From patchwork Thu Jan 6 11:07:08 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Liu, Hongtao" X-Patchwork-Id: 1576045 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: bilbo.ozlabs.org; dkim=pass (1024-bit key; unprotected) header.d=gcc.gnu.org header.i=@gcc.gnu.org header.a=rsa-sha256 header.s=default header.b=Bt6CcMkf; dkim-atps=neutral Authentication-Results: ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=gcc.gnu.org (client-ip=8.43.85.97; helo=sourceware.org; envelope-from=gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org; receiver=) Received: from sourceware.org (ip-8-43-85-97.sourceware.org [8.43.85.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by bilbo.ozlabs.org (Postfix) with ESMTPS id 4JV3V106jsz9t6S for ; Thu, 6 Jan 2022 22:07:36 +1100 (AEDT) Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id F0175385843A for ; Thu, 6 Jan 2022 11:07:32 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org F0175385843A DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1641467253; bh=RvwvMxoIocQ0938yNDbL1rKQDeotrSYug/egHGENMI0=; h=To:Subject:Date:List-Id:List-Unsubscribe:List-Archive:List-Post: List-Help:List-Subscribe:From:Reply-To:From; b=Bt6CcMkfn7UYtf+QtNIlKh07yzrbkdDpckd0v+TtZcfDazBAmGmdsS/6jiePPZdJ5 ViWQAUUpIPr/mo4INIVEEFHgPmwRkOd3ekEZbmKFdGR5xE09j/mOCi2YVWTUcVwdxT CzU0tJdAiPvKs5cn4hRZ3kfOzqdI4uU6267bx7pY= X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from mga05.intel.com (mga05.intel.com [192.55.52.43]) by sourceware.org (Postfix) with ESMTPS id 0749B385840C for ; Thu, 6 Jan 2022 11:07:10 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 0749B385840C X-IronPort-AV: E=McAfee;i="6200,9189,10217"; a="328984402" X-IronPort-AV: E=Sophos;i="5.88,266,1635231600"; d="scan'208";a="328984402" Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by fmsmga105.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 06 Jan 2022 03:07:10 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.88,266,1635231600"; d="scan'208";a="526932228" Received: from scymds01.sc.intel.com ([10.148.94.138]) by fmsmga007.fm.intel.com with ESMTP; 06 Jan 2022 03:07:10 -0800 Received: from shliclel320.sh.intel.com (shliclel320.sh.intel.com [10.239.236.50]) by scymds01.sc.intel.com with ESMTP id 206B78uW003076; Thu, 6 Jan 2022 03:07:08 -0800 To: gcc-patches@gcc.gnu.org Subject: [PATCH] [i386] Support commutative alternative for AVX512 vpcmpeq{b, w, d, q} Date: Thu, 6 Jan 2022 19:07:08 +0800 Message-Id: <20220106110708.2915-1-hongtao.liu@intel.com> X-Mailer: git-send-email 2.18.1 X-Spam-Status: No, score=-12.6 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_SHORT, SPF_HELO_NONE, SPF_NONE, TXREP autolearn=ham autolearn_force=no version=3.4.4 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on server2.sourceware.org X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: liuhongt via Gcc-patches From: "Liu, Hongtao" Reply-To: liuhongt Errors-To: gcc-patches-bounces+incoming=patchwork.ozlabs.org@gcc.gnu.org Sender: "Gcc-patches" Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}. Ready to push to trunk. gcc/ChangeLog: * config/i386/sse.md (*_eq3_1): Extend to UNSPEC_PCMP_UNSIGNED. gcc/testsuite/ChangeLog: * gcc.target/i386/pr103774.c: New test. * gcc.target/i386/avx512bw-vpcmpequb-1.c: Adjust scan assembler from vpcmpub to (?:vpcmpub|vpcmpeqb). * gcc.target/i386/avx512bw-vpcmpequw-1.c: Ditto. * gcc.target/i386/avx512bw-vpcmpub-1.c: Ditto. * gcc.target/i386/avx512bw-vpcmpuw-1.c: Ditto. * gcc.target/i386/avx512f-vpcmpequd-1.c: Ditto. * gcc.target/i386/avx512f-vpcmpequq-1.c: Ditto. * gcc.target/i386/avx512f-vpcmpud-1.c: Ditto. * gcc.target/i386/avx512vl-vpcmpequd-1.c: Ditto. * gcc.target/i386/avx512vl-vpcmpequq-1.c: Ditto. * gcc.target/i386/avx512vl-vpcmpuq-1.c: Ditto. --- gcc/config/i386/sse.md | 64 +++++++++---------- .../gcc.target/i386/avx512bw-vpcmpequb-1.c | 12 ++-- .../gcc.target/i386/avx512bw-vpcmpequw-1.c | 12 ++-- .../gcc.target/i386/avx512bw-vpcmpub-1.c | 2 +- .../gcc.target/i386/avx512bw-vpcmpuw-1.c | 2 +- .../gcc.target/i386/avx512f-vpcmpequd-1.c | 4 +- .../gcc.target/i386/avx512f-vpcmpequq-1.c | 4 +- .../gcc.target/i386/avx512f-vpcmpud-1.c | 2 +- .../gcc.target/i386/avx512vl-vpcmpequd-1.c | 8 +-- .../gcc.target/i386/avx512vl-vpcmpequq-1.c | 8 +-- .../gcc.target/i386/avx512vl-vpcmpuq-1.c | 2 +- gcc/testsuite/gcc.target/i386/pr103774.c | 25 ++++++++ 12 files changed, 85 insertions(+), 60 deletions(-) create mode 100644 gcc/testsuite/gcc.target/i386/pr103774.c diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index fa1d56ae3e3..35360004818 100644 --- a/gcc/config/i386/sse.md +++ b/gcc/config/i386/sse.md @@ -3895,6 +3895,22 @@ (define_insn_and_split "*_cmp3" UNSPEC_PCMP_ITER))] "operands[4] = GEN_INT (INTVAL (operands[3]) ^ 4);") +(define_insn "*_eq3_1" + [(set (match_operand: 0 "register_operand" "=k,k") + (unspec: + [(match_operand:VI12_AVX512VL 1 "nonimm_or_0_operand" "%v,v") + (match_operand:VI12_AVX512VL 2 "nonimm_or_0_operand" "vm,C") + (const_int 0)] + UNSPEC_PCMP_ITER))] + "TARGET_AVX512BW && !(MEM_P (operands[1]) && MEM_P (operands[2]))" + "@ + vpcmpeq\t{%2, %1, %0|%0, %1, %2} + vptestnm\t{%1, %1, %0|%0, %1, %1}" + [(set_attr "type" "ssecmp") + (set_attr "prefix_extra" "1") + (set_attr "prefix" "evex") + (set_attr "mode" "")]) + (define_insn "_ucmp3" [(set (match_operand: 0 "register_operand" "=k") (unspec: @@ -3977,6 +3993,22 @@ (define_insn_and_split "*_ucmp3_zero_extend")]) +(define_insn "*_eq3_1" + [(set (match_operand: 0 "register_operand" "=k,k") + (unspec: + [(match_operand:VI48_AVX512VL 1 "nonimm_or_0_operand" "%v,v") + (match_operand:VI48_AVX512VL 2 "nonimm_or_0_operand" "vm,C") + (const_int 0)] + UNSPEC_PCMP_ITER))] + "TARGET_AVX512F && !(MEM_P (operands[1]) && MEM_P (operands[2]))" + "@ + vpcmpeq\t{%2, %1, %0|%0, %1, %2} + vptestnm\t{%1, %1, %0|%0, %1, %1}" + [(set_attr "type" "ssecmp") + (set_attr "prefix_extra" "1") + (set_attr "prefix" "evex") + (set_attr "mode" "")]) + (define_insn "_ucmp3" [(set (match_operand: 0 "register_operand" "=k") (unspec: @@ -16248,38 +16280,6 @@ (define_expand "_eq3" "TARGET_AVX512F" "ix86_fixup_binary_operands_no_copy (EQ, mode, operands);") -(define_insn "*_eq3_1" - [(set (match_operand: 0 "register_operand" "=k,k") - (unspec: - [(match_operand:VI12_AVX512VL 1 "nonimm_or_0_operand" "%v,v") - (match_operand:VI12_AVX512VL 2 "nonimm_or_0_operand" "vm,C") - (const_int 0)] - UNSPEC_PCMP))] - "TARGET_AVX512BW && !(MEM_P (operands[1]) && MEM_P (operands[2]))" - "@ - vpcmpeq\t{%2, %1, %0|%0, %1, %2} - vptestnm\t{%1, %1, %0|%0, %1, %1}" - [(set_attr "type" "ssecmp") - (set_attr "prefix_extra" "1") - (set_attr "prefix" "evex") - (set_attr "mode" "")]) - -(define_insn "*_eq3_1" - [(set (match_operand: 0 "register_operand" "=k,k") - (unspec: - [(match_operand:VI48_AVX512VL 1 "nonimm_or_0_operand" "%v,v") - (match_operand:VI48_AVX512VL 2 "nonimm_or_0_operand" "vm,C") - (const_int 0)] - UNSPEC_PCMP))] - "TARGET_AVX512F && !(MEM_P (operands[1]) && MEM_P (operands[2]))" - "@ - vpcmpeq\t{%2, %1, %0|%0, %1, %2} - vptestnm\t{%1, %1, %0|%0, %1, %1}" - [(set_attr "type" "ssecmp") - (set_attr "prefix_extra" "1") - (set_attr "prefix" "evex") - (set_attr "mode" "")]) - (define_insn "*sse4_1_eqv2di3" [(set (match_operand:V2DI 0 "register_operand" "=Yr,*x,x") (eq:V2DI diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequb-1.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequb-1.c index 31a0afc2f17..9a85753c9fb 100644 --- a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequb-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequb-1.c @@ -1,11 +1,11 @@ /* { dg-do compile } */ /* { dg-options "-mavx512bw -mavx512vl -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequw-1.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequw-1.c index da5c4015797..a6c7776b52f 100644 --- a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequw-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpequw-1.c @@ -1,11 +1,11 @@ /* { dg-do compile } */ /* { dg-options "-mavx512bw -mavx512vl -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpub-1.