- 28 Aug, 2019 6 commits
-
-
Dominik Inführ authored
Instead of fully reverting the CL that introduced the old-to-new invalidated set, simply do not filter recorded slots yet. Bug: v8:9454 Change-Id: I2b880f64f29e319056ad49e2284dca26eb8770f4 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1773252Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> Cr-Commit-Position: refs/heads/master@{#63423}
-
Jakob Kummerow authored
The previous pseudo-smi storage scheme for caching call target addresses in a struct without requiring a custom visitor only works on uncompressed 64-bit platforms. This patch fixes other platforms (natural or compressed 32-bit) by boxing the address in a Foreign. Change-Id: I3c182c1d9ccae4858cac2757fc3daa40d1520998 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771780 Commit-Queue: Jakob Kummerow <jkummerow@chromium.org> Reviewed-by:
Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#63422}
-
Maya Lekova authored
Remove unnecessary call to Serialize. Change a Handle cast to a Ref cast. Add a CHECK to JSHeapBroker::SetFeedback that the source is valid. Bug: v8:7790 Change-Id: I6a2f10d45456c6b128beff009016009457dc109c Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1773248 Commit-Queue: Maya Lekova <mslekova@chromium.org> Reviewed-by:
Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#63421}
-
Primiano Tucci authored
Doing some refactoring to perfetto build files. Rolling to ensure that we are not breaking embedders in the process. Cq-Include-Trybots: luci.v8.try:v8_linux64_perfetto_dbg_ng Bug: v8:8339 Change-Id: I5c603680cf6dd295e9d11a4eba70785e6dc512ba Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1773244 Auto-Submit: Primiano Tucci <primiano@chromium.org> Commit-Queue: Tamer Tas <tmrts@chromium.org> Reviewed-by:
Peter Marshall <petermarshall@chromium.org> Reviewed-by:
Tamer Tas <tmrts@chromium.org> Cr-Commit-Position: refs/heads/master@{#63420}
-
Andreas Haas authored
At the moment we only run the js-api spec tests of the core API on our try bots. With the new staging process we want to introduce for WebAssembly language features, see https://docs.google.com/document/d/1hB8mpWmzmtaxZ8PuJEkAWLwFqXTjrw7mJ3Ge9W1dB4E, we also want to run the js-api spec tests of proposals for which we already staged the implementation. With this CL I do the following changes: 1) The tools/wasm/update-wasm-spec-tests.sh now copies the js-api spec tests of the main spec and of the proposals to test/wasm-js/tests, and then uploads this directory to google cloud storage. The main spec tests are in test/wasm-js/tests, the proposal tests are in test/wasm-js/tests/proposals/PROPOSAL_NAME/. 2) Adjust the test-runner in test/wasm-js to run tests in tests/* instead of data/test/js-api/*. Thereby it also runs the proposal tests in test/wasm-js/tests/proposals/PROPOSAL_NAME/. For the proposal tests, the test runner now also adds d8 flags. 3) Remove the dependency to https://github.com/WebAssembly/spec from DEPS. 4) Cleanup .gitignore and wasm-js.status 5) Disable spec tests we don't pass with the new proposal. R=tmrts@chromium.org Bug: v8:9653 Change-Id: Ib3420871f17cb146d6cc7868f5613942a7f79d84 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768372 Commit-Queue: Tamer Tas <tmrts@chromium.org> Reviewed-by:
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/3f22131..3fe260c Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/7ad424d..c9fdf26 Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/0e5fff1..217195c Rolling v8/third_party/googletest/src: https://chromium.googlesource.com/external/github.com/google/googletest/+log/ed2eef6..6a3d632 TBR=machenbach@chromium.org,tmrts@chromium.org Change-Id: I2994dedaf7a48656dd2a05860ef6c8f3efc6e69c Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1772412Reviewed-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@{#63418}
-
- 27 Aug, 2019 11 commits
-
-
Joshua Litt authored
