From patchwork Wed Dec 11 06:24:41 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: zhichang.yuan@linaro.org X-Patchwork-Id: 3322451 Return-Path: X-Original-To: patchwork-linux-arm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 8EAD4C0D4A for ; Wed, 11 Dec 2013 06:28:36 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id E43AE20412 for ; Wed, 11 Dec 2013 06:28:34 +0000 (UTC) Received: from casper.infradead.org (casper.infradead.org [85.118.1.10]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id C9231205E5 for ; Wed, 11 Dec 2013 06:28:32 +0000 (UTC) Received: from merlin.infradead.org ([2001:4978:20e::2]) by casper.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1VqdG4-0003Ms-MU; Wed, 11 Dec 2013 06:26:57 +0000 Received: from localhost ([::1] helo=merlin.infradead.org) by merlin.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1VqdFZ-0000op-Nd; Wed, 11 Dec 2013 06:26:25 +0000 Received: from mail-pd0-f173.google.com ([209.85.192.173]) by merlin.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1VqdFR-0000lx-CW for linux-arm-kernel@lists.infradead.org; Wed, 11 Dec 2013 06:26:23 +0000 Received: by mail-pd0-f173.google.com with SMTP id p10so8864114pdj.18 for ; Tue, 10 Dec 2013 22:25:53 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=6LjSuJOI5uyD0v0LT/hReyHkXeH0fWurrI64Aex+EcI=; b=a5qm3IasXkaSusah8cxFCmafhJqo/w7CTbMWU1SrDi2abZHbjRPthXQRUdXShoP366 7cLEGIc3qhVcjJY+dw+2t0GWonS3barrmNlMXJ21vblQXf6+iGMnkGqwd7fzMUrsjXYN vbP+pTW+nEdSh4FRfd6BGkVakPzfL1VrSGQm0H1G//5zkHhg899Ygam8+1fw/mVdOQ16 M4eCzpNp2Xc/aKXFQ/DLPWBdVHcWmhlJfZAL1VyGbv4EFVifyVVsoYJG9sJ8kCsqi14B PvAat3Nhok1BG51imRTgjO1MWm60I5M8f+eUAmJTtA2dbKEbmxl5QrNei6gCNbQ9jq25 PcFg== X-Gm-Message-State: ALoCoQltwl8g4uPhSnMdF0I48UuusjI54fJllS/ONQdxfbPE7Yqzb9GN4tpbdXEnO8u1bsPCIUpg X-Received: by 10.69.31.97 with SMTP id kl1mr32213554pbd.127.1386743151973; Tue, 10 Dec 2013 22:25:51 -0800 (PST) Received: from localhost ([58.251.159.252]) by mx.google.com with ESMTPSA id xs1sm41633913pac.7.2013.12.10.22.25.48 for (version=TLSv1.2 cipher=RC4-SHA bits=128/128); Tue, 10 Dec 2013 22:25:51 -0800 (PST) From: zhichang.yuan@linaro.org To: linux-arm-kernel@lists.infradead.org, catalin.marinas@arm.com, will.deacon@arm.com Subject: [PATCH 5/6] arm64: lib: Implement optimized string compare routines Date: Wed, 11 Dec 2013 14:24:41 +0800 Message-Id: <1386743082-5231-6-git-send-email-zhichang.yuan@linaro.org> X-Mailer: git-send-email 1.7.9.5 In-Reply-To: <1386743082-5231-1-git-send-email-zhichang.yuan@linaro.org> References: <1386743082-5231-1-git-send-email-zhichang.yuan@linaro.org> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20131211_012617_969292_145E1982 X-CRM114-Status: GOOD ( 24.62 ) X-Spam-Score: -1.9 (-) Cc: Deepak Saxena , liguozhu@huawei.com, "zhichang.yuan" X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Spam-Status: No, score=-4.4 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: "zhichang.yuan" This patch, based on Linaro's Cortex Strings library, adds an assembly optimized strcmp() and strncmp() functions. Signed-off-by: Zhichang Yuan Signed-off-by: Deepak Saxena --- arch/arm64/include/asm/string.h | 6 + arch/arm64/kernel/arm64ksyms.c | 2 + arch/arm64/lib/Makefile | 2 +- arch/arm64/lib/strcmp.S | 256 +++++++++++++++++++++++++++++ arch/arm64/lib/strncmp.S | 340 +++++++++++++++++++++++++++++++++++++++ 5 files changed, 605 insertions(+), 1 deletion(-) create mode 100644 arch/arm64/lib/strcmp.S create mode 100644 arch/arm64/lib/strncmp.S diff --git a/arch/arm64/include/asm/string.h