Re: [PATCH] [ARM] force dcache flush if dcache_dirty bit set

[Paul Mundt <lethal@linux-sh.org>]

On Mon, Oct 12, 2009 at 06:03:12PM +0100, Russell King wrote:
> On Mon, Oct 12, 2009 at 05:09:53PM +0100, Hugh Dickins wrote:
> > Sorry to muddy the waters on this, if you and Dave are sure that
> > you have the right fix, down in your architectures, and that fix
> > isn't going to hurt your performance significantly.
> 
> If I look at the issue from this point of view:
> 
> - we are using PG_arch_1 to delay cache handling for the page
> 
> - if PG_arch_1 is set on a page, we set it explicitly because we
>   didn't do some flushing between the allocation of the page and
>   mapping it into userspace
> 
> - if a page with PG_arch_1 set ever gets to userspace, this can
>   only be because we did the lazy flushing thing
> 
> I don't see that there should have been any bearing on whether a page
> has a mapping or not when we get to update_mmu_cache.  The issue here
> is that > if PG_arch_1 is set on a page, then we didn't flush it at
> the time when we believed it was appropriate to do so. <
> 
> Tell me I'm wrong (having only just sent it to Linus...)
> 
Having looked at the ARM fix, in the !mapping case do you not need the
I-cache flush on vma->vm_flags & VM_EXEC? Or is the presumption that
flush_icache_page()-type action doesn't need to be undertaken by
flush_dcache_page()/update_mmu_cache() when there is no page_mapping()?