Updates:
Status: Accepted
Owner: [email protected]
Comment #1 on issue 2097 by [email protected]: sha1 compiled with
Emscripten runs 10X slower than with SpiderMonkey
http://code.google.com/p/v8/issues/detail?id=2097
I did not look closely, but my hypothesis that we are suffering from the
lack of uint32 support on optimized frames.
We deoptimize a _lot_ on @load-keyed-specialized-array-element (I suspect
load from Uint32Array) and @shift-i (I suspect: x >>> 0) inside
__Z13SHA1TransformPjPKh function which uses uint32 excessively.
routing to danno for triage
--
v8-dev mailing list
[email protected]
http://groups.google.com/group/v8-dev
