f21991b104
Reason for revert: This also changes verylargebitmap, and the difference appears to be meaningful. Henrik, I have emailed you the images that differ. Original issue's description: > Modify sample buffer size for larger displays. > > Increases the intermediate buffer size for sample pixel indexes, > used in the sample proc function calls. If the operation is bigger > than the buffer it's split into multiple calls, creating overhead. > This would especially impact the performance of SIMD optimizations. > Also, aligns the start address of the buffer to 16 bytes, to enable > more efficient SIMD optimizations. > > Author: henrik.smiding@intel.com > > Signed-off-by: Henrik Smiding <henrik.smiding@intel.com> > > Committed: http://code.google.com/p/skia/source/detail?r=14825 > > Committed: http://code.google.com/p/skia/source/detail?r=14872 R=reed@google.com, mtklein@google.com, tomhudson@google.com, djsollen@google.com, joakim.landberg@intel.com, bsalomon@chromium.org, bsalomon@google.com, henrik.smiding@intel.com TBR=reed@google.com NOTREECHECKS=true NOTRY=true Author: scroggo@google.com Review URL: https://codereview.chromium.org/294023016 git-svn-id: http://skia.googlecode.com/svn/trunk@14878 2bbb7eff-a529-9590-31e7-b0007b416f81 |
||
|---|---|---|
| .. | ||
| bench | ||
| gm | ||
| skimage | ||