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpub-1.c index 120fe8fd19c..6d8a442d79f 100644 --- a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpub-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpub-1.c @@ -1,6 +1,6 @@ /* { dg-do compile } */ /* { dg-options "-O2 -mavx512bw -mavx512vl" } */ -/* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpub|vpcmpeqb)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpub\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ diff --git a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpuw-1.c b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpuw-1.c index f17c73b05c1..c0d84ded11b 100644 --- a/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpuw-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512bw-vpcmpuw-1.c @@ -5,7 +5,7 @@ /* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuw\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuw|vpcmpeqw)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequd-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequd-1.c index c02f2b1acef..3c567c648b6 100644 --- a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequd-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequd-1.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-mavx512f -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[0-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[0-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequq-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequq-1.c index 1a443d64cd6..f3f35ec62ec 100644 --- a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequq-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpequq-1.c @@ -1,7 +1,7 @@ /* { dg-do compile } */ /* { dg-options "-mavx512f -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[0-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[0-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpud-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpud-1.c index 331f95c231d..be0a8feb025 100644 --- a/gcc/testsuite/gcc.target/i386/avx512f-vpcmpud-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512f-vpcmpud-1.c @@ -1,6 +1,6 @@ /* { dg-do compile } */ /* { dg-options "-O2 -mavx512f" } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequd-1.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequd-1.c index 7c87035afab..a0bee114c15 100644 --- a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequd-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequd-1.c @@ -1,9 +1,9 @@ /* { dg-do compile } */ /* { dg-options "-mavx512vl -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpud|vpcmpeqd)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequq-1.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequq-1.c index 4d672a54698..c925d3226ec 100644 --- a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequq-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpequq-1.c @@ -1,9 +1,9 @@ /* { dg-do compile } */ /* { dg-options "-mavx512vl -O2" } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpuq-1.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpuq-1.c index a5491c92b18..f93fca2b655 100644 --- a/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpuq-1.c +++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpcmpuq-1.c @@ -2,7 +2,7 @@ /* { dg-options "-O2 -mavx512vl" } */ /* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ -/* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ +/* { dg-final { scan-assembler-times "(?:vpcmpuq|vpcmpeqq)\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\](?:\n|\[ \\t\]+#)" 1 } } */ /* { dg-final { scan-assembler-times "vpcmpuq\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n^k\]*%k\[0-7\]\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */ #include diff --git a/gcc/testsuite/gcc.target/i386/pr103774.c b/gcc/testsuite/gcc.target/i386/pr103774.c new file mode 100644 index 00000000000..0dafb1b0117 --- /dev/null +++ b/gcc/testsuite/gcc.target/i386/pr103774.c @@ -0,0 +1,25 @@ +/* { dg-do compile } */ +/* { dg-options "-march=cannonlake -O2" } */ +/* { dg-final { scan-assembler-not {(?n)vmovdq} } } */ +/* There should be no load + vpcmp, just vpcmp with memory operand. */ + +#include +__mmask16 +foo (__m256i* n, __m256i mch256) +{ + __m256i data1 = _mm256_loadu_si256(n); + __m256i data2 = _mm256_loadu_si256(n+1); + __mmask16 mask1 = _mm256_cmpeq_epu16_mask(data1, mch256); + __mmask16 mask2 = _mm256_cmpeq_epu16_mask(data2, mch256); + return mask1 + mask2; +} + +__mmask16 +foo1 (__m256i* n, __m256i mch256) +{ + __m256i data1 = _mm256_loadu_si256(n); + __m256i data2 = _mm256_loadu_si256(n+1); + __mmask16 mask1 = _mm256_cmpeq_epu16_mask(mch256, data1); + __mmask16 mask2 = _mm256_cmpeq_epu16_mask(mch256, data2); + return mask1 + mask2; +}