From patchwork Wed Feb 3 05:38:59 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: develop--- via Libc-alpha X-Patchwork-Id: 41911 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 817A4398C806; Wed, 3 Feb 2021 05:39:15 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 817A4398C806 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1612330755; bh=vgHzeZifLTLI8PaH6NdCi8fDFONWFdXXaFqFvtahDiY=; h=To:Subject:Date:List-Id:List-Unsubscribe:List-Archive:List-Post: List-Help:List-Subscribe:From:Reply-To:Cc:From; b=pwLIzd2M0OdcDlPw/QQWbugNMfiWqMmDtEaL4qZsaIqYvSsGzsy0SPHvVbekTZ1CM a+jErrumCWWyguSSw0NjSB0GUou+08yIaLkSI4xy9yKe8KPCDNuYjh4WOJzi1WstwL s//2jfx+6BBSX0kX3w+ENHWIvFGP+PzrdcyBvce4= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from mail-pj1-x1035.google.com (mail-pj1-x1035.google.com [IPv6:2607:f8b0:4864:20::1035]) by sourceware.org (Postfix) with ESMTPS id 5545A386F83F for ; Wed, 3 Feb 2021 05:39:12 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 5545A386F83F Received: by mail-pj1-x1035.google.com with SMTP id g15so3891617pjd.2 for ; Tue, 02 Feb 2021 21:39:12 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=vgHzeZifLTLI8PaH6NdCi8fDFONWFdXXaFqFvtahDiY=; b=kcGs0GPy2hu3Vbu+i+4Vzm8IZbMTVo3U3fwvEYeci83TWWUh7P28hri8cl0f/zJw7c gjerdKB5BuI+KSzD5h1xrimbBj9KpV7m0cP0bxqBAE6glP/KHfl15+k5We9EkRRtlsco lBaxgM8qUAvh29bfoH8/hlhlyRPV+mPcC57xTiBXg5lPuIqfJKPFpsYe3HECohV2SzIT uCqkid0Lz5iHYe6y6Q7DGnJEK0HwJrrEo9/7DGCLmCmJAte9/kF5y/Pj84yjduIwaZ6w w4aGGzGeMv44rSXUG5wnyDe9oL+Lp5EjUEv3pumrpCnvHb1LiuV2PEJZVIFRTTi7d+zV tC5Q== X-Gm-Message-State: AOAM533medgjDYNnBQlvgdq2PVTuOCU7AO7pE14jP1JPKxMgAR3MP064 M3otSyw5kb5Os8+d8Y87fNiQL/Juu5evzQ== X-Google-Smtp-Source: ABdhPJyr336SnAH6gPyBKKzhyyJczzk2s7VP4+os0ySgLmAEUOL1SCOxYLVdZs+WmtMv+1M1s2ZXqg== X-Received: by 2002:a17:90a:15cf:: with SMTP id w15mr1498585pjd.171.1612330751102; Tue, 02 Feb 2021 21:39:11 -0800 (PST) Received: from localhost.localdomain (c-73-241-149-213.hsd1.ca.comcast.net. [73.241.149.213]) by smtp.googlemail.com with ESMTPSA id s9sm713378pfd.38.2021.02.02.21.39.10 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 02 Feb 2021 21:39:10 -0800 (PST) To: libc-alpha@sourceware.org Subject: [PATCH v4 1/2] x86: Refactor and improve performance of strchr-avx2.S Date: Wed, 3 Feb 2021 00:38:59 -0500 Message-Id: <20210203053900.4125403-1-goldstein.w.n@gmail.com> X-Mailer: git-send-email 2.29.2 MIME-Version: 1.0 X-Spam-Status: No, score=-12.