Hello,
I have tried perfmon2 with linux-2.6.30 (from git) /as well as with perf_events
linux-2.6.39 and libpfm-4.1.0 on AMD Phenom9650.
Part of /proc/cpuinfo shown below:
cpu family : 16
model : 2
model name : AMD Phenom(tm) 9650 Quad-Core Processor
Availability of event wi
Hi
I could install the perfmon2 along with linux-2.6.30 and libpfm-3.9 on xeon
x5550. But getting error while trying to compile pfmon-3.9.
compiling pfmon for x86_64 architecture
compiling pfmon as a shared binary
make[1]: Entering directory `/home/jk/XeonX5550/linuxsrc/pfmon-3.9/pfmon'
cc -g
> > I use perfmon2 with linux-2.6.30, libpfm-3.9 on intel
> xeon x5482.
> >
> > I am trying to get asynchronous overflow notification
> (instrn retired) in my controlling process, (for child
> process ls) by doing fcntl on perfmon2 context fd. (sampling
> period = 1000)
> >
> > I do standard signal
Hi
I use perfmon2 with linux-2.6.30, libpfm-3.9 on intel xeon x5482.
I am trying to get asynchronous overflow notification (instrn retired) in my
controlling process, (for child process ls) by doing fcntl on perfmon2 context
fd. (sampling period = 1000)
I do standard signal handler installatio
Hi Nagy
You seem to compare two things which are not compatible.
In first instance by using time you measured the time spent by the program in
user and kernel mode.
While in second instance you are doing sampling, that too in kernel mode only.
Regards,
JK
--- On Sun, 19/4/09, Nagy Mostafa wr
Hi Dennis,
Both events are not exactly same. For more details look into intel manual
available on net.
Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3B: System
Programming Guide, Part 2
Moreover the event LAST_LEVEL_CACHE_MISSES is an alias as shown by
showeventinfo.
tel core microarch based machine.
Thanks again,
Regards,
JK
________
From: Kristian Nielsen
To: J K Rai
Cc: perfmon2
Sent: Monday, 23 February, 2009 7:43:04 PM
Subject: Re: [perfmon2] Difference in LAST_LEVEL_CACHE_MISSES and
L2_LINES_IN:SELF values for kernel +user
Hi All
I get same values for the two events (i.e. LAST_LEVEL_CACHE_MISSES and
L2_LINES_IN:SELF) in user mode while the two change when the same are collected
for kernel+user mode. E.g.
pfmon -u -eLAST_LEVEL_CACHE_MISSES,L2_LINES_IN:SELF /bin/ls
cpu_2006_mix_lim_random_order_pfmon_t.sh
pair_l
Thanks a lot, I could compile successfully.
regards,
JK
- Original Message
From: stephane eranian <[EMAIL PROTECTED]>
To: J K Rai <[EMAIL PROTECTED]>
Cc: perfmon2
Sent: Saturday, August 2, 2008 12:48:15 PM
Subject: Re: [perfmon2] Compilation error with 2.6.26 kernel pat
__NR_ia32_rt_sigreturn 173
#define __NR_ia32_pfm_create_context 327
#endif /* _ASM_X86_64_IA32_UNISTD_H_ */
JK
- Original Message
From: stephane eranian <[EMAIL PROTECTED]>
To: J K Rai <[EMAIL PROTECTED]>
Cc: perfmon2
Sent: Friday, August 1, 2008 10:56:41 AM
Subject: Re: [perfmon2] Compi
r 2
make: *** [sub-make] Error 2
Regards
JK
- Original Message
From: stephane eranian <[EMAIL PROTECTED]>
To: J K Rai <[EMAIL PROTECTED]>
Cc: perfmon2
Sent: Friday, August 1, 2008 6:32:06 AM
Subject: Re: [perfmon2] Compilation error with 2.6.26 kernel patch on x86_64
(XeonX5482
Hello
I got following error while compiling linux 2.6.26 kernel on x86_64:
In file included from
/home/jk/linuxsrc/linux-2.6.26/include/linux/perfmon_kern.h:351,
from
/home/jk/linuxsrc/linux-2.6.26/arch/x86/kernel/process_64.c:39:
include2/asm/perfmon_kern.h: In function ‘pfm_a
Hello,
I am using perfmon2 29042008 full release with linux 2.6.25.
I could compile the kernel with perfmon2 patch as well as libpfm-3.4 and
install these.
After installing pfmon when I try to use it I get the following error:--->
[EMAIL PROTECTED] pfmon-3.4]$ pfmon -v
pfmon: error while loading
Hello
Long back I reported problem on Pentium4.
I find same on perfmon2 release of 2.6.25 kernel also.
May I get some direction on where to look into perfmon2 to fix it.
Mentioning below again the problem encountered.
Thanks & regards
JK
Problems encountered while using two events togethe
Hello
Which event on Core2 Duo counts L1 data cache misses. I am using perfmon2
release of linux-2.6.24.
Is it L1D_REPL ?
Regards,
JK
-
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.--
iod only instead of cumulative one. The first event is for deciding sampling
period and rest of the events are additional, which I want to sample for the
sampling period.
Regards
JK
stephane eranian <[EMAIL PROTECTED]> wrote: JK,
On Wed, Apr 2, 2008 at 3:23 PM, J K Rai wrote:
> Hi all
eranian <[EMAIL PROTECTED]> wrote: Hello,
On Wed, Apr 2, 2008 at 1:04 PM, J K Rai wrote:
> Thanks Dan And Stephen
>
> I want to sample no. of accesses on TLB (I+D) and L2 cache misses
> simultaneously for a given process / thread.
>
> You said Pentium4 is complex. What
ht that the second counter assignment is probably
messed up. Pentium4's a beast...
- d
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:perfmon2-devel-
> [EMAIL PROTECTED] On Behalf Of stephane eranian
> Sent: Tuesday, April 01, 2008 3:20 AM
> To: J K Rai
> Cc
Hi all,
Seeking help for the following problem on using perfmon2 on Pentium4.
Thanks & regards
JK
Problems encountered while using two events together:
**
19 matches
Mail list logo