Launching nullish behind a flag resulted in a small performance regression in the adwords parsing benchmark. From local tests, doing a little manual PGO seemed to improve performance slightly. Parse.duration on this benchmark dropped from 1,639.188 ms to 1,535.312 ms Bug: chromium:997652 Change-Id: I537985793cdf310a0dda5a69ded9f0ea2c0a7fb0 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1773098 Commit-Queue: Joshua Litt <joshualitt@chromium.org> Reviewed-by:
Leszek Swirski <leszeks@chromium.org> Cr-Commit-Position: refs/heads/master@{#63417}
-
Ulan Degenbaev authored
This writes 0 to all trimmed slots and thus ensures that the invariant that all recorded slots in the trimmed area are valid, which will allow us to simplify invalidated slots tracking. Specifically: 1) Arrays are never inserted into the invalidated objects set. 2) The recorded slots outside an invalidated object are valid. Bug: v8:9454 Change-Id: Ifbef3752d52b5b47f2b694bd2b6c0a4c122abb7a Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771793Reviewed-by:
Dominik Inführ <dinfuehr@chromium.org> Commit-Queue: Ulan Degenbaev <ulan@chromium.org> Cr-Commit-Position: refs/heads/master@{#63416}
-
Michael Starzinger authored
R=ahaas@chromium.org TEST=mjsunit/wasm/type-reflection-with-exnref BUG=v8:7742,v8:8091 Change-Id: Ib8bd7b7cfafa3509db743e4404c2f1b573253881 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771790Reviewed-by:
Andreas Haas <ahaas@chromium.org> Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#63415}
-
Michael Starzinger authored
R=ahaas@chromium.org TEST=mjsunit/wasm/exceptions-global BUG=v8:8091 Change-Id: I9eb4c92cca087719afda2da5d11c206aaed28c95 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771788Reviewed-by:
-
Joyee Cheung authored
Previously variations of #constructor can be parsed when they are static. This patch throws early errors for them always. Bug: v8:8330 Change-Id: I51ab9b83f713c70d0896c0e8cab3282ef9a105f0 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1770332Reviewed-by:
Sathya Gunasekaran <gsathya@chromium.org> Commit-Queue: Joyee Cheung <joyee@igalia.com> Cr-Commit-Position: refs/heads/master@{#63413}
-
Dominik Inführ authored
InvalidatedSlotsCleanup::Free assumed that the size of invalidated objects was updated after sweeping a page and before allowing allocations again. This is now not the case anymore, so remove those CHECKs. Bug: chromium:997901 Change-Id: Ic9af7136bb336464b8cc023c7c1022a233f4e887 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771786 Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> Reviewed-by:
-
Jakob Gruber authored
Prior to this CL, collected shared function infos with identical source ranges were sorted non-deterministically during coverage collection. This lead to non-deterministically incorrectly-reported coverage due to an optimization which depended on the sort order later on. With this CL, we now sort shared function infos by the source range *and* call count. Bug: v8:6000,v8:9212 Change-Id: If8bf900727591e71dbd0df621e472a4303f3a353 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771776Reviewed-by:
Yang Guo <yangguo@chromium.org> Commit-Queue: Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#63411}
-
Leszek Swirski authored
Since we use Word32Equal with a (truncating) ReinterpretCast for pointer compressed TaggedEqual, we also have to allow it in the machine verifier. Change-Id: Ic16af837f03ebc51dde5bdc7f0c574b2aaf11909 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771784 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Auto-Submit: Leszek Swirski <leszeks@chromium.org> Cr-Commit-Position: refs/heads/master@{#63410}
-
Leszek Swirski authored
Using the tool again, the previous iteration accidentally ignored Node/TNode behind a typedef. Automatic replacement of types with manual cleanup/addition of CASTs where necessary. Bug: v8:9396 Change-Id: I33b6d229669cb80586d5d8e82c04542df671f0b9 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768367 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Michael Starzinger authored
This makes sure that all encodings of a table index are based upon a single implementation in the {TableIndexImmediate} class. It also fixes one encoding that wasn't extended to support u32v yet. R=ahaas@chromium.org TEST=unittests/WasmOpcodeLengthTest.VariableLength Change-Id: If24b6dc5e303d2d9e1e91cb2640c7c13eac40198 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768375Reviewed-by: -