b/arch/arm64/include/asm/string.h index 3a43305..6133f49 100644 --- a/arch/arm64/include/asm/string.h +++ b/arch/arm64/include/asm/string.h @@ -22,6 +22,12 @@ extern char *strrchr(const char *, int c); #define __HAVE_ARCH_STRCHR extern char *strchr(const char *, int c); +#define __HAVE_ARCH_STRCMP +extern int strcmp(const char *, const char *); + +#define __HAVE_ARCH_STRNCMP +extern int strncmp(const char *, const char *, __kernel_size_t); + #define __HAVE_ARCH_MEMCPY extern void *memcpy(void *, const void *, __kernel_size_t); diff --git a/arch/arm64/kernel/arm64ksyms.c b/arch/arm64/kernel/arm64ksyms.c index af02a25..d61fa39 100644 --- a/arch/arm64/kernel/arm64ksyms.c +++ b/arch/arm64/kernel/arm64ksyms.c @@ -47,6 +47,8 @@ EXPORT_SYMBOL(memstart_addr); /* string / mem functions */ EXPORT_SYMBOL(strchr); EXPORT_SYMBOL(strrchr); +EXPORT_SYMBOL(strcmp); +EXPORT_SYMBOL(strncmp); EXPORT_SYMBOL(memset); EXPORT_SYMBOL(memcpy); EXPORT_SYMBOL(memmove); diff --git a/arch/arm64/lib/Makefile b/arch/arm64/lib/Makefile index a6a8d3d..5ded21f 100644 --- a/arch/arm64/lib/Makefile +++ b/arch/arm64/lib/Makefile @@ -3,4 +3,4 @@ lib-y := bitops.o delay.o \ copy_from_user.o copy_to_user.o copy_in_user.o \ copy_page.o clear_page.o \ memchr.o memcpy.o memmove.o memset.o memcmp.o \ - strchr.o strrchr.o + strchr.o strrchr.o strcmp.o strncmp.o diff --git a/arch/arm64/lib/strcmp.S b/arch/arm64/lib/strcmp.S new file mode 100644 index 0000000..e07ea0d --- /dev/null +++ b/arch/arm64/lib/strcmp.S @@ -0,0 +1,256 @@ +/* + * Copyright (C) 2013 ARM Ltd. + * Copyright (C) 2013 Linaro. + * + * This code is based on glibc cortex strings work originally authored by Linaro + * and re-licensed under GPLv2 for the Linux kernel. The original code can + * be found @ + * + * http://bazaar.launchpad.net/~linaro-toolchain-dev/cortex-strings/trunk/ + * files/head:/src/aarch64/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#include +#include + +/* + * compare two strings + * + * Parameters: + * x0 - const string 1 pointer + * x1 - const string 2 pointer + * Returns: + * x0 - an integer less than, equal to, or greater than zero + * if s1 is found, respectively, to be less than, to match, + * or be greater than s2. + */ + +#define REP8_01 0x0101010101010101 +#define REP8_7f 0x7f7f7f7f7f7f7f7f +#define REP8_80 0x8080808080808080 + +/* Parameters and result. */ +#define src1 x0 +#define src2 x1 +#define result x0 + +/* Internal variables. */ +#define data1 x2 +#define data1w w2 +#define data2 x3 +#define data2w w3 +#define has_nul x4 +#define diff x5 +#define syndrome x6 +#define tmp1 x7 +#define tmp2 x8 +#define tmp3 x9 +#define zeroones x10 +#define pos x11 + +ENTRY(strcmp) + eor tmp1, src1, src2 + mov zeroones, #REP8_01 + tst tmp1, #7 + b.ne .Lmisaligned8 + ands tmp1, src1, #7 + b.ne .Lmutual_align + + /* + * NUL detection works on the principle that (X - 1) & (~X) & 0x80 + * (=> (X - 1) & ~(X | 0x7f)) is non-zero iff a byte is zero, and + * can be done in parallel across the entire word. + */ +.Lloop_aligned: + ldr data1, [src1], #8 + ldr data2, [src2], #8 +.Lstart_realigned: + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + eor diff, data1, data2 /* Non-zero if differences found. */ + bic has_nul, tmp1, tmp2 /* Non-zero if NUL terminator. */ + orr syndrome, diff, has_nul + cbz syndrome, .Lloop_aligned + b .Lcal_cmpresult + +.Lmutual_align: + /* + * Sources are mutually aligned, but are not currently at an + * alignment boundary. Round down the addresses and then mask off + * the bytes that preceed the start point. + */ + bic