- 25 Nov, 2020 26 commits
-
-
Etienne Pierre-Doray authored
This reverts commit 873e5aa3. Reason for revert: Flaky cctest/test-incremental-marking/IncrementalMarkingUsingTasks Original change's description: > Reland "Reland "[Heap]: Convert Sweep to Job"" > > This is a reland of b16c7e5b > > Issue: ShouldYield is called multiple time. > Fix: ConcurrentSweepSpace returns false if not done (yielding), to avoid > calling it again. > > Issue: failing test-streaming-compilation > Safe to reland after > https://chromium-review.googlesource.com/c/v8/v8/+/2507379 > > Original change's description: > > Reland "[Heap]: Convert Sweep to Job" > > > > This is a reland of 795c0b1c > > Reason for revert: > > TSAN failures https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/33884 > > Safe to reland as-is with fix to EagerUnmappingInCollectAllAvailableGarbage > > https://chromium-review.googlesource.com/c/v8/v8/+/2502809 > > > > Original change's description: > > > [Heap]: Convert Sweep to Job > > > > > > max concurrency is inferred from queue size for OLD_SPACE & MAP_SPACE. > > > Extra Sweeper::TearDown() in MarkCompactCollector::TearDown() is needed > > > to cancel job. > > > > > > Change-Id: Iafba7d7d24e8f6e5c5a1d5c0348dea731f0ac224 > > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2480783 > > > Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org> > > > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > > > Cr-Commit-Position: refs/heads/master@{#70767} > > > > Change-Id: Id9a5baceed4664f53da39597af56a2067e4f3c6f > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2502808 > > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > > Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org> > > Cr-Commit-Position: refs/heads/master@{#70845} > > Change-Id: I32de9faebdbd2f7f6d7f9a9525871fc691fb3f2c > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2507378 > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org> > Cr-Commit-Position: refs/heads/master@{#71347} TBR=ulan@chromium.org,etiennep@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Issue: ShouldYield is called multiple time. Issue: failing test-streaming-compilation Change-Id: Ic82221d118d7a1f6ea98021e03f4d4b95368b01c Bug: v8:11198 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2561198Reviewed-by:
Leszek Swirski <leszeks@chromium.org> Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Commit-Queue: Leszek Swirski <leszeks@chromium.org> Cr-Commit-Position: refs/heads/master@{#71413}
-
Clemens Backes authored
Function tables have been removed from the scope object in https://crrev.com/c/2507696, hence the code for printing them is dead now. R=bmeurer@chromium.org Change-Id: Ib36fb314ae54468239737f100a6594d8d2031218 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557982Reviewed-by:
Benedikt Meurer <bmeurer@chromium.org> Commit-Queue: Clemens Backes <clemensb@chromium.org> Cr-Commit-Position: refs/heads/master@{#71412}
-
Clemens Backes authored
After preparing Liftoff, TurboFan, and the interpreter for this change, we now store the memory offset as uint64_t. {LoadLane} and {StoreLane} were added after the TurboFan refactoring, so those two are adapted similar to the other memory operations. TBR=manoskouk@chromium.org Bug: v8:10949 Cq-Include-Trybots: luci.v8.try:v8_win64_msvc_rel_ng Change-Id: I8f3084c21a7d99f72df1bc18c2b507c4e84570cd Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2560720 Commit-Queue: Clemens Backes <clemensb@chromium.org> Reviewed-by:Clemens Backes <clemensb@chromium.org> Cr-Commit-Position: refs/heads/master@{#71411}
-
Vasili Skurydzin authored
Change-Id: I63446404789a33b97db2bacee2faf3a04e137ed5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558973Reviewed-by:
Milad Fa <mfarazma@redhat.com> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Commit-Queue: Vasili Skurydzin <vasili.skurydzin@ibm.com> Cr-Commit-Position: refs/heads/master@{#71410}
-
Leszek Swirski authored
