gcc/testsuite/g++.target/loongarch/pr106828.C
index 190c1db715f..74590891a67 100644
--- a/gcc/testsuite/g++.target/loongarch/pr106828.C
+++ b/gcc/testsuite/g++.target/loongarch/pr106828.C
@@ -1,4 +1,4 @@
 /* { dg-do-preprocess } */
-/* { dg-options "-mabi=lp64d -fsanitize=address" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -fsanitize=address" } */
 
 /* Tests whether the compiler supports compile option '-fsanitize=address'.  */
diff --git a/gcc/testsuite/gcc.target/loongarch/add-const.c b/gcc/testsuite/gcc.target/loongarch/add-const.c
index 7b6a7cb92aa..04286afad49 100644
--- a/gcc/testsuite/gcc.target/loongarch/add-const.c
+++ b/gcc/testsuite/gcc.target/loongarch/add-const.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O -mabi=lp64d" } */
+/* { dg-options "-O -march=loongarch64 -mabi=lp64d" } */
 
 /* None of these functions should load the const operand into a temp
    register.  */
diff --git a/gcc/testsuite/gcc.target/loongarch/attr-model-1.c b/gcc/testsuite/gcc.target/loongarch/attr-model-1.c
index 916d715b98b..005c7cd727e 100644
--- a/gcc/testsuite/gcc.target/loongarch/attr-model-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/attr-model-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mexplicit-relocs -mcmodel=normal -O2" } */
+/* { dg-options "-mexplicit-relocs -mcmodel=normal -O2 -march=loongarch64" } */
 /* { dg-final { scan-assembler-times "%pc64_hi12" 2 } } */
 
 #define ATTR_MODEL_TEST
diff --git a/gcc/testsuite/gcc.target/loongarch/attr-model-2.c b/gcc/testsuite/gcc.target/loongarch/attr-model-2.c
index a74c795ac3e..242c40161e9 100644
--- a/gcc/testsuite/gcc.target/loongarch/attr-model-2.c
+++ b/gcc/testsuite/gcc.target/loongarch/attr-model-2.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mexplicit-relocs -mcmodel=extreme -O2" } */
+/* { dg-options "-mexplicit-relocs -mcmodel=extreme -O2 -march=loongarch64" } */
 /* { dg-final { scan-assembler-times "%pc64_hi12" 3 } } */
 