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/e3f4298..3f22131 Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/1078fdd..7ad424d Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/31f187e..0e5fff1 TBR=machenbach@chromium.org,tmrts@chromium.org Change-Id: Idd7c0eed9e9e090d9d45448e0b6dfd52583f0928 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771344Reviewed-by:
-
- 26 Aug, 2019 16 commits
-
-
Z Nguyen-Huu authored
RegExpPrototypeDotAllGetter, RegExpPrototypeGlobalGetter, RegExpPrototypeIgnoreCaseGetter, RegExpPrototypeMultilineGetter, RegExpPrototypeStickyGetter, RegExpPrototypeUnicodeGetter. Bug: v8:8976 Change-Id: I2a5c19256cacc2438a6b40516565960f5c847205 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1764491 Commit-Queue: Z Nguyen-Huu <duongn@microsoft.com> Reviewed-by:
Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#63406}
-
Ng Zhi An authored
This off-by-1 error surfaces when the load/store opcodes take up 2 bytes, which is the case for v128.load and v128.store SIMD operations. Bug: v8:9015 Change-Id: Ife17375ed3450a95399b326bc6415dbc3ed3773b Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769480 Commit-Queue: Zhi An Ng <zhin@chromium.org> Reviewed-by:
-
Frank Tang authored
Bug: v8:9642 Change-Id: I2dcd1c0e3c208b15b5c0ec0f08880744134f7474 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769479Reviewed-by:
Jakob Kummerow <jkummerow@chromium.org> Commit-Queue: Frank Tang <ftang@chromium.org> Cr-Commit-Position: refs/heads/master@{#63404}
-
Toon Verwaest authored
By marking maps detached from the transition tree as prototypes, we'll automatically stop tracking transitions from those detached fast maps. That allows us to quickly check whether a map is detached (or the initial map anyway); and saves memory. We can use this information to ignore sibling type feedback when parsing a JSON array with many distinctly shaped json objects. Bug: chromium:993980 Change-Id: I86d493ac2cabec2c31c6e322ad5c5a7ace059dfc Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1771778Reviewed-by:
-
Leszek Swirski authored
For stores with Double feedback, StoreIC needs to check that the representation is still Double before doing the store, in case it accidentally tries to write to an object or worse, mutate a non-mutable HeapNumber. Bug: v8:9606 Bug: chromium:997485 Change-Id: I51e0953b40f752648c5e86b8644c23baf636367e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768373 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Zhou, Zhiguo authored
This CL adds a new class WasmModuleSourceMap for source map support of WasmModule, which maps C/C++ source code to WASM bytecode. Via this support, V8 can build a direct map of source code and JITted code and inform profilers of it, thus the source-code-level profiling information is presented. Change-Id: I346f6216809ce4f3bf8b27f1e839dd4efdb00ead Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1708029Reviewed-by:
-
Jakob Kummerow authored
Host info used to be stored on the global reference underlying a Ref; now it is stored in a JSWeakMap and hence tied to the lifetime of the actual object on V8's heap. Additionally, the internal metadata needed for C-API functions is now stored on the SharedFunctionInfo and no longer overlaps with the host info mechanism. Bonus content: Roll 6db391e: Remove a few more leftover uses of _enum types Change-Id: Ibb1fa4b0dd5157fef15c030bac705a11aa3beaea Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768368Reviewed-by:
-
Yang Guo authored
R=sigurds@chromium.org Bug: chromium:997469 Change-Id: I83c8a50a5626b3e4679ff7977474d495cdbf7e90 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768369 Commit-Queue: Yang Guo <yangguo@chromium.org> Reviewed-by:
Sigurd Schneider <sigurds@chromium.org> Cr-Commit-Position: refs/heads/master@{#63399}
-
Leszek Swirski authored
