Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-29 Thread Nick Piggin
On Saturday 29 September 2007 03:57, Mathieu Desnoyers wrote: > Isn't the actual instrumentation present in the VM subsystem consisting > mostly of event counters ? This kind of profiling provides limited help in > following specific delays in the kernel. Martin Bligh's paper "Linux > Kernel

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-29 Thread Nick Piggin
On Saturday 29 September 2007 03:57, Mathieu Desnoyers wrote: Isn't the actual instrumentation present in the VM subsystem consisting mostly of event counters ? This kind of profiling provides limited help in following specific delays in the kernel. Martin Bligh's paper Linux Kernel Debugging

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Frank Ch. Eigler
Mathieu Desnoyers <[EMAIL PROTECTED]> writes: >> [...] >> > I totally agree with Ingo here. Having a basic instrumentation that is >> > enabled by default will help to identify code paths causing unexpected >> > delays in the kernel. [...] > >> It is. See: CONFIG_VM_EVENT_COUNTERS and all the

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Mathieu Desnoyers
* Nick Piggin ([EMAIL PROTECTED]) wrote: > On Saturday 29 September 2007 02:23, Mathieu Desnoyers wrote: > > * Ingo Molnar ([EMAIL PROTECTED]) wrote: > > > * Andrew Morton <[EMAIL PROTECTED]> wrote: > > > > This is a pretty major bugfix. > > > > > > > > GFP_NOIO and GFP_NOFS callers should have

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Saturday 29 September 2007 02:23, Mathieu Desnoyers wrote: > * Ingo Molnar ([EMAIL PROTECTED]) wrote: > > * Andrew Morton <[EMAIL PROTECTED]> wrote: > > > This is a pretty major bugfix. > > > > > > GFP_NOIO and GFP_NOFS callers should have been spending really large > > > amounts of time stuck

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Mathieu Desnoyers
* Ingo Molnar ([EMAIL PROTECTED]) wrote: > > * Andrew Morton <[EMAIL PROTECTED]> wrote: > > > This is a pretty major bugfix. > > > > GFP_NOIO and GFP_NOFS callers should have been spending really large > > amounts of time stuck in that sleep. > > > > I wonder why nobody noticed this

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Friday 28 September 2007 18:02, Ingo Molnar wrote: > * Andrew Morton <[EMAIL PROTECTED]> wrote: > > This is a pretty major bugfix. > > > > GFP_NOIO and GFP_NOFS callers should have been spending really large > > amounts of time stuck in that sleep. > > > > I wonder why nobody noticed this

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Thursday 27 September 2007 11:50, Fengguang Wu wrote: > We don't want to introduce pointless delays in throttle_vm_writeout() > when the writeback limits are not yet exceeded, do we? I don't think so (ie. I agree with you). IIRC, Marcelo initially did the throttle_vm_writeout? > Cc: Nick

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Ingo Molnar
* Andrew Morton <[EMAIL PROTECTED]> wrote: > This is a pretty major bugfix. > > GFP_NOIO and GFP_NOFS callers should have been spending really large > amounts of time stuck in that sleep. > > I wonder why nobody noticed this happening. Either a) it turns out > that kswapd is doing a good

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Ingo Molnar
* Andrew Morton [EMAIL PROTECTED] wrote: This is a pretty major bugfix. GFP_NOIO and GFP_NOFS callers should have been spending really large amounts of time stuck in that sleep. I wonder why nobody noticed this happening. Either a) it turns out that kswapd is doing a good job and

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Thursday 27 September 2007 11:50, Fengguang Wu wrote: We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? I don't think so (ie. I agree with you). IIRC, Marcelo initially did the throttle_vm_writeout? Cc: Nick Piggin

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Friday 28 September 2007 18:02, Ingo Molnar wrote: * Andrew Morton [EMAIL PROTECTED] wrote: This is a pretty major bugfix. GFP_NOIO and GFP_NOFS callers should have been spending really large amounts of time stuck in that sleep. I wonder why nobody noticed this happening. Either

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Nick Piggin
On Saturday 29 September 2007 02:23, Mathieu Desnoyers wrote: * Ingo Molnar ([EMAIL PROTECTED]) wrote: * Andrew Morton [EMAIL PROTECTED] wrote: This is a pretty major bugfix. GFP_NOIO and GFP_NOFS callers should have been spending really large amounts of time stuck in that sleep.

