On Tue, Jun 26, 2007 at 10:14:40AM -0700, Andrew Morton wrote:
> On my machine, time(2) doesn't do any syscall at all - it uses the vsyscall
> page. I'd be surprised if a database uses sys_time() either.
Large boxes unfortunately can't always use vsyscalls... that's a real
pity. I also had to
On Tue, Jun 26, 2007 at 10:13:31AM -0700, Ray Lee wrote:
> faster? Weird. It shouldn't. They must be doing something wrong,
> therefore the patch is stupid."
Just in case it's not obvious the above are Ray Lee words, mine not.
---
#!/usr/bin/env stap
# edited top.stp from systemtap
On Mon, 25 Jun 2007 15:15:08 -0700 Andrew Morton <[EMAIL PROTECTED]> wrote:
> > What did you test?
> > There can be many ways to read the clock, do you want to put this hook
> > everywhere?
>
> Yeah, it isn't immediately obvious (to this little black duck) why similar
> fixups weren't needed in
On Tue, 26 Jun 2007 19:08:27 +0200 (CEST) Roman Zippel <[EMAIL PROTECTED]>
wrote:
> > This current ... interesting piece of Roman about a _single_ trivial
> > unlikely() branch in do_gettimeofday() borders on the ridiculous. My
> > patch might be wrong for various reasons, but that single
> >
On Tue, 26 Jun 2007 17:26:29 +0200 Ingo Molnar <[EMAIL PROTECTED]> wrote:
>
> * Andrew Morton <[EMAIL PROTECTED]> wrote:
>
> > > > the patch improves the sysbench OLTP macrobenchmark significantly:
> > >
> > > Has that any real practical relevance?
> >
> > Interesting question. [...]
>
> i'm
On 6/26/07, Andrea Arcangeli <[EMAIL PROTECTED]> wrote:
On Tue, Jun 26, 2007 at 06:18:57PM +0200, Ingo Molnar wrote:
> My patch improves MySQL wall-clock performance by ~10% on a dual-core
> box [see the numbers i cited in my initial mail, they were ran on a T60
> with a 1.83 GHz Core2Duo] and
Hi,
On Tue, 26 Jun 2007, Ingo Molnar wrote:
Another BTW before someone takes this seriously:
> ( whether there is any correlation between a decade long fundamental
> suckage and stagnation of the Linux time and NTP subsystem and Roman's
> decade long negative feedback presence in that area
On Tue, 2007-06-26 at 16:58 +0200, Ingo Molnar wrote:
> * Mark Lord <[EMAIL PROTECTED]> wrote:
>
> > Ingo Molnar wrote:
> > >Subject: [patch] sys_time() speedup
> > >From: Ingo Molnar <[EMAIL PROTECTED]>
> > >
> > >improve performance of sys_time(). sys_time() returns time in seconds,
> > >but
On Tue, Jun 26, 2007 at 06:18:57PM +0200, Ingo Molnar wrote:
> My patch improves MySQL wall-clock performance by ~10% on a dual-core
> box [see the numbers i cited in my initial mail, they were ran on a T60
> with a 1.83 GHz Core2Duo] and by 7% on an 8-way box:
mysql isn't froznen in stone like
Hi,
On Tue, 26 Jun 2007, Ingo Molnar wrote:
> if you are curious why Roman's reaction to this patch was so negative:
Instead of answering all the open questions, pretty much the second thing
you do is to discredit me personally. :-(
BTW there is a difference between critical and negative...
* Jesper Juhl <[EMAIL PROTECTED]> wrote:
> > Ingo provided no information about his test setup and his patch was
> > a little strange, so I can't say that yet.
>
> What did you find strange about it? I'm currious. Sure it needs
> testing and of course it would be nice with some more details on
On Mon, Jun 25, 2007 at 03:15:08PM -0700, Andrew Morton wrote:
> Interesting question. The patch adds a new test-n-branch to gettimeofday()
> so if gettimeofday() is used much more frequently than time(), we lose.
I think gettimeofday is generally used much more frequently than
time. Real db
* Andrew Morton <[EMAIL PROTECTED]> wrote:
> > > the patch improves the sysbench OLTP macrobenchmark significantly:
> >
> > Has that any real practical relevance?