Forbid using LoadRoot in CSA (with a bailout via CodeAssembler), so that users are forced to use helper macros for roots, which have statically known types. Convert all current uses of LoadRoot to use these macros, introducing new ones where necessary. Bug: v8:9396 Change-Id: I91214fca6e5ace7554d79605706a8a60117468fa Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1762526 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Sigurd Schneider authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/8528a36..e3f4298 Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/2662d14..1078fdd Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/940c282..31f187e Rolling v8/third_party/googletest/src: https://chromium.googlesource.com/external/github.com/google/googletest/+log/fb49e6c..ed2eef6 Rolling v8/third_party/icu: https://chromium.googlesource.com/chromium/deps/icu/+log/2b2ee71..952ccb9 Rolling v8/tools/clang: https://chromium.googlesource.com/chromium/src/tools/clang/+log/656fd11..2ddb2b2 This required porting fuchsia related changes from https://chromium.googlesource.com/chromium/src/+/9c4748c8106196fa5f8a13cbd958fbf9e9aadddd over to this CL. TBR=machenbach@chromium.org,tmrts@chromium.org Change-Id: I2b3b81151c3023d67e970170191e6d5078970a7e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768795Reviewed-by:
-
Dominik Inführ authored
Introduce list of invalidated objects for old-to-new slots. Objects are registered as invalidated in NotifyObjectLayoutChange, however no slots are filtered right now. Slots are still deleted, so all recorded slots are valid. Always treat old-to-new slots in free space as valid. With old-to-new slots, invalid ones are still cleared. Bug: v8:9454 Change-Id: I18c65fdaccdb651f4124d36861105ce4e8af0fad Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768357Reviewed-by:
-
Tobias Tebbi authored
Changes in the reland: Rebased and added a check that JavaScript-linkage builtins use JSAny in parameters and return type, plus the necessary cleanups for this test to pass. Design Doc: https://docs.google.com/document/d/1z6j0pWHnNIfId0v00uWN2HBrGRDJxJfYuCr5K7Kr1xA This reverts commit 4418a7b9. Original change's description: > Revert "[torque] introduce JSAny type for user-accessible JavaScript values" > > This reverts commit 79b00555. > > Reason for revert: needs more discussion > > Original change's description: > > [torque] introduce JSAny type for user-accessible JavaScript values > > > > This CL introduces a JSAny type for user-exposed JavaScript values and > > a few new types to define it. Especially, it splits Symbol into > > PrivateSymbol (not exposed) and PublicSymbol (JavaScript exposed > > symbols). > > > > The change is mostly mechanical, but a few things are interesting: > > - PropertyKey and JSPrimitive were designed to coincide with the spec > > notions of IsPropertyKey() and primitive value, respectively. > > - Since Name is an open type, we define AnyName to be the known > > subtypes of Name. This is not too elegant, but by using AnyName > > instead of Name, typeswitch can properly conclude something if a > > subtype of Name is excluded. > > > > Small drive-by changes, which were necessary: > > - Allow subtyping on label parameters. > > - Fix the formatting of typeswitch, it was broken with union types > > in case types. > > > > Bug: v8:7793 > > Change-Id: I14b10507f8cf316ad85e048fe8d53d1df5e0bb13 > > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1735322 > > Commit-Queue: Tobias Tebbi <tebbi@chromium.org> > > Reviewed-by: Jakob Gruber <jgruber@chromium.org> > > Cr-Commit-Position: refs/heads/master@{#63114} > > TBR=neis@chromium.org,jgruber@chromium.org,tebbi@chromium.org > > Change-Id: Ifde7881d74afe407628f40047997339d54cb2424 > No-Presubmit: true > No-Tree-Checks: true > No-Try: true > Bug: v8:7793 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1741652 > Reviewed-by: Tobias Tebbi <tebbi@chromium.org> > Commit-Queue: Tobias Tebbi <tebbi@chromium.org> > Cr-Commit-Position: refs/heads/master@{#63115} TBR=neis@chromium.org,jgruber@chromium.org,tebbi@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: v8:7793 Change-Id: Icca34e3824f55009b984d9348fd21884400f0081 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769316 Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