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Mathieu Desnoyers
* Ingo Molnar ([EMAIL PROTECTED]) wrote: * Andrew Morton [EMAIL PROTECTED] wrote: This is a pretty major bugfix. GFP_NOIO and GFP_NOFS callers should have been spending really large amounts of time stuck in that sleep. I wonder why nobody noticed this happening. Either a) it

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Mathieu Desnoyers
* Nick Piggin ([EMAIL PROTECTED]) wrote: On Saturday 29 September 2007 02:23, Mathieu Desnoyers wrote: * Ingo Molnar ([EMAIL PROTECTED]) wrote: * Andrew Morton [EMAIL PROTECTED] wrote: This is a pretty major bugfix. GFP_NOIO and GFP_NOFS callers should have been spending really

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-28 Thread Frank Ch. Eigler
Mathieu Desnoyers [EMAIL PROTECTED] writes: [...] I totally agree with Ingo here. Having a basic instrumentation that is enabled by default will help to identify code paths causing unexpected delays in the kernel. [...] It is. See: CONFIG_VM_EVENT_COUNTERS and all the other vm specific

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Fengguang Wu
On Thu, Sep 27, 2007 at 11:16:10AM -0400, Rik van Riel wrote: > On Thu, 27 Sep 2007 09:50:16 +0800 > Fengguang Wu <[EMAIL PROTECTED]> wrote: > > > We don't want to introduce pointless delays in throttle_vm_writeout() > > when the writeback limits are not yet exceeded, do we? > > Good catch.

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Andrew Morton
On Thu, 27 Sep 2007 09:50:16 +0800 Fengguang Wu <[EMAIL PROTECTED]> wrote: > We don't want to introduce pointless delays in throttle_vm_writeout() > when the writeback limits are not yet exceeded, do we? > > Cc: Nick Piggin <[EMAIL PROTECTED]> > Cc: OGAWA Hirofumi <[EMAIL PROTECTED]> > Cc: Kumar

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Rik van Riel
On Thu, 27 Sep 2007 09:50:16 +0800 Fengguang Wu <[EMAIL PROTECTED]> wrote: > We don't want to introduce pointless delays in throttle_vm_writeout() > when the writeback limits are not yet exceeded, do we? Good catch. > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> Reviewed-by: Rik van Riel

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Rik van Riel
On Thu, 27 Sep 2007 09:50:16 +0800 Fengguang Wu [EMAIL PROTECTED] wrote: We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? Good catch. Signed-off-by: Fengguang Wu [EMAIL PROTECTED] Reviewed-by: Rik van Riel [EMAIL

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Andrew Morton
On Thu, 27 Sep 2007 09:50:16 +0800 Fengguang Wu [EMAIL PROTECTED] wrote: We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? Cc: Nick Piggin [EMAIL PROTECTED] Cc: OGAWA Hirofumi [EMAIL PROTECTED] Cc: Kumar Gala [EMAIL

Re: [PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-27 Thread Fengguang Wu
On Thu, Sep 27, 2007 at 11:16:10AM -0400, Rik van Riel wrote: On Thu, 27 Sep 2007 09:50:16 +0800 Fengguang Wu [EMAIL PROTECTED] wrote: We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? Good catch. Thank you.

[PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-26 Thread Fengguang Wu
We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? Cc: Nick Piggin <[EMAIL PROTECTED]> Cc: OGAWA Hirofumi <[EMAIL PROTECTED]> Cc: Kumar Gala <[EMAIL PROTECTED]> Cc: Pete Zaitcev <[EMAIL PROTECTED]> Cc: Greg KH <[EMAIL

[PATCH] writeback: remove unnecessary wait in throttle_vm_writeout()

2007-09-26 Thread Fengguang Wu
We don't want to introduce pointless delays in throttle_vm_writeout() when the writeback limits are not yet exceeded, do we? Cc: Nick Piggin [EMAIL PROTECTED] Cc: OGAWA Hirofumi [EMAIL PROTECTED] Cc: Kumar Gala [EMAIL PROTECTED] Cc: Pete Zaitcev [EMAIL PROTECTED] Cc: Greg KH [EMAIL PROTECTED]