- 28 Sep, 2017 6 commits
-
-
Michael Starzinger authored
This is a reland of 9d3c4b4b Original change's description: > [turbofan] Implement lowering of {JSCreateClosure}. > > This adds support for inline allocation of {JSFunction} objects as part > of closures instantiation for {JSCreateClosure} nodes. The lowering is > limited to instantiation sites which have already seen more than one > previous instantiation, this avoids the need to increment the respective > counter. > > R=jarin@chromium.org > > Change-Id: I462c557453fe58bc5f09020a3d5ebdf11c2ea68b > Reviewed-on: https://chromium-review.googlesource.com/594287 > Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> > Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> > Cr-Commit-Position: refs/heads/master@{#48176} Change-Id: I3ec3880bea89798a34a3878e6122b95db1014151 Reviewed-on: https://chromium-review.googlesource.com/686834Reviewed-by:
Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48198}
-
Peter Marshall authored
The members of HandleScopeImplementer are copied with memcpy when the isolate is transferred to another thread. List contained some primitives which allowed us to manually free the backing store, which was needed in order to ensure that threads would not hold on to old pointers and use them later. With std::vector, we can't do that. Here we change the HandleScopeImplementer to instead use a custom structure DetachableVector, which contains a std::vector but allows manual detaching and freeing of the backing store. This allows us to maintain the old behavior. Bug: v8:6333 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: I6361d161cdb19878ba19ed51d6ba2fae99e8cdc0 Reviewed-on: https://chromium-review.googlesource.com/660125Reviewed-by:
Yang Guo <yangguo@chromium.org> Commit-Queue: Peter Marshall <petermarshall@chromium.org> Cr-Commit-Position: refs/heads/master@{#48197}
-
Michael Achenbach authored
Bug: v8:6857 Change-Id: Icab007681753c6f8143147ea8b2c1b06bf325d92 Reviewed-on: https://chromium-review.googlesource.com/686900 Commit-Queue: Michael Achenbach <machenbach@chromium.org> Reviewed-by:
Sergiy Byelozyorov <sergiyb@chromium.org> Cr-Commit-Position: refs/heads/master@{#48196}
-
Toon Verwaest authored
Bug: Change-Id: I6894c097a994527b5e749527966c5de6c42d964b Reviewed-on: https://chromium-review.googlesource.com/668405 Commit-Queue: Toon Verwaest <verwaest@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Cr-Commit-Position: refs/heads/master@{#48195}
-
Yang Guo authored
Both flags are obsolete. R=cbruni@chromium.org Change-Id: I78f5980f2d0bd569b8e5f6e29539cf87293cab94 Reviewed-on: https://chromium-review.googlesource.com/684296Reviewed-by:
Camillo Bruni <cbruni@chromium.org> Commit-Queue: Yang Guo <yangguo@chromium.org> Cr-Commit-Position: refs/heads/master@{#48194}
-
Jakob Kummerow authored
- The fallback path for "digit_mul" was missing shifts - RightTrim() could end up reading digit(-1) (i.e. the bit field) - Add SLOW_DCHECKs to digit()/set_digit() to catch any further invalid-index issues Bug: v8:6791 Change-Id: I901edfce90bb6ac11e74c3c3f577225f12bad177 Reviewed-on: https://chromium-review.googlesource.com/683796Reviewed-by:
Daniel Ehrenberg <littledan@chromium.org> Commit-Queue: Jakob Kummerow <jkummerow@chromium.org> Cr-Commit-Position: refs/heads/master@{#48193}
-
- 27 Sep, 2017 18 commits
-
-
Clemens Hammacher authored
Since code objects cannot grow larger than 2GB anyway, it's enough to store the instruction offset and landing pad offset as 32-bit values. This reduces the size of the ProtectedInstructionData struct by 50%. R=eholk@chromium.org Bug: v8:5277 Change-Id: I4d2e0dc76b8a853fb50d51d70d5ec4038ee594ac Reviewed-on: https://chromium-review.googlesource.com/686757 Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Reviewed-by:
