* [PATCH] c++: Handle '#pragma GCC target optimize' early [PR48026]
@ 2023-12-06 15:48 Gwenole Beauchesne
0 siblings, 0 replies; only message in thread
From: Gwenole Beauchesne @ 2023-12-06 15:48 UTC (permalink / raw)
To: gcc-patches; +Cc: Gwenole Beauchesne
Tested on x86_64-pc-linux-gnu with --enable-languages=c,c++
-- >8 --
Handle '#pragma GCC optimize' earlier as the __OPTIMIZE__ macro may need
to be defined as well for certain usages.
This is a follow-up to r14-4967-g8697d3a1. Also add more tests for the
'#pragma GCC target' case with auto-vectorization enabled and multiple
combinations of namespaces and/or class member functions.
gcc/c-family/ChangeLog:
PR c++/48026
* c-pragma.cc (init-pragma): Register `#pragma GCC optimize' in
preprocess-only mode, and enable early handling.
gcc/testsuite/ChangeLog:
PR c++/48026
PR c++/41201
* g++.target/i386/vect-pragma-target-1.C: New test.
* g++.target/i386/vect-pragma-target-2.C: New test.
* gcc.target/i386/vect-pragma-target-1.c: New test.
* gcc.target/i386/vect-pragma-target-1.c: New test.
Signed-off-by: Gwenole Beauchesne <gb.devel@gmail.com>
---
gcc/c-family/c-pragma.cc | 4 +-
.../g++.target/i386/vect-pragma-target-1.C | 6 +
.../g++.target/i386/vect-pragma-target-2.C | 6 +
.../gcc.target/i386/vect-pragma-target-1.c | 194 ++++++++++++++++++
.../gcc.target/i386/vect-pragma-target-2.c | 7 +
5 files changed, 216 insertions(+), 1 deletion(-)
create mode 100644 gcc/testsuite/g++.target/i386/vect-pragma-target-1.C
create mode 100644 gcc/testsuite/g++.target/i386/vect-pragma-target-2.C
create mode 100644 gcc/testsuite/gcc.target/i386/vect-pragma-target-1.c
create mode 100644 gcc/testsuite/gcc.target/i386/vect-pragma-target-2.c
diff --git a/gcc/c-family/c-pragma.cc b/gcc/c-family/c-pragma.cc
index 849f8ac8c8b..26d4c0c71e0 100644
--- a/gcc/c-family/c-pragma.cc
+++ b/gcc/c-family/c-pragma.cc
@@ -1852,7 +1852,9 @@ init_pragma (void)
c_register_pragma_with_early_handler ("GCC", "target",
handle_pragma_target,
handle_pragma_target);
- c_register_pragma ("GCC", "optimize", handle_pragma_optimize);
+ c_register_pragma_with_early_handler ("GCC", "optimize",
+ handle_pragma_optimize,
+ handle_pragma_optimize);
c_register_pragma_with_early_handler ("GCC", "push_options",
handle_pragma_push_options,
handle_pragma_push_options);
diff --git a/gcc/testsuite/g++.target/i386/vect-pragma-target-1.C b/gcc/testsuite/g++.target/i386/vect-pragma-target-1.C
new file mode 100644
index 00000000000..2f360cf50e1
--- /dev/null
+++ b/gcc/testsuite/g++.target/i386/vect-pragma-target-1.C
@@ -0,0 +1,6 @@
+/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */
+/* { dg-options "-O0" } */
+/* { dg-final { scan-assembler-times "paddd.+xmm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vfmadd132ps.+ymm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vpaddw.+zmm\[0-9]+" 1 } } */
+#include "../../gcc.target/i386/vect-pragma-target-1.c"
diff --git a/gcc/testsuite/g++.target/i386/vect-pragma-target-2.C b/gcc/testsuite/g++.target/i386/vect-pragma-target-2.C
new file mode 100644
index 00000000000..b85bc93d845
--- /dev/null
+++ b/gcc/testsuite/g++.target/i386/vect-pragma-target-2.C
@@ -0,0 +1,6 @@
+/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */
+/* { dg-options "-O0" } */
+/* { dg-final { scan-assembler-times "paddd.+xmm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vfmadd132ps.+ymm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vpaddw.+zmm\[0-9]+" 1 } } */
+#include "../../gcc.target/i386/vect-pragma-target-2.c"
diff --git a/gcc/testsuite/gcc.target/i386/vect-pragma-target-1.c b/gcc/testsuite/gcc.target/i386/vect-pragma-target-1.c
new file mode 100644
index 00000000000..f5e71e453ec
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/vect-pragma-target-1.c
@@ -0,0 +1,194 @@
+/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */
+/* { dg-options "-O0" } */
+/* { dg-final { scan-assembler-times "paddd.+xmm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vfmadd132ps.+ymm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vpaddw.+zmm\[0-9]+" 1 } } */
+#ifndef CHECK_DEFINES
+#define CHECK_DEFINES 0
+#endif
+
+#define N 1024
+
+/* Optimization flags and tree vectorizer shall be disabled at this point */
+#if CHECK_DEFINES && defined(__OPTIMIZE__)
+#error "__OPTIMIZE__ is defined (not compiled with -O0?)"
