summaryrefslogtreecommitdiffstats
path: root/V8Binding
Commit message (Collapse)AuthorAgeFilesLines
* Merge commit 'goog/master' into mergeAndrei Popescu2009-08-192-0/+13
|\
| * Fix the sim-eng build.Feng Qian2009-08-182-0/+13
| | | | | | | | The same change is been upstreaming to V8.
* | Merge commit 'goog/master' into mergeBen Murdoch2009-08-1883-1750/+7872
|\ \ | |/ | | | | | | Conflicts: WEBKIT_MERGE_REVISION
| * Update V8 to r2703.Feng Qian2009-08-1883-1750/+7872
| | | | | | | | New revision has improvement of scanner.
* | Merge commit 'goog/master' into mergeBen Murdoch2009-08-183-1/+18
|\ \ | |/ | | | | | | | | | | Conflicts: WebCore/bindings/v8/ScriptController.cpp WebCore/page/Geolocation.cpp WebCore/platform/android/GeolocationServiceAndroid.cpp
| * Add GC support in V8 and calling GC in WebCoreFrameBridge::clearCache.Feng Qian2009-08-132-0/+14
| | | | | | | | This is experimental, we'd like to see the impact on page cycler intl1 and intl2.
| * Avoid calling GC in Context::NewFeng Qian2009-08-131-0/+5
| | | | | | | | TestShellActivity now sets V8 flag --expose_gc, and it should have the same effect to page cycler performance.
| * Remove experimental nolazy flag.Feng Qian2009-08-131-5/+0
| | | | | | | | I didn't see performance changes.
| * EXPERIMENTAL: turn off lazy compilation to see the impact on page cycler.Feng Qian2009-08-121-0/+5
| | | | | | | | Will turn it back after getting data from page cycler. (ETA 1 day)
| * Merge change 20814Android (Google) Code Review2009-08-111-4/+0
| |\ | | | | | | | | | | | | * changes: Enable GC call before creating a new context.
| | * Enable GC call before creating a new context.Feng Qian2009-08-111-4/+0
| | | | | | | | | | | | I think this is very likely the cause of big memory jump and slowdown of page cycler.
| * | Don't install v8shell by default in kila-testsSriram Raman2009-08-101-0/+2
| |/
* | sort out inspector linking errorsAndrei Popescu2009-08-171-10/+0
| |
* | Continue with linker error fixes.Ben Murdoch2009-08-175-3/+90
| |
* | Fix first set of linker errors.Ben Murdoch2009-08-173-5/+7
| |
* | Fix ReadWriteLock class problem. Will upstream this asapAndrei Popescu2009-08-141-7/+1
| |
* | Fix a bunch more bindings errors.Ben Murdoch2009-08-131-1/+14
| |
* | Fix some build errors.Ben Murdoch2009-08-131-359/+359
| |
* | More changes to get V8 bindings to build.Ben Murdoch2009-08-131-30/+26
| |
* | Update generated files in V8 derived sources makefile.Ben Murdoch2009-08-121-7/+21
| |
* | makefile changes part 1Andrei Popescu2009-08-125-3115/+39
| |
* | Merge change 20934 into master-webkit-mergeAndroid (Google) Code Review2009-08-1224-3046/+0
|\ \ | | | | | | | | | | | | * changes: Merge half of V8Binding/binding
| * | Merge half of V8Binding/bindingAndrei Popescu2009-08-1224-3046/+0
| |/
* | r47029: Merge V8Index and V8Proxy.Ben Murdoch2009-08-124-5177/+0
|/
* Fix sim-eng build.Feng Qian2009-08-071-0/+1
| | | | Forgot to add cfg-ia32.cc to the make file.
* Get a new V8 revision with performance improvement in ARM compiler.Feng Qian2009-08-07111-2501/+9272
| | | | | | http://v8.googlecode.com/svn/branches/bleeding_edge@2654 This is a clean drop without local changes.
* Add V8 locks at the v8 entry points for use with workers.Andrei Popescu2009-08-053-2/+30
|
* Enable workers for V8Andrei Popescu2009-08-031-2/+18
|
* Merge changes 8489,8490Android (Google) Code Review2009-07-243-4/+0
|\ | | | | | | | | | | * changes: Android specifics for 'Removes uneccessary IDL files for Geolocation permissions'. Removes uneccessary IDL files for Geolocation callbacks.
| * Android specifics for 'Removes uneccessary IDL files for Geolocation ↵Steve Block2009-07-243-4/+0
| | | | | | | | permissions'.