Eric Holk <eholk@chromium.org> Cr-Commit-Position: refs/heads/master@{#48192}
-
Eric Holk (eholk) authored
This is the first of a series of refactoring CLs to make way for Windows trap handling support. See https://chromium-review.googlesource.com/c/v8/v8/+/626558 as well. Bug: Change-Id: I5fe9ef9c1cec58a81e51fcffbbe4419e0e298ab7 Reviewed-on: https://chromium-review.googlesource.com/644104Reviewed-by:
Brad Nelson <bradnelson@chromium.org> Commit-Queue: Eric Holk <eholk@chromium.org> Cr-Commit-Position: refs/heads/master@{#48191}
-
Paul Irish authored
Bug: NONE Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel Change-Id: Iaf820f8235ea85c21fb8e904c132c8c2039bf941 Reviewed-on: https://chromium-review.googlesource.com/653237Reviewed-by:
Aleksey Kozyatinskiy <kozyatinskiy@chromium.org> Reviewed-by:
Pavel Feldman <pfeldman@chromium.org> Commit-Queue: Pavel Feldman <pfeldman@chromium.org> Cr-Commit-Position: refs/heads/master@{#48190}
-
Georgia Kouveli authored
Bug: v8:6644 Change-Id: I601b78ba681925521005acabed1ee6e91df98ed0 Reviewed-on: https://chromium-review.googlesource.com/686762Reviewed-by:
Benedikt Meurer <bmeurer@chromium.org> Commit-Queue: Georgia Kouveli <georgia.kouveli@arm.com> Cr-Commit-Position: refs/heads/master@{#48189}
-
Andreas Haas authored
Async compilation causes problem when the isolate shuts down concurrently. R=hablich@chromium.org Bug: chromium:764313 Change-Id: I35b5ba9bdb70291b53e87a5323f5408930f01ddd Reviewed-on: https://chromium-review.googlesource.com/687076Reviewed-by:
Clemens Hammacher <clemensh@chromium.org> Commit-Queue: Andreas Haas <ahaas@chromium.org> Cr-Commit-Position: refs/heads/master@{#48188}
-
Clemens Hammacher authored
By reusing a single vector instead of allocating one per code object, we can save lots of heap allocations. R=eholk@chromium.org Bug: v8:5277 Change-Id: Ia462c97293cd00607d9f2faf29e265ea78d49394 Reviewed-on: https://chromium-review.googlesource.com/686819Reviewed-by:
-
Eric Holk authored
This CL includes validation code for the trap handler data structures in debug mode to help catch issues like v8:6841 sooner in the future. We also now eagerly initialize the free list pointers to make the logic of finding the next free entry more obvious. Bug: v8:5277 Change-Id: I13c3180c59b6152508c480e2042072a91e6ca977 Reviewed-on: https://chromium-review.googlesource.com/674128 Commit-Queue: Eric Holk <eholk@chromium.org> Reviewed-by:
Mircea Trofin <mtrofin@chromium.org> Cr-Commit-Position: refs/heads/master@{#48186}
-
Sergiy Byelozyorov authored
TBR=machenbach@chromium.org Bug: chromium:747960 No-Try: true No-Presubmit: true No-Tree-Checks: true Change-Id: I06af6e09c76a8006717d10a669812840c4f58b24 Reviewed-on: https://chromium-review.googlesource.com/687498 Commit-Queue: Sergiy Byelozyorov <sergiyb@chromium.org> Reviewed-by:
-
Yang Guo authored
We used to only mark top-level SFIs with the 'deserialized' bit. Now we do it for every SFI that has cached code. This is the first step to surface caching information in the future. R=cbruni@chromium.org Bug: chromium:769166 Change-Id: I12f21511419ce54fd07a2cc277a65866660c366a Reviewed-on: https://chromium-review.googlesource.com/686715Reviewed-by:
-
Ben L. Titzer authored
These functions clearly belong closer to respective classes in wasm-objects.h. R=clemensh@chromium.org Bug: Change-Id: I46117b15181d33b237f4cdf318ce66ddb18d96bd Reviewed-on: https://chromium-review.googlesource.com/684594Reviewed-by:
-
