Merging 64 bit changes to -HEAD

Jayachandran C. c.jayachandran at gmail.com
Wed Jun 30 23:05:27 UTC 2010


On Thu, Jul 1, 2010 at 4:17 AM, Juli Mallett <jmallett at freebsd.org> wrote:
> On Wed, Jun 30, 2010 at 15:40, Jayachandran C. <c.jayachandran at gmail.com> wrote:
>> I've attached the final(final.diff) version I want to check-in, can
>> you please quickly test it?
>>
>> If that does not work, can you tell me if the attached alt1.diff or
>> alt2.diff works? The try.diff had three changes: handle case of
>> index>0, remove pagemask operation, restore full entryhi instead of
>> asid. So if the first does not work, this will help narrow down the
>> rest of the cases.
>>
>> Hopefully this is the last iteration :)
>
> Hi JC,
>
> I think you're right that the issue is not masking enough bits in
> EntryHi.  I suspect the other changes are not necessary and I feel
> that the change to always add the entry if it doesn't exist is
> undesirable if it's unnecessary.  When a lot of otherwise-unused
> mappings are changing, this has the effect of spilling the TLB which
> kills performance.  I hope someone is willing to try just the
> TLBHI_ENTRY changes to make sure that fixes it.

I had a patch that did just the EntryHi thing (the ealier mail
pte.h-fix.diff), and that did not fix the issue, that is why I just
ended up doing what the original code did.

Also, I don't think the pagemask read/write is necessary, it should
not have any effect, as far i can see.

JC.


More information about the freebsd-mips mailing list