Messages in this thread | | | Date | Sat, 16 Oct 2021 15:22:49 +0200 | From | Borislav Petkov <> | Subject | Re: 回复:[PATCH] perf : optimiz e clear page in Intel specified model with movq instruction |
| |
On Sat, Oct 16, 2021 at 08:58:32PM +0800, JY Ni wrote: > I rebased this patch on linux-next repo and measured the time of building a > same kernel in original/add-movq-patch version on the same intel CPX server.
...
> delta = (*build_original_time* - *build_movq_time*) / (*build_movq_time*) > > This set of data shows that movq-patch version has better performance than > original version in most cases.
First of all, please do not top-post but put your answer under the quoted text like everyone else.
Then, please explain how exactly you ran that measurement so that I can try to reproduce it here too.
And just to make sure we're talking about the same thing: the patch in question is this one:
https://lore.kernel.org/r/1631177151-53723-1-git-send-email-wujinhua@linux.alibaba.com
correct?
Thx.
-- Regards/Gruss, Boris.
https://people.kernel.org/tglx/notes-about-netiquette
| |