Michael Achenbach authored
TBR=sergiyb@chromium.org Bug: chromium:747960 No-Try: true No-Presubmit: true No-Tree-Checks: true Change-Id: I4acaa2b825585ba41b70de58bed727b558c3b1ac Reviewed-on: https://chromium-review.googlesource.com/687215 Commit-Queue: Michael Achenbach <machenbach@chromium.org> Reviewed-by:
Michael Achenbach <machenbach@chromium.org> Cr-Commit-Position: refs/heads/master@{#48182}
-
Clemens Hammacher authored
There was an issue with passing float32 parameters, if the value was spilled on the stack and passed as stack parameter. First, we sometimes reduced the stack pointer by 8 bytes instead of 4, and second, there was a mismatch between movsd and movss. R=titzer@chromium.org Bug: chromium:718858 Change-Id: Ia884df369ddd95adeff3733f9715f589996f0b65 Also-By: ahaas@chromium.org Reviewed-on: https://chromium-review.googlesource.com/684738Reviewed-by:
Andreas Haas <ahaas@chromium.org> Reviewed-by:
Ben Titzer <titzer@chromium.org> Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#48181}
-
Michael Starzinger authored
R=clemensh@chromium.org Change-Id: I03fb9ae37a7950fdb8a819383b114dfc2c13d618 Reviewed-on: https://chromium-review.googlesource.com/686914Reviewed-by:
-
Sergiy Byelozyorov authored
TBR=machenbach@chromium.org Bug: 747960 No-Try: true No-Presubmit: true No-Tree-Checks: true Change-Id: I0cab4da0bb6070e43e636fe9b5f2991f31b93d35 Reviewed-on: https://chromium-review.googlesource.com/686815 Commit-Queue: Sergiy Byelozyorov <sergiyb@chromium.org> Reviewed-by:
-
Michael Starzinger authored
This reverts commit 9d3c4b4b. Reason for revert: Breaks cctest/test-debug/NoBreakWhenBootstrapping in no-snap mode. Original change's description: > [turbofan] Implement lowering of {JSCreateClosure}. > > This adds support for inline allocation of {JSFunction} objects as part > of closures instantiation for {JSCreateClosure} nodes. The lowering is > limited to instantiation sites which have already seen more than one > previous instantiation, this avoids the need to increment the respective > counter. > > R=jarin@chromium.org > > Change-Id: I462c557453fe58bc5f09020a3d5ebdf11c2ea68b > Reviewed-on: https://chromium-review.googlesource.com/594287 > Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> > Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> > Cr-Commit-Position: refs/heads/master@{#48176} TBR=mstarzinger@chromium.org,jarin@chromium.org Change-Id: Id52281f6a3c0b7c2603053ecf002777d5b0d6f1f No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/686534Reviewed-by:
Michael Starzinger <mstarzinger@chromium.org> Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#48178}
-
Martyn Capewell authored
Abstract some stack slot copies through a macro assembler function. This eliminates some non-paired stack operations. Bug: v8:6644 Change-Id: Icaa3ebb1bec42587ad461bd548a4225c7b4e5d15 Reviewed-on: https://chromium-review.googlesource.com/685238Reviewed-by:
-
Michael Starzinger authored
This adds support for inline allocation of {JSFunction} objects as part of closures instantiation for {JSCreateClosure} nodes. The lowering is limited to instantiation sites which have already seen more than one previous instantiation, this avoids the need to increment the respective counter. R=jarin@chromium.org Change-Id: I462c557453fe58bc5f09020a3d5ebdf11c2ea68b Reviewed-on: https://chromium-review.googlesource.com/594287 Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Reviewed-by: -
Benedikt Meurer authored
Bug: chromium:768367, v8:6819, v8:6820, v8:6831 Change-Id: I90538217f794c91a83ae5cfb12e0d0347d5f8574 Reviewed-on: https://chromium-review.googlesource.com/685240Reviewed-by:
-
