Re: [PATCH v2 11/17] wifi: cc33xx: Add init.c, init.h

[Simon Horman <horms@kernel.org>]

On Thu, Jun 20, 2024 at 11:40:31AM +0300, Nemanov, Michael wrote:
> On 6/15/2024 11:51 AM, Simon Horman wrote:
> ...
> 
> > 
> > Hi Michael,
> > 
> > allmodconfig builds on x86_64 with gcc-13 flag the following:
> > 
> > In file included from ./include/linux/string.h:374,
> >                   from ./include/linux/bitmap.h:13,
> >                   from ./include/linux/cpumask.h:13,
> >                   from ./arch/x86/include/asm/paravirt.h:21,
> >                   from ./arch/x86/include/asm/irqflags.h:60,
> >                   from ./include/linux/irqflags.h:18,
> >                   from ./include/linux/spinlock.h:59,
> >                   from ./include/linux/mmzone.h:8,
> >                   from ./include/linux/gfp.h:7,
> >                   from ./include/linux/firmware.h:8,
> >                   from drivers/net/wireless/ti/cc33xx/init.c:6:
> > In function 'fortify_memcpy_chk',
> >      inlined from 'cc33xx_init_vif_specific' at drivers/net/wireless/ti/cc33xx/init.c:156:2:
> > ./include/linux/fortify-string.h:580:25: warning: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Wattribute-warning]
> >    580 |                         __read_overflow2_field(q_size_field, size);
> >        |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> > In function 'fortify_memcpy_chk',
> >      inlined from 'cc33xx_init_vif_specific' at drivers/net/wireless/ti/cc33xx/init.c:157:2:
> > ./include/linux/fortify-string.h:580:25: warning: call to '__read_overflow2_field' declared with attribute warning: detected read beyond size of field (2nd parameter); maybe use struct_group()? [-Wattribute-warning]
> >    580 |                         __read_overflow2_field(q_size_field, size);
> >        |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >    CC [M]  drivers/net/wireless/ti/cc33xx/rx.o
> > 
> > I believe that this is because the destination for each of the two memcpy()
> > calls immediately above is too narrow - 1 structure wide instead of 4 or 8.
> > 
> > I think this can be resolved by either using:
> > 1. struct_group in .../cc33xx/conf.h:struct conf_tx_settings
> >     to wrap ac_conf0 ... ac_conf3, and separately tid_conf0 ... tid_conf7.
> > 2. Using arrays for ac_conf and tid_conf in
> >     .../cc33xx/conf.h:struct conf_tx_settings, in which case perhaps
> >     .../wlcore/conf.h:struct conf_tx_settings can be reused somehow
> >     (I did not check closely)?
> > 
> 
> Thank you for checking. I agree this code should be rewritten so it is more
> clear and w/o any warnings. Will fix.
> 
> I was unsuccessful reproducing the warning on my end. Tried with GCC 13.2.0
> (ARCH=x86_64, allmodconfig) and Arm GNU Toolchain 13.2 (ARCH=arm,
> allmodconfig) and only got errors in scan.c which I assume you refer to
> below (will also be fixed).

Hi Michael,

I tried this again with GCC 13.2.0 on x86_64 with allmodconfig.
And I was able to see this with a W=1 (make W=1) build.

I don't think it is an important detail, but for reference,
I am using the compiler here, on an x86_64 host.
https://mirrors.edge.kernel.org/pub/tools/crosstool/

> > Similar errors are flagged elsewhere in this series.
> > Please take a look at allmodconfig builds and make sure
> > no warnings are introduced.
> > 
> > Lastly, more related to the series as a whole than this patch in
> > particular, please consider running checkpatch.pl --codespell
> 
> Sure, will add checkpatch.pl --codespell to my tests.

Great, thanks.