Re: [PATCH RESEND 0/5] arm: sa1100: add cpu clock for fbdev and pcmcia

Dear Russell,

2014-11-06 10:41 GMT+03:00 Dmitry Eremin-Solenikov [off-list ref]:

These patches are required to make SL-5500 (collie) to work properly.
Framebuffer maintainer added his ack to the respective patch. PCMCIA
subsystem seems to be unmaintained at this moment. Russell, could you
please review and hopefully ack these patches?

On SA-1100 framebuffer and PCMCIA drivers make use of cpufreq_get(0)
function call to determine the cpu frequency. Russell's commit
1937f5b91833e2e8e53bcc821fc7a5fbe6ccb9b5 (ARM: fix sa1100 build) fixed
the build issues, but broke two devices (Collie and Jornada720). For
those two boards the cpufreq code gets compiled but is not enabled (as
board files do not provide timing information for the CPUFREQ driver).
Thus cpufreq_get(0) returns incorrect value and incorrect timings get
programmed into the hardware.

PXA2xx (the very similar platform) uses Clock API to determine CPU
frequency both in framebuffer and PCMCIA drivers. These patches make
similar changes to StrongARM drivers.

We can continue to carry over this patch set in local tree. Having a grave bug
in upstream kernels. What is the purpose of upstream kernel then? Why did
we have so many talks about bad practice of 'vendor/local/private' Linux trees?

Do you see anything wrong with this patchset? Why do you keep on ignoring it?
Did you stop caring about sa11x0?

-- 
With best wishes
Dmitry