From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:46218) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bGSiQ-0005DB-RQ for qemu-devel@nongnu.org; Fri, 24 Jun 2016 11:08:19 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bGSiM-0001Ws-FZ for qemu-devel@nongnu.org; Fri, 24 Jun 2016 11:08:17 -0400 Received: from mailhub.sw.ru ([195.214.232.25]:37344 helo=relay.sw.ru) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bGSiL-0001WM-4G for qemu-devel@nongnu.org; Fri, 24 Jun 2016 11:08:14 -0400 From: "Denis V. Lunev" Date: Fri, 24 Jun 2016 18:06:40 +0300 Message-Id: <1466780802-30424-2-git-send-email-den@openvz.org> In-Reply-To: <1466780802-30424-1-git-send-email-den@openvz.org> References: <1466780802-30424-1-git-send-email-den@openvz.org> Subject: [Qemu-devel] [PATCH 1/3] block: ignore flush requests when storage is clean List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-block@nongnu.org, qemu-devel@nongnu.org Cc: den@openvz.org, Evgeny Yakovlev , Kevin Wolf , Max Reitz , Stefan Hajnoczi , Fam Zheng , John Snow From: Evgeny Yakovlev Some guests (win2008 server for example) do a lot of unnecessary flushing when underlying media has not changed. This adds additional overhead on host when calling fsync/fdatasync. This change introduces a dirty flag in BlockDriverState which is set in bdrv_set_dirty and is checked in bdrv_co_flush. This allows us to avoid unnesessary flushing when storage is clean. The problem with excessive flushing was found by a performance test which does parallel directory tree creation (from 2 processes). Results improved from 0.424 loops/sec to 0.432 loops/sec. Each loop creates 10^3 directories with 10 files in each. Signed-off-by: Evgeny Yakovlev Signed-off-by: Denis V. Lunev CC: Kevin Wolf CC: Max Reitz CC: Stefan Hajnoczi CC: Fam Zheng CC: John Snow --- block.c | 1 + block/dirty-bitmap.c | 3 +++ block/io.c | 19 +++++++++++++++++++ include/block/block_int.h | 2 ++ 4 files changed, 25 insertions(+) diff --git a/block.c b/block.c index f4648e9..e36f148 100644 --- a/block.c +++ b/block.c @@ -2582,6 +2582,7 @@ int bdrv_truncate(BlockDriverState *bs, int64_t offset) ret = refresh_total_sectors(bs, offset >> BDRV_SECTOR_BITS); bdrv_dirty_bitmap_truncate(bs); bdrv_parent_cb_resize(bs); + bs->dirty = true; /* file node sync is needed after truncate */ } return ret; } diff --git a/block/dirty-bitmap.c b/block/dirty-bitmap.c index 4902ca5..54e0413 100644 --- a/block/dirty-bitmap.c +++ b/block/dirty-bitmap.c @@ -370,6 +370,9 @@ void bdrv_set_dirty(BlockDriverState *bs, int64_t cur_sector, } hbitmap_set(bitmap->bitmap, cur_sector, nr_sectors); } + + /* Set global block driver dirty flag even if bitmap is disabled */ + bs->dirty = true; } /** diff --git a/block/io.c b/block/io.c index 7cf3645..8078af2 100644 --- a/block/io.c +++ b/block/io.c @@ -2239,6 +2239,25 @@ int coroutine_fn bdrv_co_flush(BlockDriverState *bs) goto flush_parent; } + /* Check if storage is actually dirty before flushing to disk */ + if (!bs->dirty) { + /* Flush requests are appended to tracked request list in order so that + * most recent request is at the head of the list. Following code uses + * this ordering to wait for the most recent flush request to complete + * to ensure that requests return in order */ + BdrvTrackedRequest *prev_req; + QLIST_FOREACH(prev_req, &bs->tracked_requests, list) { + if (prev_req == &req || prev_req->type != BDRV_TRACKED_FLUSH) { + continue; + } + + qemu_co_queue_wait(&prev_req->wait_queue); + break; + } + goto flush_parent; + } + bs->dirty = false; + BLKDBG_EVENT(bs->file, BLKDBG_FLUSH_TO_DISK); if (bs->drv->bdrv_co_flush_to_disk) { ret = bs->drv->bdrv_co_flush_to_disk(bs); diff --git a/include/block/block_int.h b/include/block/block_int.h index 2057156..616058b 100644 --- a/include/block/block_int.h +++ b/include/block/block_int.h @@ -418,6 +418,8 @@ struct BlockDriverState { int sg; /* if true, the device is a /dev/sg* */ int copy_on_read; /* if true, copy read backing sectors into image note this is a reference count */ + + bool dirty; bool probed; BlockDriver *drv; /* NULL means no media */ -- 2.1.4