v8/test
jarin@chromium.org 713aa33f2a Fix of argument materialization of captured heap numbers.
The escape analysis calculates the number of slots in an object as
no-of-slots = object-size / pointer-size.  This gives 3 slots for
heap numbers on 32-bit architectures (one slot for the map, two for
the double value); however, my argument materialization code assumed
just two slots (map + value). Since Hydrogen allocates heap numbers
quite rarely, it is hard to produce a more meaningful repro than the
one provided by Clusterfuzz. Any suggestions are welcome.

The fix is simple - we just read out all extra slots (beyond the map
and the double) for heap numbers.

R=mstarzinger@chromium.org
BUG=351315
LOG=N

Review URL: https://codereview.chromium.org/196283004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19874 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2014-03-13 07:17:37 +00:00
..
benchmarks Merge experimental/a64 to bleeding_edge. 2014-02-12 09:19:30 +00:00
cctest Implement PersistentValueMap, a map that stores UniquePersistent values. 2014-03-13 07:10:59 +00:00
intl Merge experimental/a64 to bleeding_edge. 2014-02-12 09:19:30 +00:00
message Clean up some A64 specific code in common code that was introduced by A64 merge 2014-02-12 13:27:13 +00:00
mjsunit Fix of argument materialization of captured heap numbers. 2014-03-13 07:17:37 +00:00
mozilla When upgrading the test data twice, don't bail out because of an existing backup 2014-02-28 12:32:54 +00:00
preparser Merge experimental/a64 to bleeding_edge. 2014-02-12 09:19:30 +00:00
test262 Update README about where the test262 files are downloaded from. 2014-02-28 14:16:50 +00:00
webkit Reland "Enable Object.observe by default" again 2014-03-13 00:20:06 +00:00