This reverts commit cf9a28b6. Reason for revert: TSAN failures: https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/34374? Original change's description: > [wasm] Reduce job priority once baseline compilation finishes > > This Cl changes the priority of baseline compilation from kUserVisible > to kUserBlocking. Once baseline compilation finishes, the priority is > reduced to kUserVisible. The reason for using kUserBlocking is that > thereby TurboFan compilation cannot block Liftoff compilation anymore. > Additionally, kUserBlocking is quite appropriate, as the initial > compilation does block a whole section of a web app from execution. > > R=clemensb@chromium.org > > Bug: v8:11088 > Change-Id: Ifde42d20f36d4c0a5122b0008311ccdffbb60e48 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2519559 > Commit-Queue: Andreas Haas <ahaas@chromium.org> > Reviewed-by: Clemens Backes <clemensb@chromium.org> > Cr-Commit-Position: refs/heads/master@{#71404} TBR=ahaas@chromium.org,clemensb@chromium.org Change-Id: I9a975c4c43189015491b08d3a98de991d8167daf No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:11088 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2560200Reviewed-by:
-
Vasili Skurydzin authored
platform Change-Id: I644970a5dfe3794d2b9c90719b3a1b2b87c2f5e3 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558971Reviewed-by:
-
Michael Lippautz authored
Refactor write barriers and split calls, as e.g. DijkstraWriteBarrier also contained logic for recording slots (cards) for the young generation. The new API exposes the following: - GetWriteBarrierType(): Retrieving the type of barrier that must be emitted; - DijkstraWriteBarrier(), DijkstraWriteBarrierRange(): Dijkstra-style write barriers; - SteeleWriteBarrier(): Steele-style write barrier; - GenerationalBarrier(): Barrier for recording slots when using multiple generations; Compilers running with -O3 optimize the DijkstraWriteBarrierPolicy down to the same instructions as before the split. Change-Id: If68839cc6357b2f568986c9ce8ca753b1e96a70a Bug: chromium:1056170 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557514 Commit-Queue: Michael Lippautz <mlippautz@chromium.org> Reviewed-by:
Omer Katz <omerkatz@chromium.org> Reviewed-by:
-
Andreas Haas authored
In the generic wrapper we popped the wrong number of parameters off the stack. We always popped the number of parameters needed by the generic wrapper, according to the signature. The correct number though is max(parameters provide, parameters needed). R=victorgomes@chromium.org, thibaudm@chromium.org CC=vkont@google.com Bug: v8:10982 Change-Id: If9b8d4dbe093eb6df08ddf9f3594d5c60b9be33f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558317 Commit-Queue: Andreas Haas <ahaas@chromium.org> Reviewed-by:
Thibaud Michaud <thibaudm@chromium.org> Reviewed-by:
Victor Gomes <victorgomes@chromium.org> Cr-Commit-Position: refs/heads/master@{#71406}
-
Mythri A authored
With concurrent inlining, the inlining phase happens on the background thread and the data needed for the inlining phase is serialized on the main thread. The serialization phase tries to gather data about functions called which is sometimes more expensive than inlining phase itself. So it's better not to use concurrent inlining for TurboFan compilations when tiering up from Turboprop to TurboFan. Turboprop compilations don't inline and hence it is OK to continue using concurrent inlining for Turboprop compilations. Bug: v8:9684 Change-Id: Ib529905213fa7f0df84ee52218adc27f7c219f60 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557504 Commit-Queue: Mythri Alle <mythria@chromium.org> Reviewed-by:
Ross McIlroy <rmcilroy@chromium.org> Cr-Commit-Position: refs/heads/master@{#71405}
-
Andreas Haas authored
This Cl changes the priority of baseline compilation from kUserVisible to kUserBlocking. Once baseline compilation finishes, the priority is reduced to kUserVisible. The reason for using kUserBlocking is that thereby TurboFan compilation cannot block Liftoff compilation anymore. Additionally, kUserBlocking is quite appropriate, as the initial compilation does block a whole section of a web app from execution. R=clemensb@chromium.org Bug: v8:11088 Change-Id: Ifde42d20f36d4c0a5122b0008311ccdffbb60e48 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2519559 Commit-Queue: Andreas Haas <ahaas@chromium.org> Reviewed-by:
-
Sylvain Defresne authored
The gn variable ios_use_goma_rbe is ignored since the CL https://crrev.com/c/2555117 landed, so stop overriding it on the bots (it is now always enabled which is what the bots want). Bug: none Change-Id: Iaa085dd1fd0559a41372744ed4c4491c4b5d9908 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558218Reviewed-by:
Michael Achenbach <machenbach@chromium.org> Commit-Queue: Sylvain Defresne <sdefresne@chromium.org> Cr-Commit-Position: refs/heads/master@{#71403}
-
Hannes Payer authored
Change-Id: If01a247f2bae3c90708f9f8355325fa5b7113913 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2560196Reviewed-by:
-
Eric Seckler authored
The dependency on v8_tracing supplies include paths & dependencies on the tracing library when built with v8_use_perfetto. This is an attempt to fix the linux-perfetto-rel builder [1], which is currently erroring: FAILED: obj/v8/cppgc_base/sweeper.o /b/s/w/ir/cache/goma/client/gomacc ../../third_party/llvm-build/Release+Asserts/bin/clang++ -MMD -MF...(too long) In file included from ../../v8/src/heap/cppgc/sweeper.cc:24: In file included from ../../v8/src/heap/cppgc/stats-collector.h:17: In file included from ../../v8/src/heap/cppgc/trace-event.h:9: In file included from ../../v8/src/tracing/trace-event.h:12: gen/third_party/perfetto/protos/perfetto/trace/track_event/debug_annotation.pbzero.h:9:10: fatal error: 'perfetto/protozero/message.h' file not found ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ [1] https://ci.chromium.org/p/chromium/builders/ci/linux-perfetto-rel Bug: chromium:1056170 Change-Id: Id5a382d472139f7abe5ead67ec6eed2f8395e6b5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2560257 Commit-Queue: Eric Seckler <eseckler@chromium.org> Reviewed-by:
Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#71401}
-
Manos Koukoutos authored
The logic to detect if a 'br' instruction is a branch or a return was duplicated in function-body-decoder-impl.h and in both interfaces. Apart from code duplication, this structure also made it hard to implement planned compiler improvements. This CL removes this duplication by upgrading BrOrRet (that already existed in both Liftoff and Turbofan interfaces) to an interface function and using it for unconditional branches. Change-Id: Ia04952cce621335268fc40ef9544a99b61dc7da3 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557515 Commit-Queue: Manos Koukoutos <manoskouk@chromium.org> Reviewed-by:
-
Clemens Backes authored
This is a minor refactoring before fixing actual issues. 1) The update of the {per_isolate_data_} is moved into {FloodWithBreakpoints}, which is already taking the mutex. 2) The {PrepareStep} method takes a {WasmFrame*} directly instead of its ID. In most cases, this prevents the creation of an additional stack frame iterator. R=thibaudm@chromium.org Bug: chromium:1145176 Change-Id: I1a6cd15550bbb4ef78ba522427bed1c23185569e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558318Reviewed-by:Simon Zünd <szuend@chromium.org> Reviewed-by:
-
Leszek Swirski authored
Pass the Isolate/LocalIsolate through to StringTable matchers and WriteToFlat, so avoid having to get the Isolate via the String, and to avoid locking on the main thread entirely. This allows us to remove the String overload of the SharedStringAccessGuardIfNeeded constructor entirely, to avoid this anti-pattern in the future. Bug: chromium:1146972 Change-Id: I53bba126b105e1c9629d6e64d8bb574e62e3ad45 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557988 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
Santiago Aboy Solanes <solanes@chromium.org> Reviewed-by:
Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#71398}
-
Santiago Aboy Solanes authored
This allows us to assert at compile time that a class instance is assigned, which is particularly useful for Guard classes. Change-Id: Id16b2bb70d29573566e821c908c1169d49ec57af Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2552415 Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org> Reviewed-by:
-
Milad Fa authored
Port 3836aeb0 Original Commit Message: Apart from removing Min and Max (utils.h), this is mostly a renaming. In a few cases I had to add a cast. In a bunch of cases I had to use initializer lists to force call-by-value for static member constants because call-by-reference wouldn't compile (like in the previous CL). In a few places I used initializer lists in place of nested min/max operations. R=neis@chromium.org, joransiu@ca.ibm.com, junyan@redhat.com, midawson@redhat.com BUG= LOG=N Change-Id: Iecb43c19b8e16721e942553d7d811daf74bedc02 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557570 Commit-Queue: Georg Neis <neis@chromium.org> Reviewed-by:
Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#71396}
-