 #define ATTR_MODEL_TEST
diff --git a/gcc/testsuite/gcc.target/loongarch/flt-abi-isa-1.c b/gcc/testsuite/gcc.target/loongarch/flt-abi-isa-1.c
index 1c9490f6a87..9cc2d439bf1 100644
--- a/gcc/testsuite/gcc.target/loongarch/flt-abi-isa-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/flt-abi-isa-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -mfpu=64 -march=loongarch64 -O2" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -mfpu=64 -march=loongarch64 -O2" } */
 /* { dg-final { scan-assembler "frecip\\.d" } } */
 /* { dg-final { scan-assembler-not "movgr2fr\\.d" } } */
 /* { dg-final { scan-assembler-not "movfr2gr\\.d" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/fscaleb.c b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
index f18470fbb8f..d39cdc63ae8 100644
--- a/gcc/testsuite/gcc.target/loongarch/fscaleb.c
+++ b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d -mdouble-float -fno-math-errno" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d -mdouble-float -fno-math-errno" } */
 /* { dg-final { scan-assembler-times "fscaleb\\.s" 3 } } */
 /* { dg-final { scan-assembler-times "fscaleb\\.d" 4 } } */
 /* { dg-final { scan-assembler-times "slli\\.w" 1 } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c b/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
index 88b83a9c056..e98d26070bc 100644
--- a/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
+++ b/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -mdouble-float -fno-math-errno -fno-fp-int-builtin-inexact" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -mdouble-float -fno-math-errno -fno-fp-int-builtin-inexact" } */
 /* { dg-final { scan-assembler "ftint\\.l\\.s" } } */
 /* { dg-final { scan-assembler "ftint\\.l\\.d" } } */
 /* { dg-final { scan-assembler-not "ftintrm\\.l\\.s" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/ftint.c b/gcc/testsuite/gcc.target/loongarch/ftint.c
index 7a326a454d8..1002293aee9 100644
--- a/gcc/testsuite/gcc.target/loongarch/ftint.c
+++ b/gcc/testsuite/gcc.target/loongarch/ftint.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -mdouble-float -fno-math-errno -ffp-int-builtin-inexact" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -mdouble-float -fno-math-errno -ffp-int-builtin-inexact" } */
 /* { dg-final { scan-assembler "ftint\\.l\\.s" } } */
 /* { dg-final { scan-assembler "ftint\\.l\\.d" } } */
 /* { dg-final { scan-assembler "ftintrm\\.l\\.s" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-1.c b/gcc/testsuite/gcc.target/loongarch/func-call-1.c
index 76bf11b0c03..d25a71e6a4c 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fplt -mno-explicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fplt -mno-explicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*bl\t%plt\\(g\\)\n" } } */
 /* { dg-final { scan-assembler "test1:.*bl\t%plt\\(f\\)\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-2.c b/gcc/testsuite/gcc.target/loongarch/func-call-2.c
index 4b468fef8b4..ac1d3b0cec2 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-2.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-2.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fplt -mno-explicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fplt -mno-explicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*bl\t%plt\\(g\\)\n" } } */
 /* { dg-final { scan-assembler "test1:.*bl\tf\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-3.c b/gcc/testsuite/gcc.target/loongarch/func-call-3.c
index dd3a4882d60..e8156cb4eff 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-3.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-3.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fno-plt -mno-explicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fno-plt -mno-explicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*la\.global\t.*f\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-4.c b/gcc/testsuite/gcc.target/loongarch/func-call-4.c
index f8158ec349f..684475bc3bb 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-4.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-4.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fno-plt -mno-explicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fno-plt -mno-explicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*bl\tf\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-5.c b/gcc/testsuite/gcc.target/loongarch/func-call-5.c
index 37994af430d..0411d8b6421 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-5.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-5.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fplt -mexplicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fplt -mexplicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*bl\t%plt\\(g\\)\n" } } */
 /* { dg-final { scan-assembler "test1:.*bl\t%plt\\(f\\)\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-6.c b/gcc/testsuite/gcc.target/loongarch/func-call-6.c
index 8e366e376e7..beae892c8fd 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-6.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-6.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fplt -mexplicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fplt -mexplicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*bl\t%plt\\(g\\)\n" } } */
 /* { dg-final { scan-assembler "test1:.*bl\tf\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-7.c b/gcc/testsuite/gcc.target/loongarch/func-call-7.c
index 4177c3d962e..6f6d0456836 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-7.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-7.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i\t.*%got_pc_hi20\\(g\\)\n\tld\.d\t.*%got_pc_lo12\\(g\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i\t.*%got_pc_hi20\\(f\\)\n\tld\.d\t.*%got_pc_lo12\\(f\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-8.c b/gcc/testsuite/gcc.target/loongarch/func-call-8.c
index 4254eaa16d4..ebb68b368ed 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-8.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-8.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=normal" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i\t.*%got_pc_hi20\\(g\\)\n\tld\.d\t.*%got_pc_lo12\\(g\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*bl\tf\n" } } */
 /* { dg-final { scan-assembler "test2:.*bl\tl\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-extreme-1.c b/gcc/testsuite/gcc.target/loongarch/func-call-extreme-1.c
