Allow XMM16-XMM31 in vpbroadcast*

Message ID	20160513172034.GT28550@tucnak.redhat.com
State	New
Headers	show Return-Path: <gcc-patches-return-427283-incoming=patchwork.ozlabs.org@gcc.gnu.org> DomainKey-Signature: a=rsa-sha1; c=nofws; d=gcc.gnu.org; h=list-id :list-unsubscribe:list-archive:list-post:list-help:sender:date :from:to:cc:subject:message-id:reply-to:mime-version :content-type; q=dns; s=default; b=BjZ68qOKZQihTHPBQqYsazWGvfUC2 rG3XvFlUttML/OecL6Kv8EFkYLteKr1rmjARq4zLTXxkVUOJSQfrmeNw+Xqr/zS5 zyZGyw4Sj9o99MVWZziffB1Gda9tL/Bppa0LOGDlZ3LyrPUmDKv4BmgGpbHFjkwg Uvn3jGT/RyVnK4= Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk Sender: gcc-patches-owner@gcc.gnu.org Date: Fri, 13 May 2016 19:20:34 +0200 From: Jakub Jelinek <jakub@redhat.com> To: Uros Bizjak <ubizjak@gmail.com>, Kirill Yukhin <kirill.yukhin@gmail.com> Cc: gcc-patches@gcc.gnu.org Subject: [PATCH] Allow XMM16-XMM31 in vpbroadcast* Message-ID: <20160513172034.GT28550@tucnak.redhat.com> Reply-To: Jakub Jelinek <jakub@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.24 (2015-08-30)

