Re: [patch -mm series] ia64 specific /dev/mem handlers

[Jes Sorensen]

>>>>> "Andrew" == Andrew Morton <akpm@xxxxxxxx> writes:

Andrew> Jes Sorensen <jes@xxxxxxxxxxxxxxxxxx> wrote:
>>  Convert /dev/mem read/write calls to use arch_translate_mem_ptr if
>> available. Needed on ia64 for pages converted fo uncached mappings
>> to avoid it being accessed in cached mode after the conversion
>> which can lead to memory corruption. Introduces PG_uncached page
>> flag for marking pages uncached.

Andrew> For some reason this patch still gives me the creeps.  Maybe
Andrew> it's because we lose a page flag for something so obscure.

Andrew> Nothing ever clears PG_uncached.  We'll end up with every page
Andrew> in the machine marked as being uncached.

Actually there's restrictions to how many pages are getting converted
as converting pages over from cached to uncached isn't trivial on ia64.

Andrew> But then, nothing ever sets PG_uncached, either.  Is there
Andrew> some patch which you're hiding from me?

Actually I posted that earlier, but it must have gotten lost in the
noise. It's part of the genalloc/mspec patchset. I'll send it to you
directly.

Andrew> If a page is marked uncached then it'll remain marked as
Andrew> uncached even after it's unmapped.  Or will it?  Would like to
Andrew> see the other patch, please.

Coming your way in a jiffy.

Andrew> We should add PG_uncached checks to the page allocator.  Is
Andrew> this OK?

I don't see any problems with that. The way it's meant to be used is
that once pages are converted over, they don't go back into the
allocator.

Cheers,
Jes
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/