* | Update V8 to bleeding_edge@2524Andrei Popescu2009-07-2389-894/+3183
|/
* Add TODO(andreip) to those v8 binding files that have Android-specific ↵Andrei Popescu2009-07-2111-3/+30
| | | | modifications that may need fixing/upstreaming/etc
* Experiment a different GC policy.Feng Qian2009-07-201-0/+5
| | | | Smaller promotion limit makes mark-compact GC more often, but may use less memory.
* Experiment no GC before loading new page.Feng Qian2009-07-201-0/+5
| | | | I want to check this change in to see the page load time on page cyclers.
* Add indexed getter for TouchList.Feng Qian2009-07-182-1/+9
| | | | http://maps.google.com/maps/m uses TouchEvent.touches[i], which requires a customized index getter.
* Initialize context if a frame had an empty global object.Feng Qian2009-07-171-1/+3
| | | | | When a frame navigates to a new page, the global object is detached from its old context. However, if the new page has no JavaScript, its new context is not initialized. If another page kept a reference to the global object of the old page and tries to access the new page, it will fail because the context of new page is not initialized so the global object is connected to a real context.
* Merge change 7500Android (Google) Code Review2009-07-164-5/+13
|\ | | | | | | | | * changes: Convert Java string to JavaScript string.
| * Convert Java string to JavaScript string.Feng Qian2009-07-154-5/+13
| | | | | | | | This allows JavaScript call a Java function and returns a string back to JavaScript. This patch is needed by the current Gmail app.
* | Fix build error of mksnapshotFeng Qian2009-07-151-0/+2
|/ | | | After hooking up Android log facility, mksnapshot depends on liblog now.
* Make some adjustment on heap parameters on Android platform.Feng Qian2009-07-144-4/+11
| | | | | | | | 1. reduced pages per chunk to 16, so a chunk is 128K, instead of 512K 2. free all clean chunks when shrink spaces. 3. renamed some macros to if defined(ANDROID). All changes are being upstreamed to V8.
* Update V8 to bleeding_edge@2450 to pick up a few important fixes.Feng Qian2009-07-13118-2925/+6214
| | | | API stays the same, and the same revision is pushed to Chromium.
* Fix a memory leak caused by the binding code.Feng Qian2009-07-101-7/+1
| | | | | | Old code reinitialize a context when a frame loading a new page, even when the page does not have JavaScript. It also initializes an empty context when before an iframe is destroyed. A fix is made to Chromium project (not it is in WebKit), our next merge should pick up this fix.
* Pickup two ARM fixes.Feng Qian2009-07-091-14/+23
| | | | | | | http://codereview.chromium.org/155272 http://codereview.chromium.org/149266 Next drop of V8 will get both.
* Pickup two crash fixes from V8 tree:Feng Qian2009-07-081-8/+21
| | | | | | | http://code.google.com/p/v8/source/detail?r=2400 http://code.google.com/p/v8/source/detail?r=2399 Our next dropin will get both fixes.
* Add instrumentation for V8.Grace Kloba2009-07-082-0/+52
|
* Update V8 to r2361, which is a trunk revision pushed to Chrome.Feng Qian2009-07-0769-1624/+2851
| | | | Also changed compilation generation to 1 on ARM, the same change is made into V8 trunk.
* Separate V8 logging from debugging flags.Feng Qian2009-07-071-1/+5
| | | | | | Support both ENABLE_V8_LOGGING_AND_PROFILING, DEBUG_V8 flags. To enable logging, one can set JS flags to --logfile /sdcard/v8.log --log_gc (or other options).
* Avoid generating jsmin.pyc in the source directory.Feng Qian2009-07-062-4/+18
| | | | Copy js2c.py and jsmin.py to intermediate directory first, then invoke the script there. Python generates a .pyc file in that directory.
* Fix Mac build of mksnapshotGrace Kloba2009-07-063-8/+22
| | | | mksnapshot is a host program and it is platform dependent. Add rules to Makefiles so that it picks up the right file depending on the host platform.
* Add V8 snapshot to Makefile.Feng Qian2009-07-014-139/+216
| | | | | Snapshotting can be turned on by adding ENABLE_V8_SNAPSHOT in buildspec.mk file. It is off by default. When it is turned on, the binary size is increased by about 370K, it only saves the first cold startup time of browser (or first time using JavaScript).
generated by cgit v1.1 at 2025-01-21 21:24:03 +0000