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, FREEMAIL_FROM, GIT_PATCH_0, RCVD_IN_DNSWL_NONE, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: "goldstein.w.n--- via Libc-alpha" From: develop--- via Libc-alpha Reply-To: goldstein.w.n@gmail.com Cc: goldstein.w.n@gmail.com Errors-To: libc-alpha-bounces@sourceware.org Sender: "Libc-alpha" From: noah No bug. Just seemed the performance could be improved a bit. Observed and expected behavior are unchanged. Optimized body of main loop. Updated page cross logic and optimized accordingly. Made a few minor instruction selection modifications. No regressions in test suite. Both test-strchrnul and test-strchr passed. Signed-off-by: noah --- sysdeps/x86_64/multiarch/strchr-avx2.S | 235 ++++++++++++------------- sysdeps/x86_64/multiarch/strchr.c | 1 + 2 files changed, 118 insertions(+), 118 deletions(-) diff --git a/sysdeps/x86_64/multiarch/strchr-avx2.S b/sysdeps/x86_64/multiarch/strchr-avx2.S index d416558d04..8b9d78b55a 100644 --- a/sysdeps/x86_64/multiarch/strchr-avx2.S +++ b/sysdeps/x86_64/multiarch/strchr-avx2.S @@ -27,10 +27,12 @@ # ifdef USE_AS_WCSCHR # define VPBROADCAST vpbroadcastd # define VPCMPEQ vpcmpeqd +# define VPMINU vpminud # define CHAR_REG esi # else # define VPBROADCAST vpbroadcastb # define VPCMPEQ vpcmpeqb +# define VPMINU vpminub # define CHAR_REG sil # endif @@ -39,20 +41,26 @@ # endif # define VEC_SIZE 32 +# define PAGE_SIZE 4096 .section .text.avx,"ax",@progbits ENTRY (STRCHR) movl %edi, %ecx - /* Broadcast CHAR to YMM0. */ +# ifndef USE_AS_STRCHRNUL + xorl %edx, %edx +# endif + + /* Broadcast CHAR to YMM0. */ vmovd %esi, %xmm0 vpxor %xmm9, %xmm9, %xmm9 VPBROADCAST %xmm0, %ymm0 - /* Check if we may cross page boundary with one vector load. */ - andl $(2 * VEC_SIZE - 1), %ecx - cmpl $VEC_SIZE, %ecx - ja L(cros_page_boundary) - - /* Check the first VEC_SIZE bytes. Search for both CHAR and the + + /* Check if we cross page boundary with one vector load. */ + andl $(PAGE_SIZE - 1), %ecx + cmpl $(PAGE_SIZE - VEC_SIZE), %ecx + ja L(cross_page_boundary) + + /* Check the first VEC_SIZE bytes. Search for both CHAR and the null byte. */ vmovdqu (%rdi), %ymm8 VPCMPEQ %ymm8, %ymm0, %ymm1 @@ -60,50 +68,27 @@ ENTRY (STRCHR) vpor %ymm1, %ymm2, %ymm1 vpmovmskb %ymm1, %eax testl %eax, %eax - jnz L(first_vec_x0) - - /* Align data for aligned loads in the loop. */ - addq $VEC_SIZE, %rdi - andl $(VEC_SIZE - 1), %ecx - andq $-VEC_SIZE, %rdi - - jmp L(more_4x_vec) - - .p2align 4 -L(cros_page_boundary): - andl $(VEC_SIZE - 1), %ecx - andq $-VEC_SIZE, %rdi - vmovdqu (%rdi), %ymm8 - VPCMPEQ %ymm8, %ymm0, %ymm1 - VPCMPEQ %ymm8, %ymm9, %ymm2 - vpor %ymm1, %ymm2, %ymm1 - vpmovmskb %ymm1, %eax - /* Remove the leading bytes. */ - sarl %cl, %eax - testl %eax, %eax - jz L(aligned_more) - /* Found CHAR or the null byte. */ + jz