From patchwork Tue Nov 13 19:28:59 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Omar Sandoval X-Patchwork-Id: 10681405 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 066AD13B5 for ; Tue, 13 Nov 2018 19:29:12 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id EB5E92B4D5 for ; Tue, 13 Nov 2018 19:29:11 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id DEB0C2B4DA; Tue, 13 Nov 2018 19:29:11 +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=-7.9 required=2.0 tests=BAYES_00,DKIM_SIGNED, DKIM_VALID,MAILING_LIST_MULTI,RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 3BD482B4D5 for ; Tue, 13 Nov 2018 19:29:11 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727094AbeKNF2l (ORCPT ); Wed, 14 Nov 2018 00:28:41 -0500 Received: from mail-pf1-f196.google.com ([209.85.210.196]:43408 "EHLO mail-pf1-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725748AbeKNF2k (ORCPT ); Wed, 14 Nov 2018 00:28:40 -0500 Received: by mail-pf1-f196.google.com with SMTP id g7-v6so6556294pfo.10 for ; Tue, 13 Nov 2018 11:29:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=osandov-com.20150623.gappssmtp.com; s=20150623; h=from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=WTPFS2SXxtoc05Iii9kU3SAXhjoUFhmmf/9rfO8MzlE=; b=pTtvHWR5d97jjLnvacglbgo09j1hPj1UL1JoqA8MGbrrUD198SAzGxxY6IRQXwAKfq oDjMjqZMB7coYPCMNWQ7FdE/PJWq9aC9HSRmFNxD4QH0jgLYVUj0Yd1/+aQaIyjKaC37 Z60pKXNdp+Qx+1W+zrMhcKwswfJjZUzaC09wl2kZXeC+gLebtx+CogGHvVte4aKc+Xwp 4Q8Bu6WKVQxfB4EQ9EZmLea0fApYs6RAvpgH38XL9o3MWra95c3+N7EDM8DptGxCVFB+ YCC+/LpXui7eNom6GsJRKsHyNi5aiTaeZJOzdQOGWcA79IAi6pjQmZPvOA/BUpPvYj7C CsQA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding; bh=WTPFS2SXxtoc05Iii9kU3SAXhjoUFhmmf/9rfO8MzlE=; b=L5icRsch1NxWZtF8b01H1V4VZeyeR3WYLJTOMqvE1HSpfNGRtPpvcjb09NqC0phgXE fgIbizxGZA25+XPtyJHa898BY+s6skkVJuMhIofQ1yN/MOVI9Oeypt52Ilj4cRwQjgf/ dbr0XddZdCJLgZhAawnCLA+OcVg07RmhFGlFOyVS6ca+sqZgJgtIPrx5GOnrQvWrfasB 7sERsYpMeqGqGcDmdWhXvv+gRw+Q7fcJRxS56cOxC+ku2LB1Dlx6j0mDA+nCN2kZUv9v Fd5uZ25DCPGPeyEAzqZJYY3eUZzZ9FQRe764LGL8CGwIrd7sA28hTW2qhSOJ849Mg5tk j3SA== X-Gm-Message-State: AGRZ1gIuT3ypeK22AdAVKy9MzrB2GAmU1dFsoyKg5rlTcBGB0bavnoEq xgltABxuUSAy0vbfY6a2TWxqWsBta65vkA== X-Google-Smtp-Source: AJdET5e1RCYNetR8CGMnkSv676FBmNFZcdFPE0ijmHPY1BAVftUPDM9IKED4gtw4yxVtpJw8gG8dRA== X-Received: by 2002:a63:5a57:: with SMTP id k23mr5850941pgm.5.1542137347065; Tue, 13 Nov 2018 11:29:07 -0800 (PST) Received: from vader.thefacebook.com ([2620:10d:c090:180::1:a21f]) by smtp.gmail.com with ESMTPSA id 127-v6sm22910807pfx.91.2018.11.13.11.29.06 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 13 Nov 2018 11:29:06 -0800 (PST) From: Omar Sandoval To: linux-xfs@vger.kernel.org Cc: kernel-team@fb.com Subject: [PATCH v3] xfs: cache minimum realtime summary level Date: Tue, 13 Nov 2018 11:28:59 -0800 Message-Id: <95c54025014c6d5d3944924b52637cdc2b50cf4e.1542137269.git.osandov@fb.com> X-Mailer: git-send-email 2.19.1 MIME-Version: 1.0 Sender: linux-xfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-xfs@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Omar Sandoval The realtime summary is a two-dimensional array on disk, effectively: u32 rsum[log2(number of realtime extents) + 1][number of blocks in the bitmap] rsum[log][bbno] is the number of extents of size 2**log which start in bitmap block bbno. xfs_rtallocate_extent_near() uses xfs_rtany_summary() to check whether rsum[log][bbno] != 0 for any log level. However, the summary array is stored in row-major order (i.e., like an array in C), so all of these entries are not adjacent, but rather spread across the entire summary file. In the worst case (a full bitmap block), xfs_rtany_summary() has to check every level. This means that on a moderately-used realtime device, an allocation will waste a lot of time finding, reading, and releasing buffers for the realtime summary. In particular, one of our storage services (which runs on servers with 8 very slow CPUs and 15 8 TB XFS realtime filesystems) spends almost 5% of its CPU cycles in xfs_rtbuf_get() and xfs_trans_brelse() called from xfs_rtany_summary(). One solution would be to also