index db1e0f85396..5e83d3107ed 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-extreme-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-extreme-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=extreme" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=extreme" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i.*%got_pc_hi20.*\n\taddi\.d.*%got_pc_lo12.*\n\tlu32i\.d.*%got64_pc_lo20.*\n\tlu52i\.d.*%got64_pc_hi12.*\n\tldx\.d" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i.*%pc_hi20.*\n\taddi\.d.*%pc_lo12.*\n\tlu32i\.d.*%pc64_lo20.*\n\tlu52i\.d.*pc64_hi12.*\n\tadd\.d" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i.*%pc_hi20.*\n\taddi\.d.*%pc_lo12.*\n\tlu32i\.d.*%pc64_lo20.*\n\tlu52i\.d.*pc64_hi12.*\n\tadd\.d" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-extreme-2.c b/gcc/testsuite/gcc.target/loongarch/func-call-extreme-2.c
index 21bf81ae837..e09defff24b 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-extreme-2.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-extreme-2.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=extreme" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=extreme" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i.*%got_pc_hi20.*\n\taddi\.d.*%got_pc_lo12.*\n\tlu32i\.d.*%got64_pc_lo20.*\n\tlu52i\.d.*%got64_pc_hi12.*\n\tldx\.d" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i.*%got_pc_hi20.*\n\taddi\.d.*%got_pc_lo12.*\n\tlu32i\.d.*%got64_pc_lo20.*\n\tlu52i\.d.*%got64_pc_hi12.*\n\tldx\.d" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i.*%pc_hi20.*\n\taddi\.d.*%pc_lo12.*\n\tlu32i\.d.*%pc64_lo20.*\n\tlu52i\.d.*pc64_hi12.*\n\tadd\.d" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-1.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-1.c
index 6339e832fe5..e43b4b0fe40 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fplt -mno-explicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fplt -mno-explicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*la\.global\t.*f\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*la\.local\t.*l\n\tjirl" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-2.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-2.c
index a53e75e0bf9..17bfedd4d0d 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-2.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-2.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fplt -mno-explicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fplt -mno-explicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*la\.local\t.*f\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*la\.local\t.*l\n\tjirl" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-3.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-3.c
index 0da7bf98e3c..00439c6ab9e 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-3.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-3.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fno-plt -mno-explicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fno-plt -mno-explicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*la\.global\t.*f\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*la\.local\t.*l\n\tjirl" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-4.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-4.c
index 0219688ae80..ed23a4ebecc 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-4.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-4.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fno-plt -mno-explicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fno-plt -mno-explicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*la\.global\t.*g\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*la\.local\t.*f\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*la\.local\t.*l\n\tjirl" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-5.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-5.c
index 8a47b5afcba..4693517843e 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-5.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-5.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fplt -mexplicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fplt -mexplicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i.*%pc_hi20\\(g\\)\n\tjirl.*pc_lo12\\(g\\)" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i.*%pc_hi20\\(f\\)\n\tjirl.*%pc_lo12\\(f\\)" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i.*%pc_hi20\\(l\\)\n\tjirl.*%pc_lo12\\(l\\)" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-6.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-6.c
index 1e75e60e01a..ef8228ec361 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-6.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-6.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fplt -mexplicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fplt -mexplicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i.*%pc_hi20\\(g\\)\n\tjirl.*pc_lo12\\(g\\)" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i.*%pc_hi20\\(f\\)\n\tjirl.*%pc_lo12\\(f\\)" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i.*%pc_hi20\\(l\\)\n\tjirl.*%pc_lo12\\(l\\)" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-7.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-7.c
index 9e89085ca19..12f2ef37b0e 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-7.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-7.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fpic -fno-plt -mexplicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i\t.*%got_pc_hi20\\(g\\)\n\tld\.d\t.*%got_pc_lo12\\(g\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i\t.*%got_pc_hi20\\(f\\)\n\tld\.d\t.*%got_pc_lo12\\(f\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i\t.*%pc_hi20\\(l\\)\n\tjirl.*%pc_lo12\\(l\\)" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/func-call-medium-8.c b/gcc/testsuite/gcc.target/loongarch/func-call-medium-8.c
index fde9c6e0ef4..251f207dfe0 100644
--- a/gcc/testsuite/gcc.target/loongarch/func-call-medium-8.c
+++ b/gcc/testsuite/gcc.target/loongarch/func-call-medium-8.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=medium" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O0 -fno-pic -fno-plt -mexplicit-relocs -mcmodel=medium" } */
 /* { dg-final { scan-assembler "test:.*pcalau12i\t.*%got_pc_hi20\\(g\\)\n\tld\.d\t.*%got_pc_lo12\\(g\\)\n\tjirl" } } */
 /* { dg-final { scan-assembler "test1:.*pcalau12i\t.*%pc_hi20\\(f\\)\n\tjirl.*%pc_lo12\\(f\\)" } } */
 /* { dg-final { scan-assembler "test2:.*pcalau12i\t.*%pc_hi20\\(l\\)\n\tjirl.*%pc_lo12\\(l\\)" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/imm-load.c b/gcc/testsuite/gcc.target/loongarch/imm-load.c
index c04ca33996f..18102efacde 100644
--- a/gcc/testsuite/gcc.target/loongarch/imm-load.c
+++ b/gcc/testsuite/gcc.target/loongarch/imm-load.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O2 -fdump-rtl-split1" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O2 -fdump-rtl-split1" } */
 
