- 20 Jul, 2017 6 commits
-
-
Michael Lippautz authored
Scavenger and full MC now rely on the same allocation behavior for their evacuation. Bug: Change-Id: Iddb0affe171187308e5b77ab0d3cfa75211bd8b8 Reviewed-on: https://chromium-review.googlesource.com/575983Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Commit-Queue: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#46781}
-
jgruber authored
This inlines the allocation of regexp literals when a boilerplate exists. Bug: v8:6605,v8:6556 Change-Id: If0f1b9dedf8a7de1ec51c394fe39cf21d2413ac5 Reviewed-on: https://chromium-review.googlesource.com/575240 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by:
Benedikt Meurer <bmeurer@chromium.org> Reviewed-by:
Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#46780}
-
jgruber authored
In contrast to other internal fields (data, source, and flags), last_index is an in-object property. But we can still use the standard accessor macros to access it. Bug: Change-Id: If77f2bb01c6ddccebdde09d7a316c2ddaaf9b277 Reviewed-on: https://chromium-review.googlesource.com/577549Reviewed-by:
Yang Guo <yangguo@chromium.org> Commit-Queue: Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#46779}
-
Yang Guo authored
- Change indentation from 4 to 2 in accordance with Node.js guidelines. - Fix --no-review option. R=petermarshall@chromium.org Change-Id: Ia525297292b0f13776e2264d5cf82dcc7ec449ad Reviewed-on: https://chromium-review.googlesource.com/574860Reviewed-by:
Peter Marshall <petermarshall@chromium.org> Commit-Queue: Yang Guo <yangguo@chromium.org> Cr-Commit-Position: refs/heads/master@{#46778}
-
Yang Guo authored
It used to be that immortal immovable objects have to be on the first page to not be moved. This is no longer true since we flag pages wrt whether they are allowed to move. R=mlippautz@chromium.org Change-Id: I5c9c88fa358636df119108e16e871815b126ab27 Reviewed-on: https://chromium-review.googlesource.com/575976Reviewed-by:
Michael Lippautz <mlippautz@chromium.org> Commit-Queue: Yang Guo <yangguo@chromium.org> Cr-Commit-Position: refs/heads/master@{#46777}
-
v8-autoroll authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/543ec6e..37f4a78 Rolling v8/third_party/catapult: https://chromium.googlesource.com/external/github.com/catapult-project/catapult/+log/a4770ef..c7c5420 Rolling v8/tools/clang: https://chromium.googlesource.com/chromium/src/tools/clang/+log/844603c..76b8b39 TBR=machenbach@chromium.org,hablich@chromium.org Change-Id: I252c26150fe01d55107ac705cbd9d851936b6568 Reviewed-on: https://chromium-review.googlesource.com/577924Reviewed-by:
v8 autoroll <v8-autoroll@chromium.org> Commit-Queue: v8 autoroll <v8-autoroll@chromium.org> Cr-Commit-Position: refs/heads/master@{#46776}
-
- 19 Jul, 2017 27 commits
-
-
Sathya Gunasekaran authored
Check the receiver_map for the dictionary mode bit instead of comparing the properties map against the HashTableMap. Bug: Change-Id: Iebf3118f00fd0afc8f7f13e88f373282c099f682 Reviewed-on: https://chromium-review.googlesource.com/578324 Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by:
Jakob Kummerow <jkummerow@chromium.org> Cr-Commit-Position: refs/heads/master@{#46775}
-
Benedikt Meurer authored
We already have all the functionality available in the CodeStubAssembler, so this is merely connecting the dots. Drive-by-fix: Improve code generation for StringCharCodeAt to properly mark runtime entries as deferred and just use a single slow-path. Bug: v8:5049 Change-Id: I76793c823b23f676e65cdb717558473edb6b91cd Reviewed-on: https://chromium-review.googlesource.com/577533 Commit-Queue: Benedikt Meurer <bmeurer@chromium.org> Reviewed-by:
Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#46774}
-
Clemens Hammacher authored
Before the existence of "= delete", we were enforcing that the DISALLOW_* macros were used in the private: section of classes only. This is not needed any more, hence remove the comment on the macros. Also, introduce macros for making types move-only, and use them instead of our special macro in wasm. R=bmeurer@chromium.org CC=titzer@chromium.org Change-Id: Iceba456fb0a32ae67defe16e35b865db8c8da500 Reviewed-on: https://chromium-review.googlesource.com/577687Reviewed-by:
-
titzer authored