L(more_vecs) tzcntl %eax, %eax - addq %rcx, %rax -# ifdef USE_AS_STRCHRNUL + /* Found CHAR or the null byte. */ addq %rdi, %rax -# else - xorl %edx, %edx - leaq (%rdi, %rax), %rax - cmp (%rax), %CHAR_REG +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG cmovne %rdx, %rax # endif VZEROUPPER ret .p2align 4 +L(more_vecs): + /* Align data for aligned loads in the loop. */ + andq $-VEC_SIZE, %rdi L(aligned_more): - addq $VEC_SIZE, %rdi -L(more_4x_vec): - /* Check the first 4 * VEC_SIZE. Only one VEC_SIZE at a time - since data is only aligned to VEC_SIZE. */ - vmovdqa (%rdi), %ymm8 + /* Check the next 4 * VEC_SIZE. Only one VEC_SIZE at a time + since data is only aligned to VEC_SIZE. */ + vmovdqa VEC_SIZE(%rdi), %ymm8 + addq $VEC_SIZE, %rdi VPCMPEQ %ymm8, %ymm0, %ymm1 VPCMPEQ %ymm8, %ymm9, %ymm2 vpor %ymm1, %ymm2, %ymm1 @@ -125,7 +110,7 @@ L(more_4x_vec): vpor %ymm1, %ymm2, %ymm1 vpmovmskb %ymm1, %eax testl %eax, %eax - jnz L(first_vec_x2) + jnz L(first_vec_x2) vmovdqa (VEC_SIZE * 3)(%rdi), %ymm8 VPCMPEQ %ymm8, %ymm0, %ymm1 @@ -133,122 +118,136 @@ L(more_4x_vec): vpor %ymm1, %ymm2, %ymm1 vpmovmskb %ymm1, %eax testl %eax, %eax - jnz L(first_vec_x3) - - addq $(VEC_SIZE * 4), %rdi - - /* Align data to 4 * VEC_SIZE. */ - movq %rdi, %rcx - andl $(4 * VEC_SIZE - 1), %ecx - andq $-(4 * VEC_SIZE), %rdi - - .p2align 4 -L(loop_4x_vec): - /* Compare 4 * VEC at a time forward. */ - vmovdqa (%rdi), %ymm5 - vmovdqa VEC_SIZE(%rdi), %ymm6 - vmovdqa (VEC_SIZE * 2)(%rdi), %ymm7 - vmovdqa (VEC_SIZE * 3)(%rdi), %ymm8 - - VPCMPEQ %ymm5, %ymm0, %ymm1 - VPCMPEQ %ymm6, %ymm0, %ymm2 - VPCMPEQ %ymm7, %ymm0, %ymm3 - VPCMPEQ %ymm8, %ymm0, %ymm4 - - VPCMPEQ %ymm5, %ymm9, %ymm5 - VPCMPEQ %ymm6, %ymm9, %ymm6 - VPCMPEQ %ymm7, %ymm9, %ymm7 - VPCMPEQ %ymm8, %ymm9, %ymm8 - - vpor %ymm1, %ymm5, %ymm1 - vpor %ymm2, %ymm6, %ymm2 - vpor %ymm3, %ymm7, %ymm3 - vpor %ymm4, %ymm8, %ymm4 - - vpor %ymm1, %ymm2, %ymm5 - vpor %ymm3, %ymm4, %ymm6 - - vpor %ymm5, %ymm6, %ymm5 - - vpmovmskb %ymm5, %eax - testl %eax, %eax - jnz L(4x_vec_end) - - addq $(VEC_SIZE * 4), %rdi + jz L(prep_loop_4x) - jmp L(loop_4x_vec) + tzcntl %eax, %eax + leaq (VEC_SIZE * 3)(%rdi, %rax), %rax +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG + cmovne %rdx, %rax +# endif + VZEROUPPER + ret .p2align 4 L(first_vec_x0): - /* Found CHAR or the null byte. */ tzcntl %eax, %eax -# ifdef USE_AS_STRCHRNUL + /* Found CHAR or the null byte. */ addq %rdi, %rax -# else - xorl %edx, %edx - leaq (%rdi, %rax), %rax - cmp (%rax), %CHAR_REG +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG cmovne %rdx, %rax # endif VZEROUPPER ret - + .p2align 4 L(first_vec_x1): tzcntl %eax, %eax -# ifdef USE_AS_STRCHRNUL - addq $VEC_SIZE, %rax - addq %rdi, %rax -# else - xorl %edx, %edx leaq VEC_SIZE(%rdi, %rax), %rax - cmp (%rax), %CHAR_REG +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG cmovne %rdx, %rax # endif VZEROUPPER - ret - + ret + .p2align 4 L(first_vec_x2): tzcntl %eax, %eax -# ifdef USE_AS_STRCHRNUL - addq $(VEC_SIZE * 2), %rax - addq %rdi, %rax -# else - xorl %edx, %edx + /* Found CHAR or the null byte. */ leaq (VEC_SIZE * 2)(%rdi, %rax), %rax - cmp (%rax), %CHAR_REG +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG cmovne %rdx, %rax # endif VZEROUPPER ret + +L(prep_loop_4x): + /* Align data to 4 * VEC_SIZE. */ + andq $-(VEC_SIZE * 4), %rdi .p2align 4 -L(4x_vec_end): +L(loop_4x_vec): + /* Compare 4 * VEC at a time forward. */ + vmovdqa (VEC_SIZE * 4)(%rdi), %ymm5 + vmovdqa (VEC_SIZE * 5)(%rdi), %ymm6 + vmovdqa (VEC_SIZE * 6)(%rdi), %ymm7 + vmovdqa (VEC_SIZE * 7)(%rdi), %ymm8 + + /* Leaves only CHARS matching esi as 0. */ + vpxor %ymm5, %ymm0, %ymm1 + vpxor %ymm6, %ymm0, %ymm2 + vpxor %ymm7, %ymm0, %ymm3 + vpxor %ymm8, %ymm0, %ymm4 + + VPMINU %ymm1, %ymm5, %ymm1 + VPMINU %ymm2, %ymm6, %ymm2 + VPMINU %ymm3, %ymm7, %ymm3 + VPMINU %ymm4, %ymm8, %ymm4 + + VPMINU %ymm1, %ymm2, %ymm5 + VPMINU %ymm3, %ymm4, %ymm6 + + VPMINU %ymm5, %ymm6, %ymm5 + + VPCMPEQ %ymm5, %ymm9, %ymm5 + vpmovmskb %ymm5, %eax + + addq $(VEC_SIZE * 4), %rdi + testl %eax, %eax + jz L(loop_4x_vec) + + VPCMPEQ %ymm1, %ymm9, %ymm1 vpmovmskb %ymm1, %eax testl %eax, %eax jnz L(first_vec_x0) + + VPCMPEQ %ymm2, %ymm9, %ymm2 vpmovmskb %ymm2, %eax testl %eax, %eax jnz L(first_vec_x1) - vpmovmskb %ymm3, %eax - testl %eax, %eax - jnz L(first_vec_x2) + + VPCMPEQ %ymm3, %ymm9, %ymm3 + VPCMPEQ %ymm4, %ymm9, %ymm4 + vpmovmskb %ymm3, %ecx vpmovmskb %ymm4, %eax + salq $32, %rax + orq %rcx, %rax + tzcntq %rax, %rax + leaq (VEC_SIZE * 2)(%rdi, %rax), %rax +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG + cmovne %rdx, %rax +# endif + VZEROUPPER + ret + + /* Cold case for crossing page with first load. */ + .p2align 4 +L(cross_page_boundary): + andq $-VEC_SIZE, %rdi + andl $(VEC_SIZE - 1), %ecx + + vmovdqa (%rdi), %ymm8 + VPCMPEQ %ymm8, %ymm0, %ymm1 + VPCMPEQ %ymm8, %ymm9, %ymm2 + vpor %ymm1, %ymm2, %ymm1 + vpmovmskb %ymm1, %eax + /* Remove the leading bits. */ + sarxl %ecx, %eax, %eax testl %eax, %eax -L(first_vec_x3): + jz L(aligned_more) tzcntl %eax, %eax -# ifdef USE_AS_STRCHRNUL - addq $(VEC_SIZE * 3), %rax + addq %rcx, %rdi addq %rdi, %rax -# else - xorl %edx, %edx - leaq (VEC_SIZE * 3)(%rdi, %rax), %rax - cmp (%rax), %CHAR_REG +# ifndef USE_AS_STRCHRNUL + cmp (%rax), %CHAR_REG cmovne %rdx, %rax # endif VZEROUPPER ret END (STRCHR) -#endif +# endif diff --git a/sysdeps/x86_64/multiarch/strchr.c b/sysdeps/x86_64/multiarch/strchr.c index 583a152794..4dfbe3b58b 100644 --- a/sysdeps/x86_64/multiarch/strchr.c +++ b/sysdeps/x86_64/multiarch/strchr.c @@ -37,6 +37,7 @@ IFUNC_SELECTOR (void) if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER) && CPU_FEATURE_USABLE_P (cpu_features, AVX2) + && CPU_FEATURE_USABLE_P (cpu_features, BMI2) && CPU_FEATURES_ARCH_P (cpu_features, AVX_Fast_Unaligned_Load)) return OPTIMIZE (avx2); From patchwork Wed Feb 3 05:39:00 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: develop--- via Libc-alpha X-Patchwork-Id: 41912 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 0F9D9398C83B; Wed, 3 Feb 2021 05:39:17 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 0F9D9398C83B DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1612330757; bh=VzBg7K7i7cjvzZEyY3fkxaMtC57krHJo0I7p9tsKRl4=; h=To:Subject:Date:In-Reply-To:References:List-Id:List-Unsubscribe: List-Archive:List-Post:List-Help:List-Subscribe:From:Reply-To:Cc: From; b=HXdbVfiIZQhSyXi2KZumq8D4ar+7nHy9uJdTTCrF/vcm+W2sEfjwfxeVOi1oFIION DYrKrCF0Pndrgj22SI9cD+YNAMdic1ZsEAEOp9gHyDowxz3EUDvOPg1otVGf42Mxdn EoGaENGkWpbSSOYMa2HD6mP7mTS0aOthiU6IoLBM= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from mail-pj1-x1034.google.com (mail-pj1-x1034.google.com [IPv6:2607:f8b0:4864:20::1034]) by sourceware.org (Postfix) with ESMTPS id 8A20D386F83F for ; Wed, 3 Feb 2021 05:39:14 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 8A20D386F83F Received: by mail-pj1-x1034.google.com with SMTP id q72so2321618pjq.2 for ; Tue, 02 Feb 2021 21:39:14 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=VzBg7K7i7cjvzZEyY3fkxaMtC57krHJo0I7p9tsKRl4=; b=JPt7BCqlY6SttWjvuVegoorrjOXgy6kNHcD/xmbisav9sxvzkCYcDsLlOBGyDxpA1Y Wv00rmL8MJAyYCr2bUVJEy06W89qy1IsXBdoeDQ4anmhZHxzYoaLMC/y9DRy1HoEloEO 9fpIJJD/+3uTtXdaTcznQFK7u+mwplUzC4E4U4GfRW4TjE0x4AerT4IAhO3hMEbVmBf7 g6M2PsCBEn9GxiJZyUPUc/sFM2YPYhAmG5s6RrdapOPQA73VaSyqUCUkMS4hwZ4IQQit Yq12n9ZTrojmVscgLGf6G3aimT1LwVEmEb6j7bZ9Q3UF9hCv9YvdA3rf5O1P0/lYRfvj pzfg== X-Gm-Message-State: AOAM532lfOo1E6wHH8V988uQMG5h7mLPjT0KAH/WewykEjxNNjR6OdLO QC0WoNCytZiIfV2pabGevuyL3jmlS5tIww== X-Google-Smtp-Source: ABdhPJwnglMOJEkHZr8eee22zlnDFSOk32YdX1OtvsIxoxtXgdM8LsUvICWziKkCxVsYHwjs1L0D5w== X-Received: by 2002:a17:902:ea0d:b029:e1:54c:6344 with SMTP id s13-20020a170902ea0db02900e1054c6344mr1754931plg.22.1612330753525; Tue, 02 Feb 2021 21:39:13 -0800 (PST) Received: from localhost.localdomain (c-73-241-149-213.hsd1.ca.comcast.net. [73.241.149.213]) by smtp.googlemail.com with ESMTPSA id