Michael Lippautz authored
Bug: chromium:1056170 Change-Id: I303715630c79213d5b0dc34ff7107e8ffa973539 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557991 Commit-Queue: Michael Lippautz <mlippautz@chromium.org> Reviewed-by:
-
Zhi An Ng authored
Prototype v128.load{8,16,32,64}_lane on IA32 (stores will come later). This is pretty similar to x64 version, except that there is no signal handler for OOB access, so kProtected is not a valid access mode. Left some TODOs for myself to merge the new instruction codes (kIA32Pinsrb) with the replace lane Wasm instructions. Bug: v8:10975 Change-Id: I5c9f9a45e2e7f06e8fab4a28cdfe1857ccc35880 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557063 Commit-Queue: Zhi An Ng <zhin@chromium.org> Reviewed-by:Bill Budge <bbudge@chromium.org> Cr-Commit-Position: refs/heads/master@{#71394}
-
Zhi An Ng authored
This reverts commit 21e47944. Reason for revert: Broke build https://ci.chromium.org/p/v8/builders/ci/V8%20Arm%20-%20debug/16872? Original change's description: > [wasm-simd][arm] Prototype i64x2.bitmask > > Cleanup to simulator to remove repetitive logic to get instruction > fields. > > Bug: v8:10997 > Change-Id: I01f0b99f85788b41e4cab505fc94362d637c396f > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2554256 > Commit-Queue: Zhi An Ng <zhin@chromium.org> > Reviewed-by: Jakob Kummerow <jkummerow@chromium.org> > Reviewed-by: Bill Budge <bbudge@chromium.org> > Cr-Commit-Position: refs/heads/master@{#71391} TBR=bbudge@chromium.org,jkummerow@chromium.org,v8-arm-ports@googlegroups.com,zhin@chromium.org Change-Id: I15513cc5923db7d189d08a9faf5051e57af7190b No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:10997 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558260Reviewed-by:
Zhi An Ng <zhin@chromium.org> Commit-Queue: Zhi An Ng <zhin@chromium.org> Cr-Commit-Position: refs/heads/master@{#71393}
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/356ef25..29207aa Rolling v8/third_party/aemu-linux-x64: qDJOg4W2RuPZ92H6d33I9kLLWjqfYuMr_gFsPRodSQAC..b5ckZyVJ3XwwvnxV2J_ybKfLyiHfOj81r9Llym22_UsC Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/a629d81..ae003f5 Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/260eb0f..8820ab8 Rolling v8/tools/luci-go: git_revision:6cbe3f56e9f00b8f65eae21f01838a8b58191a47..git_revision:1a022d3a4c50be4207ee93451255d71896416596 Rolling v8/tools/luci-go: git_revision:6cbe3f56e9f00b8f65eae21f01838a8b58191a47..git_revision:1a022d3a4c50be4207ee93451255d71896416596 Rolling v8/tools/luci-go: git_revision:6cbe3f56e9f00b8f65eae21f01838a8b58191a47..git_revision:1a022d3a4c50be4207ee93451255d71896416596 TBR=machenbach@chromium.org,tmrts@chromium.org,v8-waterfall-sheriff@grotations.appspotmail.com Change-Id: I56abc6880884805075c73201c3c871c1ceedf284 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558979Reviewed-by:
v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> Commit-Queue: v8-ci-autoroll-builder <v8-ci-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com> Cr-Commit-Position: refs/heads/master@{#71392}
-
Zhi An Ng authored
Cleanup to simulator to remove repetitive logic to get instruction fields. Bug: v8:10997 Change-Id: I01f0b99f85788b41e4cab505fc94362d637c396f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2554256 Commit-Queue: Zhi An Ng <zhin@chromium.org> Reviewed-by:
Jakob Kummerow <jkummerow@chromium.org> Reviewed-by:
-
Zhi An Ng authored
Bug: v8:11074 Change-Id: I41a62490b3abe9ff00905fd9ea7e4bfff5f23a48 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557064Reviewed-by:
-
Zhi An Ng authored
Bug: v8:11074 Change-Id: I01ffd94290aa5deafa0953a86ff555147b813bf4 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557069Reviewed-by:
-
Zhi An Ng authored
Also remove a comment referring to using the macro. Bug: v8:11074 Change-Id: Ib56a0360b28812833b372738f4956ef41c59a97b Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557058Reviewed-by:
-
- 24 Nov, 2020 14 commits
-
-
Daniel Clark authored