 long int
 test (void)
diff --git a/gcc/testsuite/gcc.target/loongarch/imm-load1.c b/gcc/testsuite/gcc.target/loongarch/imm-load1.c
index 2ff02971239..4a0fe4fd30c 100644
--- a/gcc/testsuite/gcc.target/loongarch/imm-load1.c
+++ b/gcc/testsuite/gcc.target/loongarch/imm-load1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -O2" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -O2" } */
 /* { dg-final { scan-assembler "test:.*lu52i\.d.*\n\taddi\.w.*\n\.L2:" } } */
 
 
diff --git a/gcc/testsuite/gcc.target/loongarch/mulw_d_w.c b/gcc/testsuite/gcc.target/loongarch/mulw_d_w.c
index 4ab7df8836b..0abe7b014b9 100644
--- a/gcc/testsuite/gcc.target/loongarch/mulw_d_w.c
+++ b/gcc/testsuite/gcc.target/loongarch/mulw_d_w.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d" } */
 /* { dg-final { scan-assembler "mulw.d.w" } } */
 
 /* This should be optimized to mulw.d.w for LA64.  */
diff --git a/gcc/testsuite/gcc.target/loongarch/pr109465-1.c b/gcc/testsuite/gcc.target/loongarch/pr109465-1.c
index 4cd35d13904..1f36f4e0ab6 100644
--- a/gcc/testsuite/gcc.target/loongarch/pr109465-1.c
+++ b/gcc/testsuite/gcc.target/loongarch/pr109465-1.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d -mno-strict-align" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d -mno-strict-align" } */
 /* { dg-final { scan-assembler-times "st\\.d|stptr\\.d" 1 } } */
 /* { dg-final { scan-assembler-times "st\\.w|stptr\\.w" 1 } } */
 /* { dg-final { scan-assembler-times "st\\.h" 1 } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/pr109465-2.c b/gcc/testsuite/gcc.target/loongarch/pr109465-2.c
index 703eb951c6d..3e934567d0e 100644
--- a/gcc/testsuite/gcc.target/loongarch/pr109465-2.c
+++ b/gcc/testsuite/gcc.target/loongarch/pr109465-2.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d -mstrict-align" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d -mstrict-align" } */
 /* { dg-final { scan-assembler-times "st\\.d|stptr\\.d" 1 } } */
 /* { dg-final { scan-assembler-times "st\\.w|stptr\\.w" 1 } } */
 /* { dg-final { scan-assembler-times "st\\.h" 1 } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/pr109465-3.c b/gcc/testsuite/gcc.target/loongarch/pr109465-3.c
index d6a80659b31..7c684b2a5bf 100644
--- a/gcc/testsuite/gcc.target/loongarch/pr109465-3.c
+++ b/gcc/testsuite/gcc.target/loongarch/pr109465-3.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d -mstrict-align" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d -mstrict-align" } */
 
 /* Three loop iterations each contains 4 st.b, and 3 st.b after the loop */
 /* { dg-final { scan-assembler-times "st\\.b" 7 } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/prolog-opt.c b/gcc/testsuite/gcc.target/loongarch/prolog-opt.c
index e6a64263384..682d34ae750 100644
--- a/gcc/testsuite/gcc.target/loongarch/prolog-opt.c
+++ b/gcc/testsuite/gcc.target/loongarch/prolog-opt.c
@@ -1,7 +1,7 @@
 /* Test that LoongArch backend stack drop operation optimized.  */
 