R=ishell@chromium.org,clemensh@chromium.org BUG=chromium:742659 Review-Url: https://codereview.chromium.org/2977113002 Cr-Commit-Position: refs/heads/master@{#46772}
-
Mythri authored
Inlining heuristics in Turbofan used ast node count. Bytecode size is a better approximation of the size of the graph than the ast node count. This cl changes the heuristics to use the bytecode size instead. Also removing the ast_node_count filed in the shared function info. It was used only for the inlining heuristics. Also removed the max_inlined_source_size flag which is no longer used. Bug: Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: I8a2d2509c8e8d2779b33b817bb217de203d54ec3 Reviewed-on: https://chromium-review.googlesource.com/570055 Commit-Queue: Mythri Alle <mythria@chromium.org> Reviewed-by:
Michael Starzinger <mstarzinger@chromium.org> Reviewed-by:
Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#46771}
-
Daniel Ehrenberg authored
Async functions and generator declarations are only permitted as StatementListItems, not as ExpressionStatements, and therefore not as the entire body of an if statement, etc. Previously, they were incorrectly permitted. However, ChakraCore and SpiderMonkey seem to ban them in this context, and the feature was introduced relatively recently, so it is likely to be web-compatible to ship the prohibition. This patch also unifies the error message wording of async functions and generators to ordinary functions, explaining more clearly what the issue is. Bug: v8:4483 Cq-Include-Trybots: master.tryserver.v8:v8_linux_noi18n_rel_ng Change-Id: I31ed7818d6ab3e7e325031bfabb933dbf4512143 Reviewed-on: https://chromium-review.googlesource.com/568979 Commit-Queue: Daniel Ehrenberg <littledan@chromium.org> Reviewed-by:
Adam Klein <adamk@chromium.org> Cr-Commit-Position: refs/heads/master@{#46770}
-
Ross McIlroy authored
There remained a few of regressions and we didn't see any significant improvement in the real world with this turned on. This CL reverts all the StringConcat bytecode work which landed. BUG=v8:6243 Change-Id: I832eb72e880ad41411dbec8fe29f71ef0f2025c8 Reviewed-on: https://chromium-review.googlesource.com/575130 Commit-Queue: Ross McIlroy <rmcilroy@chromium.org> Reviewed-by:
Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Daniel Clifford authored
Change-Id: I1a89833e51ad92102dece1aed850144d18380df0 Reviewed-on: https://chromium-review.googlesource.com/576178 Commit-Queue: Daniel Clifford <danno@chromium.org> Reviewed-by:
-
Michael Achenbach authored
This copies a necessary suppression from Chromium's blacklist. NOTRY=true TBR=ishell@chromium.org Bug: chromium:726584 Change-Id: I7dfa13ad4a929b28bccb782be0a0faa9d29be1eb Reviewed-on: https://chromium-review.googlesource.com/576001Reviewed-by:
Michael Achenbach <machenbach@chromium.org> Commit-Queue: Michael Achenbach <machenbach@chromium.org> Cr-Commit-Position: refs/heads/master@{#46767}
-
Ulan Degenbaev authored
This patch adds a new intrinsic: %DebugTrackRetainingPath(object). Calling the intrinsic in JS code saves a weak reference to the given object in GC internal table of tracked objects. Each subsequent full GC prints to stdout the retaining path for each tracked object (if it is still alive). The retaining path is the real path that the marker took from the root set to the tracked object. This is useful for investigating of memory leaks: 1) Add %DebugTrackRetainingPath(leaking_object) in JS code. For example: function foo() { let x = { bar: "bar"}; %DebugTrackRetainingPath(x); return () => { return x; } } let closure = foo(); gc(); 2) Run d8 with --allow-natives-syntax --track-retaining-path --expose-gc. 3) Check the retaining path in stdout. For more detailed inspection, run d8 in gdb and set breakpoint in v8: :internal::Heap::PrintRetainingPath. Change-Id: I01a0faac1e009bc6c321fa75613900b49d2b036f Reviewed-on: https://chromium-review.googlesource.com/575972 Commit-Queue: Ulan Degenbaev <ulan@chromium.org> Reviewed-by: -
Andreas Haas authored
The outer_zone_ is only used to determine if the graph may be unverifiable. R=bmeurer@chromium.org Change-Id: Idad2bbb0d2a4ba9006c852276651e6780c1128c5 Reviewed-on: https://chromium-review.googlesource.com/566821Reviewed-by:
-
Dusan Simicic authored
