From patchwork Fri Oct 14 21:14:59 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Noah Goldstein X-Patchwork-Id: 58868 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 79890386074D for ; Fri, 14 Oct 2022 21:15:37 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 79890386074D DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1665782137; bh=oZS6mmDDuZKPjO7oMGxh4f73UtXkib6YgIKhfsMk+js=; h=To:Subject:Date:In-Reply-To:References:List-Id:List-Unsubscribe: List-Archive:List-Post:List-Help:List-Subscribe:From:Reply-To: From; b=RufaKgHG7RLUJbIyVCrOa2/Ah/VZx8jC9MFGathlqWMFdbipxAHOaXeyYL/yY018B 9FI+KoX9NTDyWEt5100wn1oSKCjmoipKpxIglpKl/8L5ss5Mnyj0lcfG4bIK1B9LXn 6eThmq3SgQhcGV8PcKj2GYdCBkZjf5sQaprtJdVs= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from mail-ej1-x62d.google.com (mail-ej1-x62d.google.com [IPv6:2a00:1450:4864:20::62d]) by sourceware.org (Postfix) with ESMTPS id 537563858D38 for ; Fri, 14 Oct 2022 21:15:10 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 537563858D38 Received: by mail-ej1-x62d.google.com with SMTP id a26so13083052ejc.4 for ; Fri, 14 Oct 2022 14:15:10 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=oZS6mmDDuZKPjO7oMGxh4f73UtXkib6YgIKhfsMk+js=; b=cTZJ4CSVlDTjr8OmT4DlUg02iPWv8YEzIR0+MhvSWpx3ysAx2tY2mrH/U2W+HcjwGt 4WpYeu9aS/ILX9AIZZDYLT46sO9p5KSqldivY2BB49FVQsk3gVlrPYpiltHsOr14GKy1 Z+LgbKrNTUUPwKA2+LxVgT/SPSCi91En2lDPcQCICR4ZWUnP7i6lMhD9DWdgdpGSUjON D2QigyEwBBynX7qEl3PsTIA9FdiRtJ4MJnry6S2aCaVIQpQ43jjGNGTuoHMep88yJbxL riapC2fusPCY8OsicH/grWjaPYx9PZV2knc+nX7mCKhXA7Bmn/IiZwUJOJ8/UsW8GHaj 6K2Q== X-Gm-Message-State: ACrzQf24ttaarOnSUYrBqvvuZZbflXzp5o7B8xnswRVEeQ7bo7zYmqh/ HbV1fZfbTiCHgAngYC8R4njquiHpK4liPg== X-Google-Smtp-Source: AMsMyM5qHFrt7WoipXJx06kCGFxJsc06SyQrFpbxD8Y2GT3ardy0xMVQeRXsp77omUJVdYxN+zy1dQ== X-Received: by 2002:a17:906:9bf8:b0:78d:41aa:5eea with SMTP id de56-20020a1709069bf800b0078d41aa5eeamr4754680ejc.137.1665782108033; Fri, 14 Oct 2022 14:15:08 -0700 (PDT) Received: from noahgold-desk.an.intel.com ([192.55.60.38]) by smtp.gmail.com with ESMTPSA id j16-20020aa7c410000000b0045769dee1fdsm2465749edq.29.2022.10.14.14.15.05 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 14 Oct 2022 14:15:07 -0700 (PDT) To: libc-alpha@sourceware.org Subject: [PATCH v5 1/3] x86: Update evex256/512 vec macros Date: Fri, 14 Oct 2022 16:14:59 -0500 Message-Id: <20221014211501.524094-1-goldstein.w.n@gmail.com> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20221014164008.1325863-1-goldstein.w.n@gmail.com> References: <20221014164008.1325863-1-goldstein.w.n@gmail.com> MIME-Version: 1.0 X-Spam-Status: No, score=-11.1 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, URIBL_BLACK autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) 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: Noah Goldstein via Libc-alpha From: Noah Goldstein Reply-To: Noah Goldstein Errors-To: libc-alpha-bounces+patchwork=sourceware.org@sourceware.org Sender: "Libc-alpha" 1) Make section only define if there is not a previous definition 2) Add `VEC_lo` definition for proper reg-width but in the ymm/zmm0-15 range. This commit does not change libc.so Tested build on x86-64 --- sysdeps/x86_64/multiarch/avx-vecs.h | 4 +- sysdeps/x86_64/multiarch/evex-vecs-common.h | 6 +- sysdeps/x86_64/multiarch/evex256-vecs.h | 7 +- sysdeps/x86_64/multiarch/evex512-vecs.h | 7 +- .../memmove-avx-unaligned-erms-rtm.S | 15 +- .../multiarch/memmove-avx-unaligned-erms.S | 9 +- .../multiarch/memmove-avx512-unaligned-erms.S | 30 +- .../multiarch/memmove-evex-unaligned-erms.S | 30 +- .../multiarch/memmove-sse2-unaligned-erms.S | 11 +- .../multiarch/memmove-vec-unaligned-erms.S | 262 +++++++++--------- sysdeps/x86_64/multiarch/memrchr-evex.S | 40 +-- .../memset-avx2-unaligned-erms-rtm.S | 8 +- .../multiarch/memset-avx2-unaligned-erms.S | 14 +- .../multiarch/memset-avx512-unaligned-erms.S | 20 +- .../multiarch/memset-evex-unaligned-erms.S | 20 +- .../multiarch/memset-sse2-unaligned-erms.S | 10 +- .../multiarch/memset-vec-unaligned-erms.S | 70 +++-- sysdeps/x86_64/multiarch/sse2-vecs.h | 4 +- sysdeps/x86_64/multiarch/vec-macros.h | 112 ++++---- 19 files changed, 271 insertions(+), 408 deletions(-) diff --git a/sysdeps/x86_64/multiarch/avx-vecs.h b/sysdeps/x86_64/multiarch/avx-vecs.h index 89680f5db8..e0c865eaef 100644 --- a/sysdeps/x86_64/multiarch/avx-vecs.h +++ b/sysdeps/x86_64/multiarch/avx-vecs.h @@ -41,7 +41,7 @@ #define VMOVNT vmovntdq /* Often need to access xmm portion. */ -#define VEC_xmm VEC_any_xmm -#define VEC VEC_any_ymm +#define VMM_128 VMM_any_xmm +#define VMM VMM_any_ymm #endif diff --git a/sysdeps/x86_64/multiarch/evex-vecs-common.h b/sysdeps/x86_64/multiarch/evex-vecs-common.h index 99806ebcd7..5468de8d42 100644 --- a/sysdeps/x86_64/multiarch/evex-vecs-common.h +++ b/sysdeps/x86_64/multiarch/evex-vecs-common.h @@ -32,8 +32,8 @@ #define VMOVA vmovdqa64 #define VMOVNT vmovntdq -#define VEC_xmm VEC_hi_xmm -#define VEC_ymm VEC_hi_ymm -#define VEC_zmm VEC_hi_zmm +#define VMM_128 VMM_hi_xmm +#define VMM_256 VMM_hi_ymm +#define VMM_512 VMM_hi_zmm #endif diff --git a/sysdeps/x86_64/multiarch/evex256-vecs.h b/sysdeps/x86_64/multiarch/evex256-vecs.h index 222ba46dc7..497020fbbc 100644 --- a/sysdeps/x86_64/multiarch/evex256-vecs.h +++ b/sysdeps/x86_64/multiarch/evex256-vecs.h @@ -28,8 +28,11 @@ #include "evex-vecs-common.h" #define USE_WITH_EVEX256 1 -#define SECTION(p) p##.evex -#define VEC VEC_ymm +#ifndef SECTION +# define SECTION(p) p##.evex +#endif +#define