- 23 Jan, 2012 4 commits
-
-
svenpanne@chromium.org authored
Yak shaving for map sharing with accessor properties contd.: When CALLBACKS can have map transitions, simply looking at the property type is not sufficient anymore to decide if a property is there or not. One has to look at the actual contents of the descriptor entry then, but this breaks down sometimes when the lookup is being done with a NULL holder. Luckily enough, we can oftren replace IsProperty by the simpler IsFound, because we inspect the type immediately afterwards, anyway. Review URL: https://chromiumcodereview.appspot.com/9280007 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10474 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
svenpanne@chromium.org authored
The main goal is to cleanly separate between the several parts involved in the traversal: * iterating over all transitions in a descriptor array * iterating over all prototype transitions * storing the parent and the current local traversal position in a map * the iterative traversal algorithm itself The previous algorithm for iterating over prototype transitions did a little bit too much here, iterating over the whole array instead only the filled part. This has been fixed on the way, too. With this CL, it will be much easier to make the necessary changes to the descriptor array iterator to correctly handle map transitions for accessor properties. Furthermore, perhaps we represent transitions a bit different in the future, making finding them a bit easier. This would make some code in this CL (and elsewhere) quite a bit shorter and more efficient. Review URL: https://chromiumcodereview.appspot.com/9252007 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10473 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/9193007 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10472 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
Removing experimental i18n code from v8 repository. Internationalization support is hosted under code.google.com/p/v8-i18n for couple months now. TEST=Chromium/WebKit builds should pass. Review URL: https://chromiumcodereview.appspot.com/9016034 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10471 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
- 20 Jan, 2012 13 commits
-
-
vegorov@chromium.org authored
This allows to avoid infinite loops in pathalogical cases e.g. when OS refuses to give new pages to V8. R=erik.corry@gmail.com BUG=99027 Review URL: https://chromiumcodereview.appspot.com/8286020 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10466 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
BUG=v8:1653 Review URL: https://chromiumcodereview.appspot.com/9192010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10465 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
danno@chromium.org authored
TBR=mstarzinger@chromium.org BUG=none TEST=build turns green Review URL: https://chromiumcodereview.appspot.com/9270010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10464 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
BUG=v8:1761 TEST= Review URL: https://chromiumcodereview.appspot.com/9233002 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10463 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
svenpanne@chromium.org authored
It had only 2 uses: One use could be rewritten in a simpler way, and the other is actually more readable after inlining and fixing the comments. Review URL: https://chromiumcodereview.appspot.com/9233006 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10462 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
danno@chromium.org authored
Port r10356 (69da81). Original commit message: Loosen the requirement for Map equivalency on several map checks, including checks up the prototype chain, that are not sensitive to ElementsKinds. These selected map checks should also match against FAST_DOUBLE_ELEMENT and FAST_ELEMENT transitions of the original map. This specifically helps all variants of transitioned JSArrays to still efficiently call builtins like push, pop and sort. BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9265007 Patch from Daniel Kalmar <kalmard@homejinni.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10461 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
danno@chromium.org authored
BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9233005 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10460 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
rossberg@chromium.org authored
which use yet another way to determine strict vs non-strict function receivers. R=kmillikin@chromium.org BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9270004 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10459 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
erik.corry@gmail.com authored
of one in order to reduce collisions. Review URL: https://chromiumcodereview.appspot.com/9269004 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10458 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
vegorov@chromium.org authored
Port r10380 (46f646). Original commit message: The stub was directly patching caller's code without issuing write barrier which violated incremental marking invariants. BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9159008 Patch from Daniel Kalmar <kalmard@homejinni.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10457 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
vegorov@chromium.org authored
for call-sites with mismatched number of arguments. Port r10424 (92a05c9c). Original commit message: Adjust InvokeFunction to avoid generating dead code when number when arity mismatch is detected in compile time. BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9150026 Patch from Daniel Kalmar <kalmard@homejinni.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10456 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
mstarzinger@chromium.org authored
R=vegorov@chromium.org TEST=cctest/test-heap/PrototypeTransitionClearing Review URL: https://chromiumcodereview.appspot.com/9138016 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10455 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