src1, src1, #7 + bic src2, src2, #7 + lsl tmp1, tmp1, #3 /* Bytes beyond alignment -> bits. */ + ldr data1, [src1], #8 + neg tmp1, tmp1 /* Bits to alignment -64. */ + ldr data2, [src2], #8 + mov tmp2, #~0 +#ifdef __ARM64EB__ + /* Big-endian. Early bytes are at MSB. */ + lsl tmp2, tmp2, tmp1 /* Shift (tmp1 & 63). */ +#else + /* Little-endian. Early bytes are at LSB. */ + lsr tmp2, tmp2, tmp1 /* Shift (tmp1 & 63). */ +#endif + orr data1, data1, tmp2 + orr data2, data2, tmp2 + b .Lstart_realigned + +.Lmisaligned8: + /* + * Get the align offset length to compare per byte first. + * After this process, one string's address will be aligned. + */ + and tmp1, src1, #7 + neg tmp1, tmp1 + add tmp1, tmp1, #8 + and tmp2, src2, #7 + neg tmp2, tmp2 + add tmp2, tmp2, #8 + subs tmp3, tmp1, tmp2 + csel pos, tmp1, tmp2, hi /*Choose the maximum. */ +.Ltinycmp: + ldrb data1w, [src1], #1 + ldrb data2w, [src2], #1 + subs pos, pos, #1 + ccmp data1w, #1, #0, ne /* NZCV = 0b0000. */ + ccmp data1w, data2w, #0, cs /* NZCV = 0b0000. */ + b.eq .Ltinycmp + cbnz pos, 1f /*find the null or unequal...*/ + cmp data1w, #1 + ccmp data1w, data2w, #0, cs + b.eq .Lstart_align /*the last bytes are equal....*/ +1: + sub result, data1, data2 + ret + +.Lstart_align: + ands xzr, src1, #7 + /* + * eq means tmp1 src1 is aligned now; + * tmp3 is positive for this branch. + */ + b.eq .Lrecal_offset + add src1, src1, tmp3 + add src2, src2, tmp3 + ldr data1, [src1], #8 + ldr data2, [src2], #8 + + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + bic has_nul, tmp1, tmp2 + eor diff, data1, data2 /* Non-zero if differences found. */ + orr syndrome, diff, has_nul + cbnz syndrome, .Lcal_cmpresult + and tmp3, tmp3, #7 /*tmp3 = 8 + tmp3 ( old tmp3 is negative)*/ + /* + * src1 is aligned and src1 is in the right of src2. + * start to compare the next 8 bytes of two strings. + */ +.Lrecal_offset: + neg pos, tmp3 +.Lloopcmp_proc: + /* + * Fall back pos bytes, get the first bytes segment of one Dword of src1. + * pos is negative here. We also can use : + * ldr data1, [src1] + * ldr data2, [src2, pos] + * These two instructions will read data with aligned address. + * But if we adapt this method, have to add some shift and mask out + * some bits from these two Dwords to construct two new Dwords. + * Some more instructions will be added,and most important, + * it will need more time cost. + */ + ldr data1, [src1,pos] + ldr data2, [src2,pos] + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + bic has_nul, tmp1, tmp2 + eor diff, data1, data2 /* Non-zero if differences found. */ + orr syndrome, diff, has_nul + cbnz syndrome, .Lcal_cmpresult + + /*The second part process*/ + ldr data1, [src1], #8 + ldr data2, [src2], #8 + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + bic has_nul, tmp1, tmp2 + eor diff, data1, data2 /* Non-zero if differences found. */ + orr syndrome, diff, has_nul + cbz syndrome, .Lloopcmp_proc + +.Lcal_cmpresult: +#ifndef __ARM64EB__ + rev syndrome, syndrome + rev data1, data1 + /* + * The MS-non-zero bit of the syndrome marks either the first bit + * that is different, or the top bit of the first zero byte. + * Shifting left now will bring the critical information into the + * top bits. + */ + clz pos, syndrome + rev data2, data2 + lsl data1, data1, pos + lsl data2, data2, pos + /* + * But we need to zero-extend (char is unsigned) the value and then + * perform a signed 32-bit subtraction. + */ + lsr data1, data1, #56 + sub result, data1, data2, lsr #56 + ret +#else + /* + * For big-endian we cannot use the trick with