Re: Bogus stats when running SMP programs

2006-09-29 Thread Tomasz Zielonka
On Thu, Sep 28, 2006 at 12:42:50PM +0100, Ian Lynagh wrote: On Thu, Sep 28, 2006 at 09:07:33AM +0200, Tomasz Zielonka wrote: Great! Which snapshot can I use to test this change? Tonight's or later - there isn't one yet. As I wrote in another message to glasgow-haskell-bugs, there are some

Re: Bogus stats when running SMP programs

2006-09-29 Thread Tomasz Zielonka
On Fri, Sep 29, 2006 at 11:00:25AM +0200, Tomasz Zielonka wrote: I will compile GHC from sources now. The stats seem to be OK now. Thanks! Best regards Tomasz ___ Glasgow-haskell-bugs mailing list Glasgow-haskell-bugs@haskell.org

Re: Bogus stats when running SMP programs

2006-09-28 Thread Tomasz Zielonka
On Thu, Sep 28, 2006 at 01:57:18AM +0100, Ian Lynagh wrote: I've found a machine which had the same problem. I think I've fixed it with Thu Sep 28 00:46:30 BST 2006 Ian Lynagh [EMAIL PROTECTED] * Handle clock_gettime failing Great! Which snapshot can I use to test this change? Best

Re: Bogus stats when running SMP programs

2006-09-28 Thread Ian Lynagh
On Thu, Sep 28, 2006 at 09:07:33AM +0200, Tomasz Zielonka wrote: On Thu, Sep 28, 2006 at 01:57:18AM +0100, Ian Lynagh wrote: I've found a machine which had the same problem. I think I've fixed it with Thu Sep 28 00:46:30 BST 2006 Ian Lynagh [EMAIL PROTECTED] * Handle clock_gettime

Re: Bogus stats when running SMP programs

2006-09-27 Thread Simon Marlow
Tomasz Zielonka wrote: On Tue, Sep 26, 2006 at 01:46:30AM +0100, Ian Lynagh wrote: I couldn't reproduce this with a simple program that forked off 4 threads to do computation on a single CPU machine, and neither could someone on IRC with head from Sep 15 on a Core Duo. Is it possible to send

Re: Bogus stats when running SMP programs

2006-09-27 Thread Tomasz Zielonka
On Wed, Sep 27, 2006 at 09:25:39AM +0100, Simon Marlow wrote: Perhaps clock_gettime() is returning strange results on your system. Could you try compiling with -threaded -debug, and run the program under gdb. When I compile with -threaded -debug, the stats are OK :-/ Best regards Tomasz

Re: Bogus stats when running SMP programs

2006-09-27 Thread Simon Marlow
Tomasz Zielonka wrote: On Wed, Sep 27, 2006 at 09:25:39AM +0100, Simon Marlow wrote: Perhaps clock_gettime() is returning strange results on your system. Could you try compiling with -threaded -debug, and run the program under gdb. When I compile with -threaded -debug, the stats are OK

Re: Bogus stats when running SMP programs

2006-09-27 Thread Tomasz Zielonka
On Wed, Sep 27, 2006 at 12:06:34PM +0100, Simon Marlow wrote: When I compile with -threaded -debug, the stats are OK :-/ Ok, maybe try strace? Nothing suspicious, at least for me. Strace logs attached. I'll try to compile GHC from sources and put some debugging prints in the RTS. There

Re: Bogus stats when running SMP programs

2006-09-27 Thread Ian Lynagh
On Thu, Sep 21, 2006 at 08:03:53PM +0200, Tomasz Zielonka wrote: Hello! I am getting nonsensical execution statistics (+RTS -Sstderr) when running programs in SMP mode (+RTS -N2). I've found a machine which had the same problem. I think I've fixed it with Thu Sep 28 00:46:30 BST 2006 Ian

Re: Bogus stats when running SMP programs

2006-09-26 Thread Ian Lynagh
Hi Tomasz, On Thu, Sep 21, 2006 at 08:03:53PM +0200, Tomasz Zielonka wrote: I am getting nonsensical execution statistics (+RTS -Sstderr) when running programs in SMP mode (+RTS -N2). Example: Task 0 (worker) : MUT time: 401572821.14s ( 21.89s elapsed) GC

Bogus stats when running SMP programs

2006-09-21 Thread Tomasz Zielonka
Hello! I am getting nonsensical execution statistics (+RTS -Sstderr) when running programs in SMP mode (+RTS -N2). Example: Task 0 (worker) : MUT time: 401572821.14s ( 21.89s elapsed) GC time: 0.00s ( 0.00s elapsed) Task 1 (worker) : MUT time: 296.00s (