BUG=v8:1691 TEST= Review URL: https://chromiumcodereview.appspot.com/9138015 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10454 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
- 19 Jan, 2012 16 commits
-
-
mstarzinger@chromium.org authored
R=vegorov@chromium.org Review URL: https://chromiumcodereview.appspot.com/9150041 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10451 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/9265010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10450 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/9200006 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10449 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
ulan@chromium.org authored
R=jkummerow@chromium.org Review URL: https://chromiumcodereview.appspot.com/9264011 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10445 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
jkummerow@chromium.org authored
Due to unexpected performance implications. This reverts r10429. Review URL: https://chromiumcodereview.appspot.com/9150032 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10444 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
erik.corry@gmail.com authored
in-snapshot code on IA32. Review URL: https://chromiumcodereview.appspot.com/9254027 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10443 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
erik.corry@gmail.com authored
MIPS: Use shift and add instead of mul on ARM for integer hash. This is a commit of http://codereview.chromium.org/9192011/ for Daniel Kalmar git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10442 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
erik.corry@gmail.com authored
Review URL: https://chromiumcodereview.appspot.com/9159010 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10441 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
kmillikin@chromium.org authored
Port r10374 (03bbe39). Note: the MIPS version was not affected by the bug. This is merely a style cleanup. Original commit message: An off-by-one in the register allocator could lead to allocating (and clobbering) the reserved 0.0 double register. This required a function with 14 or more live double values. BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9150025 Patch from Daniel Kalmar <kalmard@homejinni.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10440 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
mstarzinger@chromium.org authored
R=vegorov@chromium.org Review URL: https://chromiumcodereview.appspot.com/9195008 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10439 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
erik.corry@gmail.com authored
Review URL: https://chromiumcodereview.appspot.com/9159009 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10438 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
fschneider@chromium.org authored
Port r10379 (3399d6). Original commit message: The input register does not need to be made writable. We already allocate enough temp registers and the input register will always be preserved. BUG= TEST= Review URL: http://codereview.chromium.org/9254025 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10437 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
Port r10391 (7900f3c). BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9195009 Patch from Daniel Kalmar <kalmard@homejinni.com>. git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10436 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
fschneider@chromium.org authored
Review URL: http://codereview.chromium.org/9195005 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10435 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
vegorov@chromium.org authored
R=ulan@chromium.org BUG=http://crbug.com/110509 TEST=test/mjsunit/regress/regress-110509.js Review URL: https://chromiumcodereview.appspot.com/9265003 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10434 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
yangguo@chromium.org authored
BUG=v8:1897 TEST= Review URL: https://chromiumcodereview.appspot.com/9249013 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10433 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
- 18 Jan, 2012 6 commits
-
-
rossberg@chromium.org authored
R=kmillikin@chromium.org BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9231025 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10432 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
jkummerow@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/9255014 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10431 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
fschneider@chromium.org authored
In the case where we generate a generic load or store, we don't need to emit a non-smi-check and a map-check before it. Review URL: http://codereview.chromium.org/9252008 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10430 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
jkummerow@chromium.org authored
Review URL: https://chromiumcodereview.appspot.com/9254005 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10429 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
jkummerow@chromium.org authored
BUG=v8:1887 Review URL: https://chromiumcodereview.appspot.com/9231019 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10427 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
vegorov@chromium.org authored
R=mstarzinger@chromium.org BUG= TEST= Review URL: https://chromiumcodereview.appspot.com/9178021 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10426 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-
- 17 Jan, 2012 1 commit
-
-
fschneider@chromium.org authored
When calling a constant function property from optimized code, we need to check that there is no interceptor on the receiver map. TEST=cctest/InterceptorCallICConstantFunctionNotNeededWrapped Review URL: http://codereview.chromium.org/9240006 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10425 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
-