the syndrome value + * as carry-propagation can corrupt the upper bits if the trailing + * bytes in the string contain 0x01. + * However, if there is no NUL byte in the dword, we can generate + * the result directly. We can't just subtract the bytes as the + * MSB might be significant. + */ + cbnz has_nul, 1f + cmp data1, data2 + cset result, ne + cneg result, result, lo + ret +1: + /*Re-compute the NUL-byte detection, using a byte-reversed value. */ + rev tmp3, data1 + sub tmp1, tmp3, zeroones + orr tmp2, tmp3, #REP8_7f + bic has_nul, tmp1, tmp2 + rev has_nul, has_nul + orr syndrome, diff, has_nul + clz pos, syndrome + /* + * The MS-non-zero bit of the syndrome marks either the first bit + * that is different, or the top bit of the first zero byte. + * Shifting left now will bring the critical information into the + * top bits. + */ + lsl data1, data1, pos + lsl data2, data2, pos + /* + * But we need to zero-extend (char is unsigned) the value and then + * perform a signed 32-bit subtraction. + */ + lsr data1, data1, #56 + sub result, data1, data2, lsr #56 + ret +#endif +ENDPROC(strcmp) diff --git a/arch/arm64/lib/strncmp.S b/arch/arm64/lib/strncmp.S new file mode 100644 index 0000000..e883fd3 --- /dev/null +++ b/arch/arm64/lib/strncmp.S @@ -0,0 +1,340 @@ +/* + * Copyright (C) 2013 ARM Ltd. + * Copyright (C) 2013 Linaro. + * + * This code is based on glibc cortex strings work originally authored by Linaro + * and re-licensed under GPLv2 for the Linux kernel. The original code can + * be found @ + * + * http://bazaar.launchpad.net/~linaro-toolchain-dev/cortex-strings/trunk/ + * files/head:/src/aarch64/ + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see . + */ + +#include +#include + +/* + * compare two strings + * + * Parameters: + * x0 - const string 1 pointer + * x1 - const string 2 pointer + * x2 - the maximal length to be compared + * Returns: + * x0 - an integer less than, equal to, or greater than zero if s1 is found, + * respectively, to be less than, to match, or be greater than s2. + */ + +#define REP8_01 0x0101010101010101 +#define REP8_7f 0x7f7f7f7f7f7f7f7f +#define REP8_80 0x8080808080808080 + +/* Parameters and result. */ +#define src1 x0 +#define src2 x1 +#define limit x2 +#define result x0 + +/* Internal variables. */ +#define data1 x3 +#define data1w w3 +#define data2 x4 +#define data2w w4 +#define has_nul x5 +#define diff x6 +#define syndrome x7 +#define tmp1 x8 +#define tmp2 x9 +#define tmp3 x10 +#define zeroones x11 +#define pos x12 +#define limit_wd x13 +#define mask x14 +#define endloop x15 + +ENTRY(strncmp) + cbz limit, .Lret0 + eor tmp1, src1, src2 + mov zeroones, #REP8_01 + tst tmp1, #7 + b.ne .Lmisaligned8 + ands tmp1, src1, #7 + b.ne .Lmutual_align + /* Calculate the number of full and partial words -1. */ + /* + * when limit is mulitply of 8, if not sub 1, + * the judgement of last dword will wrong. + */ + sub limit_wd, limit, #1 /* limit != 0, so no underflow. */ + lsr limit_wd, limit_wd, #3 /* Convert to Dwords. */ + + /* + * NUL detection works on the principle that (X - 1) & (~X) & 0x80 + * (=> (X - 1) & ~(X | 0x7f)) is non-zero iff a byte is zero, and + * can be done in parallel across the entire word. + */ +.Lloop_aligned: + ldr data1, [src1], #8 + ldr data2, [src2], #8 +.Lstart_realigned: + subs limit_wd, limit_wd, #1 + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + eor diff, data1, data2 /* Non-zero if differences found. */ + csinv endloop, diff, xzr, pl /* Last Dword or differences.