diff options
| author | liuhongt <hongtao.liu@intel.com> | 2021-08-17 13:11:26 +0800 |
|---|---|---|
| committer | liuhongt <hongtao.liu@intel.com> | 2021-08-18 11:26:36 +0800 |
| commit | 97d51c1764554fcef05fe94ee6445f5d2252bcff (patch) | |
| tree | a7a05130c7237ed77774d58f6a606ea5facc8ab2 /gcc | |
| parent | 2d14d64bf2d42a87ec58dd3760be12aeaa4a4279 (diff) | |
| download | gcc-97d51c1764554fcef05fe94ee6445f5d2252bcff.zip gcc-97d51c1764554fcef05fe94ee6445f5d2252bcff.tar.gz gcc-97d51c1764554fcef05fe94ee6445f5d2252bcff.tar.bz2 | |
Add x86 tune to enable v2df vector reduction by paddpd.
The tune is disabled by default.
gcc/ChangeLog:
PR target/97147
* config/i386/i386.h (TARGET_V2DF_REDUCTION_PREFER_HADDPD):
New macro.
* config/i386/sse.md (*sse3_haddv2df3_low): Add
TARGET_V2DF_REDUCTION_PREFER_HADDPD.
(*sse3_hsubv2df3_low): Ditto.
* config/i386/x86-tune.def
(X86_TUNE_V2DF_REDUCTION_PREFER_HADDPD): New tune.
gcc/testsuite/ChangeLog:
PR target/97147
* gcc.target/i386/pr54400.c: Adjust testcase.
* gcc.target/i386/pr94147.c: New test.
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/config/i386/i386.h | 2 | ||||
| -rw-r--r-- | gcc/config/i386/sse.md | 4 | ||||
| -rw-r--r-- | gcc/config/i386/x86-tune.def | 5 | ||||
| -rw-r--r-- | gcc/testsuite/gcc.target/i386/pr54400.c | 2 | ||||
| -rw-r--r-- | gcc/testsuite/gcc.target/i386/pr94147.c | 22 |
5 files changed, 32 insertions, 3 deletions
diff --git a/gcc/config/i386/i386.h b/gcc/config/i386/i386.h index 21fe51b..b3e57a8 100644 --- a/gcc/config/i386/i386.h +++ b/gcc/config/i386/i386.h @@ -418,6 +418,8 @@ extern unsigned char ix86_tune_features[X86_TUNE_LAST]; ix86_tune_features[X86_TUNE_EMIT_VZEROUPPER] #define TARGET_EXPAND_ABS \ ix86_tune_features[X86_TUNE_EXPAND_ABS] +#define TARGET_V2DF_REDUCTION_PREFER_HADDPD \ + ix86_tune_features[X86_TUNE_V2DF_REDUCTION_PREFER_HADDPD] /* Feature tests against the various architecture variations. */ enum ix86_arch_indices { diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index 27e25cc..1388968 100644 --- a/gcc/config/i386/sse.md +++ b/gcc/config/i386/sse.md @@ -2771,7 +2771,7 @@ (vec_select:DF (match_dup 1) (parallel [(match_operand:SI 3 "const_0_to_1_operand")]))))] - "TARGET_SSE3 + "TARGET_SSE3 && TARGET_V2DF_REDUCTION_PREFER_HADDPD && INTVAL (operands[2]) != INTVAL (operands[3])" "@ haddpd\t{%0, %0|%0, %0} @@ -2790,7 +2790,7 @@ (vec_select:DF (match_dup 1) (parallel [(const_int 1)]))))] - "TARGET_SSE3" + "TARGET_SSE3 && TARGET_V2DF_REDUCTION_PREFER_HADDPD" "@ hsubpd\t{%0, %0|%0, %0} vhsubpd\t{%1, %1, %0|%0, %1, %1}" diff --git a/gcc/config/i386/x86-tune.def b/gcc/config/i386/x86-tune.def index eb057a6..8f55da8 100644 --- a/gcc/config/i386/x86-tune.def +++ b/gcc/config/i386/x86-tune.def @@ -452,6 +452,11 @@ DEF_TUNE (X86_TUNE_AVOID_128FMA_CHAINS, "avoid_fma_chains", m_ZNVER) smaller FMA chain. */ DEF_TUNE (X86_TUNE_AVOID_256FMA_CHAINS, "avoid_fma256_chains", m_ZNVER2 | m_ZNVER3) +/* X86_TUNE_V2DF_REDUCTION_PREFER_PHADDPD: Prefer haddpd + for v2df vector reduction. */ +DEF_TUNE (X86_TUNE_V2DF_REDUCTION_PREFER_HADDPD, + "v2df_reduction_prefer_haddpd", m_NONE) + /*****************************************************************************/ /* AVX instruction selection tuning (some of SSE flags affects AVX, too) */ /*****************************************************************************/ diff --git a/gcc/testsuite/gcc.target/i386/pr54400.c b/gcc/testsuite/gcc.target/i386/pr54400.c index 5ed5ba0..3a45037 100644 --- a/gcc/testsuite/gcc.target/i386/pr54400.c +++ b/gcc/testsuite/gcc.target/i386/pr54400.c @@ -1,5 +1,5 @@ /* { dg-do compile } */ -/* { dg-options "-O2 -msse3 -mfpmath=sse" } */ +/* { dg-options "-O2 -msse3 -mfpmath=sse -mtune-ctrl=v2df_reduction_prefer_haddpd" } */ #include <x86intrin.h> diff --git a/gcc/testsuite/gcc.target/i386/pr94147.c b/gcc/testsuite/gcc.target/i386/pr94147.c new file mode 100644 index 0000000..8ff5c34 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr94147.c @@ -0,0 +1,22 @@ +/* { dg-do compile } */ +/* { dg-options "-O2 -msse3 -mfpmath=sse" } */ + +#include <x86intrin.h> + +double f (__m128d p) +{ + return p[0] - p[1]; +} + +double g1 (__m128d p) +{ + return p[0] + p[1]; +} + +double g2 (__m128d p) +{ + return p[1] + p[0]; +} + +/* { dg-final { scan-assembler-not "hsubpd" } } */ +/* { dg-final { scan-assembler-not "haddpd" } } */ |