Messages in this thread | | | Date | Fri, 23 Sep 2011 21:46:54 -0600 | From | David Ahern <> | Subject | Re: [GIT PULL 00/12] perf tools fixes |
| |
On 09/23/2011 05:46 PM, Arnaldo Carvalho de Melo wrote: > Em Fri, Sep 23, 2011 at 04:12:54PM -0700, Linus Torvalds escreveu: >> On Fri, Sep 23, 2011 at 2:25 PM, Arnaldo Carvalho de Melo >> <acme@infradead.org> wrote: >>> >>> Sorry about the noise, there is a problem in the way the python binding >>> is built, i.e. it should notice that the header changed and rebuild the >>> python binding, I bet this was what made David not notice it :-\ >> >> Can you please make sure that gets fixed at some point too? No huge >> hurry, it can wait for the next merge window, but bugs like that not >> only cause this kind of stupid compile errors, they also cause really >> odd "why doesn't it work" problems that people can spend tons and tons >> of time debugging, because they don't really "exist" - they're just a >> result of a stale object file that didn't get properly recompiled. > > Will get that fixed. Its a hell of a lot annoying to me.
My sincere apologies for introducing such a dumb failure; I always test the hell out of my patches before submitting them.
In this case I did not have the python-devel rpm installed, so the python path in perf was not traversed and I did not trip on this. Situation remedied on my end.
Yes, the correct arg is 0.
David
> >>> You noticed it probably because it was the first build of the tool on >>> your local repo or used a new O= output dir. >> >> I do "git clean -dqfx" every once in a while (especially after having >> done "make allmodconfig" builds before -rc releases etc), and I don't >> recompile the perf tools very often. So yeah, this was the first build >> after a cleanout event. > > Thanks for testing it that way. I'll do my best not to fail in such last > minute -rc tests. <expletives about world-as-whe-known-it-falling-apart> > removed. :-P > > - Arnaldo
| |