On Fri, 2006-08-11 at 09:53 +0300, Jorn Baayen wrote: > On Tue, 2006-08-08 at 11:03 +0100, Richard Purdie wrote: > > On Tue, 2006-08-08 at 10:24 +0100, Michael Meeks wrote: > > > Ah - interesting; so poking at: > > > > > > http://www.o-hand.com/~jorn/pango-benchmarks/28/full-report.txt > > > > > > I see a ton of vmlinux stuff apparently emulating the FPU - but, > > > unfortunately, it doesn't appear to tell me where those calls came from > > > (?) is there a pretty picture that shows that ? [ something like > > > 'sysprof' output would do that I guess ]. > > > > Floating point instructions generate CPU exceptions on ARM and I don't > > think oprofile or sysprof support back tracing through the CPU exception > > handler. > > > > One solution would be to run the tests against a soft floating point > > image, then the floating point instructions would show up as any other > > library instead of generating exceptions. They have slightly less > > overhead in that case but it would show them in context in the stack > > back traces. > > Done: http://www.o-hand.com/~jorn/pango-benchmarks/210-softfloat/
Now if we look at the cairo profile here (cairo.txt), we see that 7.7% and 3.8% are you used by __muldf3() and __adddf3(), respectively. These are softfloat functions, confirming suspicions that FP is a problem. Jorn -- OpenedHand Ltd. http://o-hand.com/ _______________________________________________ Performance-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/performance-list