This change refactors the v8.h API as discussed in https://docs.google.com/document/d/1yuXgNHSbTAPubT1Mg0JXp5uTrfirkvO1g5cHHCe-LmY/edit#heading=h.q0c9h4p928mn such that a v8::Module exposes module requests as a FixedArray of ModuleRequest objects, which can then be used to obtain their module specifier and source code offset. This replaces the old functions that passed back individual specifier Strings and Locations via repeated calls to getters that take an index. These are marked as deprecated. The new ModuleRequest interface includes a getter for an ImportAssertions FixedArray, which will contain the import assertions for the request if --harmony-import-assertions is set, and will be empty otherwise. One notable change here is that the APIs now return source code offsets rather than v8::Locations. The host must then call the new Module::SourceOffsetToLocation to convert these offsets into line/column numbers. This requires a bit more back-and-forth, but allows the host to defer the cost of converting from source offset to line/column numbers until an error needs to be reported, potentially skipping the work altogether. Bug: v8:10958 Change-Id: I181639737c701e467324e6c781aa4d7bdd87ae8c Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2545577 Commit-Queue: Dan Clark <daniec@microsoft.com> Reviewed-by:
Camillo Bruni <cbruni@chromium.org> Reviewed-by:
Marja Hölttä <marja@chromium.org> Cr-Commit-Position: refs/heads/master@{#71387}
-
Sara Tang authored
Removed wall_clock_time_in_us field. Bug: v8:11109 Change-Id: Ib3efa74fc741b793ef064c1d12ca1797be3a10c1 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2550280Reviewed-by:
-
Leszek Swirski authored
Add a CompareCharsEqual to complement CompareChars, where we only care about equality and not ordering. For such cases, we can memcmp for two- byte as well as one-byte strings (we can't for CompareChars because the ordering would be incorrect on little-endian systems). Replace uses of CompareChars that only compare the result against zero, with CompareCharsEqual. Additionally, use some template magic to simplify the "make unsigned" operation in these methods. Change-Id: I0d65bee81b98d3938d15daa4af331c90558ea84f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557980 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Camillo Bruni authored
- Use C++ primitives (int, bool) for the ScriptOrigin constructor. - Deprecate the old accessors and constructor Bug: v8:11195 Change-Id: I739edd6b4c58e19a8a16ddce863eea14ec933697 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2555005Reviewed-by:
-
Bill Budge authored
This reverts commit 44efa00b. Reason for revert: Breaks MSVC with warning as error: https://ci.chromium.org/p/v8/builders/ci/V8%20Win64%20-%20msvc/15903 Original change's description: > [wasm][memory64] Decode memory offset as 64-bit LEB > > After preparing Liftoff, TurboFan, and the interpreter for this change, > we now store the memory offset as uint64_t. {LoadLane} and {StoreLane} > were added after the TurboFan refactoring, so those two are adapted > similar to the other memory operations. > > R=manoskouk@chromium.org > > Bug: v8:10949 > Change-Id: Iba66ce448904e23b152fcb8612d171124e615473 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2555006 > Commit-Queue: Clemens Backes <clemensb@chromium.org> > Reviewed-by: Manos Koukoutos <manoskouk@chromium.org> > Cr-Commit-Position: refs/heads/master@{#71382} TBR=clemensb@chromium.org,manoskouk@chromium.org Change-Id: Ia0f46a0b6fd2102a61c7664d7cdd86a2cf8ddb24 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:10949 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558752Reviewed-by:
-
Clemens Backes authored
After preparing Liftoff, TurboFan, and the interpreter for this change, we now store the memory offset as uint64_t. {LoadLane} and {StoreLane} were added after the TurboFan refactoring, so those two are adapted similar to the other memory operations. R=manoskouk@chromium.org Bug: v8:10949 Change-Id: Iba66ce448904e23b152fcb8612d171124e615473 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2555006 Commit-Queue: Clemens Backes <clemensb@chromium.org> Reviewed-by:Manos Koukoutos <manoskouk@chromium.org> Cr-Commit-Position: refs/heads/master@{#71382}
-
Michael Lippautz authored
Exposes an opaque handle for uniformly (cppgc and V8) referring to an instance of a heap. Exposes a set of raw write barriers for advances embedders through subtle::HeapConsistency which is a mirror into write barrier internals. The following barriers are exposed: - DijkstraWriteBarrier: Regular Dijkstra-style write barrier (add to wavefront); - DijkstraWriteBarrierRange: Same as DijkstraWriteBarrier but operating on a range of slots that are composite (inlined) objects; - SteeleWriteBarrier: Regular Steele-style write barrier (retreating wavefront); Change-Id: Ib5ac280204686bf887690f72df1cdb506ea6ef70 Bug: chromium:1056170 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2554601Reviewed-by:
