[PATCH V3 2/2] ARM: Handle user space mapped pages in flush_kernel_dcache_page

[Ming Lei]

On Wed, May 8, 2013 at 11:08 PM, Catalin Marinas
<catalin.marinas at arm.com> wrote:
> On Sat, May 04, 2013 at 09:21:27AM +0100, Ming Lei wrote:
>> On Fri, May 3, 2013 at 6:02 PM, Catalin Marinas <catalin.marinas at arm.com> wrote:
>> >>
>> >> I assume that you inhibited the call to flush_dcache_page() in
>> >> __get_user_pages() for anon pages.  Otherwise, you will be flooded
>> >> with warnings.
>> >
>> > I haven't done any stress testing so I don't think I hit this code path,
>> > so no warning. But yes, it should have triggered. Anyway, in this case
>> > flush_dcache_page() should have just ignored (clearing PG_arch_1 is
>> > harmless anyway if we also ignore this bit in __sync_icache_dcache for
>> > non-aliasing caches).
>>
>> Yes, maybe we can do a little optimization for O_DIRECT since no
>> dcache alias and I/D coherency problem in this case on ARMv7, how
>> about below change?
>>
>> diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
>> index 1c8f7f5..962a657 100644
>> --- a/arch/arm/mm/flush.c
>> +++ b/arch/arm/mm/flush.c
>> @@ -287,6 +287,8 @@ void flush_dcache_page(struct page *page)
>>           mapping && !mapping_mapped(mapping))
>>               clear_bit(PG_dcache_clean, &page->flags);
>>       else {
>> +             if (!mapping && cache_is_vipt_nonaliasing())
>> +                     return;
>>               __flush_dcache_page(mapping, page);
>>               if (mapping && cache_is_vivt())
>>                       __flush_dcache_aliases(mapping, page);
>
> I wonder whether we could move the:
>
>         if (!mapping)
>                 return
>
> at the top of this function, IOW don't touch any anonymous pages. Would
> anything be broken (apart from wrong API use)?

In case of O_DIRECT, the anonymous page may be shared, so looks it
should be flushed for vipt_aliasing cache.

Catalin, btw, would you mind give comments on the patch below(it might
be a 'fix' on one commit you submitted)?

      http://marc.info/?l=linux-arm-kernel&m=136757247311694&w=2

Thanks
-- 
Ming Lei