+#endif
+
+#pragma GCC push_options
+#pragma GCC optimize ("O2", "tree-vectorize")
+
+/* Optimization flags and tree vectorizer shall be enabled at this point */
+#if CHECK_DEFINES && !defined(__OPTIMIZE__)
+#error "__OPTIMIZE__ is not defined"
+#endif
+
+#pragma GCC push_options
+#pragma GCC target ("sse4.2")
+#ifdef __cplusplus
+namespace {
+#endif
+
+/* Target flags up to including SSE4.2 shall be enabled at this point */
+#if CHECK_DEFINES && !defined(__SSE3__)
+#error "Target flag (SSE3) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSSE3__)
+#error "Target flag (SSSE3) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSE4_1__)
+#error "Target flag (SSE4.1) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSE4_2__)
+#error "Target flag (SSE4.2) is not defined"
+#endif
+
+void
+__attribute__((__noinline__, __used__))
+vec_saxpy_i32(int y[N], const int a[N], const int x[N])
+{
+ int i;
+ for (i = 0; i < N; i++)
+ y[i] += a[i] * x[i];
+}
+
+#ifdef __cplusplus
+}
+#endif
+#pragma GCC pop_options
+
+/* Target flags up to including SSE4.2 shall be disabled at this point */
+#if CHECK_DEFINES && defined(__SSE3__)
+#error "Target flag (SSE3) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSSE3__)
+#error "Target flag (SSSE3) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSE4_1__)
+#error "Target flag (SSE4.1) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSE4_2__)
+#error "Target flag (SSE4.2) is still defined"
+#endif
+
+#pragma GCC push_options
+#pragma GCC target ("avx2", "fma")
+#ifdef __cplusplus
+struct A {
+#endif
+
+/* Target flags up to including AVX2+FMA shall be enabled at this point */
+#if CHECK_DEFINES && !defined(__SSE3__)
+#error "Target flag (SSE3) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSSE3__)
+#error "Target flag (SSSE3) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSE4_1__)
+#error "Target flag (SSE4.1) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__SSE4_2__)
+#error "Target flag (SSE4.2) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__AVX__)
+#error "Target flag (AVX) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__AVX2__)
+#error "Target flag (AVX2) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__FMA__)
+#error "Target flag (FMA) is not defined"
+#endif
+
+void
+__attribute__((__noinline__, __used__))
+vec_saxpy_f32(float y[N], const float a[N], const float x[N])
+{
+ int i;
+ for (i = 0; i < N; i++)
+ y[i] += a[i] * x[i];
+}
+
+#ifdef __cplusplus
+};
+#endif
+#pragma GCC pop_options
+
+/* Target flags up to including AVX2+FMA shall be disabled at this point */
+#if CHECK_DEFINES && defined(__SSE3__)
+#error "Target flag (SSE3) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSSE3__)
+#error "Target flag (SSSE3) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSE4_1__)
+#error "Target flag (SSE4.1) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__SSE4_2__)
+#error "Target flag (SSE4.2) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__AVX__)
+#error "Target flag (AVX) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__AVX2__)
+#error "Target flag (AVX2) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__FMA__)
+#error "Target flag (FMA) is still defined"
+#endif
+
+#pragma GCC push_options
+#pragma GCC target ("arch=x86-64-v4")
+#ifdef __cplusplus
+namespace avx512 {
+struct A {
+#endif
+
+/* Essential AVX512 target flags shall be enabled at this point */
+#if CHECK_DEFINES && !defined(__AVX512F__)
+#error "Target flag (AVX512F) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__AVX512VL__)
+#error "Target flag (AVX512VL) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__AVX512DQ__)
+#error "Target flag (AVX512DQ) is not defined"
+#endif
+#if CHECK_DEFINES && !defined(__AVX512BW__)
+#error "Target flag (AVX512BW) is not defined"
+#endif
+
+void
+__attribute__((__noinline__, __used__))
+vec_saxpy_i16(short y[N], const short a[N], const short x[N])
+{
+ int i;
+ for (i = 0; i < N; i++)
+ y[i] += a[i] * x[i];
+}
+
+#ifdef __cplusplus
+};
+}
+#endif
+#pragma GCC pop_options
+
+/* Essential AVX512 target flags shall be disabled at this point */
+#if CHECK_DEFINES && defined(__AVX512F__)
+#error "Target flag (AVX512F) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__AVX512VL__)
+#error "Target flag (AVX512VL) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__AVX512DQ__)
+#error "Target flag (AVX512DQ) is still defined"
+#endif
+#if CHECK_DEFINES && defined(__AVX512BW__)
+#error "Target flag (AVX512BW) is still defined"
+#endif
+
+#pragma GCC pop_options
+
+/* Optimization flags and tree vectorizer shall be disabled at this point */
+#if CHECK_DEFINES && defined(__OPTIMIZE__)
+#error "__OPTIMIZE__ is still defined"
+#endif
diff --git a/gcc/testsuite/gcc.target/i386/vect-pragma-target-2.c b/gcc/testsuite/gcc.target/i386/vect-pragma-target-2.c
new file mode 100644
index 00000000000..349680453a4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/vect-pragma-target-2.c
@@ -0,0 +1,7 @@
+/* { dg-do compile { target { i?86-*-* x86_64-*-* } } } */
+/* { dg-options "-O0" } */
+/* { dg-final { scan-assembler-times "paddd.+xmm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vfmadd132ps.+ymm\[0-9]+" 1 } } */
+/* { dg-final { scan-assembler-times "vpaddw.+zmm\[0-9]+" 1 } } */
+#define CHECK_DEFINES 1
+#include "vect-pragma-target-1.c"
--
2.39.2
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2023-12-06 15:48 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-12-06 15:48 [PATCH] c++: Handle '#pragma GCC target optimize' early [PR48026] Gwenole Beauchesne
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).