From nobody Fri Nov 14 00:48:47 2025 Delivered-To: importer@patchew.org Received-SPF: pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; envelope-from=qemu-devel-bounces+importer=patchew.org@nongnu.org; helo=lists.gnu.org; Authentication-Results: mx.zohomail.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=fail(p=none dis=none) header.from=virtuozzo.com ARC-Seal: i=1; a=rsa-sha256; t=1583480577; cv=none; d=zohomail.com; s=zohoarc; b=QINU9jbFj1n2wua2RgMh/aAmdnXVCg1saIsIFZovbWM+t7GewXIQs5oXmFCYjCvMzyumsiorC9pY20xTYwJhCk2xspo8ZO1umqollrfwYfM+0Eenh5zNd16fzmpOtmou7qL3Jki51wmQNzeknCvMLPiFRLInRQUDdAcdL6d6gCU= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=zohomail.com; s=zohoarc; t=1583480577; h=Content-Transfer-Encoding:Cc:Date:From:In-Reply-To:List-Subscribe:List-Post:List-Id:List-Archive:List-Help:List-Unsubscribe:MIME-Version:Message-ID:References:Sender:Subject:To; bh=vGSXrcC+ofK2HbHefYMZNyfkh+ZbUSQIzw1puvyM5T4=; b=AK7pZrmPsZK4xu6Z4kl+mO4rCnq/mWcBxEms6LibWhImQM6q8zexJtZ1oy/rGLY8+WqZZ7JtdansQ29Sh+pCSPJ7lS5hVConSOUBDnhH8Xz7If8pjx7rZ3auTDuRIawaHgwNVkOhjpu/jPTCidwU1WCvDLo9Ci4k5PqeovWDNjU= ARC-Authentication-Results: i=1; mx.zohomail.com; spf=pass (zohomail.com: domain of gnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom=qemu-devel-bounces+importer=patchew.org@nongnu.org; dmarc=fail header.from= (p=none dis=none) header.from= Return-Path: Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) by mx.zohomail.com with SMTPS id 1583480577418197.99213558413715; Thu, 5 Mar 2020 23:42:57 -0800 (PST) Received: from localhost ([::1]:60544 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jA7dI-0005Zy-DN for importer@patchew.org; Fri, 06 Mar 2020 02:42:56 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]:46342) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1jA7ZI-00056i-JI for qemu-devel@nongnu.org; Fri, 06 Mar 2020 02:38:50 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1jA7ZG-0008Nm-PD for qemu-devel@nongnu.org; Fri, 06 Mar 2020 02:38:48 -0500 Received: from relay.sw.ru ([185.231.240.75]:37982) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1jA7ZC-0007wn-KX; Fri, 06 Mar 2020 02:38:42 -0500 Received: from vovaso.qa.sw.ru ([10.94.3.0] helo=kvm.qa.sw.ru) by relay.sw.ru with esmtp (Exim 4.92.3) (envelope-from ) id 1jA7Z8-0002Qu-ON; Fri, 06 Mar 2020 10:38:38 +0300 From: Vladimir Sementsov-Ogievskiy To: qemu-block@nongnu.org Subject: [PATCH v3 8/9] block/block-copy: reduce intersecting request lock Date: Fri, 6 Mar 2020 10:38:30 +0300 Message-Id: <20200306073831.7737-9-vsementsov@virtuozzo.com> X-Mailer: git-send-email 2.21.0 In-Reply-To: <20200306073831.7737-1-vsementsov@virtuozzo.com> References: <20200306073831.7737-1-vsementsov@virtuozzo.com> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-detected-operating-system: by eggs.gnu.org: GNU/Linux 3.x [fuzzy] X-Received-From: 185.231.240.75 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: kwolf@redhat.com, vsementsov@virtuozzo.com, qemu-devel@nongnu.org, mreitz@redhat.com, andrey.shinkevich@virtuozzo.com, jsnow@redhat.com Errors-To: qemu-devel-bounces+importer=patchew.org@nongnu.org Sender: "Qemu-devel" Content-Type: text/plain; charset="utf-8" Currently, block_copy operation lock the whole requested region. But there is no reason to lock clusters, which are already copied, it will disturb other parallel block_copy requests for no reason. Let's instead do the