[Letux-kernel] jz4730-i2c - clocksource

[H. Nikolaus Schaller]

> Am 06.03.2021 um 17:07 schrieb Paul Boddie <paul at boddie.org.uk>:
> 
> On Saturday, 6 March 2021 16:22:50 CET H. Nikolaus Schaller wrote:
>> latest code is now here:
>> 
>> https://git.goldelico.com/?p=letux-kernel.git;a=shortlog;h=refs/heads/letux/
> jz4730-v9

Now there is also letux-current which is the latest complete tree that I did build
and test.

> 
> I'll try and build this, at least.
> 
> As for the futex support, I did notice that the Ingenic CPUs supposedly have 
> cpu_has_llsc defined as 1 in...
> 
> arch/mips/include/asm/mach-ingenic/cpu-feature-overrides.h
> 
> In the legacy kernel, there is no similar overriding for Ingenic CPUs, and 
> instead it seems that arch/mips/kernel/cpu-probe.c does some work testing the 
> processor identifier against known values and setting MIPS_CPU_LLSC 
> accordingly, this later being used in include/asm-mips/cpu-features.h to 
> define cpu_has_llsc.

Ok, that is really intersting. Either the jz4730 is more special than all
other Ingenic CPUs. Or we can in addition have some bug in the CONFIG which
should be different for CONFIG_JZ4730.

But what would happen if cpu_has_llsc is wrongly defined as "1"? Shouldn't
this trigger some ILLEGAL INSTRUCTION fault in kernel and perhaps panic
the kernel?

> 
> So it is possible that LL/SC support is missing and that this flag might be 
> unset for better performance. But that is just an idea.

We should give it a try...

Best regards,
Nikolaus