*/ + bics has_nul, tmp1, tmp2 /* Non-zero if NUL terminator. */ + ccmp endloop, #0, #0, eq + b.eq .Lloop_aligned + + /*Not reached the limit, must have found the end or a diff. */ + tbz limit_wd, #63, .Lnot_limit + + /* Limit % 8 == 0 => all bytes significant. */ + ands limit, limit, #7 + b.eq .Lnot_limit + + lsl limit, limit, #3 /* Bits -> bytes. */ + mov mask, #~0 +#ifdef __ARM64EB__ + lsr mask, mask, limit +#else + lsl mask, mask, limit +#endif + bic data1, data1, mask + bic data2, data2, mask + + /* Make sure that the NUL byte is marked in the syndrome. */ + orr has_nul, has_nul, mask + +.Lnot_limit: + orr syndrome, diff, has_nul + b .Lcal_cmpresult + +.Lmutual_align: + /* + * Sources are mutually aligned, but are not currently at an + * alignment boundary. Round down the addresses and then mask off + * the bytes that precede the start point. + * We also need to adjust the limit calculations, but without + * overflowing if the limit is near ULONG_MAX. + */ + bic src1, src1, #7 + bic src2, src2, #7 + ldr data1, [src1], #8 + neg tmp3, tmp1, lsl #3 /* 64 - bits(bytes beyond align). */ + ldr data2, [src2], #8 + mov tmp2, #~0 + sub limit_wd, limit, #1 /* limit != 0, so no underflow. */ +#ifdef __ARM64EB__ + /* Big-endian. Early bytes are at MSB. */ + lsl tmp2, tmp2, tmp3 /* Shift (tmp1 & 63). */ +#else + /* Little-endian. Early bytes are at LSB. */ + lsr tmp2, tmp2, tmp3 /* Shift (tmp1 & 63). */ +#endif + and tmp3, limit_wd, #7 + lsr limit_wd, limit_wd, #3 + /* Adjust the limit. Only low 3 bits used, so overflow irrelevant.*/ + add limit, limit, tmp1 + add tmp3, tmp3, tmp1 + orr data1, data1, tmp2 + orr data2, data2, tmp2 + add limit_wd, limit_wd, tmp3, lsr #3 + b .Lstart_realigned + +/*when src1 offset is not equal to src2 offset...*/ +.Lmisaligned8: + cmp limit, #8 + b.lo .Ltiny8proc /*limit < 8... */ + /* + * Get the align offset length to compare per byte first. + * After this process, one string's address will be aligned.*/ + and tmp1, src1, #7 + neg tmp1, tmp1 + add tmp1, tmp1, #8 + and tmp2, src2, #7 + neg tmp2, tmp2 + add tmp2, tmp2, #8 + subs tmp3, tmp1, tmp2 + csel pos, tmp1, tmp2, hi /*Choose the maximum. */ + /* + * Here, limit is not less than 8, so directly run .Ltinycmp + * without checking the limit.*/ + sub limit, limit, pos +.Ltinycmp: + ldrb data1w, [src1], #1 + ldrb data2w, [src2], #1 + subs pos, pos, #1 + ccmp data1w, #1, #0, ne /* NZCV = 0b0000. */ + ccmp data1w, data2w, #0, cs /* NZCV = 0b0000. */ + b.eq .Ltinycmp + cbnz pos, 1f /*find the null or unequal...*/ + cmp data1w, #1 + ccmp data1w, data2w, #0, cs + b.eq .Lstart_align /*the last bytes are equal....*/ +1: + sub result, data1, data2 + ret + +.Lstart_align: + lsr limit_wd, limit, #3 + cbz limit_wd, .Lremain8 + ands xzr, src1, #7 + /*eq means src1 is aligned now,tmp3 is positive in this branch.*/ + b.eq .Lrecal_offset + add src1, src1, tmp3 + add src2, src2, tmp3 + ldr data1, [src1], #8 + ldr data2, [src2], #8 + /* + * since tmp3 is negative and limit is not less than 8,so cbz is not + * needed.. + */ + sub limit, limit, tmp3 + lsr limit_wd, limit, #3 + subs limit_wd, limit_wd, #1 + + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + eor diff, data1, data2 /* Non-zero if differences found. */ + csinv endloop, diff, xzr, ne/*if limit_wd is 0,will finish the cmp*/ + bics has_nul, tmp1, tmp2 + ccmp endloop, #0, #0, eq /*has_null is ZERO: no null byte*/ + b.ne .Lunequal_proc + and tmp3, tmp3, #7 /*tmp3 = 8 + tmp3 ( old tmp3 is negative)*/ + /* + * src1 is aligned and src1 is in the right of src2. + * start the next 8 bytes compare.. + */ +.Lrecal_offset: + neg pos, tmp3 +.Lloopcmp_proc: + /* + * Fall back pos bytes, get the first bytes segment of one Dword of src1. + * pos is negative here. We also can use : + * ldr data1, [src1] + * ldr data2, [src2, pos] + * These two instructions will read data with aligned address. + * But if we adapt this method, have to add some shift and mask out + * some bits from these two Dword to construct two new Dwords. + * Some more instructions will be added, + * and most important, it will need more time cost. + */ + ldr data1, [src1,pos] + ldr data2, [src2,pos] + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + bics has_nul, tmp1, tmp2 /* Non-zero if NUL terminator. */ + eor diff, data1, data2 /* Non-zero if differences found. */ + csinv endloop, diff, xzr, eq + cbnz endloop, .Lunequal_proc + + /*The second part process*/ + ldr data1, [src1], #8 + ldr data2, [src2], #8 + subs limit_wd, limit_wd, #1 + sub tmp1, data1, zeroones + orr tmp2, data1, #REP8_7f + eor diff, data1, data2 /* Non-zero if differences found. */ + csinv endloop, diff, xzr, ne/*if limit_wd is 0,will finish the cmp*/ + bics has_nul, tmp1, tmp2 + ccmp endloop, #0, #0, eq /*has_null is ZERO: no null byte*/ + b.eq .Lloopcmp_proc + +.Lunequal_proc: + orr syndrome, diff, has_nul + cbz syndrome, .Lremain8 +.Lcal_cmpresult: +#ifndef __ARM64EB__ + rev syndrome, syndrome + rev data1, data1 + /* + * The MS-non-zero bit of the syndrome marks either the first bit + * that is different, or the top bit of the first zero byte. + * Shifting left now will bring the critical information into the + * top bits. + */ + clz pos, syndrome + rev data2, data2 + lsl data1, data1, pos + lsl data2, data2, pos + /* But we need to zero-extend (char is unsigned) the value and then + perform a signed 32-bit subtraction. */ + lsr data1, data1, #56 + sub result, data1, data2, lsr #56 + ret +#else + /* + * For big-endian we cannot use the trick with the syndrome value + * as carry-propagation can corrupt the upper bits if the trailing + * bytes in the string contain 0x01. + * However, if there is no NUL byte in the dword, we can generate + * the result directly. We can't just subtract the bytes as the + * MSB might be significant. + */ + cbnz has_nul, 1f + cmp data1, data2 + cset result, ne + cneg result, result, lo + ret +1: + /* Re-compute the NUL-byte detection, using a byte-reversed value.*/ + rev tmp3, data1 + sub tmp1, tmp3, zeroones + orr tmp2, tmp3, #REP8_7f + bic has_nul, tmp1, tmp2 + rev has_nul, has_nul + orr syndrome, diff, has_nul + clz pos, syndrome + /* + * The MS-non-zero bit of the syndrome marks either the first bit + * that is different, or the top bit of the first zero byte. + * Shifting left now will bring the critical information into the + * top bits. + */ + lsl data1, data1, pos + lsl data2, data2, pos + /* + * But we need to zero-extend (char is unsigned) the value and then + * perform a signed 32-bit subtraction. + */ + lsr data1, data1, #56 + sub result, data1, data2, lsr #56 + ret +#endif + +.Lremain8: + /* Limit % 8 == 0 => all bytes significant. */ + ands limit, limit, #7 + b.eq .Lret0 +.Ltiny8proc: + /* Perhaps we can do better than this. */ + ldrb data1w, [src1], #1 + ldrb data2w, [src2], #1 + subs limit, limit, #1 + /* + * nz satisfied means current limit > 0. + * Z=1 will make cs =0, lead to next ccmp use ZERO to set flags + */ + ccmp data1w, #1, #0, ne /* NZCV = 0b0000. */ + ccmp data1w, data2w, #0, cs /* NZCV = 0b0000. */ + b.eq .Ltiny8proc + sub result, data1, data2 + ret + +.Lret0: + mov result, #0 + ret +ENDPROC(strncmp)