It was the change to add the vsyscall page. atomic didn't change, and I think timing changed just because the layout of memory changed a little. It should be fine to update the stats.
Gabe Gabe Black wrote: > I'm looking at it. I think there were three x86 changes since these last > passed, "X86: Hook up time syscall on X86", "X86: Enable x86_64 vsyscall > support", and "X86: Fix problem with movhps instruction". I figure out > which one, and look at the stats changes. > > Gabe > > nathan binkert wrote: > >>> ***** build/X86_SE/tests/fast/long/70.twolf/x86/linux/simple-timing FAILED! >>> ***** build/X86_SE/tests/fast/long/00.gzip/x86/linux/simple-timing FAILED! >>> ***** build/X86_SE/tests/fast/long/10.mcf/x86/linux/simple-timing FAILED! >>> ***** build/X86_SE/tests/fast/long/20.parser/x86/linux/simple-timing FAILED! >>> ***** build/X86_SE/tests/fast/long/60.bzip2/x86/linux/simple-timing FAILED! >>> >>> >> Looks like you changed timing somewhere. Vince, Gabe? Can someone >> run regressions? (I can if you guys just want to sign off.) >> >> Nate >> _______________________________________________ >> m5-dev mailing list >> m5-dev@m5sim.org >> http://m5sim.org/mailman/listinfo/m5-dev >> >> > > _______________________________________________ > m5-dev mailing list > m5-dev@m5sim.org > http://m5sim.org/mailman/listinfo/m5-dev > _______________________________________________ m5-dev mailing list m5-dev@m5sim.org http://m5sim.org/mailman/listinfo/m5-dev