[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20190809151921.GB93481@jaegeuk-macbookpro.roam.corp.google.com>
Date: Fri, 9 Aug 2019 08:19:21 -0700
From: Jaegeuk Kim <jaegeuk@...nel.org>
To: Chao Yu <yuchao0@...wei.com>
Cc: linux-f2fs-devel@...ts.sourceforge.net,
linux-kernel@...r.kernel.org, chao@...nel.org
Subject: Re: [PATCH v2] f2fs: fix wrong available node count calculation
On 08/06, Chao Yu wrote:
> On 2019/8/5 18:27, Chao Yu wrote:
> > In mkfs, we have counted quota file's node number in cp.valid_node_count,
> > so we have to avoid wrong substraction of quota node number in
> > .available_nid/.avail_node_count calculation.
> >
> > f2fs_write_check_point_pack()
> > {
> > ..
> > set_cp(valid_node_count, 1 + c.quota_inum + c.lpf_inum);
> >
> > Fixes: 292c196a3695 ("reserve nid resource for quota sysfile")
>
> Jaegeuk,
>
> Could you help to add prefix "f2fs: " into commit tile in your branch, I missed
> to add it. :P
Done. :)
>
> Thanks,
>
> > Fixes: 7b63f72f73af ("f2fs: fix to do sanity check on valid node/block count")
> > Signed-off-by: Chao Yu <yuchao0@...wei.com>
> > ---
> > fs/f2fs/node.c | 2 +-
> > fs/f2fs/super.c | 6 ++----
> > 2 files changed, 3 insertions(+), 5 deletions(-)
> >
> > diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
> > index a18b2a895771..d9ba1db2d01e 100644
> > --- a/fs/f2fs/node.c
> > +++ b/fs/f2fs/node.c
> > @@ -2964,7 +2964,7 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
> >
> > /* not used nids: 0, node, meta, (and root counted as valid node) */
> > nm_i->available_nids = nm_i->max_nid - sbi->total_valid_node_count -
> > - sbi->nquota_files - F2FS_RESERVED_NODE_NUM;
> > + F2FS_RESERVED_NODE_NUM;
> > nm_i->nid_cnt[FREE_NID] = 0;
> > nm_i->nid_cnt[PREALLOC_NID] = 0;
> > nm_i->nat_cnt = 0;
> > diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
> > index 6a7f1166d068..118a31f90a37 100644
> > --- a/fs/f2fs/super.c
> > +++ b/fs/f2fs/super.c
> > @@ -1297,8 +1297,7 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
> > else
> > buf->f_bavail = 0;
> >
> > - avail_node_count = sbi->total_node_count - sbi->nquota_files -
> > - F2FS_RESERVED_NODE_NUM;
> > + avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
> >
> > if (avail_node_count > user_block_count) {
> > buf->f_files = user_block_count;
> > @@ -2750,8 +2749,7 @@ int f2fs_sanity_check_ckpt(struct f2fs_sb_info *sbi)
> > }
> >
> > valid_node_count = le32_to_cpu(ckpt->valid_node_count);
> > - avail_node_count = sbi->total_node_count - sbi->nquota_files -
> > - F2FS_RESERVED_NODE_NUM;
> > + avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
> > if (valid_node_count > avail_node_count) {
> > f2fs_err(sbi, "Wrong valid_node_count: %u, avail_node_count: %u",
> > valid_node_count, avail_node_count);
> >
Powered by blists - more mailing lists