VMM VMM_256 +#define VMM_lo VMM_any_ymm #endif diff --git a/sysdeps/x86_64/multiarch/evex512-vecs.h b/sysdeps/x86_64/multiarch/evex512-vecs.h index d1784d5368..2ddc7cb41e 100644 --- a/sysdeps/x86_64/multiarch/evex512-vecs.h +++ b/sysdeps/x86_64/multiarch/evex512-vecs.h @@ -28,8 +28,11 @@ #include "evex-vecs-common.h" #define USE_WITH_EVEX512 1 -#define SECTION(p) p##.evex512 -#define VEC VEC_zmm +#ifndef SECTION +# define SECTION(p) p##.evex512 +#endif +#define VMM VMM_512 +#define VMM_lo VMM_any_zmm #endif diff --git a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms-rtm.S b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms-rtm.S index 67a55f0c85..54b29bcf66 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms-rtm.S +++ b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms-rtm.S @@ -1,16 +1,9 @@ -#if IS_IN (libc) -# define VEC_SIZE 32 -# define VEC(i) ymm##i -# define VMOVNT vmovntdq -# define VMOVU vmovdqu -# define VMOVA vmovdqa -# define MOV_SIZE 4 -# define ZERO_UPPER_VEC_REGISTERS_RETURN \ - ZERO_UPPER_VEC_REGISTERS_RETURN_XTEST +#include -# define VZEROUPPER_RETURN jmp L(return) +#if ISA_SHOULD_BUILD (3) + +# include "avx-rtm-vecs.h" -# define SECTION(p) p##.avx.rtm # define MEMMOVE_SYMBOL(p,s) p##_avx_##s##_rtm # include "memmove-vec-unaligned-erms.S" diff --git a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S index a14b155667..6960571779 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-avx-unaligned-erms.S @@ -2,14 +2,7 @@ #if ISA_SHOULD_BUILD (3) -# define VEC_SIZE 32 -# define VEC(i) ymm##i -# define VMOVNT vmovntdq -# define VMOVU vmovdqu -# define VMOVA vmovdqa -# define MOV_SIZE 4 - -# define SECTION(p) p##.avx +# include "avx-vecs.h" # ifndef MEMMOVE_SYMBOL # define MEMMOVE_SYMBOL(p,s) p##_avx_##s diff --git a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S index 8d1568a7ba..1794c7de6d 100644 --- a/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-avx512-unaligned-erms.S @@ -2,35 +2,7 @@ #if ISA_SHOULD_BUILD (4) -# define VEC_SIZE 64 -# define XMM0 xmm16 -# define XMM1 xmm17 -# define YMM0 ymm16 -# define YMM1 ymm17 -# define VEC0 zmm16 -# define VEC1 zmm17 -# define VEC2 zmm18 -# define VEC3 zmm19 -# define VEC4 zmm20 -# define VEC5 zmm21 -# define VEC6 zmm22 -# define VEC7 zmm23 -# define VEC8 zmm24 -# define VEC9 zmm25 -# define VEC10 zmm26 -# define VEC11 zmm27 -# define VEC12 zmm28 -# define VEC13 zmm29 -# define VEC14 zmm30 -# define VEC15 zmm31 -# define VEC(i) VEC##i -# define VMOVNT vmovntdq -# define VMOVU vmovdqu64 -# define VMOVA vmovdqa64 -# define VZEROUPPER -# define MOV_SIZE 6 - -# define SECTION(p) p##.evex512 +# include "evex512-vecs.h" # ifndef MEMMOVE_SYMBOL # define MEMMOVE_SYMBOL(p,s) p##_avx512_##s diff --git a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S index 2373017358..200e7042a0 100644 --- a/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-evex-unaligned-erms.S @@ -2,35 +2,7 @@ #if ISA_SHOULD_BUILD (4) -# define VEC_SIZE 32 -# define XMM0 xmm16 -# define XMM1 xmm17 -# define YMM0 ymm16 -# define YMM1 ymm17 -# define VEC0 ymm16 -# define VEC1 ymm17 -# define VEC2 ymm18 -# define VEC3 ymm19 -# define VEC4 ymm20 -# define VEC5 ymm21 -# define VEC6 ymm22 -# define VEC7 ymm23 -# define VEC8 ymm24 -# define VEC9 ymm25 -# define VEC10 ymm26 -# define VEC11 ymm27 -# define VEC12 ymm28 -# define VEC13 ymm29 -# define VEC14 ymm30 -# define VEC15 ymm31 -# define VEC(i) VEC##i -# define VMOVNT vmovntdq -# define VMOVU vmovdqu64 -# define VMOVA vmovdqa64 -# define VZEROUPPER -# define MOV_SIZE 6 - -# define SECTION(p) p##.evex +# include "evex256-vecs.h" # ifndef MEMMOVE_SYMBOL # define MEMMOVE_SYMBOL(p,s) p##_evex_##s diff --git a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S index 422a079902..a2fe816270 100644 --- a/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-sse2-unaligned-erms.S @@ -22,18 +22,9 @@ so we need this to build for ISA V2 builds. */ #if ISA_SHOULD_BUILD (2) -# include +# include "sse2-vecs.h" -# define VEC_SIZE 16 -# define VEC(i) xmm##i # define PREFETCHNT prefetchnta -# define VMOVNT movntdq -/* Use movups and movaps for smaller code sizes. */ -# define VMOVU movups -# define VMOVA movaps -# define MOV_SIZE 3 - -# define SECTION(p) p # ifndef MEMMOVE_SYMBOL # define MEMMOVE_SYMBOL(p,s) p##_sse2_##s diff --git a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S index 04747133b7..5b758cae5e 100644 --- a/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S @@ -60,14 +60,6 @@ # define MEMMOVE_CHK_SYMBOL(p,s) MEMMOVE_SYMBOL(p, s) #endif -#ifndef XMM0 -# define XMM0 xmm0 -#endif - -#ifndef YMM0 -# define YMM0 ymm0 -#endif - #ifndef VZEROUPPER # if VEC_SIZE > 16 # define VZEROUPPER vzeroupper @@ -225,13 +217,13 @@ L(start): cmp $VEC_SIZE, %RDX_LP jb L(less_vec) /* Load regardless. */ - VMOVU (%rsi), %VEC(0) + VMOVU (%rsi), %VMM(0) cmp $(VEC_SIZE * 2), %RDX_LP ja L(more_2x_vec) /* From VEC and to 2 * VEC. No branch when size == VEC_SIZE. */ - VMOVU -VEC_SIZE(%rsi,%rdx), %VEC(1) - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(1), -VEC_SIZE(%rdi,%rdx) + VMOVU -VEC_SIZE(%rsi,%rdx), %VMM(1) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(1), -VEC_SIZE(%rdi,%rdx) #if !