Hi Steve, > Good catch (I hate typos!). You should review your own patches before pushing them to upstream:-) And in addition a review from someone else is also very useful... > Fix attached. I verified that the optimization does work now. Should > cut traffic in some cases with default smb3 mounts. I'd split this into two commits, otherwise it looks good. metze > On Thu, Aug 2, 2018 at 11:00 AM Stefan Metzmacher <metze@xxxxxxxxx> wrote: >> >> Hi Steve, >> >> I'm wondering if this patch: >> >>> commit 18dd8e1a65ddae2351d0f0d6dd4a334f441fc5fa >>> Author: Steve French <smfrench@xxxxxxxxx> >>> Date: Mon Sep 26 14:23:08 2016 -0500 >>> >>> Do not send SMB3 SET_INFO request if nothing is changing >>> >>> [CIFS] We had cases where we sent a SMB2/SMB3 setinfo request with >all >>> timestamp (and DOS attribute) fields marked as 0 (ie do not change) >>> e.g. on chmod or chown. >>> >>> Signed-off-by: Steve French <steve.french@xxxxxxxxxxxxxxx> >>> CC: Stable <stable@xxxxxxxxxxxxxxx> >>> --- >>> fs/cifs/smb2inode.c | 6 ++++++ >>> 1 file changed, 6 insertions(+) >>> >>> diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c >>> index 4f0231e685a92..1238cd3552f9c 100644 >>> --- a/fs/cifs/smb2inode.c >>> +++ b/fs/cifs/smb2inode.c >>> @@ -266,9 +266,15 @@ smb2_set_file_info(struct inode *inode, const char >>> *full_path, >>> struct tcon_link *tlink; >>> int rc; >>> >>> + if ((buf->CreationTime == 0) && (buf->LastAccessTime == 0) && >>> + (buf->LastWriteTime == 0) && (buf->ChangeTime) && >>> + (buf->Attributes == 0)) >>> + return 0; /* would be a no op, no sense sending this */ >>> + >> >> Should this contain (buf->ChangeTime == 0) instead of >> (buf->ChangeTime). >> >> Am I missing something? >> >> metze >> > >
Attachment:
signature.asc
Description: OpenPGP digital signature
