From patchwork Thu May 25 21:04:59 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Aurelien Jarno X-Patchwork-Id: 9749177 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 4067D60246 for ; Thu, 25 May 2017 21:19:26 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 278FE2835B for ; Thu, 25 May 2017 21:19:26 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 1AC4E283CD; Thu, 25 May 2017 21:19:26 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-6.9 required=2.0 tests=BAYES_00,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 869E32835B for ; Thu, 25 May 2017 21:19:25 +0000 (UTC) Received: from localhost ([::1]:33741 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dE0AG-0002dG-NU for patchwork-qemu-devel@patchwork.kernel.org; Thu, 25 May 2017 17:19:24 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:55236) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dDzwc-0005ef-MZ for qemu-devel@nongnu.org; Thu, 25 May 2017 17:05:20 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dDzwa-0001Lq-Er for qemu-devel@nongnu.org; Thu, 25 May 2017 17:05:18 -0400 Received: from hall.aurel32.net ([2001:bc8:30d7:100::1]:37428) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dDzwa-0001Kg-6D for qemu-devel@nongnu.org; Thu, 25 May 2017 17:05:16 -0400 Received: from [2001:bc8:30d7:121:cc44:2c6c:4d9c:42c0] (helo=ohm.rr44.fr) by hall.aurel32.net with esmtpsa (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.89) (envelope-from ) id 1dDzwY-0008RT-Hc; Thu, 25 May 2017 23:05:14 +0200 Received: from aurel32 by ohm.rr44.fr with local (Exim 4.89) (envelope-from ) id 1dDzwW-0001KS-M8; Thu, 25 May 2017 23:05:12 +0200 From: Aurelien Jarno To: qemu-devel@nongnu.org Date: Thu, 25 May 2017 23:04:59 +0200 Message-Id: <20170525210508.4910-18-aurelien@aurel32.net> X-Mailer: git-send-email 2.11.0 In-Reply-To: <20170525210508.4910-1-aurelien@aurel32.net> References: <20170525210508.4910-1-aurelien@aurel32.net> X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 2001:bc8:30d7:100::1 Subject: [Qemu-devel] [PATCH 17/26] target/s390x: improve MOVE LONG and MOVE LONG EXTENDED X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Alexander Graf , Aurelien Jarno , Richard Henderson Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP As MVCL and MVCLE only differ by their operands, use a common do_mvcl helper. Optimize it calling fast_memmove and fast_memset. Correctly write back addresses. Signed-off-by: Aurelien Jarno Reviewed-by: Richard Henderson --- target/s390x/mem_helper.c | 90 +++++++++++++++++++++-------------------------- 1 file changed, 40 insertions(+), 50 deletions(-) diff --git a/target/s390x/mem_helper.c b/target/s390x/mem_helper.c index aa665fe0f7..6add413531 100644 --- a/target/s390x/mem_helper.c +++ b/target/s390x/mem_helper.c @@ -621,49 +621,60 @@ void HELPER(stam)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3) } } -/* move long */ -uint32_t HELPER(mvcl)(CPUS390XState *env, uint32_t r1, uint32_t r2) +/* move long helper */ +static inline uint32_t do_mvcl(CPUS390XState *env, + uint64_t *dest, uint64_t *destlen, + uint64_t *src, uint64_t *srclen, + uint8_t pad, uintptr_t ra) { - uintptr_t ra = GETPC(); - uint64_t destlen = env->regs[r1 + 1] & 0xffffff; - uint64_t dest = get_address(env, r1); - uint64_t srclen = env->regs[r2 + 1] & 0xffffff; - uint64_t src = get_address(env, r2); - uint8_t pad = env->regs[r2 + 1] >> 24; - uint8_t v; + uint64_t len = MIN(*srclen, *destlen); uint32_t cc; - if (destlen == srclen) { + if (*destlen == *srclen) { cc = 0; - } else if (destlen < srclen) { + } else if (*destlen < *srclen) { cc = 1; } else { cc = 2; } - if (srclen > destlen) { - srclen = destlen; - } + /* Copy the src array */ + fast_memmove(env, *dest, *src, len, ra); + *src += len; + *srclen -= len; + *dest += len; + *destlen -= len; - for (; destlen && srclen; src++, dest++, destlen--, srclen--) { - v = cpu_ldub_data_ra(env, src, ra); - cpu_stb_data_ra(env, dest, v, ra); - } + /* Pad the remaining area */ + fast_memset(env, *dest, pad, *destlen, ra); + *dest += *destlen; + *destlen = 0; - for (; destlen; dest++, destlen--) { - cpu_stb_data_ra(env, dest, pad, ra); - } + return cc; +} - env->regs[r1 + 1] = destlen; - /* can't use srclen here, we trunc'ed it */ - env->regs[r2 + 1] -= src - env->regs[r2]; +/* move long */ +uint32_t HELPER(mvcl)(CPUS390XState *env, uint32_t r1, uint32_t r2) +{ + uintptr_t ra = GETPC(); + uint64_t destlen = env->regs[r1 + 1] & 0xffffff; + uint64_t dest = get_address(env, r1); + uint64_t srclen = env->regs[r2 + 1] & 0xffffff; + uint64_t src = get_address(env, r2); + uint8_t pad = env->regs[r2 + 1] >> 24; + uint32_t cc; + + cc = do_mvcl(env, &dest, &destlen, &src, &srclen, pad, ra); + + env->regs[r1 + 1] = deposit64(env->regs[r1 + 1], 0, 24, destlen); + env->regs[r2 + 1] = deposit64(env->regs[r2 + 1], 0, 24, srclen); set_address(env, r1, dest); set_address(env, r2, src); return cc; } -/* move long extended another memcopy insn with more bells and whistles */ +/* move long extended */ uint32_t HELPER(mvcle)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint32_t r3) { @@ -672,34 +683,13 @@ uint32_t HELPER(mvcle)(CPUS390XState *env, uint32_t r1, uint64_t a2, uint64_t dest = get_address(env, r1); uint64_t srclen = get_length(env, r3 + 1); uint64_t src = get_address(env, r3); - uint8_t pad = a2 & 0xff; - uint8_t v; + uint8_t pad = a2; uint32_t cc; - if (destlen == srclen) { - cc = 0; - } else if (destlen < srclen) { - cc = 1; - } else { - cc = 2; - } - - if (srclen > destlen) { - srclen = destlen; - } - - for (; destlen && srclen; src++, dest++, destlen--, srclen--) { - v = cpu_ldub_data_ra(env, src, ra); - cpu_stb_data_ra(env, dest, v, ra); - } - - for (; destlen; dest++, destlen--) { - cpu_stb_data_ra(env, dest, pad, ra); - } + cc = do_mvcl(env, &dest, &destlen, &src, &srclen, pad, ra); - set_length(env, r1 + 1 , destlen); - /* can't use srclen here, we trunc'ed it */ - set_length(env, r3 + 1, env->regs[r3 + 1] - src - env->regs[r3]); + set_length(env, r1 + 1, destlen); + set_length(env, r3 + 1, srclen); set_address(env, r1, dest); set_address(env, r3, src);