(defined USE_MULTIARCH && IS_IN (libc)) ZERO_UPPER_VEC_REGISTERS_RETURN #else @@ -270,15 +262,15 @@ L(start_erms): cmp $VEC_SIZE, %RDX_LP jb L(less_vec) /* Load regardless. */ - VMOVU (%rsi), %VEC(0) + VMOVU (%rsi), %VMM(0) cmp $(VEC_SIZE * 2), %RDX_LP ja L(movsb_more_2x_vec) /* From VEC and to 2 * VEC. No branch when size == VEC_SIZE. */ - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(1) - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(1), -VEC_SIZE(%rdi, %rdx) -L(return): + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(1) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(1), -VEC_SIZE(%rdi, %rdx) +L(return_vzeroupper): # if VEC_SIZE > 16 ZERO_UPPER_VEC_REGISTERS_RETURN # else @@ -359,10 +351,10 @@ L(between_16_31): .p2align 4,, 10 L(between_32_63): /* From 32 to 63. No branch when size == 32. */ - VMOVU (%rsi), %YMM0 - VMOVU -32(%rsi, %rdx), %YMM1 - VMOVU %YMM0, (%rdi) - VMOVU %YMM1, -32(%rdi, %rdx) + VMOVU (%rsi), %VMM_256(0) + VMOVU -32(%rsi, %rdx), %VMM_256(1) + VMOVU %VMM_256(0), (%rdi) + VMOVU %VMM_256(1), -32(%rdi, %rdx) VZEROUPPER_RETURN #endif @@ -380,12 +372,12 @@ L(last_4x_vec): /* Copy from 2 * VEC + 1 to 4 * VEC, inclusively. */ /* VEC(0) and VEC(1) have already been loaded. */ - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(2) - VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VEC(3) - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(1), VEC_SIZE(%rdi) - VMOVU %VEC(2), -VEC_SIZE(%rdi, %rdx) - VMOVU %VEC(3), -(VEC_SIZE * 2)(%rdi, %rdx) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(2) + VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VMM(3) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(1), VEC_SIZE(%rdi) + VMOVU %VMM(2), -VEC_SIZE(%rdi, %rdx) + VMOVU %VMM(3), -(VEC_SIZE * 2)(%rdi, %rdx) VZEROUPPER_RETURN .p2align 4 @@ -400,24 +392,24 @@ L(more_2x_vec): cmpq $(VEC_SIZE * 8), %rdx ja L(more_8x_vec) /* Load VEC(1) regardless. VEC(0) has already been loaded. */ - VMOVU VEC_SIZE(%rsi), %VEC(1) + VMOVU VEC_SIZE(%rsi), %VMM(1) cmpq $(VEC_SIZE * 4), %rdx jbe L(last_4x_vec) /* Copy from 4 * VEC + 1 to 8 * VEC, inclusively. */ - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(3) - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(4) - VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VEC(5) - VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VEC(6) - VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VEC(7) - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(1), VEC_SIZE(%rdi) - VMOVU %VEC(2), (VEC_SIZE * 2)(%rdi) - VMOVU %VEC(3), (VEC_SIZE * 3)(%rdi) - VMOVU %VEC(4), -VEC_SIZE(%rdi, %rdx) - VMOVU %VEC(5), -(VEC_SIZE * 2)(%rdi, %rdx) - VMOVU %VEC(6), -(VEC_SIZE * 3)(%rdi, %rdx) - VMOVU %VEC(7), -(VEC_SIZE * 4)(%rdi, %rdx) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(3) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(4) + VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VMM(5) + VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VMM(6) + VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VMM(7) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(1), VEC_SIZE(%rdi) + VMOVU %VMM(2), (VEC_SIZE * 2)(%rdi) + VMOVU %VMM(3), (VEC_SIZE * 3)(%rdi) + VMOVU %VMM(4), -VEC_SIZE(%rdi, %rdx) + VMOVU %VMM(5), -(VEC_SIZE * 2)(%rdi, %rdx) + VMOVU %VMM(6), -(VEC_SIZE * 3)(%rdi, %rdx) + VMOVU %VMM(7), -(VEC_SIZE * 4)(%rdi, %rdx) VZEROUPPER_RETURN .p2align 4,, 4 @@ -466,14 +458,14 @@ L(more_8x_vec_forward): */ /* First vec was already loaded into VEC(0). */ - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(5) - VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VEC(6) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(5) + VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VMM(6) /* Save begining of dst. */ movq %rdi, %rcx /* Align dst to VEC_SIZE - 1. */ orq $(VEC_SIZE - 1), %rdi - VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VEC(7) - VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VEC(8) + VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VMM(7) + VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VMM(8) /* Subtract dst from src. Add back after dst aligned. */ subq %rcx, %rsi @@ -488,25 +480,25 @@ L(more_8x_vec_forward): .p2align 4,, 11 L(loop_4x_vec_forward): /* Copy 4 * VEC a time forward. */ - VMOVU (%rsi), %VEC(1) - VMOVU VEC_SIZE(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(3) - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(4) + VMOVU (%rsi), %VMM(1) + VMOVU VEC_SIZE(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(3) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(4) subq $-(VEC_SIZE * 4), %rsi - VMOVA %VEC(1), (%rdi) - VMOVA %VEC(2), VEC_SIZE(%rdi) - VMOVA %VEC(3), (VEC_SIZE * 2)(%rdi) - VMOVA %VEC(4), (VEC_SIZE * 3)(%rdi) + VMOVA %VMM(1), (%rdi) + VMOVA %VMM(2), VEC_SIZE(%rdi) + VMOVA %VMM(3), (VEC_SIZE * 2)(%rdi) + VMOVA %VMM(4), (VEC_SIZE * 3)(%rdi) subq $-(VEC_SIZE * 4), %rdi cmpq %rdi, %rdx ja L(loop_4x_vec_forward) /* Store the last 4 * VEC. */ - VMOVU %VEC(5), (VEC_SIZE * 3)(%rdx) - VMOVU %VEC(6), (VEC_SIZE * 2)(%rdx) - VMOVU %VEC(7), VEC_SIZE(%rdx) - VMOVU %VEC(8), (%rdx) + VMOVU %VMM(5), (VEC_SIZE * 3)(%rdx) + VMOVU %VMM(6), (VEC_SIZE * 2)(%rdx) + VMOVU %VMM(7), VEC_SIZE(%rdx) + VMOVU %VMM(8), (%rdx) /* Store the first VEC. */ - VMOVU %VEC(0), (%rcx) + VMOVU %VMM(0), (%rcx) /* Keep L(nop_backward) target close to jmp for 2-byte encoding. */ L(nop_backward): @@ -523,12 +515,12 @@ L(more_8x_vec_backward): addresses. */ /* First vec was also loaded into VEC(0). */ - VMOVU VEC_SIZE(%rsi), %VEC(5) - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(6) + VMOVU VEC_SIZE(%rsi), %VMM(5) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(6) /* Begining of region for 4x backward copy stored in