Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Linus Torvalds
On Fri, 25 May 2007, Ingo Molnar wrote: > > sure enough, i'll work something sane out - i already have it partly > split up. It will probably be something along the lines of: 'remove > stuff that we can remove and still have functional scheduling' followed > by an 'add minimal CFS patch' and

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Ingo Molnar
* Linus Torvalds <[EMAIL PROTECTED]> wrote: > Ingo: to make things easier, the _best_ split-up would be not "this is > the historical series of patches leading up to CFS v15" or something > like that, but it would be nice to get that final CFS version as a > series of patches that do some spec

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Andi Kleen
On Fri, May 25, 2007 at 09:49:38AM -0700, Linus Torvalds wrote: > > > On Fri, 25 May 2007, Ingo Molnar wrote: > > > > ok - then please merge that single hunk into the paravirtops patch - and > > leave the other 6 hunks in this patch. > > Note: I'm actually much more interested in applying the

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Linus Torvalds
On Fri, 25 May 2007, Ingo Molnar wrote: > > ok - then please merge that single hunk into the paravirtops patch - and > leave the other 6 hunks in this patch. Note: I'm actually much more interested in applying the scheduler changes than the paravirt-ops changes. It looks like CFS is getting s

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Andi Kleen
> I wonder if this was the source of the lockdep selftest failures, or the > mystery hang this patch caused (IIRC).. No, the inbalance was just on the ftp server for an hour or so. I doubt anybody except Ingo ever saw that code. I introduced it while changing the callbacks around to make the code

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Daniel Walker
On Fri, 2007-05-25 at 09:17 -0700, Andrew Morton wrote: > On Fri, 25 May 2007 14:15:40 +0200 Andi Kleen <[EMAIL PROTECTED]> wrote: > > > -Andi (who hopes this thread will end soon now and we can all go > > back to more important issues) > > fyi, the thread has been damn useful for me. If Ingo ha

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Andrew Morton
On Fri, 25 May 2007 14:15:40 +0200 Andi Kleen <[EMAIL PROTECTED]> wrote: > -Andi (who hopes this thread will end soon now and we can all go > back to more important issues) fyi, the thread has been damn useful for me. If Ingo hadn't spotted that preempt_count() imbalance then there's a decent ch

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Andi Kleen
On Fri, May 25, 2007 at 02:02:28PM +0200, Ingo Molnar wrote: > > * Andi Kleen <[EMAIL PROTECTED]> wrote: > > > > find below the cleanups from my first patch that didnt make it into > > > your cleanups. (plus one more cleanup i noticed while merging the > > > missing bits from my first patch) Go

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Ingo Molnar
* Andi Kleen <[EMAIL PROTECTED]> wrote: > > find below the cleanups from my first patch that didnt make it into > > your cleanups. (plus one more cleanup i noticed while merging the > > missing bits from my first patch) Goes after the bugfix i just sent. > > Please apply. > > I cannot apply i

Re: [patch] sched_clock(): cleanups, #2

2007-05-25 Thread Andi Kleen
On Fri, May 25, 2007 at 01:50:04PM +0200, Ingo Molnar wrote: > > * Ingo Molnar <[EMAIL PROTECTED]> wrote: > > > * Andi Kleen <[EMAIL PROTECTED]> wrote: > > > > > > please send me your current sched-clock.c, i'll redo any remaining > > > > cleanups. > > > > > > It needs at least one new prelimi