s9sm713378pfd.38.2021.02.02.21.39.12 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 02 Feb 2021 21:39:13 -0800 (PST) To: libc-alpha@sourceware.org Subject: [PATCH v4 2/2] x86: Add additional benchmarks and tests for strchr Date: Wed, 3 Feb 2021 00:39:00 -0500 Message-Id: <20210203053900.4125403-2-goldstein.w.n@gmail.com> X-Mailer: git-send-email 2.29.2 In-Reply-To: <20210203053900.4125403-1-goldstein.w.n@gmail.com> References: <20210203053900.4125403-1-goldstein.w.n@gmail.com> MIME-Version: 1.0 X-Spam-Status: No, score=-12.5 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, FREEMAIL_FROM, GIT_PATCH_0, RCVD_IN_DNSWL_NONE, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: "goldstein.w.n--- via Libc-alpha" From: develop--- via Libc-alpha Reply-To: goldstein.w.n@gmail.com Cc: goldstein.w.n@gmail.com Errors-To: libc-alpha-bounces@sourceware.org Sender: "Libc-alpha" From: noah This patch adds additional benchmarks and tests for string size of 4096 and several benchmarks for string size 256 with different alignments. Signed-off-by: noah --- Added 2 additional benchmark and test sizes: 4096: Just feels like a natural "large" size to test 256 with multiple alignments: This essentially is to test how expensive the initial work prior to the 4x loop is depending on different alignments. results from bench-strchr: All times are in seconds and the medium of 100 runs. Old is current strchr-avx2.S implementation. New is this patch. Summary: New is definetly faster for medium -> large sizes. Once the 4x loop is hit there is a 10%+ speedup and New always wins out. For smaller sizes there is more variance as to which is faster and the differences are small. Generally it seems the New version wins out. This is likely because 0 - 31 sized strings are the fast path for new (no jmp). Also something that is neat is the significant performance improved for alignment 96 and 112. This is because the 5x vectors before 4x loop really favor that alignment. Benchmarking CPU: Icelake: Intel(R) Core(TM) i7-1065G7 CPU @ 1.30GHz size, algn, Old T , New T -------- Win Dif 0 , 0 , 2.54 , 2.52 -------- New -0.02 1 , 0 , 2.57 , 2.52 -------- New -0.05 2 , 0 , 2.56 , 2.52 -------- New -0.04 3 , 0 , 2.58 , 2.54 -------- New -0.04 4 , 0 , 2.61 , 2.55 -------- New -0.06 5 , 0 , 2.65 , 2.62 -------- New -0.03 6 , 0 , 2.73 , 2.74 -------- Old -0.01 7 , 0 , 2.75 , 2.74 -------- New -0.01 8 , 0 , 2.62 , 2.6 -------- New -0.02 9 , 0 , 2.73 , 2.75 -------- Old -0.02 10 , 0 , 2.74 , 2.74 -------- Eq N/A 11 , 0 , 2.76 , 2.72 -------- New -0.04 12 , 0 , 2.74 , 2.72 -------- New -0.02 13 , 0 , 2.75 , 2.72 -------- New -0.03 