>
> Interesting question. [...]
i'm missing the tag i guess ;-)
Oh my, does database macro-performance have any relevance to
* Mark Lord <[EMAIL PROTECTED]> wrote:
> Ingo Molnar wrote:
> >Subject: [patch] sys_time() speedup
> >From: Ingo Molnar <[EMAIL PROTECTED]>
> >
> >improve performance of sys_time(). sys_time() returns time in seconds,
> >but it does so by calling do_gettimeofday() and then returning the
>
* Mark Lord [EMAIL PROTECTED] wrote:
Ingo Molnar wrote:
Subject: [patch] sys_time() speedup
From: Ingo Molnar [EMAIL PROTECTED]
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of
* Andrew Morton [EMAIL PROTECTED] wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
Interesting question. [...]
i'm missing the sarcastic tag i guess ;-)
sarcastic Oh my, does database macro-performance have any
On Mon, Jun 25, 2007 at 03:15:08PM -0700, Andrew Morton wrote:
Interesting question. The patch adds a new test-n-branch to gettimeofday()
so if gettimeofday() is used much more frequently than time(), we lose.
I think gettimeofday is generally used much more frequently than
time. Real db calls
* Jesper Juhl [EMAIL PROTECTED] wrote:
Ingo provided no information about his test setup and his patch was
a little strange, so I can't say that yet.
What did you find strange about it? I'm currious. Sure it needs
testing and of course it would be nice with some more details on
Ingo's
Hi,
On Tue, 26 Jun 2007, Ingo Molnar wrote:
if you are curious why Roman's reaction to this patch was so negative:
Instead of answering all the open questions, pretty much the second thing
you do is to discredit me personally. :-(
BTW there is a difference between critical and negative...
On Tue, Jun 26, 2007 at 06:18:57PM +0200, Ingo Molnar wrote:
My patch improves MySQL wall-clock performance by ~10% on a dual-core
box [see the numbers i cited in my initial mail, they were ran on a T60
with a 1.83 GHz Core2Duo] and by 7% on an 8-way box:
mysql isn't froznen in stone like
On Tue, 2007-06-26 at 16:58 +0200, Ingo Molnar wrote:
* Mark Lord [EMAIL PROTECTED] wrote:
Ingo Molnar wrote:
Subject: [patch] sys_time() speedup
From: Ingo Molnar [EMAIL PROTECTED]
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling
Hi,
On Tue, 26 Jun 2007, Ingo Molnar wrote:
Another BTW before someone takes this seriously:
( whether there is any correlation between a decade long fundamental
suckage and stagnation of the Linux time and NTP subsystem and Roman's
decade long negative feedback presence in that area of
On 6/26/07, Andrea Arcangeli [EMAIL PROTECTED] wrote:
On Tue, Jun 26, 2007 at 06:18:57PM +0200, Ingo Molnar wrote:
My patch improves MySQL wall-clock performance by ~10% on a dual-core
box [see the numbers i cited in my initial mail, they were ran on a T60
with a 1.83 GHz Core2Duo] and by 7%
On Tue, 26 Jun 2007 17:26:29 +0200 Ingo Molnar [EMAIL PROTECTED] wrote:
* Andrew Morton [EMAIL PROTECTED] wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
Interesting question. [...]
i'm missing the sarcastic
On Tue, 26 Jun 2007 19:08:27 +0200 (CEST) Roman Zippel [EMAIL PROTECTED]
wrote:
This current ... interesting piece of Roman about a _single_ trivial
unlikely() branch in do_gettimeofday() borders on the ridiculous. My
patch might be wrong for various reasons, but that single
'if
On Mon, 25 Jun 2007 15:15:08 -0700 Andrew Morton [EMAIL PROTECTED] wrote:
What did you test?
There can be many ways to read the clock, do you want to put this hook
everywhere?
Yeah, it isn't immediately obvious (to this little black duck) why similar
fixups weren't needed in
On Tue, Jun 26, 2007 at 10:13:31AM -0700, Ray Lee wrote:
faster? Weird. It shouldn't. They must be doing something wrong,
therefore the patch is stupid.
Just in case it's not obvious the above are Ray Lee words, mine not.
---
#!/usr/bin/env stap
# edited top.stp from systemtap
On Tue, Jun 26, 2007 at 10:14:40AM -0700, Andrew Morton wrote:
On my machine, time(2) doesn't do any syscall at all - it uses the vsyscall
page. I'd be surprised if a database uses sys_time() either.
Large boxes unfortunately can't always use vsyscalls... that's a real
pity. I also had to
On Mon, 25 Jun 2007 15:15:08 -0700 Andrew Morton <[EMAIL PROTECTED]> wrote:
>
> On Mon, 25 Jun 2007 23:09:46 +0200
> Roman Zippel <[EMAIL PROTECTED]> wrote:
>
> > On Monday 25 June 2007, Ingo Molnar wrote:
> >
> > > the patch improves the sysbench OLTP macrobenchmark significantly:
> >
> > Has
Ingo Molnar wrote:
Subject: [patch] sys_time() speedup
From: Ingo Molnar <[EMAIL PROTECTED]>
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure
Hi,
On Tue, 26 Jun 2007, Jesper Juhl wrote:
> Even if it is not faster, what would make it slower? Have you spotted
> something I have not?