rcx. */ leaq (VEC_SIZE * -4 + -1)(%rdi, %rdx), %rcx - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(7) - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(8) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(7) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(8) /* Subtract dst from src. Add back after dst aligned. */ subq %rdi, %rsi /* Align dst. */ @@ -540,25 +532,25 @@ L(more_8x_vec_backward): .p2align 4,, 11 L(loop_4x_vec_backward): /* Copy 4 * VEC a time backward. */ - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(1) - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 1)(%rsi), %VEC(3) - VMOVU (VEC_SIZE * 0)(%rsi), %VEC(4) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(1) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 1)(%rsi), %VMM(3) + VMOVU (VEC_SIZE * 0)(%rsi), %VMM(4) addq $(VEC_SIZE * -4), %rsi - VMOVA %VEC(1), (VEC_SIZE * 3)(%rcx) - VMOVA %VEC(2), (VEC_SIZE * 2)(%rcx) - VMOVA %VEC(3), (VEC_SIZE * 1)(%rcx) - VMOVA %VEC(4), (VEC_SIZE * 0)(%rcx) + VMOVA %VMM(1), (VEC_SIZE * 3)(%rcx) + VMOVA %VMM(2), (VEC_SIZE * 2)(%rcx) + VMOVA %VMM(3), (VEC_SIZE * 1)(%rcx) + VMOVA %VMM(4), (VEC_SIZE * 0)(%rcx) addq $(VEC_SIZE * -4), %rcx cmpq %rcx, %rdi jb L(loop_4x_vec_backward) /* Store the first 4 * VEC. */ - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(5), VEC_SIZE(%rdi) - VMOVU %VEC(6), (VEC_SIZE * 2)(%rdi) - VMOVU %VEC(7), (VEC_SIZE * 3)(%rdi) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(5), VEC_SIZE(%rdi) + VMOVU %VMM(6), (VEC_SIZE * 2)(%rdi) + VMOVU %VMM(7), (VEC_SIZE * 3)(%rdi) /* Store the last VEC. */ - VMOVU %VEC(8), -VEC_SIZE(%rdx, %rdi) + VMOVU %VMM(8), -VEC_SIZE(%rdx, %rdi) VZEROUPPER_RETURN #if defined USE_MULTIARCH && IS_IN (libc) @@ -568,7 +560,7 @@ L(loop_4x_vec_backward): # if ALIGN_MOVSB L(skip_short_movsb_check): # if MOVSB_ALIGN_TO > VEC_SIZE - VMOVU VEC_SIZE(%rsi), %VEC(1) + VMOVU VEC_SIZE(%rsi), %VMM(1) # endif # if MOVSB_ALIGN_TO > (VEC_SIZE * 2) # error Unsupported MOVSB_ALIGN_TO @@ -597,9 +589,9 @@ L(skip_short_movsb_check): rep movsb - VMOVU %VEC(0), (%r8) + VMOVU %VMM(0), (%r8) # if MOVSB_ALIGN_TO > VEC_SIZE - VMOVU %VEC(1), VEC_SIZE(%r8) + VMOVU %VMM(1), VEC_SIZE(%r8) # endif VZEROUPPER_RETURN # endif @@ -640,7 +632,7 @@ L(movsb): # endif # if ALIGN_MOVSB # if MOVSB_ALIGN_TO > VEC_SIZE - VMOVU VEC_SIZE(%rsi), %VEC(1) + VMOVU VEC_SIZE(%rsi), %VMM(1) # endif # if MOVSB_ALIGN_TO > (VEC_SIZE * 2) # error Unsupported MOVSB_ALIGN_TO @@ -664,9 +656,9 @@ L(movsb_align_dst): rep movsb /* Store VECs loaded for aligning. */ - VMOVU %VEC(0), (%r8) + VMOVU %VMM(0), (%r8) # if MOVSB_ALIGN_TO > VEC_SIZE - VMOVU %VEC(1), VEC_SIZE(%r8) + VMOVU %VMM(1), VEC_SIZE(%r8) # endif VZEROUPPER_RETURN # else /* !ALIGN_MOVSB. */ @@ -701,18 +693,18 @@ L(large_memcpy_2x): /* First vec was also loaded into VEC(0). */ # if VEC_SIZE < 64 - VMOVU VEC_SIZE(%rsi), %VEC(1) + VMOVU VEC_SIZE(%rsi), %VMM(1) # if VEC_SIZE < 32 - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(3) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(3) # endif # endif - VMOVU %VEC(0), (%rdi) + VMOVU %VMM(0), (%rdi) # if VEC_SIZE < 64 - VMOVU %VEC(1), VEC_SIZE(%rdi) + VMOVU %VMM(1), VEC_SIZE(%rdi) # if VEC_SIZE < 32 - VMOVU %VEC(2), (VEC_SIZE * 2)(%rdi) - VMOVU %VEC(3), (VEC_SIZE * 3)(%rdi) + VMOVU %VMM(2), (VEC_SIZE * 2)(%rdi) + VMOVU %VMM(3), (VEC_SIZE * 3)(%rdi) # endif # endif @@ -761,12 +753,12 @@ L(loop_large_memcpy_2x_inner): PREFETCH_ONE_SET(1, (%rsi), PAGE_SIZE + PREFETCHED_LOAD_SIZE) PREFETCH_ONE_SET(1, (%rsi), PAGE_SIZE + PREFETCHED_LOAD_SIZE * 2) /* Load vectors from rsi. */ - LOAD_ONE_SET((%rsi), 0, %VEC(0), %VEC(1), %VEC(2), %VEC(3)) - LOAD_ONE_SET((%rsi), PAGE_SIZE, %VEC(4), %VEC(5), %VEC(6), %VEC(7)) + LOAD_ONE_SET((%rsi), 0, %VMM(0), %VMM(1), %VMM(2), %VMM(3)) + LOAD_ONE_SET((%rsi), PAGE_SIZE, %VMM(4), %VMM(5), %VMM(6), %VMM(7)) subq $-LARGE_LOAD_SIZE, %rsi /* Non-temporal store vectors to rdi. */ - STORE_ONE_SET((%rdi), 0, %VEC(0), %VEC(1), %VEC(2), %VEC(3)) - STORE_ONE_SET((%rdi), PAGE_SIZE, %VEC(4), %VEC(5), %VEC(6), %VEC(7)) + STORE_ONE_SET((%rdi), 0, %VMM(0), %VMM(1), %VMM(2), %VMM(3)) + STORE_ONE_SET((%rdi), PAGE_SIZE, %VMM(4), %VMM(5), %VMM(6), %VMM(7)) subq $-LARGE_LOAD_SIZE, %rdi decl %ecx jnz L(loop_large_memcpy_2x_inner) @@ -785,31 +777,31 @@ L(loop_large_memcpy_2x_tail): /* Copy 4 * VEC a time forward with non-temporal stores. */ PREFETCH_ONE_SET (1, (%rsi), PREFETCHED_LOAD_SIZE) PREFETCH_ONE_SET (1, (%rdi), PREFETCHED_LOAD_SIZE) - VMOVU (%rsi), %VEC(0) - VMOVU VEC_SIZE(%rsi), %VEC(1) - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(3) + VMOVU (%rsi), %VMM(0) + VMOVU VEC_SIZE(%rsi), %VMM(1) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(3) subq $-(VEC_SIZE * 4), %rsi addl $-(VEC_SIZE * 4), %edx - VMOVA %VEC(0), (%rdi) - VMOVA %VEC(1), VEC_SIZE(%rdi) - VMOVA %VEC(2), (VEC_SIZE * 2)(%rdi) - VMOVA %VEC(3), (VEC_SIZE * 3)(%rdi) + VMOVA %VMM(0), (%rdi) + VMOVA %VMM(1), VEC_SIZE(%rdi) + VMOVA %VMM(2), (VEC_SIZE * 2)(%rdi) + VMOVA %VMM(3), (VEC_SIZE * 3)(%rdi) subq $-(VEC_SIZE * 4), %rdi cmpl $(VEC_SIZE * 4), %edx ja L(loop_large_memcpy_2x_tail) L(large_memcpy_2x_end): /* Store the last 4 * VEC. */ - VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VEC(0) - VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VEC(1) - VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VEC(2) - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(3) - - VMOVU %VEC(0), -(VEC_SIZE * 4)(%rdi, %rdx) - VMOVU %VEC(1), -(VEC_SIZE * 3)(%rdi, %rdx) - VMOVU %VEC(2), -(VEC_SIZE * 2)(%rdi, %rdx) - VMOVU %VEC(3), -VEC_SIZE(%rdi, %rdx) + VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VMM(0) + VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VMM(1) + VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VMM(2) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(3) + + VMOVU %VMM(0), -(VEC_SIZE * 4)(%rdi, %rdx) + VMOVU %VMM(1), -(VEC_SIZE * 3)(%rdi, %rdx) + VMOVU %VMM(2), -(VEC_SIZE * 2)(%rdi, %rdx) + VMOVU %VMM(3), -VEC_SIZE(%rdi, %rdx) VZEROUPPER_RETURN .p2align 4 @@ -831,16 +823,16 @@ L(loop_large_memcpy_4x_inner): PREFETCH_ONE_SET(1, (%rsi), PAGE_SIZE * 2 + PREFETCHED_LOAD_SIZE) PREFETCH_ONE_SET(1, (%rsi), PAGE_SIZE * 3 + PREFETCHED_LOAD_SIZE) /* Load vectors from rsi. */ - LOAD_ONE_SET((%rsi), 0, %VEC(0), %VEC(1), %VEC(2), %VEC(3)) - LOAD_ONE_SET((%rsi), PAGE_SIZE, %VEC(4), %VEC(5), %VEC(6), %VEC(7)) - LOAD_ONE_SET((%rsi), PAGE_SIZE * 2, %VEC(8), %VEC(9), %VEC(10), %VEC(11)) - LOAD_ONE_SET((%rsi), PAGE_SIZE * 3, %VEC(12), %VEC(13), %VEC(14), %VEC(15)) + LOAD_ONE_SET((%rsi), 0, %VMM(0), %VMM(1), %VMM(2), %VMM(3)) + LOAD_ONE_SET((%rsi), PAGE_SIZE, %VMM(4), %VMM(5), %VMM(6), %VMM(7)) + LOAD_ONE_SET((%rsi), PAGE_SIZE * 2, %VMM(8), %VMM(9), %VMM(10), %VMM(11)) + LOAD_ONE_SET((%rsi), PAGE_SIZE * 3, %VMM(12), %VMM(13), %VMM(14), %VMM(15)) subq $-LARGE_LOAD_SIZE, %rsi /* Non-temporal store vectors to rdi. */ - STORE_ONE_SET((%rdi), 0, %VEC(0), %VEC(1), %VEC(2), %VEC(3)) - STORE_ONE_SET((%rdi), PAGE_SIZE, %VEC(4), %VEC(5), %VEC(6), %VEC(7)) - STORE_ONE_SET((%rdi), PAGE_SIZE * 2, %VEC(8), %VEC(9), %VEC(10), %VEC(11)) - STORE_ONE_SET((%rdi), PAGE_SIZE * 3, %VEC(12), %VEC(13), %VEC(14), %VEC(15)) + STORE_ONE_SET((%rdi), 0, %VMM(0), %VMM(1), %VMM(2), %VMM(3)) + STORE_ONE_SET((%rdi), PAGE_SIZE, %VMM(4), %VMM(5), %VMM(6), %VMM(7)) + STORE_ONE_SET((%rdi), PAGE_SIZE * 2, %VMM(8), %VMM(9), %VMM(10), %VMM(11)) + STORE_ONE_SET((%rdi), PAGE_SIZE * 3, %VMM(12), %VMM(13), %VMM(14), %VMM(15)) subq $-LARGE_LOAD_SIZE, %rdi decl %ecx jnz L(loop_large_memcpy_4x_inner) @@ -858,31 +850,31 @@ L(loop_large_memcpy_4x_tail): /* Copy 4 * VEC a time forward with non-temporal stores. */ PREFETCH_ONE_SET (1, (%rsi), PREFETCHED_LOAD_SIZE) PREFETCH_ONE_SET (1, (%rdi), PREFETCHED_LOAD_SIZE) - VMOVU (%rsi), %VEC(0) - VMOVU VEC_SIZE(%rsi), %VEC(1) - VMOVU (VEC_SIZE * 2)(%rsi), %VEC(2) - VMOVU (VEC_SIZE * 3)(%rsi), %VEC(3) + VMOVU (%rsi), %VMM(0) + VMOVU VEC_SIZE(%rsi), %VMM(1) + VMOVU (VEC_SIZE * 2)(%rsi), %VMM(2) + VMOVU (VEC_SIZE * 3)(%rsi), %VMM(3) subq $-(VEC_SIZE * 4), %rsi addl $-(VEC_SIZE * 4), %edx - VMOVA %VEC(0), (%rdi) - VMOVA %VEC(1), VEC_SIZE(%rdi) - VMOVA %VEC(2), (VEC_SIZE * 2)(%rdi) - VMOVA %VEC(3), (VEC_SIZE * 3)(%rdi) + VMOVA %VMM(0), (%rdi) + VMOVA %VMM(1), VEC_SIZE(%rdi) + VMOVA %VMM(2), (VEC_SIZE * 2)(%rdi) + VMOVA %VMM(3), (VEC_SIZE * 3)(%rdi) subq $-(VEC_SIZE * 4), %rdi cmpl $(VEC_SIZE * 4), %edx ja L(loop_large_memcpy_4x_tail) L(large_memcpy_4x_end): /* Store the last 4 * VEC. */ - VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VEC(0) - VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VEC(1) - VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VEC(2) - VMOVU -VEC_SIZE(%rsi, %rdx), %VEC(3) - - VMOVU %VEC(0), -(VEC_SIZE * 4)(%rdi, %rdx) - VMOVU %VEC(1), -(VEC_SIZE * 3)(%rdi, %rdx) - VMOVU %VEC(2), -(VEC_SIZE * 2)(%rdi, %rdx) - VMOVU %VEC(3), -VEC_SIZE(%rdi, %rdx) + VMOVU -(VEC_SIZE * 4)(%rsi, %rdx), %VMM(0) + VMOVU -(VEC_SIZE * 3)(%rsi, %rdx), %VMM(1) + VMOVU -(VEC_SIZE * 2)(%rsi, %rdx), %VMM(2) + VMOVU -VEC_SIZE(%rsi, %rdx), %VMM(3) + + VMOVU %VMM(0), -(VEC_SIZE * 4)(%rdi, %rdx) + VMOVU %VMM(1), -(VEC_SIZE * 3)(%rdi, %rdx) + VMOVU %VMM(2), -(VEC_SIZE * 2)(%rdi, %rdx) + VMOVU %VMM(3), -VEC_SIZE(%rdi, %rdx) VZEROUPPER_RETURN #endif END (MEMMOVE_SYMBOL (__memmove, unaligned_erms)) diff --git a/sysdeps/x86_64/multiarch/memrchr-evex.S b/sysdeps/x86_64/multiarch/memrchr-evex.S index ea3a0a0a60..d5c535a8f0 100644 --- a/sysdeps/x86_64/multiarch/memrchr-evex.S +++ b/sysdeps/x86_64/multiarch/memrchr-evex.S @@ -31,7 +31,7 @@ # endif # define PAGE_SIZE 4096 -# define VECMATCH VEC(0) +# define VMMMATCH VMM(0) .section SECTION(.text), "ax", @progbits ENTRY_P2ALIGN(MEMRCHR, 6) @@ -47,7 +47,7 @@ ENTRY_P2ALIGN(MEMRCHR, 6) correct page cross check and 2) it correctly sets up end ptr to be subtract by lzcnt aligned. */ leaq -1(%rdi, %rdx), %rax - vpbroadcastb %esi, %VECMATCH + vpbroadcastb %esi, %VMMMATCH /* Check if we can load 1x VEC without cross a page. */ testl $(PAGE_SIZE - VEC_SIZE), %eax @@ -55,7 +55,7 @@ ENTRY_P2ALIGN(MEMRCHR, 6) /* Don't use rax for pointer here because EVEX has better encoding with offset % VEC_SIZE == 0. */ - vpcmpb $0, -(VEC_SIZE)(%rdi, %rdx), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE)(%rdi, %rdx), %VMMMATCH, %k0 kmovd %k0, %ecx /* Fall through for rdx (len) <= VEC_SIZE (expect small sizes). */ @@ -96,7 +96,7 @@ L(more_1x_vec): movq %rax, %rdx /* Need no matter what. */ - vpcmpb $0, -(VEC_SIZE)(%rax), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx subq %rdi, %rdx @@ -115,7 +115,7 @@ L(last_2x_vec): /* Don't use rax for pointer here because EVEX has better encoding with offset % VEC_SIZE == 0. */ - vpcmpb $0, -(VEC_SIZE * 2)(%rdi, %rdx), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE * 2)(%rdi, %rdx), %VMMMATCH, %k0 kmovd %k0, %ecx /* NB: 64-bit lzcnt. This will naturally add 32 to position. */ lzcntq %rcx, %rcx @@ -131,7 +131,7 @@ L(last_2x_vec): L(page_cross): movq %rax, %rsi andq $-VEC_SIZE, %rsi - vpcmpb $0, (%rsi), %VECMATCH, %k0 + vpcmpb $0, (%rsi), %VMMMATCH, %k0 kmovd %k0, %r8d /* Shift out negative alignment (because we are starting from endptr and working backwards). */ @@ -165,13 +165,13 @@ L(more_2x_vec): testl %ecx, %ecx jnz L(ret_vec_x0_dec) - vpcmpb $0, -(VEC_SIZE * 2)(%rax), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE * 2)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx testl %ecx, %ecx jnz L(ret_vec_x1) /* Need no matter what. */ - vpcmpb $0, -(VEC_SIZE * 3)(%rax), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE * 3)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx subq $(VEC_SIZE * 4), %rdx @@ -185,7 +185,7 @@ L(last_vec): /* Need no matter what. */ - vpcmpb $0, -(VEC_SIZE * 4)(%rax), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE * 4)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx lzcntl %ecx, %ecx subq $(VEC_SIZE * 3 + 1), %rax @@ -220,7 +220,7 @@ L(more_4x_vec): testl %ecx, %ecx jnz L(ret_vec_x2) - vpcmpb $0, -(VEC_SIZE * 4)(%rax), %VECMATCH, %k0 + vpcmpb $0, -(VEC_SIZE * 4)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx testl %ecx, %ecx @@ -243,17 +243,17 @@ L(more_4x_vec): L(loop_4x_vec): /* Store 1 were not-equals and 0 where equals in k1 (used to mask later on). */ - vpcmpb $4, (VEC_SIZE * 3)(%rax), %VECMATCH, %k1 + vpcmpb $4, (VEC_SIZE * 3)(%rax), %VMMMATCH, %k1 /* VEC(2/3) will have zero-byte where we found a CHAR. */ - vpxorq (VEC_SIZE * 2)(%rax), %VECMATCH, %VEC(2) - vpxorq (VEC_SIZE * 1)(%rax), %VECMATCH, %VEC(3) - vpcmpb $0, (VEC_SIZE * 0)(%rax), %VECMATCH, %k4 + vpxorq (VEC_SIZE * 2)(%rax), %VMMMATCH, %VMM(2) + vpxorq (VEC_SIZE * 1)(%rax), %VMMMATCH, %VMM(3) + vpcmpb $0, (VEC_SIZE * 0)(%rax), %VMMMATCH, %k4 /* Combine VEC(2/3) with min and maskz with k1 (k1 has zero bit where CHAR is found and VEC(2/3) have zero-byte where CHAR is found. */ - vpminub %VEC(2), %VEC(3), %VEC(3){%k1}{z} - vptestnmb %VEC(3), %VEC(3), %k2 + vpminub %VMM(2), %VMM(3), %VMM(3){%k1}{z} + vptestnmb %VMM(3), %VMM(3), %k2 /* Any 1s and we found CHAR. */ kortestd %k2, %k4 @@ -270,7 +270,7 @@ L(loop_4x_vec): L(last_4x_vec): /* Used no matter what. */ - vpcmpb $0, (VEC_SIZE * -1)(%rax), %VECMATCH, %k0 + vpcmpb $0, (VEC_SIZE * -1)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx cmpl $(VEC_SIZE * 2), %edx @@ -280,14 +280,14 @@ L(last_4x_vec): jnz L(ret_vec_x0_dec) - vpcmpb $0, (VEC_SIZE * -2)(%rax), %VECMATCH, %k0 + vpcmpb $0, (VEC_SIZE * -2)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx testl %ecx, %ecx jnz L(ret_vec_x1) /* Used no matter what. */ - vpcmpb $0, (VEC_SIZE * -3)(%rax), %VECMATCH, %k0 + vpcmpb $0, (VEC_SIZE * -3)(%rax), %VMMMATCH, %k0 kmovd %k0, %ecx cmpl $(VEC_SIZE * 3), %edx @@ -309,7 +309,7 @@ L(loop_end): testl %ecx, %ecx jnz L(ret_vec_x0_end) - vptestnmb %VEC(2), %VEC(2), %k0 + vptestnmb %VMM(2), %VMM(2), %k0 kmovd %k0, %ecx testl %ecx, %ecx jnz L(ret_vec_x1_end) diff --git a/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms-rtm.S b/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms-rtm.S index 8ac3e479bb..3bd3b34150 100644 --- a/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms-rtm.S +++ b/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms-rtm.S @@ -1,10 +1,6 @@ -#define ZERO_UPPER_VEC_REGISTERS_RETURN \ - ZERO_UPPER_VEC_REGISTERS_RETURN_XTEST +#include "avx-rtm-vecs.h" -#define VZEROUPPER_RETURN jmp L(return) - -#define SECTION(p) p##.avx.rtm #define MEMSET_SYMBOL(p,s) p##_avx2_##s##_rtm #define WMEMSET_SYMBOL(p,s) p##_avx2_##s##_rtm -#include "memset-avx2-unaligned-erms.S" +# include "memset-avx2-unaligned-erms.S" diff --git a/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms.S index a9054a9122..b3f2bbd61b 100644 --- a/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memset-avx2-unaligned-erms.S @@ -4,14 +4,9 @@ # define USE_WITH_AVX2 1 -# define VEC_SIZE 32 -# define MOV_SIZE 4 -# define RET_SIZE 4 - -# define VEC(i) ymm##i - -# define VMOVU vmovdqu -# define VMOVA vmovdqa +# ifndef VEC_SIZE +# include "avx-vecs.h" +# endif # define MEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ vmovd d, %xmm0; \ @@ -26,9 +21,6 @@ # define WMEMSET_VDUP_TO_VEC0_HIGH() vpbroadcastd %xmm0, %ymm0 # define WMEMSET_VDUP_TO_VEC0_LOW() vpbroadcastd %xmm0, %xmm0 -# ifndef SECTION -# define SECTION(p) p##.avx -# endif # ifndef MEMSET_SYMBOL # define MEMSET_SYMBOL(p,s) p##_avx2_##s # endif diff --git a/sysdeps/x86_64/multiarch/memset-avx512-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-avx512-unaligned-erms.S index 47623b8ee8..7f7a8eb33c 100644 --- a/sysdeps/x86_64/multiarch/memset-avx512-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memset-avx512-unaligned-erms.S @@ -4,26 +4,14 @@ # define USE_WITH_AVX512 1 -# define VEC_SIZE 64 -# define MOV_SIZE 6 -# define RET_SIZE 1 - -# define XMM0 xmm16 -# define YMM0 ymm16 -# define VEC0 zmm16 -# define VEC(i) VEC##i - -# define VMOVU vmovdqu64 -# define VMOVA vmovdqa64 - -# define VZEROUPPER +# include "evex512-vecs.h" # define MEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ - vpbroadcastb d, %VEC0; \ + vpbroadcastb d, %VMM(0); \ movq r, %rax # define WMEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ - vpbroadcastd d, %VEC0; \ + vpbroadcastd d, %VMM(0); \ movq r, %rax # define MEMSET_VDUP_TO_VEC0_HIGH() @@ -32,8 +20,6 @@ # define WMEMSET_VDUP_TO_VEC0_HIGH() # define WMEMSET_VDUP_TO_VEC0_LOW() -# define SECTION(p) p##.evex512 - #ifndef MEMSET_SYMBOL # define