-
Tobias Tebbi authored
The generation of TFJ declarations for non-vararg JavaScript-linkage builtins was very broken. Now it works as it should: the explicit parameters correspond to normal JavaScript parameters. Bug: v8:7793 Change-Id: I9fa0bdd45f8c4b678d4f0436af3e9b53cba69475 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769317 Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
-
Leszek Swirski authored
Wrap the obj and method registers in BuildGetIterator in a register allocation scope, so that they don't get materialised before the JumpIfJSReceiver jump if they don't have to. Bug: v8:9649 Change-Id: I8dfdd06a23c396124c495b5cb83c078080f1a7c9 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768583 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Auto-Submit: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
-
Andreas Haas authored
As described in https://docs.google.com/document/d/1hB8mpWmzmtaxZ8PuJEkAWLwFqXTjrw7mJ3Ge9W1dB4E, we want to enable --wasm-staging for libfuzzer fuzzers to increase test coverage of new feature implementations. As discussed in https://docs.google.com/a/chromium.org/document/d/1hB8mpWmzmtaxZ8PuJEkAWLwFqXTjrw7mJ3Ge9W1dB4E/edit?disco=AAAADPrjhdA, this has to be done in our implementations manually. R=thibaudm@chromium.org Bug: v8:9601 Change-Id: Iab35af75bb71b6fca697ab7e4ac37b04e1d6da35 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768581Reviewed-by:
Thibaud Michaud <thibaudm@chromium.org> Commit-Queue: Andreas Haas <ahaas@chromium.org> Cr-Commit-Position: refs/heads/master@{#63392}
-
Maya Lekova authored
This is a reland of 8ff8ddba Original change's description: > [turbofan] Further brokerize BytecodeGraphBuilder > > Replace all usages of VectorSlotPair with FeedbackSource. > > Bug: v8:7790 > Change-Id: I0ac6e9cd8f5730154cc1842e267ca1ebfdebc874 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1763536 > Commit-Queue: Maya Lekova <mslekova@chromium.org> > Reviewed-by: Georg Neis <neis@chromium.org> > Cr-Commit-Position: refs/heads/master@{#63378} Bug: v8:7790 Change-Id: I54fbf4433a65681780fda4ded09ec2d145d36ec1 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1768361Reviewed-by:
-
- 25 Aug, 2019 1 commit
-
-
Dominik Inführ authored
This reverts commit 604b0e1e. Reason for revert: Clusterfuzz found an issue. Original change's description: > [heap] Introduce old-to-new invalidation set > > Introduce list of invalidated objects for old-to-new slots. Objects > are registered as invalidated in NotifyObjectLayoutChange, however > no slots are filtered right now. Slots are still deleted, so all > recorded slots are valid. > > Bug: v8:9454 > Change-Id: Ic0ea15283c4075f4051fae6a5b148721265339f7 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1765528 > Commit-Queue: Dominik Inführ <dinfuehr@chromium.org> > Reviewed-by: Ulan Degenbaev <ulan@chromium.org> > Cr-Commit-Position: refs/heads/master@{#63367} TBR=ulan@chromium.org,dinfuehr@chromium.org # Not skipping CQ checks because original CL landed > 1 day ago. Bug: v8:9454 Change-Id: Ic898db38f297824aa54744123f85cd75df957159 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1770676Reviewed-by:
-
- 24 Aug, 2019 1 commit
-
-
Joey Gouly authored
ChangeTaggedSignedToInt32(ChangeCompressedSignedToTaggedSigned((x)) -> ChangeCompressedSignedToInt32(x) This pattern shows up in the Octane Richards benchmark (on arm64): sxtw x11, w10 asr w11, w11, #1 This patch will remove the sxtw. ChangeCompressedSignedToInt32(CheckedInt32ToCompressedSigned(x)) -> x This pattern shows up in the Octane Richards benchmark (on arm64): adds w10, w10, w10 b.vs #+0x1118 asr w11, w10, #1 stur w10, [x6, #19] cmp w11, #0x1a This patch will remove the asr, and produce: adds w11, w10, w10 b.vs #+0x1108 stur w11, [x6, #19] cmp w10, #0x1a Bug: v8:7703 Change-Id: I5843e0a4f723b202857ee86130f835cd048d7e31 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1763529Reviewed-by:Santiago Aboy Solanes <solanes@chromium.org> Commit-Queue: Rodolph Perfetta <rodolph.perfetta@arm.com> Cr-Commit-Position: refs/heads/master@{#63389}
