From: 大平怜 <rei...@gm...> - 2015-11-19 20:09:42
|
Thanks for the patch, Will, It solved the problem in my environment. Carl, I have a question about your patch. Your patch removed PM_DATA_ALL_FROM_RL2L3_MOD, but this event does not cause any error when I specify it to a previous version of OProfile. Why does your patch have to remove this event? Maybe it depends on the environment? I confirmed PM_L2_LD_MISS did cause an error, as Chencheng Ye found. I have not yet checked the other events. Thanks, Rei Odaira 2015-11-19 12:50 GMT-06:00 Carl E. Love <ce...@us...>: > On Thu, 2015-11-19 at 12:24 -0500, William Cohen wrote: > > On 11/19/2015 11:54 AM, 大平怜 wrote: > > > Carl, Will, > > > > > > After this particular commit, I cannot run any OProfile command, even > ophelp, on POWER8. > > > > > > $ ./ophelp --version > > > ./ophelp: oprofile 1.2.0git compiled on Nov 19 2015 10:43:15 > > > $ ./ophelp > > > um rc_machine is not used > > > um L1_latency is not used > > > > > > The previous version on git (commit > a58c3953a1c5f5fb00b873a5e22c878f9f4d5a2d) works fine. > > > > > > Could you help fix this problem? > > > > Hi, > > > > It is surprising that commit 34715734fd6f4b44f32206541c8a2500514c9922 is > causing things to fail. That commit only changes > > events/ppc64/power8/events. However, there seems to be some checks for > unused unit masks. The attached patch seems to address the issue. Could > you give it ia try? Carl does this patch look reasonable? > > > > -Will > > Rei, Will: > > Argh, major embarrassment, my event removal patch did break the tool. > It looks like I didn't do a complete rebuild of the tool once I removed > the events. I was able to reproduce the error Rei saw. Yes, it is due > to the unused masks in the mask file. Will's patch fixes the issue. > With the patch applied, the ophelp tool worked. > > I did a fresh pull of OProfile on to a different P8 machine. Compiled > and installed to verify the issue. Then I applied the patch and did a > rebuild and reinstall to verify the patch worked. > > Carl Love > > > |