[ Sean: don't send html email, please. It is hard to read and harder to reply to ]
On Sun, Dec 04, 2016 at 02:28:01PM -0500, Douglas Katzman wrote: > isn't it as simple as that user 'segher' is running a highly parallelized > build of something? gcc, I believe. > $ ps auxw|grep cc1plus|wc -l > 56 I often run small builds, yes. This is what the machine is *for*. > > Can anyone tell me why gcc110's load is still extremely high after so long? It is not. A few processes are stuck (this is a kernel bug, already fixed); this causes <6% CPU use. The machine is largely idle otherwise. > > It's gotten so bad that my test suites are now experiencing things like > > vfork() failures That is a completely separate issue, and likely your testsuite itself is to blame. The machine does not behave significantly different from how it always behaved. Segher _______________________________________________ Gcc-cfarm-users mailing list Gcc-cfarm-users@gna.org https://mail.gna.org/listinfo/gcc-cfarm-users