 /* { dg-do compile } */
-/* { dg-options "-O2 -mabi=lp64d -fno-stack-protector" } */
+/* { dg-options "-O2 -march=loongarch64 -mabi=lp64d -fno-stack-protector" } */
 /* { dg-final { scan-assembler "addi.d\t\\\$r3,\\\$r3,-16" } } */
 
 extern int printf (char *, ...);
diff --git a/gcc/testsuite/gcc.target/loongarch/relocs-symbol-noaddend.c b/gcc/testsuite/gcc.target/loongarch/relocs-symbol-noaddend.c
index 3ec8bd229fd..7d018ba0e4b 100644
--- a/gcc/testsuite/gcc.target/loongarch/relocs-symbol-noaddend.c
+++ b/gcc/testsuite/gcc.target/loongarch/relocs-symbol-noaddend.c
@@ -1,5 +1,5 @@
 /* { dg-do compile } */
-/* { dg-options "-mabi=lp64d -mexplicit-relocs -fno-pic -O2 -mcmodel=normal" } */
+/* { dg-options "-march=loongarch64 -mabi=lp64d -mexplicit-relocs -fno-pic -O2 -mcmodel=normal" } */
 /* { dg-final { scan-assembler "pcalau12i.*%pc_hi20\\(\.LANCHOR0\\)\n" } } */
 /* { dg-final { scan-assembler "addi\.d.*%pc_lo12\\(\.LANCHOR0\\)\n" } } */
 /* { dg-final { scan-assembler "ldptr.d\t\\\$r4,.*,0\n" } } */
diff --git a/gcc/testsuite/gcc.target/loongarch/zero-size-field-pass.c b/gcc/testsuite/gcc.target/loongarch/zero-size-field-pass.c
index 999dc913a71..7931cc2623c 100644
--- a/gcc/testsuite/gcc.target/loongarch/zero-size-field-pass.c
+++ b/gcc/testsuite/gcc.target/loongarch/zero-size-field-pass.c
@@ -2,7 +2,7 @@
    argument passing.  */
 
 /* { dg-do compile } */
-/* { dg-options "-O2 -mdouble-float -mabi=lp64d" } */
+/* { dg-options "-O2 -mdouble-float -march=loongarch64 -mabi=lp64d" } */
 /* { dg-final { scan-assembler "\\\$f1" } } */
 
 struct test
diff --git a/gcc/testsuite/gcc.target/loongarch/zero-size-field-ret.c b/gcc/testsuite/gcc.target/loongarch/zero-size-field-ret.c
index 40137d97555..ef1cb6b35c7 100644
--- a/gcc/testsuite/gcc.target/loongarch/zero-size-field-ret.c
+++ b/gcc/testsuite/gcc.target/loongarch/zero-size-field-ret.c
@@ -2,7 +2,7 @@
    returning.  */
 
 /* { dg-do compile } */
-/* { dg-options "-O2 -mdouble-float -mabi=lp64d" } */
+/* { dg-options "-O2 -mdouble-float -march=loongarch64 -mabi=lp64d" } */
 /* { dg-final { scan-assembler-not "\\\$r4" } } */
 
 struct test
-- 
2.41.0

[PATCH v2 10/14] LoongArch: Forbid ADDRESS_REG_REG in loongarch32

[Jiajie Chen]

LoongArch32 does not include LDX/STD instructions, and cannot lower
(plus (reg) (reg)) pattern. Forbid ADDRESS_REG_REG and do not emit
ldx/stx.

gcc/ChangeLog:

