Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from lists1p.gnu.org (lists1p.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 1D1E6FF885D for ; Sat, 25 Apr 2026 11:38:08 +0000 (UTC) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists1p.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1wGbK2-0002zu-RQ; Sat, 25 Apr 2026 07:37:21 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists1p.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1wGbJm-0002z9-33; Sat, 25 Apr 2026 07:37:03 -0400 Received: from mx0a-001b2d01.pphosted.com ([148.163.156.1]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1wGbJj-0003A1-QE; Sat, 25 Apr 2026 07:37:01 -0400 Received: from pps.filterd (m0353729.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.18.1.11/8.18.1.11) with ESMTP id 63P9vtxN2911456; Sat, 25 Apr 2026 11:36:48 GMT DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ibm.com; h=cc :content-transfer-encoding:content-type:date:from:in-reply-to :message-id:mime-version:references:subject:to; s=pp1; bh=pjTvNP if4KgmL+OkOT7lk8Yh4hQL082IyZ4H5HIUJFU=; b=ePAQg5/tC7IGLSqnPeRKO3 Q8IWydvDr9MncKTYWI8Eg+Wp60GHLRJ7c5iMhYqGeHaG+FUTCNALm7dFcYQxJGBe fg19w8f37OqOzH2c/JnQagsKiARuR4Xxz+BInJzFUbPl4VLG4ygffybsErMYYIcC ub2i9kthh+K01PxhPEOK00WstE4iNfU363UR2+mnx4crDKTP/i+QUA3LvSupZLi4 9tVwgtqhN3KlG/pLt818lkhvUCPKSkQnNEqAKurlWtSlq6+IJIwxn3mGuuH/Wj2x AupTCgUbVj6PF45oU1W5D+f1j2QRKAeX3SY8mFvEUOSAaoJGbBmHaKe+qt9jcQDg == Received: from ppma22.wdc07v.mail.ibm.com (5c.69.3da9.ip4.static.sl-reverse.com [169.61.105.92]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 4drn9qs5ce-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Sat, 25 Apr 2026 11:36:47 +0000 (GMT) Received: from pps.filterd (ppma22.wdc07v.mail.ibm.com [127.0.0.1]) by ppma22.wdc07v.mail.ibm.com (8.18.1.7/8.18.1.7) with ESMTP id 63PBO6kw006542; Sat, 25 Apr 2026 11:36:46 GMT Received: from smtprelay07.fra02v.mail.ibm.com ([9.218.2.229]) by ppma22.wdc07v.mail.ibm.com (PPS) with ESMTPS id 4drmabh8mt-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Sat, 25 Apr 2026 11:36:46 +0000 (GMT) Received: from smtpav04.fra02v.mail.ibm.com (smtpav04.fra02v.mail.ibm.com [10.20.54.103]) by smtprelay07.fra02v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 63PBagQV41615618 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Sat, 25 Apr 2026 11:36:42 GMT Received: from smtpav04.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id CC4DF20043; Sat, 25 Apr 2026 11:36:42 +0000 (GMT) Received: from smtpav04.fra02v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 5219E20040; Sat, 25 Apr 2026 11:36:42 +0000 (GMT) Received: from [0.0.0.0] (unknown [9.87.136.63]) by smtpav04.fra02v.mail.ibm.com (Postfix) with ESMTP; Sat, 25 Apr 2026 11:36:42 +0000 (GMT) Message-ID: Date: Sat, 25 Apr 2026 13:36:42 +0200 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH v3] target/s390x: Leverage 128-bit vector type in V{L, ST}L opcodes To: =?UTF-8?Q?Philippe_Mathieu-Daud=C3=A9?