Re: [PATCH 1/9] arm: select different compiler flags for ARM CortexM3

[Paul Osmialowski]

Hi Russell,

Thanks for the input. Seems like I trusted get_maintainer.pl blindly and 
wasn't selective - it resulted in recipient list much too long.

As for this patch, I think I went too far with adding it. The board boots 
without it properly, and since it covers more general Cortex-M related 
problems I guess it should be excluded from next iteration of the whole 
patchset.

The problems it adresses may still required further discussion, but I 
think it should not block the whole thing.

On Tue, 23 Jun 2015, Russell King - ARM Linux wrote:

> On Tue, Jun 23, 2015 at 11:19:39PM +0200, Paul Osmialowski wrote:
> > This one is inspired by two commits published on Emcraft git repo:
> >
> > https://github.com/EmcraftSystems/linux-emcraft.git
> >
> > 1) 6302b692f570ff9d5645a6e72c11f87b0c1aa409
> >  RT #62654. Fixed kernel crashes while running httpd by enabling
> >  "-mfix-cortex-m3-ldrd" compiler option, which prevents compiler from
> >  generating code like 'ldrd Ra, Rb, [Ra, #Imm]' - according to the 602117
> >  Cortex-M3 Errata it may result in incorrect base register when interrupted
> >  or faulted.
> >
> > by: Yuri Tikhonov <yur@xxxxxxxxxxx>
> >
> > 2) 359d3cda84c01c0f3fae1a519b97a31f318f57ab
> >  RT #62654. Removed "--march=..." leaving only "-mcpu=cortex-m3" to make
> >  sure only the correct instructions will be generated.
> >
> > by: Sergei Poselenov <sposelenov@xxxxxxxxxxx>
> >
> > I reworked these patches to make them less intrusive.
> >
> > Signed-off-by: Paul Osmialowski <pawelo@xxxxxxxxxxx>
>
> NAK.
>
> The EFM32 code already added core support for Cortex-M3 cores, under
> the symbol CPU_V7M.  Rather than implementing a whole new set of
> Cortex-M3 support alongside the existing code, please work with the
> EFM32 maintainer (Uwe Kleine-König) to come up with a common set of
> core changes that you can all agree on for Cortex-M3.
>
> Please also try to be a little smarter with whom you're sending your
> patches to, the Cc list looks to be excessively long - having soo many
> recipients on your message is a good way to get your message rejected
> by mailing lists as spam.
>
> --
> FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
> according to speedtest.net.