following: Lock only sub-region, which we are going to operate on. Then, after copying all dirty sub-regions, we should wait for intersecting requests block-copy, if they failed, we should retry these new dirty clusters. Signed-off-by: Vladimir Sementsov-Ogievskiy Reviewed-by: Andrey Shinkevich --- block/block-copy.c | 128 ++++++++++++++++++++++++++++++++++++--------- 1 file changed, 104 insertions(+), 24 deletions(-) diff --git a/block/block-copy.c b/block/block-copy.c index 2b29131653..d66b8eb691 100644 --- a/block/block-copy.c +++ b/block/block-copy.c @@ -39,29 +39,71 @@ static BlockCopyInFlightReq *find_conflicting_inflight_= req(BlockCopyState *s, return NULL; } =20 -static void coroutine_fn block_copy_wait_inflight_reqs(BlockCopyState *s, - int64_t offset, - int64_t bytes) +/* + * If there are no intersecting requests return false. Otherwise, wait for= the + * first found intersecting request to finish and return true. + */ +static bool coroutine_fn block_copy_wait_one(BlockCopyState *s, int64_t of= fset, + int64_t bytes) { - BlockCopyInFlightReq *req; + BlockCopyInFlightReq *req =3D find_conflicting_inflight_req(s, offset,= bytes); =20 - while ((req =3D find_conflicting_inflight_req(s, offset, bytes))) { - qemu_co_queue_wait(&req->wait_queue, NULL); + if (!req) { + return false; } + + qemu_co_queue_wait(&req->wait_queue, NULL); + + return true; } =20 +/* Called only on full-dirty region */ static void block_copy_inflight_req_begin(BlockCopyState *s, BlockCopyInFlightReq *req, int64_t offset, int64_t bytes) { + assert(!find_conflicting_inflight_req(s, offset, bytes)); + + bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, bytes); + s->in_flight_bytes +=3D bytes; + req->offset =3D offset; req->bytes =3D bytes; qemu_co_queue_init(&req->wait_queue); QLIST_INSERT_HEAD(&s->inflight_reqs, req, list); } =20 -static void coroutine_fn block_copy_inflight_req_end(BlockCopyInFlightReq = *req) +/* + * block_copy_inflight_req_shrink + * + * Drop the tail of the request to be handled later. Set dirty bits back a= nd + * wake up all requests waiting for us (may be some of them are not inters= ecting + * with shrunk request) + */ +static void coroutine_fn block_copy_inflight_req_shrink(BlockCopyState *s, + BlockCopyInFlightReq *req, int64_t new_bytes) { + if (new_bytes =3D=3D req->bytes) { + return; + } + + assert(new_bytes > 0 && new_bytes < req->bytes); + + bdrv_set_dirty_bitmap(s->copy_bitmap, + req->offset + new_bytes, req->bytes - new_bytes); + + req->bytes =3D new_bytes; + qemu_co_queue_restart_all(&req->wait_queue); +} + +static void coroutine_fn block_copy_inflight_req_end(BlockCopyState *s, + BlockCopyInFlightReq = *req, + int ret) +{ + s->in_flight_bytes -=3D req->bytes; + if (ret < 0) { + bdrv_set_dirty_bitmap(s->copy_bitmap, req->offset, req->bytes); + } QLIST_REMOVE(req, list); qemu_co_queue_restart_all(&req->wait_queue); } @@ -357,12 +399,19 @@ int64_t block_copy_reset_unallocated(BlockCopyState *= s, return ret; } =20 -int coroutine_fn block_copy(BlockCopyState *s, - int64_t offset, int64_t bytes, - bool *error_is_read) +/* + * block_copy_dirty_clusters + * + * Copy dirty clusters in @offset/@bytes range. + * Returns 1 if dirty clusters found and successfully copied, 0 if no dirty + * clusters found and -errno on failure. + */ +static int coroutine_fn block_copy_dirty_clusters(BlockCopyState *s, + int64_t offset, int64_t = bytes, + bool *error_is_read) { int ret =3D 0; - BlockCopyInFlightReq req; + bool found_dirty =3D false; =20 /* * block_copy() user is responsible for keeping source and target in s= ame @@ -374,10 +423,8 @@ int coroutine_fn block_copy(BlockCopyState *s, assert(QEMU_IS_ALIGNED(offset, s->cluster_size)); assert(QEMU_IS_ALIGNED(bytes, s->cluster_size)); =20 - block_copy_wait_inflight_reqs(s, offset, bytes); - block_copy_inflight_req_begin(s, &req, offset, bytes); - while (bytes) { + BlockCopyInFlightReq req; int64_t next_zero, cur_bytes, status_bytes; =20 if (!bdrv_dirty_bitmap_get(s->copy_bitmap, offset)) { @@ -387,6 +434,8 @@ int coroutine_fn block_copy(BlockCopyState *s, continue; /* already copied */ } =20 + found_dirty =3D true; + cur_bytes =3D MIN(bytes, s->copy_size); =20 next_zero =3D bdrv_dirty_bitmap_next_zero(s->copy_bitmap, offset, @@ -396,10 +445,14 @@ int coroutine_fn block_copy(BlockCopyState *s, assert(next_zero < offset + cur_bytes); /* no need to do MIN()= */ cur_bytes =3D next_zero - offset; } + block_copy_inflight_req_begin(s, &req, offset, cur_bytes); =20 ret =3D block_copy_block_status(s, offset, cur_bytes, &status_byte= s); + assert(ret >=3D 0); /* never fail */ + cur_bytes =3D MIN(cur_bytes, status_bytes); + block_copy_inflight_req_shrink(s, &req, cur_bytes); if (s->skip_unallocated && !(ret & BDRV_BLOCK_ALLOCATED)) { - bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, status_bytes); + block_copy_inflight_req_end(s, &req, 0); progress_set_remaining(s->progress, bdrv_get_dirty_count(s->copy_bitmap) + s->in_flight_bytes); @@ -409,21 +462,15 @@ int coroutine_fn block_copy(BlockCopyState *s, continue; } =20 - cur_bytes =3D MIN(cur_bytes, status_bytes); - trace_block_copy_process(s, offset); =20 - bdrv_reset_dirty_bitmap(s->copy_bitmap, offset, cur_bytes); - s->in_flight_bytes +=3D cur_bytes; - co_get_from_shres(s->mem, cur_bytes); ret =3D block_copy_do_copy(s, offset, cur_bytes, ret & BDRV_BLOCK_= ZERO, error_is_read); co_put_to_shres(s->mem, cur_bytes); - s->in_flight_bytes -=3D cur_bytes; + block_copy_inflight_req_end(s, &req, ret); if (ret < 0) { - bdrv_set_dirty_bitmap(s->copy_bitmap, offset, cur_bytes); - break; + return ret; } =20 progress_work_done(s->progress, cur_bytes); @@ -432,7 +479,40 @@ int coroutine_fn block_copy(BlockCopyState *s, bytes -=3D cur_bytes; } =20 - block_copy_inflight_req_end(&req); + return found_dirty; +} + +/* + * block_copy + * + * Copy requested region, accordingly to dirty bitmap. + * Collaborate with parallel block_copy requests: if they success it help = us. If + * they fail, we retry not-copied regions. So, if we return error, it mean= s that + * io operation failed in context of _this_ block_copy call, not some para= llel + * operation. + */ +int coroutine_fn block_copy(BlockCopyState *s, int64_t offset, int64_t byt= es, + bool *error_is_read) +{ + int ret; + + do { + ret =3D block_copy_dirty_clusters(s, offset, bytes, error_is_read); + + if (ret =3D=3D 0) { + ret =3D block_copy_wait_one(s, offset, bytes); + } + + /* + * We retry in two cases: + * 1. Some progress done + * Something was copied, which means that there were yield poin= ts + * and some new dirty bits may have appeared (due to failed par= allel + * block-copy requests). + * 2. We have waited for some intersecting block-copy request + * It may have failed and produced new dirty bits. + */ + } while (ret > 0); =20 return ret; } --=20 2.21.0