On Wed, 18 Sep 2013 19:40:33 -0400 Søren Sandmann Pedersen <sandm...@cs.au.dk> wrote:
> From: Søren Sandmann Pedersen <s...@redhat.com> > > The generated fetchers for NEAREST, BILINEAR, and > SEPARABLE_CONVOLUTION filters are fast paths and so they belong in > pixman-fast-path.c Just wondered why the performance improvements from http://lists.freedesktop.org/archives/pixman/2013-October/003019.html did not seem to agree with the earlier profiling reports at http://lists.freedesktop.org/archives/pixman/2013-October/003013.html (the profiling was done with a little bit stale build of pixman). This "Move generated affine fetchers into pixman-fast-path.c" commit appears to have introduced a performance regression: Before: [ 1] image firefox-canvas-scroll 4.550 4.593 0.39% 8/8 33.18% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_blt.part.10 29.17% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_composite_over_8888_8888 15.86% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_fill 8.63% cairo-perf-trac libpixman-1.so.0.31.1 [.] bits_image_fetch_bilinear_affine_pad_x8r8g8b8 3.76% cairo-perf-trac liblzo2.so.2.0.0 [.] lzo2a_decompress 1.06% cairo-perf-trac libc-2.15.so [.] __memcpy_ssse3 After: [ 1] image firefox-canvas-scroll 4.715 4.728 0.19% 7/8 31.89% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_blt.part.10 27.95% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_composite_over_8888_8888 14.91% cairo-perf-trac libpixman-1.so.0.31.1 [.] sse2_fill 8.26% cairo-perf-trac libpixman-1.so.0.31.1 [.] bits_image_fetch_affine_no_alpha 3.73% cairo-perf-trac liblzo2.so.2.0.0 [.] lzo2a_decompress 3.60% cairo-perf-trac libpixman-1.so.0.31.1 [.] fetch_pixel_x8r8g8b8 1.09% cairo-perf-trac libc-2.15.so [.] __memcpy_ssse3 0.76% cairo-perf-trac libc-2.15.so [.] __memset_sse2 -- Best regards, Siarhei Siamashka _______________________________________________ Pixman mailing list Pixman@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pixman