Anton Bikineev <bikineev@chromium.org> Reviewed-by:
-
Georg Neis authored
Apart from removing Min and Max (utils.h), this is mostly a renaming. In a few cases I had to add a cast. In a bunch of cases I had to use initializer lists to force call-by-value for static member constants because call-by-reference wouldn't compile (like in the previous CL). In a few places I used initializer lists in place of nested min/max operations. Bug: v8:11074 Change-Id: I53a5411be6334ff41e7a8517e6b87fb46f14d086 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2545523 Commit-Queue: Georg Neis <neis@chromium.org> Reviewed-by:
Hannes Payer <hpayer@chromium.org> Reviewed-by:
-
Milad Fa authored
Port 2180e20f Original Commit Message: The total frame size returned by GetTotalFrameSize includes the frame marker. However, the frame marker is pushed on the stack with a push instruction. Therefore it is not needed to allocate memory for it again on the stack. This CL therefore reduces the memory allocated on the stack by the size of the frame marker. R=ahaas@chromium.org, joransiu@ca.ibm.com, junyan@redhat.com, midawson@redhat.com BUG= LOG=N Change-Id: I76e259d54703265fcf1a84857365997af008e16a Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2558257Reviewed-by:
Junliang Yan <junyan@redhat.com> Commit-Queue: Milad Fa <mfarazma@redhat.com> Cr-Commit-Position: refs/heads/master@{#71379}
-
Leszek Swirski authored
Make String::IsEqualTo use a direct string shape dispatch and a direct call to CompareChars, rather than splitting the behaviour over IsOneByte/IsTwoByte/HasOneBytePrefix. Avoiding GetFlatContent will make this method easier to make efficient while staying string-access-lock safe. Also, redefines the sequential string table key's matcher in terms of this IsEqualTo method. Change-Id: Iab71246e12044ebaeff06f0dbc14d28b3482dcbe Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557979 Commit-Queue: Igor Sheludko <ishell@chromium.org> Auto-Submit: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Gus Caplan authored
Bug: v8:10961 Change-Id: I3746dca570de005d203a2648dcffedd81122f215 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2553157Reviewed-by:
Shu-yu Guo <syg@chromium.org> Commit-Queue: Gus Caplan <snek@chromium.org> Cr-Commit-Position: refs/heads/master@{#71377}
-
Clemens Backes authored
This CL prepares the interpreter for 64-bit addresses. The offset (in the memory immediate) can now be 64-bit, and also the index (from the stack) will be 64-bit if memory64 is enabled. memory.size will return a 64-bit constant under memory64. memory.grow is a bit more involved and requires more refactorings in the called functions. I left a TODO in the interpreter for now. R=manoskouk@chromium.org Bug: v8:10949 Change-Id: I47ab7e38a7356239c827619325f4583069e46669 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2555000 Commit-Queue: Clemens Backes <clemensb@chromium.org> Reviewed-by:
-
Vicky Kontoura authored
This reverts commit f2034bf5. Reason for revert: Issue on the generic wrapper Original change's description: > [wasm] Enable --wasm-generic-wrapper > > This CL enables the --wasm-generic-wrapper flag by default. > > Bug: v8:10982 > Change-Id: Iada906ce49810806c500ee3ca26e09e847b69a27 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2539915 > Commit-Queue: Vicky Kontoura <vkont@google.com> > Reviewed-by: Andreas Haas <ahaas@chromium.org> > Reviewed-by: Jakob Kummerow <jkummerow@chromium.org> > Cr-Commit-Position: refs/heads/master@{#71374} TBR=jkummerow@chromium.org,ahaas@chromium.org,vkont@google.com Change-Id: I809de71155a76fce503ecd75c7c4854d507c3695 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:10982 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2557990Reviewed-by:
Andreas Haas <ahaas@chromium.org> Commit-Queue: Andreas Haas <ahaas@chromium.org> Cr-Commit-Position: refs/heads/master@{#71375}
-
Vicky Kontoura authored
This CL enables the --wasm-generic-wrapper flag by default. Bug: v8:10982 Change-Id: Iada906ce49810806c500ee3ca26e09e847b69a27 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2539915 Commit-Queue: Vicky Kontoura <vkont@google.com> Reviewed-by:
-
