Re: re-inline sched functions

2005-03-24 Thread Ingo Molnar
* Chen, Kenneth W <[EMAIL PROTECTED]> wrote: > Ingo Molnar wrote on Friday, March 11, 2005 1:32 AM > > > -static unsigned int task_timeslice(task_t *p) > > > +static inline unsigned int task_timeslice(task_t *p) > > > > the patch looks good except this one - could you try to undo it and > > re-me

RE: re-inline sched functions

2005-03-24 Thread Chen, Kenneth W
Ingo Molnar wrote on Friday, March 11, 2005 1:32 AM > > -static unsigned int task_timeslice(task_t *p) > > +static inline unsigned int task_timeslice(task_t *p) > > the patch looks good except this one - could you try to undo it and > re-measure? task_timeslice() is not used in any true fastpath, i

RE: re-inline sched functions

2005-03-11 Thread Chen, Kenneth W
Ingo Molnar wrote on Friday, March 11, 2005 1:32 AM > > -static unsigned int task_timeslice(task_t *p) > > +static inline unsigned int task_timeslice(task_t *p) > > the patch looks good except this one - could you try to undo it and > re-measure? task_timeslice() is not used in any true fastpath, i

Re: re-inline sched functions

2005-03-11 Thread Ingo Molnar
* Chen, Kenneth W <[EMAIL PROTECTED]> wrote: > # size vmlinux.* >textdata bss dec hex filename > 3261844 717184 262020 4241048 40b698 vmlinux.x86.orig > 3262772 717488 262020 4242280 40bb68 vmlinux.x86.inline > Possible we can introduce them back? > -static unsigned in

Re: re-inline sched functions

2005-03-10 Thread Nick Piggin
Andrew Morton wrote: "Chen, Kenneth W" <[EMAIL PROTECTED]> wrote: This could be part of the unknown 2% performance regression with db transaction processing benchmark. The four functions in the following patch use to be inline. They are un-inlined since 2.6.7. We measured that by re-inline them ba

Re: re-inline sched functions

2005-03-10 Thread Andrew Morton
"Chen, Kenneth W" <[EMAIL PROTECTED]> wrote: > > This could be part of the unknown 2% performance regression with > db transaction processing benchmark. > > The four functions in the following patch use to be inline. They > are un-inlined since 2.6.7. > > We measured that by re-inline them back

re-inline sched functions

2005-03-10 Thread Chen, Kenneth W
This could be part of the unknown 2% performance regression with db transaction processing benchmark. The four functions in the following patch use to be inline. They are un-inlined since 2.6.7. We measured that by re-inline them back on 2.6.9, it improves performance for db transaction processi