	* config/loongarch/loongarch.cc (loongarch_valid_index_p): Check
	ADDRESS_REG_REG pattern and fail in loongarch32.
	(loongarch_output_move_index): assertion failed if generating
	ldx/stx in loongarch32.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/array-ldx.c: Add regression test for ldx
	in loongarch32.
---
 gcc/config/loongarch/loongarch.cc              | 4 +++-
 gcc/testsuite/gcc.target/loongarch/array-ldx.c | 6 ++++++
 2 files changed, 9 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/array-ldx.c

diff --git a/gcc/config/loongarch/loongarch.cc b/gcc/config/loongarch/loongarch.cc
index 49df9509ba9..1fde680ccd4 100644
--- a/gcc/config/loongarch/loongarch.cc
+++ b/gcc/config/loongarch/loongarch.cc
@@ -2016,7 +2016,8 @@ loongarch_valid_index_p (struct loongarch_address_info *info, rtx x,
       && contains_reg_of_mode[GENERAL_REGS][GET_MODE (SUBREG_REG (index))])
     index = SUBREG_REG (index);
 
-  if (loongarch_valid_base_register_p (index, mode, strict_p))
+  /* LA32 does not provide LDX/STX.  */
+  if (loongarch_valid_base_register_p (index, mode, strict_p) && !TARGET_32BIT)
     {
       info->type = ADDRESS_REG_REG;
       info->offset = index;
@@ -3853,6 +3854,7 @@ loongarch_output_move_index (rtx x, machine_mode mode, bool ldr)
       }
     };
 
+  gcc_assert (!TARGET_32BIT);
   return insn[ldr][index];
 }
 
diff --git a/gcc/testsuite/gcc.target/loongarch/array-ldx.c b/gcc/testsuite/gcc.target/loongarch/array-ldx.c
new file mode 100644
index 00000000000..0797af3bbfb
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/array-ldx.c
@@ -0,0 +1,6 @@
+/* { dg-do compile } */
+/* { dg-options "-march=loongarch32 -mabi=ilp32d -O2" } */
+long long foo(long long *arr, long long index)
+{
+	return arr[index];
+}
\ No newline at end of file
-- 
2.41.0

[PATCH v2 11/14] LoongArch: Mark am* instructions as LA64-only

[Jiajie Chen]

LoongArch32 only provides basic ll/sc instructions for atomic
operations. Mark am* atomic instructions as 64-bit only.

gcc/ChangeLog:

	* config/loongarch.sync.md: Guard am* atomic insns by
	TARGET_64BIT.
---
 gcc/config/loongarch/sync.md | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/gcc/config/loongarch/sync.md b/gcc/config/loongarch/sync.md
index 9924d522bcd..151b553bcc6 100644
--- a/gcc/config/loongarch/sync.md
+++ b/gcc/config/loongarch/sync.md
@@ -77,7 +77,7 @@
       [(match_operand:GPR 1 "reg_or_0_operand" "rJ")
        (match_operand:SI 2 "const_int_operand")]      ;; model
       UNSPEC_ATOMIC_STORE))]
-  ""
+  "TARGET_64BIT"
   "amswap%A2.<amo>\t$zero,%z1,%0"
   [(set (attr "length") (const_int 8))])
 
@@ -88,7 +88,7 @@
 			   (match_operand:GPR 1 "reg_or_0_operand" "rJ"))
 	   (match_operand:SI 2 "const_int_operand")] ;; model
 	 UNSPEC_SYNC_OLD_OP))]
-  ""
+  "TARGET_64BIT"
   "am<amop>%A2.<amo>\t$zero,%z1,%0"
   [(set (attr "length") (const_int 8))])
 
@@ -101,7 +101,7 @@
 		     (match_operand:GPR 2 "reg_or_0_operand" "rJ"))
 	   (match_operand:SI 3 "const_int_operand")] ;; model
 	 UNSPEC_SYNC_OLD_OP))]
-  ""
+  "TARGET_64BIT"
   "am<amop>%A3.<amo>\t%0,%z2,%1"
   [(set (attr "length") (const_int 8))])
 
@@ -113,7 +113,7 @@
 	  UNSPEC_SYNC_EXCHANGE))
    (set (match_dup 1)
 	(match_operand:GPR 2 "register_operand" "r"))]
-  ""
+  "TARGET_64BIT"
   "amswap%A3.<amo>\t%0,%z2,%1"
   [(set (attr "length") (const_int 8))])
 