-
- 23 Aug, 2019 5 commits
-
-
Shu-yu Guo authored
- Rename FunctionLiteral::FunctionType to FunctionSyntaxKind. - Re-express IsWrappedBit, IsDeclarationBit, IsAnonymousExpressionBit, and IsNamedExpressionBit in SFI::flags as FunctionSyntaxKind. This frees up 1 bit in SFI::flags. - Re-express the analogous bits in ParseInfo as FunctionSyntaxKind. - Simplifies some logic in the back-and-forth passing of this info between SFI and ParseInfo. - Drive-by fix parsing class member initializations as kAccessorOrMethod. Bug: v8:9644 Change-Id: I6c165d5016d968f5057a32136385ddcdc4a46ef1 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1767263Reviewed-by:
Adam Klein <adamk@chromium.org> Commit-Queue: Shu-yu Guo <syg@chromium.org> Cr-Commit-Position: refs/heads/master@{#63388}
-
Bill Budge authored
This reverts commit 8ff8ddba. Reason for revert: Causes failures on GC Stress and TSan: https://ci.chromium.org/p/v8/builders/ci/V8%20Linux%20-%20gc%20stress/24138 https://ci.chromium.org/p/v8/builders/ci/V8%20Linux64%20TSAN/27969 Original change's description: > [turbofan] Further brokerize BytecodeGraphBuilder > > Replace all usages of VectorSlotPair with FeedbackSource. > > Bug: v8:7790 > Change-Id: I0ac6e9cd8f5730154cc1842e267ca1ebfdebc874 > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1763536 > Commit-Queue: Maya Lekova <mslekova@chromium.org> > Reviewed-by: Georg Neis <neis@chromium.org> > Cr-Commit-Position: refs/heads/master@{#63378} TBR=neis@chromium.org,mslekova@chromium.org Change-Id: Ieb6fb24dd0626fd9eb2183844614ac5b3c8084e8 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:7790 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769287Reviewed-by:
Bill Budge <bbudge@chromium.org> Commit-Queue: Bill Budge <bbudge@chromium.org> Cr-Commit-Position: refs/heads/master@{#63387}
-
Joshua Litt authored
This reverts commit 9460101c. Reason for revert: Causes confusion on Blink side, as it introduces an object with >=2 internal fields that is not a wrapper (see bug). Bug: chromium:996681 Change-Id: I275b5a064a4ee8c73c05f97be322924a3bc5370e Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1769148Reviewed-by:
-
Georg Schmid authored
Even when a field is marked const, we may emit multiple consecutive in-literal stores to that field. That is, in 'JSNativeContextSpecialization::BuildPropertyStore', when the access mode is 'kStoreInLiteral' and we are accessing a const field, we may produce a StoreField node, even though another StoreField (that stores something other than 'Uninitialized') to the same const field dominates it. This appears to be sound, since earlier stores to literals cannot be observed anyways. Unfortunately this behavior conflicts with the double const store invariant in load elimination: Roughly speaking, we assume that load elimination may never observe two consecutive const stores to the same field on the same object. The apparent solution would be to treat 'kStoreInLiteral' accesses like regular 'kStore' accesses: For consecutive stores to const properties we don't emit StoreField, but instead emit code that checks whether the value about to be written is equivalent to the previously written one, and otherwise deopt ('DeoptimizeReason::kWrongValue'). Unfortunately this turns out impractical, since for 'kStoreInLiteral' accesses we can't easily decide whether we're dealing with the first such store or one of the consecutive ones. Also see this abandoned CL: https://chromium-review.googlesource.com/c/v8/v8/+/1762020. This CL instead adds an exception to the invariant in load elimination. We track whether a store arose from a 'kStoreInLiteral' access, and use this information when visiting StoreField nodes in load elimination. R=neis@chromium.org, tebbi@chromium.org Bug: chromium:987205 Change-Id: I8829752aa0637e9599677d20aad2d706d40d7fe6 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1763535Reviewed-by: -
Ng Zhi An authored
Bug: v8:8460 Change-Id: If7fd1a497b2a3b74b921e175491233be884d574b Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1749713Reviewed-by:
-