14 , 0 , 2.74 , 2.73 -------- New -0.01 15 , 0 , 2.74 , 2.73 -------- New -0.01 16 , 0 , 2.74 , 2.73 -------- New -0.01 17 , 0 , 2.74 , 2.74 -------- Eq N/A 18 , 0 , 2.73 , 2.73 -------- Eq N/A 19 , 0 , 2.73 , 2.73 -------- Eq N/A 20 , 0 , 2.73 , 2.73 -------- Eq N/A 21 , 0 , 2.73 , 2.72 -------- New -0.01 22 , 0 , 2.71 , 2.74 -------- Old -0.03 23 , 0 , 2.71 , 2.69 -------- New -0.02 24 , 0 , 2.68 , 2.67 -------- New -0.01 25 , 0 , 2.66 , 2.62 -------- New -0.04 26 , 0 , 2.64 , 2.62 -------- New -0.02 27 , 0 , 2.71 , 2.64 -------- New -0.07 28 , 0 , 2.67 , 2.69 -------- Old -0.02 29 , 0 , 2.72 , 2.72 -------- Eq N/A 30 , 0 , 2.68 , 2.69 -------- Old -0.01 31 , 0 , 2.68 , 2.68 -------- Eq N/A 32 , 0 , 3.51 , 3.52 -------- Old -0.01 32 , 1 , 3.52 , 3.51 -------- New -0.01 64 , 0 , 3.97 , 3.93 -------- New -0.04 64 , 2 , 3.95 , 3.9 -------- New -0.05 64 , 1 , 4.0 , 3.93 -------- New -0.07 64 , 3 , 3.97 , 3.88 -------- New -0.09 64 , 4 , 3.95 , 3.89 -------- New -0.06 64 , 5 , 3.94 , 3.9 -------- New -0.04 64 , 6 , 3.97 , 3.9 -------- New -0.07 64 , 7 , 3.97 , 3.91 -------- New -0.06 96 , 0 , 4.74 , 4.52 -------- New -0.22 128 , 0 , 5.29 , 5.19 -------- New -0.1 128 , 2 , 5.29 , 5.15 -------- New -0.14 128 , 3 , 5.31 , 5.22 -------- New -0.09 256 , 0 , 11.19 , 9.81 -------- New -1.38 256 , 3 , 11.19 , 9.84 -------- New -1.35 256 , 4 , 11.2 , 9.88 -------- New -1.32 256 , 16 , 11.21 , 9.79 -------- New -1.42 256 , 32 , 11.39 , 10.34 -------- New -1.05 256 , 48 , 11.88 , 10.56 -------- New -1.32 256 , 64 , 11.82 , 10.83 -------- New -0.99 256 , 80 , 11.85 , 10.86 -------- New -0.99 256 , 96 , 9.56 , 8.76 -------- New -0.8 256 , 112 , 9.55 , 8.9 -------- New -0.65 512 , 0 , 15.76 , 13.72 -------- New -2.04 512 , 4 , 15.72 , 13.74 -------- New -1.98 512 , 5 , 15.73 , 13.74 -------- New -1.99 1024, 0 , 24.85 , 21.33 -------- New -3.52 1024, 5 , 24.86 , 21.27 -------- New -3.59 1024, 6 , 24.87 , 21.32 -------- New -3.55 2048, 0 , 45.75 , 36.7 -------- New -9.05 2048, 6 , 43.91 , 35.42 -------- New -8.49 2048, 7 , 44.43 , 36.37 -------- New -8.06 4096, 0 , 96.94 , 81.34 -------- New -15.6 4096, 7 , 97.01 , 81.32 -------- New -15.69 benchtests/bench-strchr.c | 26 +++++++++++++++++++++++++- string/test-strchr.c | 26 +++++++++++++++++++++++++- 2 files changed, 50 insertions(+), 2 deletions(-) diff --git a/benchtests/bench-strchr.c b/benchtests/bench-strchr.c index bf493fe458..4ce2369d9b 100644 --- a/benchtests/bench-strchr.c +++ b/benchtests/bench-strchr.c @@ -100,7 +100,7 @@ do_test (size_t align, size_t pos, size_t len, int seek_char, int max_char) size_t i; CHAR *result; CHAR *buf = (CHAR *) buf1; - align &= 15; + align &= 127; if ((align + len) * sizeof (CHAR) >= page_size) return; @@ -151,12 +151,24 @@ test_main (void) do_test (i, 16 << i, 2048, SMALL_CHAR, MIDDLE_CHAR); } + for (i = 1; i < 8; ++i) + { + do_test (0, 16 << i, 4096, SMALL_CHAR, MIDDLE_CHAR); + do_test (i, 16 << i, 4096, SMALL_CHAR, MIDDLE_CHAR); + } + for (i = 1; i < 8; ++i) { do_test (i, 64, 256, SMALL_CHAR, MIDDLE_CHAR); do_test (i, 64, 256, SMALL_CHAR, BIG_CHAR); } + for (i = 0; i < 8; ++i) + { + do_test (16 * i, 256, 512, SMALL_CHAR, MIDDLE_CHAR); + do_test (16 * i, 256, 512, SMALL_CHAR, BIG_CHAR); + } + for (i = 0; i < 32; ++i) { do_test (0, i, i + 1, SMALL_CHAR, MIDDLE_CHAR); @@ -169,12 +181,24 @@ test_main (void) do_test (i, 16 << i, 2048, 0, MIDDLE_CHAR); } + for (i = 1; i < 8; ++i) + { + do_test (0, 16 << i, 4096, 0, MIDDLE_CHAR); + do_test (i, 16 << i, 4096, 0, MIDDLE_CHAR); + } + for (i = 1; i < 8; ++i) { do_test (i, 64, 256, 0, MIDDLE_CHAR); do_test (i, 64, 256, 0, BIG_CHAR); } + for (i = 0; i < 8; ++i) + { + do_test (16 * i, 256, 512, 0, MIDDLE_CHAR); + do_test (16 * i, 256, 512, 0, BIG_CHAR); + } + for (i = 0; i < 32; ++i) { do_test (0, i, i + 1, 0, MIDDLE_CHAR); diff --git a/string/test-strchr.c b/string/test-strchr.c index 5b6022746c..2cf4ea2add 100644 --- a/string/test-strchr.c +++ b/string/test-strchr.c @@ -130,7 +130,7 @@ do_test (size_t align, size_t pos, size_t len, int seek_char, int max_char) size_t i; CHAR *result; CHAR *buf = (CHAR *) buf1; - align &= 15; + align &= 127; if ((align + len) * sizeof (CHAR) >= page_size) return; @@ -259,12 +259,24 @@ test_main (void) do_test (i, 16 << i, 2048, SMALL_CHAR, MIDDLE_CHAR); } + for (i = 1; i < 8; ++i) + { + do_test (0, 16 << i, 4096, SMALL_CHAR, MIDDLE_CHAR); + do_test (i, 16 << i, 4096, SMALL_CHAR, MIDDLE_CHAR); + } + for (i = 1; i < 8; ++i) { do_test (i, 64, 256, SMALL_CHAR, MIDDLE_CHAR); do_test (i, 64, 256, SMALL_CHAR, BIG_CHAR); } + for (i = 0; i < 8; ++i) + { + do_test (16 * i, 256, 512, SMALL_CHAR, MIDDLE_CHAR); + do_test (16 * i, 256, 512, SMALL_CHAR, BIG_CHAR); + } + for (i = 0; i < 32; ++i) { do_test (0, i, i + 1, SMALL_CHAR, MIDDLE_CHAR); @@ -277,12 +289,24 @@ test_main (void) do_test (i, 16 << i, 2048, 0, MIDDLE_CHAR); } + for (i = 1; i < 8; ++i) + { + do_test (0, 16 << i, 4096, 0, MIDDLE_CHAR); + do_test (i, 16 << i, 4096, 0, MIDDLE_CHAR); + } + for (i = 1; i < 8; ++i) { do_test (i, 64, 256, 0, MIDDLE_CHAR); do_test (i, 64, 256, 0, BIG_CHAR); } + for (i = 0; i < 8; ++i) + { + do_test (16 * i, 256, 512, 0, MIDDLE_CHAR); + do_test (16 * i, 256, 512, 0, BIG_CHAR); + } + for (i = 0; i < 32; ++i) { do_test (0, i, i + 1, 0, MIDDLE_CHAR);