Building driver r8188eu in staging with -warray-bounds exposes two places
where arrays are too small.
Signed-off-by: Larry Finger <[email protected]>
---
v2 - get proper To and Cc
v3 - Use Dan Carpenter's suggestion for correct patch.
Larry Finger (2):
staging: r8188eu: Fix undersized array in rtw_xmit.c
staging: r8188eu: Fix warning of array overflow in ioctl_linux.c
drivers/staging/r8188eu/core/rtw_xmit.c | 17 ++++-------------
drivers/staging/r8188eu/os_dep/ioctl_linux.c | 2 +-
2 files changed, 5 insertions(+), 14 deletions(-)
--
2.36.1
On Tue, May 31, 2022 at 12:16:58PM -0500, Larry Finger wrote:
> On 5/31/22 01:40, Phillip Potter wrote:
> >
> > Hi Larry,
> >
> > Both patches look good to me, however I've already submitted a patch 10
> > days ago that does exactly what your rtw_xmit.c patch does, in
> > addition to fixing the changed error handling semantics for the relevant
> > function (in f94b47c6bde6) which broke the driver.
>
> Phil,
>
> Sorry I missed your patch. I have been really busy the past month on another
> problem, and I was not paying much attention to r8188eu.
>
> When I built a mainline kernel in the middle of the merge from 5.18 to 5.19,
> those two warnings stood out.
>
> Your patch should be ahead of mine in Greg's queue, thus mine should fail,
> and I will get the "does not apply" message. At that time, I will send the
> other patch. As both of us missed the 5.19 merge, the build warnings will
> persist until 5.20, but no harm.
>
> Larry
>
Please don't be sorry - just wanted you to know in advance :-).
Figured if I said nothing I would probably aggravate the situation. I
don't have much time to spend on r8188eu myself either so don't worry.
All the best,
Phil
On Mon, May 30, 2022 at 08:31:01PM -0500, Larry Finger wrote:
> Building driver r8188eu in staging with -warray-bounds exposes two places
> where arrays are too small.
>
> Signed-off-by: Larry Finger <[email protected]>
> ---
> v2 - get proper To and Cc
> v3 - Use Dan Carpenter's suggestion for correct patch.
>
> Larry Finger (2):
> staging: r8188eu: Fix undersized array in rtw_xmit.c
> staging: r8188eu: Fix warning of array overflow in ioctl_linux.c
>
> drivers/staging/r8188eu/core/rtw_xmit.c | 17 ++++-------------
> drivers/staging/r8188eu/os_dep/ioctl_linux.c | 2 +-
> 2 files changed, 5 insertions(+), 14 deletions(-)
>
> --
> 2.36.1
>
Hi Larry,
Both patches look good to me, however I've already submitted a patch 10
days ago that does exactly what your rtw_xmit.c patch does, in
addition to fixing the changed error handling semantics for the relevant
function (in f94b47c6bde6) which broke the driver.
Regards,
Phil
On 5/31/22 01:40, Phillip Potter wrote:
>
> Hi Larry,
>
> Both patches look good to me, however I've already submitted a patch 10
> days ago that does exactly what your rtw_xmit.c patch does, in
> addition to fixing the changed error handling semantics for the relevant
> function (in f94b47c6bde6) which broke the driver.
Phil,
Sorry I missed your patch. I have been really busy the past month on another
problem, and I was not paying much attention to r8188eu.
When I built a mainline kernel in the middle of the merge from 5.18 to 5.19,
those two warnings stood out.
Your patch should be ahead of mine in Greg's queue, thus mine should fail, and I
will get the "does not apply" message. At that time, I will send the other
patch. As both of us missed the 5.19 merge, the build warnings will persist
until 5.20, but no harm.
Larry
On Tue, May 31, 2022 at 12:16:58PM -0500, Larry Finger wrote:
> On 5/31/22 01:40, Phillip Potter wrote:
> >
> > Hi Larry,
> >
> > Both patches look good to me, however I've already submitted a patch 10
> > days ago that does exactly what your rtw_xmit.c patch does, in
> > addition to fixing the changed error handling semantics for the relevant
> > function (in f94b47c6bde6) which broke the driver.
>
> Phil,
>
> Sorry I missed your patch. I have been really busy the past month on another
> problem, and I was not paying much attention to r8188eu.
>
> When I built a mainline kernel in the middle of the merge from 5.18 to 5.19,
> those two warnings stood out.
>
> Your patch should be ahead of mine in Greg's queue, thus mine should fail,
> and I will get the "does not apply" message. At that time, I will send the
> other patch. As both of us missed the 5.19 merge, the build warnings will
> persist until 5.20, but no harm.
I've taken Phillip's patch, and your second patch here, both for
5.19-final, so all should be good now.
thanks,
greg k-h