store the summary with the dimensions swapped. However, this would require a disk format change to a very old component of XFS. Instead, we can cache the minimum size which contains any extents. We do so lazily; rather than guaranteeing that the cache contains the precise minimum, it always contains a loose lower bound which we tighten when we read or update a summary block. This only uses a few kilobytes of memory and is already serialized via the realtime bitmap and summary inode locks, so the cost is minimal. With this change, the same workload only spends 0.2% of its CPU cycles in the realtime allocator. Signed-off-by: Omar Sandoval Reviewed-by: Darrick J. Wong --- Based on Linus' master branch. Changes from v2: - Allow the cache allocation to fail, in which case we just don't use it Changes from v1: - Clarify comment in xfs_rtmount_inodes(). - Use kmem_* instead of kvmalloc/kvfree fs/xfs/libxfs/xfs_rtbitmap.c | 6 ++++++ fs/xfs/xfs_mount.h | 7 +++++++ fs/xfs/xfs_rtalloc.c | 25 +++++++++++++++++++++---- 3 files changed, 34 insertions(+), 4 deletions(-) diff --git a/fs/xfs/libxfs/xfs_rtbitmap.c b/fs/xfs/libxfs/xfs_rtbitmap.c index b228c821bae6..eaaff67e9626 100644 --- a/fs/xfs/libxfs/xfs_rtbitmap.c +++ b/fs/xfs/libxfs/xfs_rtbitmap.c @@ -505,6 +505,12 @@ xfs_rtmodify_summary_int( uint first = (uint)((char *)sp - (char *)bp->b_addr); *sp += delta; + if (mp->m_rsum_cache) { + if (*sp == 0 && log == mp->m_rsum_cache[bbno]) + mp->m_rsum_cache[bbno]++; + if (*sp != 0 && log < mp->m_rsum_cache[bbno]) + mp->m_rsum_cache[bbno] = log; + } xfs_trans_log_buf(tp, bp, first, first + sizeof(*sp) - 1); } if (sum) diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 7964513c3128..39f04aca8c3a 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -89,6 +89,13 @@ typedef struct xfs_mount { int m_logbsize; /* size of each log buffer */ uint m_rsumlevels; /* rt summary levels */ uint m_rsumsize; /* size of rt summary, bytes */ + /* + * Optional cache of rt summary level per bitmap block with the + * invariant that m_rsum_cache[bbno] <= the minimum i for which + * rsum[i][bbno] != 0. Reads and writes are serialized by the rsumip + * inode lock. + */ + uint8_t *m_rsum_cache; struct xfs_inode *m_rbmip; /* pointer to bitmap inode */ struct xfs_inode *m_rsumip; /* pointer to summary inode */ struct xfs_inode *m_rootip; /* pointer to root directory */ diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 926ed314ffba..aefd63d46397 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -64,8 +64,12 @@ xfs_rtany_summary( int log; /* loop counter, log2 of ext. size */ xfs_suminfo_t sum; /* summary data */ + /* There are no extents at levels < m_rsum_cache[bbno]. */ + if (mp->m_rsum_cache && low < mp->m_rsum_cache[bbno]) + low = mp->m_rsum_cache[bbno]; + /* - * Loop over logs of extent sizes. Order is irrelevant. + * Loop over logs of extent sizes. */ for (log = low; log <= high; log++) { /* @@ -80,13 +84,17 @@ xfs_rtany_summary( */ if (sum) { *stat = 1; - return 0; + goto out; } } /* * Found nothing, return failure. */ *stat = 0; +out: + /* There were no extents at levels < log. */ + if (mp->m_rsum_cache && log > mp->m_rsum_cache[bbno]) + mp->m_rsum_cache[bbno] = log; return 0; } @@ -1187,8 +1195,8 @@ xfs_rtmount_init( } /* - * Get the bitmap and summary inodes into the mount structure - * at mount time. + * Get the bitmap and summary inodes and the summary cache into the mount + * structure at mount time. */ int /* error */ xfs_rtmount_inodes( @@ -1211,6 +1219,14 @@ xfs_rtmount_inodes( return error; } ASSERT(mp->m_rsumip != NULL); + /* + * The rsum cache is initialized to all zeroes, which is trivially a + * lower bound on the minimum level with any free extents. We can + * continue without the cache if it couldn't be allocated. + */ + mp->m_rsum_cache = kmem_zalloc_large(sbp->sb_rbmblocks, KM_SLEEP); + if (!mp->m_rsum_cache) + xfs_warn(mp, "could not allocate realtime summary cache"); return 0; } @@ -1218,6 +1234,7 @@ void xfs_rtunmount_inodes( struct xfs_mount *mp) { + kmem_free(mp->m_rsum_cache); if (mp->m_rbmip) xfs_irele(mp->m_rbmip); if (mp->m_rsumip)