2017-06-28 18:07:12

by Long Li

[permalink] [raw]
Subject: [PATCH resend] cifs: Do not modify mid entry after submitting I/O in cifs_call_async

From: Long Li <[email protected]>

In cifs_call_async, server response may return as soon as I/O is submitted.
Because mid entry is freed on the return path, do not modify it after I/O is
submitted.

Signed-off-by: Long Li <[email protected]>
---
fs/cifs/transport.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
index 47a125e..ba62aaf 100644
--- a/fs/cifs/transport.c
+++ b/fs/cifs/transport.c
@@ -537,10 +537,10 @@ cifs_call_async(struct TCP_Server_Info *server, struct smb_rqst *rqst,
spin_unlock(&GlobalMid_Lock);


+ cifs_save_when_sent(mid);
cifs_in_send_inc(server);
rc = smb_send_rqst(server, rqst, flags);
cifs_in_send_dec(server);
- cifs_save_when_sent(mid);

if (rc < 0) {
server->sequence_number -= 2;
--
2.7.4


2017-06-28 18:23:31

by Pavel Shilovsky

[permalink] [raw]
Subject: Re: [PATCH resend] cifs: Do not modify mid entry after submitting I/O in cifs_call_async

2017-06-28 11:05 GMT-07:00 Long Li <[email protected]>:
> From: Long Li <[email protected]>
>
> In cifs_call_async, server response may return as soon as I/O is submitted.
> Because mid entry is freed on the return path, do not modify it after I/O is
> submitted.
>
> Signed-off-by: Long Li <[email protected]>
> ---
> fs/cifs/transport.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
> index 47a125e..ba62aaf 100644
> --- a/fs/cifs/transport.c
> +++ b/fs/cifs/transport.c
> @@ -537,10 +537,10 @@ cifs_call_async(struct TCP_Server_Info *server, struct smb_rqst *rqst,
> spin_unlock(&GlobalMid_Lock);
>
>
> + cifs_save_when_sent(mid);
> cifs_in_send_inc(server);
> rc = smb_send_rqst(server, rqst, flags);
> cifs_in_send_dec(server);
> - cifs_save_when_sent(mid);
>
> if (rc < 0) {
> server->sequence_number -= 2;
> --
> 2.7.4
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-cifs" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html

Looks good. Could you please add an additional comment describing the
behavior in the code?

--
Best regards,
Pavel Shilovsky

2017-06-28 18:30:53

by Long Li

[permalink] [raw]
Subject: RE: [PATCH resend] cifs: Do not modify mid entry after submitting I/O in cifs_call_async



> -----Original Message-----
> From: Pavel Shilovsky [mailto:[email protected]]
> Sent: Wednesday, June 28, 2017 11:23 AM
> To: Long Li <[email protected]>
> Cc: Steve French <[email protected]>; linux-cifs <linux-
> [email protected]>; samba-technical <[email protected]>;
> Kernel Mailing List <[email protected]>; Long Li
> <[email protected]>
> Subject: Re: [PATCH resend] cifs: Do not modify mid entry after submitting
> I/O in cifs_call_async
>
> 2017-06-28 11:05 GMT-07:00 Long Li <[email protected]>:
> > From: Long Li <[email protected]>
> >
> > In cifs_call_async, server response may return as soon as I/O is submitted.
> > Because mid entry is freed on the return path, do not modify it after
> > I/O is submitted.
> >
> > Signed-off-by: Long Li <[email protected]>
> > ---
> > fs/cifs/transport.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index
> > 47a125e..ba62aaf 100644
> > --- a/fs/cifs/transport.c
> > +++ b/fs/cifs/transport.c
> > @@ -537,10 +537,10 @@ cifs_call_async(struct TCP_Server_Info *server,
> struct smb_rqst *rqst,
> > spin_unlock(&GlobalMid_Lock);
> >
> >
> > + cifs_save_when_sent(mid);
> > cifs_in_send_inc(server);
> > rc = smb_send_rqst(server, rqst, flags);
> > cifs_in_send_dec(server);
> > - cifs_save_when_sent(mid);
> >
> > if (rc < 0) {
> > server->sequence_number -= 2;
> > --
> > 2.7.4
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-cifs"
> > in the body of a message to [email protected] More
> majordomo
> > info at
> >
> https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fvger.k
> > ernel.org%2Fmajordomo-
> info.html&data=02%7C01%7Clongli%40microsoft.com%
> >
> 7C12e72888c94049d58ceb08d4be52c28c%7C72f988bf86f141af91ab2d7cd011d
> b47%
> >
> 7C1%7C0%7C636342710042994103&sdata=h%2FkZHBQ5bwvIiS3K0HJ%2BSgH
> 7woB4EHp
> > t%2FzoWE626x3U%3D&reserved=0
>
> Looks good. Could you please add an additional comment describing the
> behavior in the code?

Sure, will do.

>
> --
> Best regards,
> Pavel Shilovsky