@@ -182,7 +182,7 @@
   [(match_operand:QI 0 "register_operand" "")     ;; bool output
    (match_operand:QI 1 "memory_operand" "+ZB")    ;; memory
    (match_operand:SI 2 "const_int_operand" "")]   ;; model
-  ""
+  "TARGET_64BIT"
 {
   /* We have no QImode atomics, so use the address LSBs to form a mask,
      then use an aligned SImode atomic.  */
-- 
2.41.0

[PATCH v2 12/14] LoongArch: Set long double width to 128 in la32

[Jiajie Chen]

According to latest loongarch procedure call standard, sizeof(long
double) == 128 in ilp32 data model regardless of target bitness.

gcc/ChangeLog:

	* config/loongarch/loongarch.h: Set LONG_DOUBLE_TYPE_SIZE to 128
	regardless of target bitness.
---
 gcc/config/loongarch/loongarch.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gcc/config/loongarch/loongarch.h b/gcc/config/loongarch/loongarch.h
index eca723293a1..ab0c80c69c1 100644
--- a/gcc/config/loongarch/loongarch.h
+++ b/gcc/config/loongarch/loongarch.h
@@ -205,7 +205,7 @@ along with GCC; see the file COPYING3.  If not see
 
 #define FLOAT_TYPE_SIZE 32
 #define DOUBLE_TYPE_SIZE 64
-#define LONG_DOUBLE_TYPE_SIZE (TARGET_64BIT ? 128 : 64)
+#define LONG_DOUBLE_TYPE_SIZE 128
 
 /* Define the sizes of fixed-point types.  */
 #define SHORT_FRACT_TYPE_SIZE 8
-- 
2.41.0

[PATCH v2 13/14] LoongArch: Fix ilp32 detection

[Jiajie Chen]

The correct ilp32 macro name is __loongarch_ilp32.

libitm/ChangeLog:

	* config/loongarch/asm.h: Fix ilp32 detection.
---
 libitm/config/loongarch/asm.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/libitm/config/loongarch/asm.h b/libitm/config/loongarch/asm.h
index 39e02b45f17..11d6d3c079e 100644
--- a/libitm/config/loongarch/asm.h
+++ b/libitm/config/loongarch/asm.h
@@ -30,7 +30,7 @@
 #  define GPR_S st.d
 #  define SZ_GPR 8
 #  define ADDSP(si)   addi.d  $sp, $sp, si
-#elif defined(__loongarch64_ilp32)
+#elif defined(__loongarch_ilp32)
 #  define GPR_L ld.w
 #  define GPR_S st.w
 #  define SZ_GPR 4
-- 
2.41.0

[PATCH v2 14/14] LoongArch: Allow ftintrz for DF->DI in loongarch32

[Jiajie Chen]

In LoongArch, signed DF->DI conversion can be done if -mfpu=64 and
-march=loongarch32.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (fix_trunc*2): Use ANYFI instead
	of GPR because it depends on fpu width instead of target bits.
---
 gcc/config/loongarch/loongarch.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index 31bdf3388f6..f6042af25b7 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -1482,11 +1482,11 @@
 
 ;; conversion of a floating-point value to a integer
 
-(define_insn "fix_trunc<ANYF:mode><GPR:mode>2"
-  [(set (match_operand:GPR 0 "register_operand" "=f")
-	(fix:GPR (match_operand:ANYF 1 "register_operand" "f")))]
+(define_insn "fix_trunc<ANYF:mode><ANYFI:mode>2"
+  [(set (match_operand:ANYFI 0 "register_operand" "=f")
+	(fix:ANYFI (match_operand:ANYF 1 "register_operand" "f")))]
   ""
-  "ftintrz.<GPR:ifmt>.<ANYF:fmt> %0,%1"
+  "ftintrz.<ANYFI:ifmt>.<ANYF:fmt> %0,%1"
   [(set_attr "type" "fcvt")
    (set_attr "mode" "<ANYF:MODE>")])
 
-- 
2.41.0

Re: [PATCH v2 01/14] LoongArch: Introduce loongarch32 target

[Xi Ruoyao]

On Wed, 2023-08-09 at 19:46 +0800, Jiajie Chen wrote:
> +      builtin_define ("_ABILP32=3");
> +      builtin_define ("_LOONGARCH_SIM=_ABILP32");

Let's remove them.  These MIPS-style definitions are deprecated:
https://github.com/loongson/LoongArch-Documentation/pull/28.