MEMSET_SYMBOL(p,s) p##_avx512_##s #endif diff --git a/sysdeps/x86_64/multiarch/memset-evex-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-evex-unaligned-erms.S index ac4b2d2d50..50fd9722f2 100644 --- a/sysdeps/x86_64/multiarch/memset-evex-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memset-evex-unaligned-erms.S @@ -4,26 +4,14 @@ # define USE_WITH_EVEX 1 -# define VEC_SIZE 32 -# define MOV_SIZE 6 -# define RET_SIZE 1 - -# define XMM0 xmm16 -# define YMM0 ymm16 -# define VEC0 ymm16 -# define VEC(i) VEC##i - -# define VMOVU vmovdqu64 -# define VMOVA vmovdqa64 - -# define VZEROUPPER +# include "evex256-vecs.h" # define MEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ - vpbroadcastb d, %VEC0; \ + vpbroadcastb d, %VMM(0); \ movq r, %rax # define WMEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ - vpbroadcastd d, %VEC0; \ + vpbroadcastd d, %VMM(0); \ movq r, %rax # define MEMSET_VDUP_TO_VEC0_HIGH() @@ -32,8 +20,6 @@ # define WMEMSET_VDUP_TO_VEC0_HIGH() # define WMEMSET_VDUP_TO_VEC0_LOW() -# define SECTION(p) p##.evex - #ifndef MEMSET_SYMBOL # define MEMSET_SYMBOL(p,s) p##_evex_##s #endif diff --git a/sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S index 44f9b8888b..11c845f6c2 100644 --- a/sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memset-sse2-unaligned-erms.S @@ -26,13 +26,7 @@ # include # define USE_WITH_SSE2 1 -# define VEC_SIZE 16 -# define MOV_SIZE 3 -# define RET_SIZE 1 - -# define VEC(i) xmm##i -# define VMOVU movups -# define VMOVA movaps +# include "sse2-vecs.h" # define MEMSET_SET_VEC0_AND_SET_RETURN(d, r) \ movd d, %xmm0; \ @@ -52,8 +46,6 @@ # define WMEMSET_VDUP_TO_VEC0_HIGH() # define WMEMSET_VDUP_TO_VEC0_LOW() -# define SECTION(p) p - # ifndef MEMSET_SYMBOL # define MEMSET_SYMBOL(p,s) p##_sse2_##s # endif diff --git a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S index 905d0fa464..03de0ab907 100644 --- a/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S +++ b/sysdeps/x86_64/multiarch/memset-vec-unaligned-erms.S @@ -34,14 +34,6 @@ # define WMEMSET_CHK_SYMBOL(p,s) WMEMSET_SYMBOL(p, s) #endif -#ifndef XMM0 -# define XMM0 xmm0 -#endif - -#ifndef YMM0 -# define YMM0 ymm0 -#endif - #ifndef VZEROUPPER # if VEC_SIZE > 16 # define VZEROUPPER vzeroupper @@ -150,8 +142,8 @@ L(entry_from_wmemset): cmpq $(VEC_SIZE * 2), %rdx ja L(more_2x_vec) /* From VEC and to 2 * VEC. No branch when size == VEC_SIZE. */ - VMOVU %VEC(0), -VEC_SIZE(%rdi,%rdx) - VMOVU %VEC(0), (%rdi) + VMOVU %VMM(0), -VEC_SIZE(%rdi,%rdx) + VMOVU %VMM(0), (%rdi) VZEROUPPER_RETURN #if defined USE_MULTIARCH && IS_IN (libc) END (MEMSET_SYMBOL (__memset, unaligned)) @@ -175,19 +167,19 @@ ENTRY_P2ALIGN (MEMSET_SYMBOL (__memset, unaligned_erms), 6) cmp $(VEC_SIZE * 2), %RDX_LP ja L(stosb_more_2x_vec) /* From VEC and to 2 * VEC. No branch when size == VEC_SIZE. */ - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(0), (VEC_SIZE * -1)(%rdi, %rdx) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(0), (VEC_SIZE * -1)(%rdi, %rdx) VZEROUPPER_RETURN #endif .p2align 4,, 4 L(last_2x_vec): #ifdef USE_LESS_VEC_MASK_STORE - VMOVU %VEC(0), (VEC_SIZE * -2)(%rdi, %rdx) - VMOVU %VEC(0), (VEC_SIZE * -1)(%rdi, %rdx) + VMOVU %VMM(0), (VEC_SIZE * -2)(%rdi, %rdx) + VMOVU %VMM(0), (VEC_SIZE * -1)(%rdi, %rdx) #else - VMOVU %VEC(0), (VEC_SIZE * -2)(%rdi) - VMOVU %VEC(0), (VEC_SIZE * -1)(%rdi) + VMOVU %VMM(0), (VEC_SIZE * -2)(%rdi) + VMOVU %VMM(0), (VEC_SIZE * -1)(%rdi) #endif VZEROUPPER_RETURN @@ -221,7 +213,7 @@ L(less_vec_from_wmemset): bzhil %edx, %ecx, %ecx kmovd %ecx, %k1 # endif - vmovdqu8 %VEC(0), (%rax){%k1} + vmovdqu8 %VMM(0), (%rax){%k1} VZEROUPPER_RETURN # if defined USE_MULTIARCH && IS_IN (libc) @@ -249,8 +241,8 @@ L(stosb_more_2x_vec): and (4x, 8x] jump to target. */ L(more_2x_vec): /* Store next 2x vec regardless. */ - VMOVU %VEC(0), (%rdi) - VMOVU %VEC(0), (VEC_SIZE * 1)(%rdi) + VMOVU %VMM(0), (%rdi) + VMOVU %VMM(0), (VEC_SIZE * 1)(%rdi) /* Two different methods of setting up pointers / compare. The two @@ -278,8 +270,8 @@ L(more_2x_vec): #endif /* Store next 2x vec regardless. */ - VMOVU %VEC(0), (VEC_SIZE * 2)(%rax) - VMOVU %VEC(0), (VEC_SIZE * 3)(%rax) + VMOVU %VMM(0), (VEC_SIZE * 2)(%rax) + VMOVU %VMM(0), (VEC_SIZE * 3)(%rax) #if defined USE_WITH_EVEX || defined USE_WITH_AVX512 @@ -304,20 +296,20 @@ L(more_2x_vec): andq $(VEC_SIZE * -2), %LOOP_REG .p2align 4 L(loop): - VMOVA %VEC(0), LOOP_4X_OFFSET(%LOOP_REG) - VMOVA %VEC(0), (VEC_SIZE + LOOP_4X_OFFSET)(%LOOP_REG) - VMOVA %VEC(0), (VEC_SIZE * 2 + LOOP_4X_OFFSET)(%LOOP_REG) - VMOVA %VEC(0), (VEC_SIZE * 3 + LOOP_4X_OFFSET)(%LOOP_REG) + VMOVA %VMM(0), LOOP_4X_OFFSET(%LOOP_REG) + VMOVA %VMM(0), (VEC_SIZE + LOOP_4X_OFFSET)(%LOOP_REG) + VMOVA %VMM(0), (VEC_SIZE * 2 + LOOP_4X_OFFSET)(%LOOP_REG) + VMOVA %VMM(0), (VEC_SIZE * 3 + LOOP_4X_OFFSET)(%LOOP_REG) subq $-(VEC_SIZE * 4), %LOOP_REG cmpq %END_REG, %LOOP_REG jb L(loop) .p2align 4,, MOV_SIZE L(last_4x_vec): - VMOVU %VEC(0), LOOP_4X_OFFSET(%END_REG) - VMOVU %VEC(0), (VEC_SIZE + LOOP_4X_OFFSET)(%END_REG) - VMOVU %VEC(0), (VEC_SIZE * 2 + LOOP_4X_OFFSET)(%END_REG) - VMOVU %VEC(0), (VEC_SIZE * 3 + LOOP_4X_OFFSET)(%END_REG) -L(return): + VMOVU %VMM(0), LOOP_4X_OFFSET(%END_REG) + VMOVU %VMM(0), (VEC_SIZE + LOOP_4X_OFFSET)(%END_REG) + VMOVU %VMM(0), (VEC_SIZE * 2 + LOOP_4X_OFFSET)(%END_REG) + VMOVU %VMM(0), (VEC_SIZE * 3 + LOOP_4X_OFFSET)(%END_REG) +L(return_vzeroupper): #if VEC_SIZE > 16 ZERO_UPPER_VEC_REGISTERS_RETURN #else @@ -355,7 +347,7 @@ L(cross_page): jge