Le 30/11/2023 à 22:30, Peter Xu a écrit :
> On Fri, Nov 24, 2023 at 11:07:51AM -0500, Peter Xu wrote:
>> On Fri, Nov 24, 2023 at 09:06:01AM +0000, Ryan Roberts wrote:
>>> I don't have any micro-benchmarks for GUP though, if that's your question. 
>>> Is
>>> there an easy-to-use test I can run to get some numbers? I'd be happy to 
>>> try it out.
>> Thanks Ryan.  Then nothing is needed to be tested if gup is not yet touched
>> from your side, afaict.  I'll see whether I can provide some rough numbers
>> instead in the next post (I'll probably only be able to test it in a VM,
>> though, but hopefully that should still reflect mostly the truth).
> An update: I finished a round of 64K cont_pte test, in the slow gup micro
> benchmark I see ~15% perf degrade with this patchset applied on a VM on top
> of Apple M1.
> Frankly that's even less than I expected, considering not only how slow gup
> THP used to be, but also on the fact that that's a tight loop over slow
> gup, which in normal cases shouldn't happen: "present" ptes normally goes
> to fast-gup, while !present goes into a fault following it.  I assume
> that's why nobody cared slow gup for THP before.  I think adding cont_pte
> support shouldn't be very hard, but that will include making cont_pte idea
> global just for arm64 and riscv Svnapot.

Is there any documentation on what cont_pte is ? I have always wondered 
if it could also fit powerpc 8xx need ?

On powerpc, for 16k pages, we have to define 4 consecutive PTEs. All 4 
PTE are flagged with the SPS bit telling it's a 16k pages, but for TLB 
misses the HW needs one entrie for each 4k fragment.

There is also a similar approach for 512k pages, we have 128 contiguous 
identical PTEs for them.

And whatever PAGE_SIZE is (either 4k or 16k), the HW needs one 'unsigned 
long' pte for each 4k fragment. So at the time being when we define 
PAGE_SIZE as 16k, we need a special pte_t which is a table of 4x 
unsigned long.

Wondering if the cont_pte concept is similar and whether it could help.


Reply via email to