It has been a fun and fulfilling weekend tracking down performance regressions when using DRI on the XO 1.5 platform. One place I was looking at specifically was blitting solids from userspace to the kernel. I found that copy_from_user was really chewing up a significant amount of cpu. Looking further I found that for VIA C7 processors these two options, X86_INTEL_USERCOPY and X86_USE_PPRO_CHECKSUM, are not enabled.
I have not done thorough testing on it, but after patch the kernel with the attached patch my gtkperf run dropped from 63 seconds down to 50 seconds. A 20% improvement is not bad, but more testing is definitely needed, and testing both options independently is also needed. I have my test kernel available here http://dev.laptop.org/~jnettlet/f14/kernel-2.6.35.9_xo1.5-20110313.2249.fc14.c06443f.i586.rpm If anyone else reports back that they are also seeing improvements I will open a bug and we can track this improvement down further. Jon
diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 2ac9069..e320d51 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -364,11 +364,11 @@ config X86_ALIGNMENT_16 config X86_INTEL_USERCOPY def_bool y - depends on MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M586MMX || X86_GENERIC || MK8 || MK7 || MEFFICEON || MCORE2 + depends on MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M586MMX || X86_GENERIC || MK8 || MK7 || MEFFICEON || MVIAC7 || MCORE2 config X86_USE_PPRO_CHECKSUM def_bool y - depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MEFFICEON || MGEODE_LX || MCORE2 || MATOM + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2 || MVIAC7 || MEFFICEON || MGEODE_LX || MCORE2 || MATOM config X86_USE_3DNOW def_bool y
_______________________________________________ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel