Ok, I have a stock gcc4 from fedora core 5 updates and this happens. gcc version 4.1.1 20060525 (Red Hat 4.1.1-1) This is on a P6 coppermine.
My debug printf is this from the stock patch... Jul 19 17:48:56 localhost kernel: perfmon: pfm_arch_ctxswout.555: CPU0 [3790]: clearing cr4.pce Jul 19 17:48:56 localhost kernel: perfmon: pfm_arch_ctxswin.371: CPU0 [3790]: setting cr4.pce In other news, PAPI/perfmon2 is almost stood up and running. Phil On Thu, 2006-07-20 at 06:11 -0700, Stephane Eranian wrote: > Phil, > > On Wed, Jul 19, 2006 at 05:53:32PM +0200, Philip Mucci wrote: > > Hi Stephane, > > > > Did you happen to get this message? > > > > I'm still having this issue with the latest CVS head. Is anyone else > > seeing this on FC5/i386 (p6)? > > Well, I don't have a machine installed with the same version of gcc. > Everything I have tried worked. > > > > > Also, I have a suggestion for PFM debugging. Is it possible that debug=1 > > could not enable debugging in the context switch routines and we have a > > separate debug option for that like overflow? > > > You should not have debug printk in context switch code, be it > debug or overflow debug. This is not safe when interrupts are masked. > This is more sensitive on some platforms (such as X86). > > > Seems like that would make a big difference in the run time of these > > tests under debugging because my syslog is a busy puppy... > > Yes. > _______________________________________________ > perfmon mailing list > [email protected] > http://www.hpl.hp.com/hosted/linux/mail-archives/perfmon/ _______________________________________________ perfmon mailing list [email protected] http://www.hpl.hp.com/hosted/linux/mail-archives/perfmon/