L(between_16_31) #endif #ifndef USE_XMM_LESS_VEC - MOVQ %XMM0, %SET_REG64 + MOVQ %VMM_128(0), %SET_REG64 #endif cmpl $8, %edx jge L(between_8_15) @@ -374,8 +366,8 @@ L(between_0_0): .p2align 4,, SMALL_MEMSET_ALIGN(MOV_SIZE, RET_SIZE) /* From 32 to 63. No branch when size == 32. */ L(between_32_63): - VMOVU %YMM0, (%LESS_VEC_REG) - VMOVU %YMM0, -32(%LESS_VEC_REG, %rdx) + VMOVU %VMM_256(0), (%LESS_VEC_REG) + VMOVU %VMM_256(0), -32(%LESS_VEC_REG, %rdx) VZEROUPPER_RETURN #endif @@ -383,8 +375,8 @@ L(between_32_63): .p2align 4,, SMALL_MEMSET_ALIGN(MOV_SIZE, 1) L(between_16_31): /* From 16 to 31. No branch when size == 16. */ - VMOVU %XMM0, (%LESS_VEC_REG) - VMOVU %XMM0, -16(%LESS_VEC_REG, %rdx) + VMOVU %VMM_128(0), (%LESS_VEC_REG) + VMOVU %VMM_128(0), -16(%LESS_VEC_REG, %rdx) ret #endif @@ -394,8 +386,8 @@ L(between_16_31): L(between_8_15): /* From 8 to 15. No branch when size == 8. */ #ifdef USE_XMM_LESS_VEC - MOVQ %XMM0, (%rdi) - MOVQ %XMM0, -8(%rdi, %rdx) + MOVQ %VMM_128(0), (%rdi) + MOVQ %VMM_128(0), -8(%rdi, %rdx) #else movq %SET_REG64, (%LESS_VEC_REG) movq %SET_REG64, -8(%LESS_VEC_REG, %rdx) @@ -408,8 +400,8 @@ L(between_8_15): L(between_4_7): /* From 4 to 7. No branch when size == 4. */ #ifdef USE_XMM_LESS_VEC - MOVD %XMM0, (%rdi) - MOVD %XMM0, -4(%rdi, %rdx) + MOVD %VMM_128(0), (%rdi) + MOVD %VMM_128(0), -4(%rdi, %rdx) #else movl %SET_REG32, (%LESS_VEC_REG) movl %SET_REG32, -4(%LESS_VEC_REG, %rdx) diff --git a/sysdeps/x86_64/multiarch/sse2-vecs.h b/sysdeps/x86_64/multiarch/sse2-vecs.h index 2b77a59d56..3574fab0aa 100644 --- a/sysdeps/x86_64/multiarch/sse2-vecs.h +++ b/sysdeps/x86_64/multiarch/sse2-vecs.h @@ -40,8 +40,8 @@ #define VMOVA movaps #define VMOVNT movntdq -#define VEC_xmm VEC_any_xmm -#define VEC VEC_any_xmm +#define VMM_128 VMM_any_xmm +#define VMM VMM_any_xmm #endif diff --git a/sysdeps/x86_64/multiarch/vec-macros.h b/sysdeps/x86_64/multiarch/vec-macros.h index 9f3ffecede..fc5aa57edc 100644 --- a/sysdeps/x86_64/multiarch/vec-macros.h +++ b/sysdeps/x86_64/multiarch/vec-macros.h @@ -25,66 +25,66 @@ #endif /* Defines so we can use SSE2 / AVX2 / EVEX / EVEX512 encoding with same - VEC(N) values. */ -#define VEC_hi_xmm0 xmm16 -#define VEC_hi_xmm1 xmm17 -#define VEC_hi_xmm2 xmm18 -#define VEC_hi_xmm3 xmm19 -#define VEC_hi_xmm4 xmm20 -#define VEC_hi_xmm5 xmm21 -#define VEC_hi_xmm6 xmm22 -#define VEC_hi_xmm7 xmm23 -#define VEC_hi_xmm8 xmm24 -#define VEC_hi_xmm9 xmm25 -#define VEC_hi_xmm10 xmm26 -#define VEC_hi_xmm11 xmm27 -#define VEC_hi_xmm12 xmm28 -#define VEC_hi_xmm13 xmm29 -#define VEC_hi_xmm14 xmm30 -#define VEC_hi_xmm15 xmm31 + VMM(N) values. */ +#define VMM_hi_xmm0 xmm16 +#define VMM_hi_xmm1 xmm17 +#define VMM_hi_xmm2 xmm18 +#define VMM_hi_xmm3 xmm19 +#define VMM_hi_xmm4 xmm20 +#define VMM_hi_xmm5 xmm21 +#define VMM_hi_xmm6 xmm22 +#define VMM_hi_xmm7 xmm23 +#define VMM_hi_xmm8 xmm24 +#define VMM_hi_xmm9 xmm25 +#define VMM_hi_xmm10 xmm26 +#define VMM_hi_xmm11 xmm27 +#define VMM_hi_xmm12 xmm28 +#define VMM_hi_xmm13 xmm29 +#define VMM_hi_xmm14 xmm30 +#define VMM_hi_xmm15 xmm31 -#define VEC_hi_ymm0 ymm16 -#define VEC_hi_ymm1 ymm17 -#define VEC_hi_ymm2 ymm18 -#define VEC_hi_ymm3 ymm19 -#define VEC_hi_ymm4 ymm20 -#define VEC_hi_ymm5 ymm21 -#define VEC_hi_ymm6 ymm22 -#define VEC_hi_ymm7 ymm23 -#define VEC_hi_ymm8 ymm24 -#define VEC_hi_ymm9 ymm25 -#define VEC_hi_ymm10 ymm26 -#define VEC_hi_ymm11 ymm27 -#define VEC_hi_ymm12 ymm28 -#define VEC_hi_ymm13 ymm29 -#define VEC_hi_ymm14 ymm30 -#define VEC_hi_ymm15 ymm31 +#define VMM_hi_ymm0 ymm16 +#define VMM_hi_ymm1 ymm17 +#define VMM_hi_ymm2 ymm18 +#define VMM_hi_ymm3 ymm19 +#define VMM_hi_ymm4 ymm20 +#define VMM_hi_ymm5 ymm21 +#define VMM_hi_ymm6 ymm22 +#define VMM_hi_ymm7 ymm23 +#define VMM_hi_ymm8 ymm24 +#define VMM_hi_ymm9 ymm25 +#define VMM_hi_ymm10 ymm26 +#define VMM_hi_ymm11 ymm27 +#define VMM_hi_ymm12 ymm28 +#define VMM_hi_ymm13 ymm29 +#define VMM_hi_ymm14 ymm30 +#define VMM_hi_ymm15 ymm31 -#define VEC_hi_zmm0 zmm16 -#define VEC_hi_zmm1 zmm17 -#define VEC_hi_zmm2 zmm18 -#define VEC_hi_zmm3 zmm19 -#define VEC_hi_zmm4 zmm20 -#define VEC_hi_zmm5 zmm21 -#define VEC_hi_zmm6 zmm22 -#define VEC_hi_zmm7 zmm23 -#define VEC_hi_zmm8 zmm24 -#define VEC_hi_zmm9 zmm25 -#define VEC_hi_zmm10 zmm26 -#define VEC_hi_zmm11 zmm27 -#define VEC_hi_zmm12 zmm28 -#define VEC_hi_zmm13 zmm29 -#define VEC_hi_zmm14 zmm30 -#define VEC_hi_zmm15 zmm31 +#define VMM_hi_zmm0 zmm16 +#define VMM_hi_zmm1 zmm17 +#define VMM_hi_zmm2 zmm18 +#define VMM_hi_zmm3 zmm19 +#define VMM_hi_zmm4 zmm20 +#define VMM_hi_zmm5 zmm21 +#define VMM_hi_zmm6 zmm22 +#define VMM_hi_zmm7 zmm23 +#define VMM_hi_zmm8 zmm24 +#define VMM_hi_zmm9 zmm25 +#define VMM_hi_zmm10 zmm26 +#define VMM_hi_zmm11 zmm27 +#define VMM_hi_zmm12 zmm28 +#define VMM_hi_zmm13 zmm29 +#define VMM_hi_zmm14 zmm30 +#define VMM_hi_zmm15 zmm31 -#define PRIMITIVE_VEC(vec, num) vec##num +#define PRIMITIVE_VMM(vec, num) vec##num -#define VEC_any_xmm(i) PRIMITIVE_VEC(xmm, i) -#define VEC_any_ymm(i) PRIMITIVE_VEC(ymm, i) -#define VEC_any_zmm(i) PRIMITIVE_VEC(zmm, i) +#define VMM_any_xmm(i) PRIMITIVE_VMM(xmm, i) +#define VMM_any_ymm(i) PRIMITIVE_VMM(ymm, i) +#define VMM_any_zmm(i) PRIMITIVE_VMM(zmm, i) -#define VEC_hi_xmm(i) PRIMITIVE_VEC(VEC_hi_xmm, i) -#define VEC_hi_ymm(i) PRIMITIVE_VEC(VEC_hi_ymm, i) -#define VEC_hi_zmm(i) PRIMITIVE_VEC(VEC_hi_zmm, i) +#define VMM_hi_xmm(i) PRIMITIVE_VMM(VMM_hi_xmm, i) +#define VMM_hi_ymm(i) PRIMITIVE_VMM(VMM_hi_ymm, i) +#define VMM_hi_zmm(i) PRIMITIVE_VMM(VMM_hi_zmm, i) #endif