Add support for slli, srai, srli, bclri, bseti, bnegi, binsli, binsri, sat_s, sat_u, srari, srlri MSA instructions in mips32 and mips64 simulators. Bug: Change-Id: I1a351a23c733f0bfbc829f25874df26295327afc Reviewed-on: https://chromium-review.googlesource.com/568020Reviewed-by:
Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com> Commit-Queue: Ivica Bogosavljevic <ivica.bogosavljevic@imgtec.com> Cr-Commit-Position: refs/heads/master@{#46764}
-
Ulan Degenbaev authored
This patch changes the semi-space size to 512K. > Original commit message: > Revert "[heap] Allow a minimum semi-space size of 512K." > This reverts commit 0d2ed6c3. > The CL introduced perf regressions: crbug.com/735649. > We are going to reland the CL in an isolated V8 roll to ensure > that perf regressions are attributed correctly. > Original commit message: > > [heap] Allow a minimum semi-space size of 512K. > > This CL also reduces the minimum semi-space size to 512K. > > BUG=chromium:716032 > BUG=chromium:735649 Change-Id: Iabc377cba2911b28d51b98bb5b85134d4e893632 Reviewed-on: https://chromium-review.googlesource.com/575066Reviewed-by:
-
Ulan Degenbaev authored
This relands parts of "[heap] Allow a minimum semi-space size of 512K." excluding the actual semi-space size change. This partially reverts commit f341bb0f > Original commit message: > Revert "[heap] Allow a minimum semi-space size of 512K." > This reverts commit 0d2ed6c3. > The CL introduced perf regressions: crbug.com/735649. > We are going to reland the CL in an isolated V8 roll to ensure > that perf regressions are attributed correctly. > Original commit message: > > [heap] Allow a minimum semi-space size of 512K. > > This CL also reduces the minimum semi-space size to 512K. > > BUG=chromium:716032 > BUG=chromium:735649 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: I5ed66b72104aa877d67fcd20bdadc807ea1551c3 Reviewed-on: https://chromium-review.googlesource.com/575065 Commit-Queue: Ulan Degenbaev <ulan@chromium.org> Reviewed-by:
-
Georgia Kouveli authored
Also remove UnsafeAcquire from UseScratchRegisterScope, which was only used for these stubs. Bug: Change-Id: Ia8648e53f1165ae489c0475dbb2d10c6978e4e84 Reviewed-on: https://chromium-review.googlesource.com/576181Reviewed-by:
-
Michael Achenbach authored
This adds a copy of tools/ubsan/vptr_blacklist.txt to V8, which is needed for ubsan compilation. NOTRY=true TBR=ishell@chromium.org Bug: chromium:726584 Change-Id: Ie06a031ce501d7f83121d45b04ac34672eb1ca9e Reviewed-on: https://chromium-review.googlesource.com/575977 Commit-Queue: Michael Achenbach <machenbach@chromium.org> Reviewed-by:
-
Igor Sheludko authored
This is a reland of b90e83f5 Original change's description: > [runtime] Add shortcuts for elements kinds transitions. > > The shortcuts ensure that field type generalization is properly > propagated in the transition graph. > > Bug: chromium:738763 > Change-Id: Id701a6f95ed6ea093c707fbe0bac228f1f856e9f > Reviewed-on: https://chromium-review.googlesource.com/567992 > Commit-Queue: Igor Sheludko <ishell@chromium.org> > Reviewed-by: Jakob Kummerow <jkummerow@chromium.org> > Cr-Commit-Position: refs/heads/master@{#46622} Bug: chromium:738763, chromium:742346, chromium:742381, chromium:745844 Change-Id: I93974e3906b2c7710bd525f15037a2dd97f263ad Reviewed-on: https://chromium-review.googlesource.com/575227 Commit-Queue: Igor Sheludko <ishell@chromium.org> Reviewed-by:
-
Michael Starzinger authored
This makes {NeedsDebugHookCheck} the default for all invocations, as there is no call-site left that doesn't perform said check. All other pieces of the {CallWrapper} are dead since Crankshafts removal. R=jgruber@chromium.org Change-Id: I158b816c089ede42972e8a7bdfc6ef0c02053a6b Reviewed-on: https://chromium-review.googlesource.com/577531Reviewed-by: -
Maya Lekova authored
Bug: v8:6559, v8:6557 Change-Id: Ide0114a5cdcce3cf948a070465460662c56bad8d Reviewed-on: https://chromium-review.googlesource.com/577527Reviewed-by:
Franziska Hinkelmann <franzih@chromium.org> Commit-Queue: Maya Lekova <mslekova@google.com> Cr-Commit-Position: refs/heads/master@{#46757}
-
Mythri authored
Remove SetForceInlineFlag from the js builtins. Bug: Change-Id: I962982509c82e4baba8dc32a0f163147c47daf34 Reviewed-on: https://chromium-review.googlesource.com/571803Reviewed-by:
-
Michael Starzinger authored
This adds handling for exceptional control projections when lowering calls to {Array.prototype.map} in the call reducer. R=mvstanton@chromium.org TEST=mjsunit/optimized-map BUG=v8:1956 Change-Id: If39ee836bbc3406a7fca4bad0d2c9321130cae2a Reviewed-on: https://chromium-review.googlesource.com/575928 Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Reviewed-by:Michael Stanton <mvstanton@chromium.org> Cr-Commit-Position: refs/heads/master@{#46755}
-
Pierre Langlois authored
This is a reland of a72b2f88 Original change's description: > [arm] Restrict grouping pushes before a TailCall to registers only > > We optimize parallel moves performed before a TailCall by grouping adjacent > pushes. This way, we may use a single instruction to push multiple registers at > once. However, we also have support for pushing immediates and stack slots for > which the benefit is questionnable therefore this patch removes support for > them. > > Concerning immediate pushes, it looks like a mistake since we do not have > support for this case in `AssembleMove` so this patch removes it. Furthermore, > if we add a test for this case, we see that a `push ip` instruction is > generated, effectively pushing whatever was in `ip` at the time instead of > pushing a constant. > > Concerning stack slot pushes, we generate a more or less equivalent sequence of > instructions. > > Finally, grouping floating point pushes is not used anywhere so this patch > removes support for this also. > > Bug: v8:6553 > Change-Id: I9b820d33361fc442dd813f66e1f96cda41009110 > Reviewed-on: https://chromium-review.googlesource.com/567191 > Reviewed-by: Benedikt Meurer <bmeurer@chromium.org> > Commit-Queue: Pierre Langlois <pierre.langlois@arm.com> > Cr-Commit-Position: refs/heads/master@{#46718} Bug: v8:6553 Change-Id: Ib9a55dae7cc5db6185d163c56088ff23426d04bb Reviewed-on: https://chromium-review.googlesource.com/576087Reviewed-by:
-
Tom Anderson authored
BUG=chromium:575778 R=machenbach@chromium.org Change-Id: Iac37a6a2f2ccf46467e4ecdbaefcc4e532491393 Reviewed-on: https://chromium-review.googlesource.com/576853 Commit-Queue: Michael Achenbach <machenbach@chromium.org> Reviewed-by:
-
Camillo Bruni authored
Empty Array literals are amongst the most commonly used literal types on our top25 page list. Using a custom bytecode we can drop the boilerplate for empty Array literals alltogether. However, we still need a proper AllocationSite to track ElementsKind transitions. Bug: v8:6211 Change-Id: Id5dbdac0ea8e24dd474e679c902c6e4a2957af1d Reviewed-on: https://chromium-review.googlesource.com/567079 Commit-Queue: Camillo Bruni <cbruni@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Reviewed-by:
-
Maya Lekova authored
Bug: v8:6558, v8:6557 Change-Id: If090be375bafc7c7c6b21ad0eef820e324c7b01c Reviewed-on: https://chromium-review.googlesource.com/574494Reviewed-by:
Camillo Bruni <cbruni@chromium.org> Reviewed-by:
-
v8-autoroll authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/1808a90..543ec6e Rolling v8/third_party/catapult: https://chromium.googlesource.com/external/github.com/catapult-project/catapult/+log/44b022b..a4770ef Rolling v8/third_party/icu: https://chromium.googlesource.com/chromium/deps/icu/+log/dfa798f..1fec0c8 TBR=machenbach@chromium.org,hablich@chromium.org Change-Id: If59abb9362c8f5136dcbf0351d172125d5fd057e Reviewed-on: https://chromium-review.googlesource.com/577187Reviewed-by:
-
Mircea Trofin authored
Temporary check ensuring no regression while we get the wasm code off the GC heap, and and until we de-contextualize wasm code. We expect the only embedded objects to be: CEntryStub, undefined, and the various builtins for throwing exceptions like OOB. These are all immovable because they are snapshotted. Additionally, we embed references to the FixedArray that backs WebAssembly.Table. That will be replaced separately with a native data structure. Once the Table is native, we can generate Wasm code off the GC heap, as long as the embedded objects are immutable, which is the property we check for here. That greatly simplifies a subsequent step, which is to replace those dependencies with a isolate-independent solution. The source of simplification is that we don't have to worry about moving pointers. Bug: Change-Id: Id1e41863a2619c2afc50f48416f422012f0c9a24 Reviewed-on: https://chromium-review.googlesource.com/574938Reviewed-by:
