before I spend the time to migrate to 7.0.
Thanks
Ravi
-Original Message-
From: Kris Kennaway [mailto:[EMAIL PROTECTED]
Sent: Monday, July 07, 2008 2:04 PM
To: [EMAIL PROTECTED]
Cc: Murty, Ravi; freebsd-hackers@freebsd.org
Subject: Re: Bug in calcru in he 6.2 and 6.3 kernels
Xin LI
On Sun, Jul 20, 2008 at 06:51:22AM -0700, Murty, Ravi wrote:
Has anyone identified the issue(s) that might be broken in the ULE
scheduler in 6.2? I am running a rather simple test - creates 8 threads
and runs it on an 8 CPU system (not a whole lot running on the system).
When I run it with
-Original Message-
From: Jeremy Chadwick [mailto:[EMAIL PROTECTED]
Sent: Sunday, July 20, 2008 6:59 AM
To: Murty, Ravi
Cc: Kris Kennaway; [EMAIL PROTECTED]; freebsd-hackers@freebsd.org
Subject: Re: Bug in calcru in he 6.2 and 6.3 kernels
On Sun, Jul 20, 2008 at 06:51:22AM -0700, Murty, Ravi wrote
Murty, Ravi wrote:
Jeremy, thanks. I look forward to switching to ULE in 7.0 and realize
that it is a completely new scheduler (I spent some time yesterday
looking at it) -- which is my porting effort is much harder than a
simple cut and paste. I just wanted to find out if there was something
On Sun, Jul 20, 2008 at 7:13 AM, Kris Kennaway [EMAIL PROTECTED] wrote:
Murty, Ravi wrote:
Jeremy, thanks. I look forward to switching to ULE in 7.0 and realize
that it is a completely new scheduler (I spent some time yesterday
looking at it) -- which is my porting effort is much harder than
Hello everyone,
Finally found what my last problem was. We were running top in a loop
and running some workloads that called sched_bind() to bind threads to
specific CPUs. The problem was that (and I am using ULE) sched_bind
calls a function to notify another CPU of a thread and then
Murty, Ravi wrote:
Hello everyone,
Finally found what my last problem was. We were running top in a loop
and running some workloads that called sched_bind() to bind threads to
specific CPUs. The problem was that (and I am using ULE) sched_bind
calls a function to notify another CPU of a
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Kris Kennaway wrote:
| Murty, Ravi wrote:
| Hello everyone,
|
|
|
| Finally found what my last problem was. We were running top in a loop
| and running some workloads that called sched_bind() to bind threads to
| specific CPUs. The problem was that
Xin LI wrote:
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Kris Kennaway wrote:
| Murty, Ravi wrote:
| Hello everyone,
|
|
|
| Finally found what my last problem was. We were running top in a loop
| and running some workloads that called sched_bind() to bind threads to
| specific CPUs. The
hihi, all -
i am currently running the experiment with both suggestions - doug's one of
simply re-ordering the equations, and my more complicated one with the
conditional (and the re-ordering) - the first significant data point will come
tomorrow, when the test cpu times pass the first error
On 26 Jan, Chris Landauer wrote:
hihi, doug -
Doug Ambrisko [EMAIL PROTECTED] wrote
...
The assumption with this calculation is that st it tend to be
small compared to tt so the 1024 X shouldn't overflow much.
...
[EMAIL PROTECTED] wrote:
|...but i'm a little
hihi, doug -
thanx for the suggestion - the notion of computing ut just with subtraction is
really good, since it is likely to be the largest of the three values in most
applications, but i'm a little worried that the 1024 multiplications aren't
large enough when tt gets really large - i'll do
hihi, doug -
Doug Ambrisko [EMAIL PROTECTED] wrote
...
The assumption with this calculation is that st it tend to be
small compared to tt so the 1024 X shouldn't overflow much.
...
[EMAIL PROTECTED] wrote:
| ...but i'm a little worried that the 1024
hihi, all -
well, i have an almost fix for the problem - read on, ...
(this is for discussin before send-pr submission)
description
in FreeBSD 5.3-RELEASE (and 5.2.1R, and 5.1R),
in file /usr/src/sys/kern/kern_resource.c,
lines 657-750 define a function calcru(),
Chris Landauer writes:
|
| hihi, all -
|
| well, i have an almost fix for the problem - read on, ...
| (this is for discussin before send-pr submission)
|
| description
|
| in FreeBSD 5.3-RELEASE (and 5.2.1R, and 5.1R),
| in file /usr/src/sys/kern/kern_resource.c,
| lines
Chris Landauer writes:
| thanx for the suggestion - the notion of computing ut just with subtraction is
| really good, since it is likely to be the largest of the three values in most
| applications, but i'm a little worried that the 1024 multiplications aren't
| large enough when tt gets really
, but in 4.1, this looks like a bug:
void
calcru(p, up, sp, ip)
struct proc *p;
struct timeval *up;
struct timeval *sp;
struct timeval *ip;
{
/* {user, system, interrupt, total} {ticks, usec}; previous tu: */
u_int64_t ut, uu, st, su, it, iu, tt, tu
17 matches
Mail list logo