There are other ways to read the clock and would require similiar
synchronization hooks.
Some archs can implement sys_time() in userspace, so there this
On 26/06/07, Roman Zippel <[EMAIL PROTECTED]> wrote:
Hi,
On Mon, 25 Jun 2007, Jesper Juhl wrote:
> > On Monday 25 June 2007, Ingo Molnar wrote:
> >
> > > the patch improves the sysbench OLTP macrobenchmark significantly:
> >
> > Has that any real practical relevance?
> >
> It seems to me that
On Mon, 25 Jun 2007 23:09:46 +0200
Roman Zippel <[EMAIL PROTECTED]> wrote:
> Hi,
>
> On Monday 25 June 2007, Ingo Molnar wrote:
>
> > the patch improves the sysbench OLTP macrobenchmark significantly:
>
> Has that any real practical relevance?
Interesting question. The patch adds a new
Ingo Molnar a écrit :
Subject: [patch] sys_time() speedup
From: Ingo Molnar <[EMAIL PROTECTED]>
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure
Hi,
On Mon, 25 Jun 2007, Jesper Juhl wrote:
> > On Monday 25 June 2007, Ingo Molnar wrote:
> >
> > > the patch improves the sysbench OLTP macrobenchmark significantly:
> >
> > Has that any real practical relevance?
> >
> It seems to me that Ingo's patch offers slightly improved performance
>
On 25/06/07, Roman Zippel <[EMAIL PROTECTED]> wrote:
Hi,
On Monday 25 June 2007, Ingo Molnar wrote:
> the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
It seems to me that Ingo's patch offers slightly improved performance
for any
Hi,
On Monday 25 June 2007, Ingo Molnar wrote:
> the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
> @@ -373,6 +376,20 @@ void do_gettimeofday (struct timeval *tv
>
> tv->tv_sec = sec;
> tv->tv_usec = usec;
> +
> + /*
> +
Subject: [patch] sys_time() speedup
From: Ingo Molnar <[EMAIL PROTECTED]>
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure "xtime", which
is updated
Subject: [patch] sys_time() speedup
From: Ingo Molnar [EMAIL PROTECTED]
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure xtime, which
is updated by
Hi,
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
@@ -373,6 +376,20 @@ void do_gettimeofday (struct timeval *tv
tv-tv_sec = sec;
tv-tv_usec = usec;
+
+ /*
+ *
On 25/06/07, Roman Zippel [EMAIL PROTECTED] wrote:
Hi,
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
It seems to me that Ingo's patch offers slightly improved performance
for any program
Hi,
On Mon, 25 Jun 2007, Jesper Juhl wrote:
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
It seems to me that Ingo's patch offers slightly improved performance
for any program
Ingo Molnar a écrit :
Subject: [patch] sys_time() speedup
From: Ingo Molnar [EMAIL PROTECTED]
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure
On Mon, 25 Jun 2007 23:09:46 +0200
Roman Zippel [EMAIL PROTECTED] wrote:
Hi,
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
Interesting question. The patch adds a new test-n-branch
On 26/06/07, Roman Zippel [EMAIL PROTECTED] wrote:
Hi,
On Mon, 25 Jun 2007, Jesper Juhl wrote:
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real practical relevance?
It seems to me that Ingo's patch
Hi,
On Tue, 26 Jun 2007, Jesper Juhl wrote:
Even if it is not faster, what would make it slower? Have you spotted
something I have not?
There are other ways to read the clock and would require similiar
synchronization hooks.
Some archs can implement sys_time() in userspace, so there this
Ingo Molnar wrote:
Subject: [patch] sys_time() speedup
From: Ingo Molnar [EMAIL PROTECTED]
improve performance of sys_time(). sys_time() returns time in seconds,
but it does so by calling do_gettimeofday() and then returning the
tv_sec portion of the GTOD time. But the data structure xtime,
On Mon, 25 Jun 2007 15:15:08 -0700 Andrew Morton [EMAIL PROTECTED] wrote:
On Mon, 25 Jun 2007 23:09:46 +0200
Roman Zippel [EMAIL PROTECTED] wrote:
On Monday 25 June 2007, Ingo Molnar wrote:
the patch improves the sysbench OLTP macrobenchmark significantly:
Has that any real
48 matches
Mail list logo
