On August 25, 2016 4:45:06 AM PDT, Borislav Petkov wrote:
>On Thu, Aug 25, 2016 at 03:05:19AM -0700, H. Peter Anvin wrote:
>> I'm wondering if one of those 23 invocations sets up some kind of
>> corrupt data that continues to get used.
>
>That could be one plausible explanation.
On August 25, 2016 4:45:06 AM PDT, Borislav Petkov wrote:
>On Thu, Aug 25, 2016 at 03:05:19AM -0700, H. Peter Anvin wrote:
>> I'm wondering if one of those 23 invocations sets up some kind of
>> corrupt data that continues to get used.
>
>That could be one plausible explanation. Look at what
On Thu, Aug 25, 2016 at 03:05:19AM -0700, H. Peter Anvin wrote:
> I'm wondering if one of those 23 invocations sets up some kind of
> corrupt data that continues to get used.
That could be one plausible explanation. Look at what calls
__sw_hweight64:
initmem_init
numa_policy_init
On Thu, Aug 25, 2016 at 03:05:19AM -0700, H. Peter Anvin wrote:
> I'm wondering if one of those 23 invocations sets up some kind of
> corrupt data that continues to get used.
That could be one plausible explanation. Look at what calls
__sw_hweight64:
initmem_init
numa_policy_init
On August 25, 2016 2:22:14 AM PDT, Borislav Petkov wrote:
>On Thu, Aug 18, 2016 at 06:11:39AM +0200, Borislav Petkov wrote:
>> So if there's no bug, alternatives should replace all "call
>> __sw_hweightXX" calls with POPCNT. So you shouldn't be even calling
>> these functions and
On August 25, 2016 2:22:14 AM PDT, Borislav Petkov wrote:
>On Thu, Aug 18, 2016 at 06:11:39AM +0200, Borislav Petkov wrote:
>> So if there's no bug, alternatives should replace all "call
>> __sw_hweightXX" calls with POPCNT. So you shouldn't be even calling
>> these functions and hitting that
On Thu, Aug 18, 2016 at 06:11:39AM +0200, Borislav Petkov wrote:
> So if there's no bug, alternatives should replace all "call
> __sw_hweightXX" calls with POPCNT. So you shouldn't be even calling
> these functions and hitting that path.
>
> Can you boot the kernel with "debug-alternative" and
On Thu, Aug 18, 2016 at 06:11:39AM +0200, Borislav Petkov wrote:
> So if there's no bug, alternatives should replace all "call
> __sw_hweightXX" calls with POPCNT. So you shouldn't be even calling
> these functions and hitting that path.
>
> Can you boot the kernel with "debug-alternative" and
On Wed, Aug 17, 2016 at 08:54:11PM -0700, Huang, Ying wrote:
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
> cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx
> pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology
>
On Wed, Aug 17, 2016 at 08:54:11PM -0700, Huang, Ying wrote:
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
> cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx
> pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology
>
On August 17, 2016 8:45:13 PM PDT, Borislav Petkov wrote:
>On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
>> branch-miss-rate decreased from ~0.30% to ~0.043%.
>>
>> So I guess there are some code alignment change, which caused
>decreased
>> branch miss rate.
>
>Hrrm,
On August 17, 2016 8:45:13 PM PDT, Borislav Petkov wrote:
>On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
>> branch-miss-rate decreased from ~0.30% to ~0.043%.
>>
>> So I guess there are some code alignment change, which caused
>decreased
>> branch miss rate.
>
>Hrrm, I still can't
Borislav Petkov writes:
> On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
>> branch-miss-rate decreased from ~0.30% to ~0.043%.
>>
>> So I guess there are some code alignment change, which caused decreased
>> branch miss rate.
>
> Hrrm, I still can't imagine how that
Borislav Petkov writes:
> On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
>> branch-miss-rate decreased from ~0.30% to ~0.043%.
>>
>> So I guess there are some code alignment change, which caused decreased
>> branch miss rate.
>
> Hrrm, I still can't imagine how that would happen
On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
> branch-miss-rate decreased from ~0.30% to ~0.043%.
>
> So I guess there are some code alignment change, which caused decreased
> branch miss rate.
Hrrm, I still can't imagine how that would happen if the machine
supports POPCNT and
On Wed, Aug 17, 2016 at 03:29:04PM -0700, Huang, Ying wrote:
> branch-miss-rate decreased from ~0.30% to ~0.043%.
>
> So I guess there are some code alignment change, which caused decreased
> branch miss rate.
Hrrm, I still can't imagine how that would happen if the machine
supports POPCNT and
Borislav Petkov writes:
> On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
>> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
>> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
>> >> Dang...
>> >
>> >Isn't 9.3% improvement a
Borislav Petkov writes:
> On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
>> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
>> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
>> >> Dang...
>> >
>> >Isn't 9.3% improvement a good thing(tm) ?
>>
>>
On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> >> Dang...
> >
> >Isn't 9.3% improvement a good thing(tm) ?
>
> Yes, it's huge. The
On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> >> Dang...
> >
> >Isn't 9.3% improvement a good thing(tm) ?
>
> Yes, it's huge. The only explanation
On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> >> Dang...
> >
> >Isn't 9.3% improvement a good thing(tm) ?
>
> Yes, it's huge. The
On Tue, Aug 16, 2016 at 04:09:19PM -0700, H. Peter Anvin wrote:
> On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
> >On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> >> Dang...
> >
> >Isn't 9.3% improvement a good thing(tm) ?
>
> Yes, it's huge. The only explanation
On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
>On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
>> Dang...
>
>Isn't 9.3% improvement a good thing(tm) ?
Yes, it's huge. The only explanation I could imagine is that scrambling %rdi
caused the scheduler to
On August 16, 2016 10:16:35 AM PDT, Borislav Petkov wrote:
>On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
>> Dang...
>
>Isn't 9.3% improvement a good thing(tm) ?
Yes, it's huge. The only explanation I could imagine is that scrambling %rdi
caused the scheduler to do completely
On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> Dang...
Isn't 9.3% improvement a good thing(tm) ?
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284
(AG Nürnberg)
--
On Tue, Aug 16, 2016 at 09:59:00AM -0700, H. Peter Anvin wrote:
> Dang...
Isn't 9.3% improvement a good thing(tm) ?
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284
(AG Nürnberg)
--
On August 16, 2016 7:26:43 AM PDT, kernel test robot
wrote:
>
>FYI, we noticed a 9.3% improvement of will-it-scale.per_process_ops due
>to commit:
>
>commit 65ea11ec6a82b1d44aba62b59e9eb20247e57c6e ("x86/hweight: Don't
>clobber %rdi")
On August 16, 2016 7:26:43 AM PDT, kernel test robot
wrote:
>
>FYI, we noticed a 9.3% improvement of will-it-scale.per_process_ops due
>to commit:
>
>commit 65ea11ec6a82b1d44aba62b59e9eb20247e57c6e ("x86/hweight: Don't
>clobber %rdi")
28 matches
Mail list logo