On Mon, Apr 14, 2014 at 09:31:16PM +0800, Ling Ma wrote:
> The kernel version 3.14 shows memcpy, memset occur 19622 and 14189
> times respectively.
> so memset is still important for us, correct?
Did you ever see it in a profile log as being hot?
I haven't. Static counts don't mean much.
-Andi
-
The kernel version 3.14 shows memcpy, memset occur 19622 and 14189
times respectively.
so memset is still important for us, correct?
Thanks
Ling
2014-04-14 6:03 GMT+08:00, Andi Kleen :
> On Sun, Apr 13, 2014 at 11:11:59PM +0800, Ling Ma wrote:
>> Any further comments ?
>
> It would be good to t
On Sun, Apr 13, 2014 at 11:11:59PM +0800, Ling Ma wrote:
> Any further comments ?
It would be good to test on some more machines that you don't cause
regressions.
But I'm not aware of any workload that is doing a lot of memset (not
counting clear_page). Copies likely matter much more.
-Andi
--
Any further comments ?
Thanks
Ling
2014-04-08 22:00 GMT+08:00, Ling Ma :
> Andi,
>
> The below is compared result on older machine(cpu info is attached):
> That shows new code get better performance up to 1.6x.
>
> Bytes: ORG_TIME: NEW_TIME: ORG vs NEW:
> 7 0.870.761.14
> 16
Andi,
The below is compared result on older machine(cpu info is attached):
That shows new code get better performance up to 1.6x.
Bytes: ORG_TIME: NEW_TIME: ORG vs NEW:
7 0.870.761.14
16 0.990.681.45
18 1.070.771.38
21 1.090.781.39
25 1.11
ling.ma.prog...@gmail.com writes:
> From: Ling Ma
>
> In this patch we manage to reduce miss branch prediction by
> avoiding using branch instructions and force destination to be aligned
> with general 64bit instruction.
> Below compared results shows we improve performance up to 1.8x
> (We mod
Append test suit
after tar, run ./test command please.
thanks
2014-04-07 22:50 GMT+08:00, ling.ma.prog...@gmail.com
:
> From: Ling Ma
>
> In this patch we manage to reduce miss branch prediction by
> avoiding using branch instructions and force destination to be aligned
> with general 64bit inst
7 matches
Mail list logo
