This looks good to me.
Let me know if you get a chance to test it out. If not, I'll test it
on my setup tomorrow.
Regards,
Shyam
On Tue, Feb 2, 2021 at 11:13 PM Aurélien Aptel <aaptel@xxxxxxxx> wrote:
>
> From: Aurelien Aptel <aaptel@xxxxxxxx>
>
> Assuming
> - //HOST/a is mounted on /mnt
> - //HOST/b is mounted on /mnt/b
>
> On a slow connection, running 'df' and killing it while it's
> processing /mnt/b can make cifs_get_inode_info() returns -ERESTARTSYS.
>
> This triggers the following chain of events:
> => the dentry revalidation fail
> => dentry is put and released
> => superblock associated with the dentry is put
> => /mnt/b is unmounted
>
> This patch makes cifs_d_revalidate() return the error instead of
> 0 (invalid) when cifs_revalidate_dentry() fails, except for ENOENT
> where that error means the dentry is invalid.
>
> Signed-off-by: Aurelien Aptel <aaptel@xxxxxxxx>
> Suggested-by: Shyam Prasad N <nspmangalore@xxxxxxxxx>
> ---
> fs/cifs/dir.c | 8 ++++++--
> 1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
> index 68900f1629bff..868c0b7263ec0 100644
> --- a/fs/cifs/dir.c
> +++ b/fs/cifs/dir.c
> @@ -737,6 +737,7 @@ static int
> cifs_d_revalidate(struct dentry *direntry, unsigned int flags)
> {
> struct inode *inode;
> + int rc;
>
> if (flags & LOOKUP_RCU)
> return -ECHILD;
> @@ -746,8 +747,11 @@ cifs_d_revalidate(struct dentry *direntry, unsigned int flags)
> if ((flags & LOOKUP_REVAL) && !CIFS_CACHE_READ(CIFS_I(inode)))
> CIFS_I(inode)->time = 0; /* force reval */
>
> - if (cifs_revalidate_dentry(direntry))
> - return 0;
> + rc = cifs_revalidate_dentry(direntry);
> + if (rc) {
> + cifs_dbg(FYI, "cifs_revalidate_dentry failed with rc=%d", rc);
> + return rc == -ENOENT ? 0 : rc;
> + }
> else {
> /*
> * If the inode wasn't known to be a dfs entry when
> --
> 2.29.2
>
--
Regards,
Shyam