--- gcc/config/i386/sse.md.jj 2016-05-13 16:12:24.631965207 +0200 +++ gcc/config/i386/sse.md 2016-05-13 17:33:32.429909899 +0200 @@ -16725,30 +16725,40 @@ (define_insn "avx_vzeroupper" (set_attr "btver2_decode" "vector") (set_attr "mode" "OI")]) +(define_mode_attr pbroadcast_evex_isa + [(V64QI "avx512bw") (V32QI "avx512bw") (V16QI "avx512bw") + (V32HI "avx512bw") (V16HI "avx512bw") (V8HI "avx512bw") + (V16SI "avx512f") (V8SI "avx512f") (V4SI "avx512f") + (V8DI "avx512f") (V4DI "avx512f") (V2DI "avx512f")]) + (define_insn "avx2_pbroadcast<mode>" - [(set (match_operand:VI 0 "register_operand" "=x") + [(set (match_operand:VI 0 "register_operand" "=x,v") (vec_duplicate:VI (vec_select:<ssescalarmode> - (match_operand:<ssexmmmode> 1 "nonimmediate_operand" "xm") + (match_operand:<ssexmmmode> 1 "nonimmediate_operand" "xm,vm") (parallel [(const_int 0)]))))] "TARGET_AVX2" "vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1}" - [(set_attr "type" "ssemov") + [(set_attr "isa" "*,<pbroadcast_evex_isa>") + (set_attr "type" "ssemov") (set_attr "prefix_extra" "1") - (set_attr "prefix" "vex") + (set_attr "prefix" "vex,evex") (set_attr "mode" "<sseinsnmode>")]) (define_insn "avx2_pbroadcast<mode>_1" - [(set (match_operand:VI_256 0 "register_operand" "=x,x") + [(set (match_operand:VI_256 0 "register_operand" "=x,x,v,v") (vec_duplicate:VI_256 (vec_select:<ssescalarmode> - (match_operand:VI_256 1 "nonimmediate_operand" "m,x") + (match_operand:VI_256 1 "nonimmediate_operand" "m,x,m,v") (parallel [(const_int 0)]))))] "TARGET_AVX2" "@ vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1} + vpbroadcast<ssemodesuffix>\t{%x1, %0|%0, %x1} + vpbroadcast<ssemodesuffix>\t{%1, %0|%0, %<iptr>1} vpbroadcast<ssemodesuffix>\t{%x1, %0|%0, %x1}" - [(set_attr "type" "ssemov") + [(set_attr "isa" "*,*,<pbroadcast_evex_isa>,<pbroadcast_evex_isa>") + (set_attr "type" "ssemov") (set_attr "prefix_extra" "1") (set_attr "prefix" "vex") (set_attr "mode" "<sseinsnmode>")]) --- gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-1.c.jj 2016-05-13 16:58:07.491988435 +0200 +++ gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-1.c 2016-05-13 17:31:29.830534782 +0200 @@ -0,0 +1,104 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mavx512bw" } */ + +#include <x86intrin.h> + +void +f1 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastb_epi8 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f2 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastw_epi16 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f3 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastd_epi32 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f4 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastq_epi64 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f5 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastb_epi8 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ + +void +f6 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastw_epi16 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ + +void +f7 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastd_epi32 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ + +void +f8 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastq_epi64 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ --- gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-2.c.jj 2016-05-13 17:23:57.412954445 +0200 +++ gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-2.c 2016-05-13 17:32:21.203853901 +0200 @@ -0,0 +1,68 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mavx512bw" } */ + +typedef char V1 __attribute__((vector_size (16))); +typedef short V2 __attribute__((vector_size (16))); +typedef char V5 __attribute__((vector_size (32))); +typedef short V6 __attribute__((vector_size (32))); +typedef int V7 __attribute__((vector_size (32))); + +void +f1 (V1 x) +{ + register V1 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V1) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f2 (V2 x) +{ + register V2 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V2) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f5 (V5 x) +{ + register V5 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V5) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */ + +void +f6 (V6 x) +{ + register V6 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V6) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */ + +void +f7 (V7 x) +{ + register V7 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V7) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */ --- gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-3.c.jj 2016-05-13 17:24:01.236896680 +0200 +++ gcc/testsuite/gcc.target/i386/avx512bw-vpbroadcast-3.c 2016-05-13 17:19:57.000000000 +0200 @@ -0,0 +1,58 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mavx512bw" } */ + +typedef char V1 __attribute__((vector_size (16))); +typedef short V2 __attribute__((vector_size (16))); +typedef char V5 __attribute__((vector_size (32))); +typedef short V6 __attribute__((vector_size (32))); +typedef int V7 __attribute__((vector_size (32))); + +void +f1 (V1 *x) +{ + register V1 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V1) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*xmm16" } } */ + +void +f2 (V2 *x) +{ + register V2 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V2) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*xmm16" } } */ + +void +f5 (V5 *x) +{ + register V5 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V5) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastb\[^\n\r]*ymm16" } } */ + +void +f6 (V6 *x) +{ + register V6 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V6) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastw\[^\n\r]*ymm16" } } */ + +void +f7 (V7 *x) +{ + register V7 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V7) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*ymm16" } } */ --- gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-1.c.jj 2016-05-13 16:58:38.167685897 +0200 +++ gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-1.c 2016-05-13 17:29:45.144162649 +0200 @@ -0,0 +1,104 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */ + +#include <x86intrin.h> + +void +f1 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastb_epi8 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */ + +void +f2 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastw_epi16 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */ + +void +f3 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastd_epi32 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f4 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = _mm_broadcastq_epi64 (a); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*xmm16\[^\n\r]*xmm16" } } */ + +void +f5 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastb_epi8 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*\[xy]mm1\[67]" } } */ + +void +f6 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastw_epi16 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*\[xy]mm1\[67]" } } */ + +void +f7 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastd_epi32 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ + +void +f8 (__m128i x) +{ + register __m128i a __asm ("xmm16"); + register __m256i b __asm ("xmm17"); + a = x; + asm volatile ("" : "+v" (a)); + b = _mm256_broadcastq_epi64 (a); + asm volatile ("" : "+v" (b)); +} + +/* { dg-final { scan-assembler "vpbroadcastq\[^\n\r]*(xmm1\[67]\[^\n\r]*ymm1\[67]|ymm1\[67]\[^\n\r]*xmm1\[67])" } } */ --- gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-2.c.jj 2016-05-13 17:24:27.606562792 +0200 +++ gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-2.c 2016-05-13 17:28:19.176240587 +0200 @@ -0,0 +1,68 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */ + +typedef char V1 __attribute__((vector_size (16))); +typedef short V2 __attribute__((vector_size (16))); +typedef char V5 __attribute__((vector_size (32))); +typedef short V6 __attribute__((vector_size (32))); +typedef int V7 __attribute__((vector_size (32))); + +void +f1 (V1 x) +{ + register V1 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V1) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */ + +void +f2 (V2 x) +{ + register V2 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V2) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */ + +void +f5 (V5 x) +{ + register V5 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V5) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*\[xy]mm16" } } */ + +void +f6 (V6 x) +{ + register V6 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V6) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*\[xy]mm16" } } */ + +void +f7 (V7 x) +{ + register V7 a __asm ("xmm16"); + a = x; + asm volatile ("" : "+v" (a)); + a = __builtin_shuffle (a, (V7) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*(xmm16\[^\n\r]*ymm16|ymm16\[^\n\r]*xmm16)" } } */ --- gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-3.c.jj 2016-05-13 17:24:32.113506886 +0200 +++ gcc/testsuite/gcc.target/i386/avx512vl-vpbroadcast-3.c 2016-05-13 17:26:31.587970561 +0200 @@ -0,0 +1,58 @@ +/* { dg-do compile { target { ! ia32 } } } */ +/* { dg-options "-O2 -mavx512vl -mno-avx512bw" } */ + +typedef char V1 __attribute__((vector_size (16))); +typedef short V2 __attribute__((vector_size (16))); +typedef char V5 __attribute__((vector_size (32))); +typedef short V6 __attribute__((vector_size (32))); +typedef int V7 __attribute__((vector_size (32))); + +void +f1 (V1 *x) +{ + register V1 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V1) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*xmm16" } } */ + +void +f2 (V2 *x) +{ + register V2 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V2) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*xmm16" } } */ + +void +f5 (V5 *x) +{ + register V5 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V5) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastb\[^\n\r]*ymm16" } } */ + +void +f6 (V6 *x) +{ + register V6 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V6) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler-not "vpbroadcastw\[^\n\r]*ymm16" } } */ + +void +f7 (V7 *x) +{ + register V7 a __asm ("xmm16"); + a = __builtin_shuffle (*x, (V7) { 0 }); + asm volatile ("" : "+v" (a)); +} + +/* { dg-final { scan-assembler "vpbroadcastd\[^\n\r]*ymm16" } } */

Allow XMM16-XMM31 in vpbroadcast*

Commit Message

Comments

Patch