On Wed, May 16, 2007 at 12:47:54PM -0700, Linus Torvalds wrote:
>
> On Wed, 16 May 2007, Hugh Dickins wrote:
>
> > > The other option of moving the bit into ->mapping hopefully avoids all
> > > the issues, and would probably be a little faster again on the P4, at the
> > > expense of being a more
On Wed, 16 May 2007, Hugh Dickins wrote:
> On Wed, 16 May 2007, Nick Piggin wrote:
> > On Wed, May 16, 2007 at 06:54:15PM +0100, Hugh Dickins wrote:
> > > On Sun, 13 May 2007, Nick Piggin wrote:
> > > >
> > > > Well I think so, but not completely sure.
> > >
> > > That's not quite enough to co
On Wed, 16 May 2007, Nick Piggin wrote:
> On Wed, May 16, 2007 at 06:54:15PM +0100, Hugh Dickins wrote:
> > On Sun, 13 May 2007, Nick Piggin wrote:
> > >
> > > Well I think so, but not completely sure.
> >
> > That's not quite enough to convince me!
>
> I did ask Linus, and he was very sure it w
On Wed, May 16, 2007 at 06:54:15PM +0100, Hugh Dickins wrote:
> On Sun, 13 May 2007, Nick Piggin wrote:
> > On Sun, May 13, 2007 at 05:39:03AM +0100, Hugh Dickins wrote:
> > > On Sun, 13 May 2007, Nick Piggin wrote:
> > > > On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
> > > >
> >
On Sun, 13 May 2007, Nick Piggin wrote:
> On Sun, May 13, 2007 at 05:39:03AM +0100, Hugh Dickins wrote:
> > On Sun, 13 May 2007, Nick Piggin wrote:
> > > On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
> > >
> > > > Hmm, well, I think that's fairly horrid, and would it even be
> > >
On Wed, May 16, 2007 at 06:21:09PM +0100, Hugh Dickins wrote:
> On Sun, 13 May 2007, Nick Piggin wrote:
> > On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
> >
> > > But again I wonder just what the gain has been, once your double
> > > unmap_mapping_range is factored in. When I sug
On Sun, 13 May 2007, Nick Piggin wrote:
> On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
>
> > But again I wonder just what the gain has been, once your double
> > unmap_mapping_range is factored in. When I suggested before that
> > perhaps the double (well, treble including the on
On Sun, May 13, 2007 at 05:39:03AM +0100, Hugh Dickins wrote:
> On Sun, 13 May 2007, Nick Piggin wrote:
> > On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
> >
> > > Hmm, well, I think that's fairly horrid, and would it even be
> > > guaranteed to work on all architectures? Playing
On Sun, 13 May 2007, Nick Piggin wrote:
> On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
>
> > Hmm, well, I think that's fairly horrid, and would it even be
> > guaranteed to work on all architectures? Playing with one char
> > of an unsigned long in one way, while playing with the
On Fri, May 11, 2007 at 02:15:03PM +0100, Hugh Dickins wrote:
> On Fri, 11 May 2007, Nick Piggin wrote:
> >
> > Don't worry, I'm only just beginning ;) Can we then do something crazy
> > like this? (working on x86-64 only, so far. It seems to eliminate
> > lat_pagefault and lat_proc regressions h
On Fri, 11 May 2007, Nick Piggin wrote:
> On Thu, May 10, 2007 at 08:14:52PM +0100, Hugh Dickins wrote:
> >
> > Well, on the x86_64 I have seen a few of your io_schedule_timeout
> > printks under load; but suspect those are no fault of your changes,
>
> Hmm, I see... well I forgot to remove those
On Thu, May 10, 2007 at 08:14:52PM +0100, Hugh Dickins wrote:
> On Thu, 10 May 2007, Nick Piggin wrote:
> >
> > OK, I found a simple bug after pulling out my hair for a while :)
> > With this, a 4-way system survives a couple of concurrent make -j250s
> > quite nicely (wheras they eventually locke
On Thu, 10 May 2007, Nick Piggin wrote:
>
> OK, I found a simple bug after pulling out my hair for a while :)
> With this, a 4-way system survives a couple of concurrent make -j250s
> quite nicely (wheras they eventually locked up before).
>
> The problem is that the bit wakeup function did not g
On Wed, May 09, 2007 at 08:33:15PM +0100, Hugh Dickins wrote:
>
> Not good enough, I'm afraid. It looks like Ben's right and you need
> a count - and counts in the page struct are a lot harder to add than
> page flags.
>
> I've now played around with the hangs on my three 4CPU machines
> (all of
> Not good enough, I'm afraid. It looks like Ben's right and you need
> a count - and counts in the page struct are a lot harder to add than
> page flags.
>
> I've now played around with the hangs on my three 4CPU machines
> (all of them in io_schedule below __lock_page, waiting on pages
> which
On Wed, 9 May 2007, Nick Piggin wrote:
> On Wed, May 09, 2007 at 12:41:24AM +0200, Nick Piggin wrote:
> > On Wed, May 09, 2007 at 07:30:27AM +1000, Benjamin Herrenschmidt wrote:
> > >
> > > Waking them all would fix it but at the risk of causing other
> > > problems... Maybe PG_waiters need to act
On Wed, May 09, 2007 at 12:41:24AM +0200, Nick Piggin wrote:
> On Wed, May 09, 2007 at 07:30:27AM +1000, Benjamin Herrenschmidt wrote:
> > On Tue, 2007-05-08 at 13:40 +0200, Nick Piggin wrote:
> > > This patch trades a page flag for a significant improvement in the
> > > unlock_page
> > > fastpath
On Wed, May 09, 2007 at 07:30:27AM +1000, Benjamin Herrenschmidt wrote:
> On Tue, 2007-05-08 at 13:40 +0200, Nick Piggin wrote:
> > This patch trades a page flag for a significant improvement in the
> > unlock_page
> > fastpath. Various problems in the previous version were spotted by Hugh and
> >
On Tue, May 08, 2007 at 01:13:35PM +0100, David Howells wrote:
>
> Nick Piggin <[EMAIL PROTECTED]> wrote:
>
> > This patch trades a page flag for a significant improvement in the
> > unlock_page
> > fastpath. Various problems in the previous version were spotted by Hugh and
> > Ben (and fixed in
On Tue, 2007-05-08 at 13:40 +0200, Nick Piggin wrote:
> This patch trades a page flag for a significant improvement in the unlock_page
> fastpath. Various problems in the previous version were spotted by Hugh and
> Ben (and fixed in this one).
>
> Comments?
>
> --
>
> Speed up unlock_page by int
On Tue, 8 May 2007, Nick Piggin wrote:
> This patch trades a page flag for a significant improvement in the unlock_page
> fastpath. Various problems in the previous version were spotted by Hugh and
> Ben (and fixed in this one).
>
> Comments?
Seems there's still a bug there. I get hangs on the
Nick Piggin <[EMAIL PROTECTED]> wrote:
> This patch trades a page flag for a significant improvement in the unlock_page
> fastpath. Various problems in the previous version were spotted by Hugh and
> Ben (and fixed in this one).
It looks reasonable at first glance, though it does consume yet ano
22 matches
Mail list logo
