Re: [PATCH 5.15 000/511] 5.15.132-rc1 review

[Greg Kroah-Hartman]

On Mon, Sep 18, 2023 at 11:40:36AM -0700, Guenter Roeck wrote:
> On Sun, Sep 17, 2023 at 09:07:07PM +0200, Greg Kroah-Hartman wrote:
> > This is the start of the stable review cycle for the 5.15.132 release.
> > There are 511 patches in this series, all will be posted as a response
> > to this one.  If anyone has any issues with these being applied, please
> > let me know.
> > 
> > Responses should be made by Tue, 19 Sep 2023 19:10:04 +0000.
> > Anything received after that time might be too late.
> > 
> 
> My tests say
> 
> Build results:
> 	total: 155 pass: 155 fail: 0
> Qemu test results:
> 	total: 509 pass: 509 fail: 0
> 
> However, trying to build with clang (as done by ChromeOS) does report
> 
> fs/ksmbd/smb2pdu.c:6256:6: error: variable 'fp' is used uninitialized whenever 'if' condition is true [-Werror,-Wsometimes-uninitialized]
>         if (work->next_smb2_rcv_hdr_off) {
>             ^~~~~~~~~~~~~~~~~~~~~~~~~~~
> fs/ksmbd/smb2pdu.c:6367:21: note: uninitialized use occurs here
>         ksmbd_fd_put(work, fp);
>                            ^~
> fs/ksmbd/smb2pdu.c:6256:2: note: remove the 'if' if its condition is always false
>         if (work->next_smb2_rcv_hdr_off) {
>         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> fs/ksmbd/smb2pdu.c:6248:23: note: initialize the variable 'fp' to silence this warning
>         struct ksmbd_file *fp;
>                              ^
>                               = NULL
> 
> I guess I should try to add clang build tests.
> 
> Anyway, smb2_write() already initializes fp with NULL, presumably to handle a
> similar problem, so maybe doing the same for the read function would be the
> easiest solution. Note that the initialization was added upstream with commit
> 2fd5dcb1c8ef9 ("ksmbd: smbd: fix missing client's memory region invalidation").

Thanks for the report, I'll go fix this up in the commit that causes the
build warning.

greg k-h