Unfortunately for LP64 ABI _ABILP64 is already a part of public API. 
I've tried to raise a deprecation warning for them, but it seems doing
so needs a major change in libcpp...  However ILP32 ABI is "fresh new"
so we should take the advantage to remove the historic burden.

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

Re: [PATCH v2 11/14] LoongArch: Mark am* instructions as LA64-only

[Xi Ruoyao]

On Wed, 2023-08-09 at 19:46 +0800, Jiajie Chen wrote:
> LoongArch32 only provides basic ll/sc instructions for atomic
> operations. Mark am* atomic instructions as 64-bit only.

I'd prefer using a different symbol, say TARGET_LOONGARCH_AM here.  Then
it would be easier to adjust the code if we have a LA32 core with am*
support in the future.  For now we can just
#define TARGET_LOONGARCH_AM TARGET_64BIT.

> gcc/ChangeLog:
> 
>         * config/loongarch.sync.md: Guard am* atomic insns by
>         TARGET_64BIT.
> ---
>  gcc/config/loongarch/sync.md | 10 +++++-----
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/gcc/config/loongarch/sync.md b/gcc/config/loongarch/sync.md
> index 9924d522bcd..151b553bcc6 100644
> --- a/gcc/config/loongarch/sync.md
> +++ b/gcc/config/loongarch/sync.md
> @@ -77,7 +77,7 @@
>        [(match_operand:GPR 1 "reg_or_0_operand" "rJ")
>         (match_operand:SI 2 "const_int_operand")]      ;; model
>        UNSPEC_ATOMIC_STORE))]
> -  ""
> +  "TARGET_64BIT"
>    "amswap%A2.<amo>\t$zero,%z1,%0"
>    [(set (attr "length") (const_int 8))])
>  
> @@ -88,7 +88,7 @@
>                            (match_operand:GPR 1 "reg_or_0_operand" "rJ"))
>            (match_operand:SI 2 "const_int_operand")] ;; model
>          UNSPEC_SYNC_OLD_OP))]
> -  ""
> +  "TARGET_64BIT"
>    "am<amop>%A2.<amo>\t$zero,%z1,%0"
>    [(set (attr "length") (const_int 8))])
>  
> @@ -101,7 +101,7 @@
>                      (match_operand:GPR 2 "reg_or_0_operand" "rJ"))
>            (match_operand:SI 3 "const_int_operand")] ;; model
>          UNSPEC_SYNC_OLD_OP))]
> -  ""
> +  "TARGET_64BIT"
>    "am<amop>%A3.<amo>\t%0,%z2,%1"
>    [(set (attr "length") (const_int 8))])
>  
> @@ -113,7 +113,7 @@
>           UNSPEC_SYNC_EXCHANGE))
>     (set (match_dup 1)
>         (match_operand:GPR 2 "register_operand" "r"))]
> -  ""
> +  "TARGET_64BIT"
>    "amswap%A3.<amo>\t%0,%z2,%1"
>    [(set (attr "length") (const_int 8))])
>  
> @@ -182,7 +182,7 @@
>    [(match_operand:QI 0 "register_operand" "")     ;; bool output
>     (match_operand:QI 1 "memory_operand" "+ZB")    ;; memory
>     (match_operand:SI 2 "const_int_operand" "")]   ;; model
> -  ""
> +  "TARGET_64BIT"
>  {
>    /* We have no QImode atomics, so use the address LSBs to form a mask,
>       then use an aligned SImode atomic.  */

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University