- 26 Sep, 2017 16 commits
-
-
Jakob Kummerow authored
- Move things to conversions.cc that don't need to be in headers - Turn InternalStringToInt into a subclassable helper class so we can re-use it for BigInt.parseInt - Bonus: play a round of IWYU with all the .cc files who thought that #including conversions-inl.h would give them nice Unicode things Bug: v8:6791 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: I64022543a9b83002e2b78416c7e87b40a1a016e6 Reviewed-on: https://chromium-review.googlesource.com/673725 Commit-Queue: Jakob Kummerow <jkummerow@chromium.org> Reviewed-by:
Marja Hölttä <marja@chromium.org> Reviewed-by:
Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#48174}
-
Jakob Kummerow authored
This CL teaches the respective bytecode handlers and standalone stubs about BigInts, and collects "kBigInt" feedback for them. However, Turbofan does not yet care about such feedback, so it is simply converted to "any" for now (making TF emit stub calls for BigInt operations). Bug: v8:6791 Change-Id: I6440c108ccd79058d77adc2a6041251db9d5f81d Reviewed-on: https://chromium-review.googlesource.com/683758Reviewed-by:
-
Jakob Kummerow authored
This is a reland of r48152 / 2f88c9b2, originally reviewed on https://chromium-review.googlesource.com/678037, with a small fix for Clang on Windows. TBR=littledan@chromium.org Bug: v8:6791 Change-Id: I70bc950f82682f40486540d2ac6e10540888d663 Reviewed-on: https://chromium-review.googlesource.com/685255Reviewed-by:
Jakob Kummerow <jkummerow@chromium.org> Commit-Queue: Jakob Kummerow <jkummerow@chromium.org> Cr-Commit-Position: refs/heads/master@{#48172}
-
Michael Achenbach authored
This reverts commit e0b76c9a. Reason for revert: Changes layout tests: https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/18522 See also: https://github.com/v8/v8/wiki/Blink-layout-tests Original change's description: > Add fast path to ObjectGetOwnPropertyDescriptor > > Bug: v8:6557 > Change-Id: I384e9f36058c73d40be1faf5ae1bf8c01d068f0e > Reviewed-on: https://chromium-review.googlesource.com/682059 > Commit-Queue: Maya Lekova <mslekova@google.com> > Reviewed-by: Igor Sheludko <ishell@chromium.org> > Cr-Commit-Position: refs/heads/master@{#48166} TBR=ishell@chromium.org,mslekova@google.com Change-Id: I27aa52786f79dd617faea6336df684821c1720d2 No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: v8:6557 Reviewed-on: https://chromium-review.googlesource.com/685314Reviewed-by:
-
Clemens Hammacher authored
With --trace-wasm-decoder, we were printing the local variable index for values on the stack generated by get_local instructions. This CL adds the same feature for globals. R=titzer@chromium.org Change-Id: Ie113ebcc2d1abcd67df2d01bf4bdb452635732c8 Reviewed-on: https://chromium-review.googlesource.com/684737Reviewed-by:
-
Michael Starzinger authored
This fixes the signature of "Math.abs" from "(signed) -> signed" to "(signed) -> unsigned" and hence fixes cases where the absolute value would overflow the range of signed 32-bit values. This is in sync with spec erratas (and ECMAScript semantics). Note that this also switches the underlying implementation of the above absolute value function to a branch-free version. R=clemensh@chromium.org TEST=mjsunit/regress/regress-6838-3 BUG=v8:6838 Change-Id: Ib13b7ecd336ae386cbde7c574e727bf52f841e00 Reviewed-on: https://chromium-review.googlesource.com/684181 Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Reviewed-by:
-
Andreas Haas authored
