[llvm-testresults] bwilson__llvm-gcc_PROD__i386 nightly tester results

A big compile time regression. Any ideas?

Ciao, Duncan.

A big compile time regression. Any ideas?

Ciao, Duncan.

False alarm. For some reason that I have not yet been able to figure out, these tests run significantly more slowly when I run them during the daytime, which I did for that run. I checked a few of the worst regressions reported here and they all recovered in subsequent runs.

Hi Bob,

A big compile time regression. Any ideas?

Ciao, Duncan.

False alarm. For some reason that I have not yet been able to figure out, these tests run significantly more slowly when I run them during the daytime, which I did for that run. I checked a few of the worst regressions reported here and they all recovered in subsequent runs.

here is a crazy idea: during the day the room the computer is in is hotter,
and to avoid overheating the CPU frequency is reduced, resulting in the machine
running slower.

Ciao, Duncan.

Yeah, I had actually considered that. The air conditioning in my office is pretty good, though. Until I find a solution, I'll try to remember to comment on the list about daytime runs that show regressions so that they won't raise any false alarms.

Hi,

Did you disable hyper-threading and Turbo ? Turbo alone can boost/hinder the performance of kernels by 30%, and make the perf results very unstable.

Nadav