[U-Boot] [PATCH v5 0/4] Buffer overruns in printf

Simon Glass sjg at chromium.org
Wed Nov 2 02:18:31 CET 2011


Hi Wolfgang,

On Tue, Nov 1, 2011 at 2:21 PM, Wolfgang Denk <wd at denx.de> wrote:
> Dear Simon Glass,
>
> In message <CAPnjgZ2wW3tvGjxWgkVr0PHLr2Ttj8abtPc4aiyk=tgnpsSk3g at mail.gmail.com> you wrote:
>>
>> I have suggested making these functions available by default, and
>> having CONFIG_NO_SYS_VSNPRINT to remove them. The rationale is that it
>
> This makes little sense to me.  We don't need an opt out here, but an
> opt in instead.  Otherwise you willprobably see a number of builds
> breaking due to the increased code size.

OK I can change this to CONFIG_SYS_VSNPRINT easily enough.

>
>> There is therefore a code size increase by default with this series.
>> We can instead use CONFIG_SYS_VSNPRINT if people think it is safe.
>
> Which boards were build tested with these patches applied?

I didn't see much difference with MAKEALL (but a lot of boards give
warnings still). It would be great if I could get a clean build with
no USB warnings, etc. At present I see about 180 with warnings out of
249, but I didn't see any that seemed to be affected by this patch.
Perhaps they don't use printf() very early in the init when code size
matters?

I am going to try to apply a few fix-up patches and see if I can get a
clearer picture.

Regards,
Simon

>
> Best regards,
>
> Wolfgang Denk
>
> --
> DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
> HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
> Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
> Sex is like air.  It's only a big deal if you can't get any.
>


More information about the U-Boot mailing list