= , qemu-devel@nongnu.org Cc: Eric Farman , qemu-s390x@nongnu.org, Richard Henderson , David Hildenbrand , Matthew Rosato , Cornelia Huck References: <20260424104325.81484-1-philmd@linaro.org> <56bfecd4-6169-4512-b019-2c9d6367525e@linux.ibm.com> Content-Language: en-US From: Ilya Leoshkevich In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit X-TM-AS-GCONF: 00 X-Proofpoint-GUID: 2T-XlAE-V-rE_t3H0MG3MCzSys2oLuh6 X-Proofpoint-Spam-Details-Enc: AW1haW4tMjYwNDI1MDExMyBTYWx0ZWRfX37lXEggRR+6b SvmeqZzyRsDJB0vFaspqrQK4l6bYiorg3ysOTQl70LEiEBfjyFWHYMNImamZEsEQQGDKb7AwZOD jrXvP109c7aSmfPnNQR9AkBhyx7/0i9RQqsBVmrkng/Scd2fWmgyM9MqNJl8LcqvbW683ISnBUe cGiaggaLB9oYLRvpwt3rrLbT4bamWE29kA8jZyIK/VT0YKRdfUl/tIi+12y8takvP66V066EjdJ LE9/eGir7dReXhbWOWje54+mPTsis53kdCXI9CXBVqLdr97FdEDZyhQ9rIr/tNs+o7i53nLyevV H7oH4Nt0xWEbZ3UmGxSXYplI0Ag7z3/WOfDJEWyExYDTbmB3XjjfPgY/SmOH32B00I8QlSr0yhm 5gEPO48Mo7lyAX4duxhouCMzxE4ae88nkNb0vihvvgM6T3COuxmF/tKvOuQiTXRzFIU7QyUziet b7bccylPZ9gTBxtnH8Q== X-Authority-Analysis: v=2.4 cv=Kc7idwYD c=1 sm=1 tr=0 ts=69eca750 cx=c_pps a=5BHTudwdYE3Te8bg5FgnPg==:117 a=5BHTudwdYE3Te8bg5FgnPg==:17 a=IkcTkHD0fZMA:10 a=A5OVakUREuEA:10 a=VkNPw1HP01LnGYTKEx00:22 a=RnoormkPH1_aCDwRdu11:22 a=uAbxVGIbfxUO_5tXvNgY:22 a=KKAkSRfTAAAA:8 a=gbpRu0Ca7s765whaVOgA:9 a=3ZKOabzyN94A:10 a=QEXdDO2ut3YA:10 a=cvBusfyB2V15izCimMoJ:22 X-Proofpoint-ORIG-GUID: 2T-XlAE-V-rE_t3H0MG3MCzSys2oLuh6 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.293,Aquarius:18.0.1143,Hydra:6.1.51,FMLib:17.12.100.49 definitions=2026-04-25_02,2026-04-21_02,2025-10-01_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 clxscore=1015 phishscore=0 bulkscore=0 adultscore=0 spamscore=0 malwarescore=0 impostorscore=0 priorityscore=1501 lowpriorityscore=0 suspectscore=0 classifier=typeunknown authscore=0 authtc= authcc= route=outbound adjust=0 reason=mlx scancount=1 engine=8.22.0-2604200000 definitions=main-2604250113 Received-SPF: pass client-ip=148.163.156.1; envelope-from=iii@linux.ibm.com; helo=mx0a-001b2d01.pphosted.com X-Spam_score_int: -26 X-Spam_score: -2.7 X-Spam_bar: -- X-Spam_report: (-2.7 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H4=0.001, RCVD_IN_MSPIKE_WL=0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: qemu development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org On 4/25/26 11:09, Philippe Mathieu-Daudé wrote: > On 25/4/26 00:05, Ilya Leoshkevich wrote: >> >> On 4/24/26 11:43, Philippe Mathieu-Daudé wrote: >>> The access is unaligned (MO_UNALN), we can easily replace the 2 >>> consecutive 64-bit calls by a single 128-bit one. >>> Considering atomicity of the new 128-bit access, the both opcodes >>> VLL (Vector Load with Length) and VSTL (Vector STore with Length) >>> access byte granularity, so use MO_ATOM_NONE. >>> >>> Signed-off-by: Philippe Mathieu-Daudé >>> --- >>> Based-on: <20260423135035.50126-1-philmd@linaro.org> >>> >>> v3: >>> - do not increase addr (Ilya) >>> - MO_ATOM_NONE (Richard) >>> --- >>>   target/s390x/tcg/vec.h        | 12 ++++++++++++ >>>   target/s390x/tcg/vec_helper.c | 17 ++++++----------- >>>   2 files changed, 18 insertions(+), 11 deletions(-) >>> >>> diff --git a/target/s390x/tcg/vec.h b/target/s390x/tcg/vec.h >>> index 8d095efcfc6..34413f7e029 100644 >>> --- a/target/s390x/tcg/vec.h >>> +++ b/target/s390x/tcg/vec.h >>> @@ -13,8 +13,10 @@ >>>   #define S390X_VEC_H >>>   #include "tcg/tcg.h" >>> +#include "qemu/int128.h" >>>   typedef union S390Vector { >>> +    Int128 quadword; >>>       uint64_t doubleword[2]; >>>       uint32_t word[4]; >>>       uint16_t halfword[8]; >>> @@ -72,6 +74,11 @@ static inline uint64_t >>> s390_vec_read_element64(const S390Vector *v, uint8_t enr) >>>       return v->doubleword[enr]; >>>   } >>> +static inline Int128 s390_vec_read_element128(const