-
- 18 Jul, 2017 7 commits
-
-
Michael Lippautz authored
- Create items for each page containing OLD_TO_NEW references. - Introduce the flags for parallel scavenge since this forces us to introduce Tasks. Bug: chromium:738865 Change-Id: Idad63f4318bdb3786117441e5413eb5e8594b7fb Reviewed-on: https://chromium-review.googlesource.com/575052 Commit-Queue: Michael Lippautz <mlippautz@chromium.org> Reviewed-by:
-
Clemens Hammacher authored
This makes all data structures containing non-trivially-copyable fields move-only, to prevent security and performance bugs. Drive-by: Fix smaller performance bugs found by this refactoring. R=titzer@chromium.org Change-Id: I6802ac3591534c2ab5cacb2ca42b737f3b7fa801 Reviewed-on: https://chromium-review.googlesource.com/576170Reviewed-by:
Ben Titzer <titzer@chromium.org> Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#46747}
-
Jakob Kummerow authored
Bug: v8:6550 Change-Id: I888f91db1fd842d1fef8a5fb749da229dfb6ab97 Reviewed-on: https://chromium-review.googlesource.com/575756Reviewed-by:
Daniel Clifford <danno@chromium.org> Reviewed-by:
-
Ulan Degenbaev authored
This reverts commit 5d8840e2. Reason for revert: performance regressions. BUG=chromium:737472,chromium:736640 Original change's description: > [runtime] Copy descriptors on prototype map transition. > > Currently the descriptors are moved from the old map to the new map, > which is unsafe for the concurrent marker. This patch removes the map > mutation. > > Change-Id: I3f7ce455c7344148a122c7443cf32a4eef0307be > Reviewed-on: https://chromium-review.googlesource.com/535480 > Commit-Queue: Ulan Degenbaev <ulan@chromium.org> > Reviewed-by: Toon Verwaest <verwaest@chromium.org> > Cr-Commit-Position: refs/heads/master@{#46131} TBR=ulan@chromium.org,verwaest@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Change-Id: I2525a8101a01f39ccad1a6e49aae1e3f28da5b5d Reviewed-on: https://chromium-review.googlesource.com/555551Reviewed-by:
-
Sathya Gunasekaran authored
Add SetProperties as the generic interface to set properties. In the future, this will switch based on the input properties type and correctly store the hash code. This patch also updates tests to check against empty_property_array instead of empty_fixed_array. Bug: v8:6404 Change-Id: I39d324ea3ab3cc2c2223b6f4be64139bb88edd94 Reviewed-on: https://chromium-review.googlesource.com/574761Reviewed-by:
-
Benedikt Meurer authored
The test for CanTreatHoleAsUndefined on keyed element access was checking for stability of Object.prototype and Array.prototype and even adding stability dependencies on both, which is too restrictive and leads to unnecessary deoptimizations (and might disable further optimization of the keyed access depending on the state of the prototype objects during optimization). This was not intended and is considered a (performance) bug. Instead use the correct approach of checking whether the receiver's prototype is one of the current Object.prototype or Array.prototype objects (since the Array protector works isolate-wide), and then check the Array protector and install an appropriate code dependency on the protector only. Bug: v8:6607 Change-Id: I0bcfe32813ca3693e7b22de31b03edb3509d0a27 Reviewed-on: https://chromium-review.googlesource.com/574849Reviewed-by:
-
Alexey Kozyatinskiy authored
Nop bytecodes are required only for break locations in debugger. Since nop bytecode doesn't change program state we can remove all of them. There are at least two changes which this CL produce: - we don't provide break position when we load local variable (still provide when load variable from global), - we don't provide break position for statements without actual break positions (e.g. "a;") - these expressions should be super rare and user always can set breakpoint before or after this statement. More details in one pager: https://docs.google.com/a/google.com/document/d/1JXlQpfMa9vRojbE272b6GMBbrfh6m_00135iAUOJEz8/edit?usp=sharing Bug: v8:6425 Change-Id: I4aee73d497a84f7b5d89caa6dda6d3060567dfda Reviewed-on: https://chromium-review.googlesource.com/543161Reviewed-by:
-