S390Vector *v) >>> +{ >>> +    return v->quadword; >>> +} >>> + >>>   static inline uint64_t s390_vec_read_element(const S390Vector *v, >>> uint8_t enr, >>>                                                uint8_t es) >>>   { >>> @@ -117,6 +124,11 @@ static inline void >>> s390_vec_write_element64(S390Vector *v, uint8_t enr, >>>       v->doubleword[enr] = data; >>>   } >>> +static inline void s390_vec_write_element128(S390Vector *v, Int128 >>> data) >>> +{ >>> +    v->quadword = data; >>> +} >>> + >>>   static inline void s390_vec_write_element(S390Vector *v, uint8_t enr, >>>                                             uint8_t es, uint64_t data) >>>   { >>> diff --git a/target/s390x/tcg/vec_helper.c b/target/s390x/tcg/ >>> vec_helper.c >>> index 98eecd9fde6..a86819f61b6 100644 >>> --- a/target/s390x/tcg/vec_helper.c >>> +++ b/target/s390x/tcg/vec_helper.c >>> @@ -48,14 +48,11 @@ void HELPER(vll)(CPUS390XState *env, void *v1, >>> uint64_t addr, uint64_t bytes) >>>       MemOpIdx oi; >>>       if (likely(bytes >= 16)) { >>> -        uint64_t t0, t1; >>> +        Int128 t; >>> -        oi = make_memop_idx(MO_BE | MO_64 | MO_UNALN, mmu_idx); >>> -        t0 = cpu_ldq_mmu(env, addr, oi, ra); >>> -        addr = wrap_address(env, addr + 8); >>> -        t1 = cpu_ldq_mmu(env, addr, oi, ra); >>> -        s390_vec_write_element64(v1, 0, t0); >>> -        s390_vec_write_element64(v1, 1, t1); >>> +        oi = make_memop_idx(MO_128 | MO_UNALN | MO_ATOM_NONE | >>> MO_BE, mmu_idx); >>> +        t = cpu_ld16_mmu(env, addr, oi, ra); >>> +        s390_vec_write_element128(v1, t); >>>       } else { >>>           S390Vector tmp = {}; >>> @@ -206,10 +203,8 @@ void HELPER(vstl)(CPUS390XState *env, const >>> void *v1, uint64_t addr, >>>       probe_write_access(env, addr, MIN(bytes, 16), GETPC()); >>>       if (likely(bytes >= 16)) { >>> -        oi = make_memop_idx(MO_BE | MO_64 | MO_UNALN, mmu_idx); >>> -        cpu_stq_mmu(env, addr, s390_vec_read_element64(v1, 0), oi, >>> ra); >>> -        addr = wrap_address(env, addr + 8); >>> -        cpu_stq_mmu(env, addr, s390_vec_read_element64(v1, 1), oi, >>> ra); >>> +        oi = make_memop_idx(MO_128 | MO_UNALN | MO_ATOM_NONE | >>> MO_BE, mmu_idx); >>> +        cpu_st16_mmu(env, addr, s390_vec_read_element128(v1), oi, ra); >>>       } else { >>>           oi = make_memop_idx(MO_8, mmu_idx); >> >> >> In general LGTM, but I just tried this on top of >> 20260423135035.50126-1- philmd@linaro.org and I'm consistently getting: >> >> >> make[2]: *** [Makefile:236: run-precise-smc-softmmu] Error 1 >> >> >> The test uses VSTL, so it's definitely related. > > Is this on a big-endian host? (Asking because this is the single > configuration I didn't tested) No, that's my x86_64 laptop. Configure flags: '--target-list=s390x-linux-user,s390x-softmmu' '--disable-docs' $ git log -1 commit 2fc3bdf329c54489438408161ebc907a6fa62329 (HEAD) Author: Philippe Mathieu-Daudé Date:   Fri Apr 24 12:43:25 2026 +0200     target/s390x: Leverage 128-bit vector type in V{L, ST}L opcodes $ uname -a && make -j24 && make -j24 check-tcg ; echo $? Linux heavy 6.19.10-100.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Wed Mar 25 17:18:37 UTC 2026 x86_64 GNU/Linux [...] make[2]: *** [Makefile:236: run-precise-smc-softmmu] Error 1 $ git revert HEAD $ uname -a && make -j24 && make -j24 check-tcg ; echo $? [...] 0