With this CL we use the same optimizations for the trap handler ool code which we already use for trap-if. * Call a builtin instead of calling the runtime directly. * Use one call per ool code instead of a source position parameter. * Do not pass the trap reason as parameter. R=titzer@chromium.org, eholk@chromium.org Change-Id: Ieef6da96f340269c3e91efd21ac24e61a42193f4 Reviewed-on: https://chromium-review.googlesource.com/684436Reviewed-by:
-
Tobias Tebbi authored
Bug: Change-Id: I52e757aa2de951ff40660545472321c7dec84241 Reviewed-on: https://chromium-review.googlesource.com/632156Reviewed-by:
Daniel Clifford <danno@chromium.org> Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Cr-Commit-Position: refs/heads/master@{#48167}
-
Maya Lekova authored
Bug: v8:6557 Change-Id: I384e9f36058c73d40be1faf5ae1bf8c01d068f0e Reviewed-on: https://chromium-review.googlesource.com/682059 Commit-Queue: Maya Lekova <mslekova@google.com> Reviewed-by:
-
Marja Hölttä authored
Follow up to https://chromium-review.googlesource.com/671020 We still didn't return the correct amount of invalid characters, according to the Encoding spec ( https://encoding.spec.whatwg.org/#utf-8-decoder ), when we saw a byte sequence which was as start of an overlong / invalid sequence, but there weren't enough continuation bytes. A more rigorous test will follow in https://chromium-review.googlesource.com/c/v8/v8/+/681217 BUG=chromium:765608 Change-Id: I535670edc14d3bae144e5a9ca373f12eec78a934 Reviewed-on: https://chromium-review.googlesource.com/681674 Commit-Queue: Marja Hölttä <marja@chromium.org> Reviewed-by:
Daniel Vogelheim <vogelheim@chromium.org> Reviewed-by:
-
Yang Guo authored
This is a reland of 629406d1 Original change's description: > [snapshot] include version string in the startup snapshot. > > This is to easier diagnose build issues involving the snapshot. > Sample error message for mismatching snapshot: > > # > # Fatal error in ../../src/snapshot/snapshot-common.cc, line 286 > # Version mismatch between V8 binary and snapshot. > # V8 binary version: 6.3.1 (candidate) > # Snapshot version: 6.3.0 (candidate) > # The snapshot consists of 2820444 bytes and contains 1 contexts. > # > > > R=machenbach@chromium.org > > Bug: chromium:764327 > Change-Id: Icdc7aeac77819b113985b424feda814a072d5406 > Reviewed-on: https://chromium-review.googlesource.com/684295 > Reviewed-by: Michael Achenbach <machenbach@chromium.org> > Commit-Queue: Yang Guo <yangguo@chromium.org> > Cr-Commit-Position: refs/heads/master@{#48161} Bug: chromium:764327 Change-Id: I3721689824e0a6909eede86d0829dc258ae40c4d Reviewed-on: https://chromium-review.googlesource.com/684494Reviewed-by:
-
Tobias Tebbi authored
On big endian 64 bit architectures, kHashFieldOffset is not word-aligned. This breaks the assumption in escape analysis that all fields are word-aligned. Fix this by not dematerializing such objects. Alternative fix for https://chromium-review.googlesource.com/c/v8/v8/+/681335 Change-Id: I7d8e4c7934d9306cc06a614ae110e7cf7235394f Reviewed-on: https://chromium-review.googlesource.com/681714 Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
-
Yang Guo authored
This reverts commit 629406d1. Reason for revert: cross platform builds break. Original change's description: > [snapshot] include version string in the startup snapshot. > > This is to easier diagnose build issues involving the snapshot. > Sample error message for mismatching snapshot: > > # > # Fatal error in ../../src/snapshot/snapshot-common.cc, line 286 > # Version mismatch between V8 binary and snapshot. > # V8 binary version: 6.3.1 (candidate) > # Snapshot version: 6.3.0 (candidate) > # The snapshot consists of 2820444 bytes and contains 1 contexts. > # > > > R=machenbach@chromium.org > > Bug: chromium:764327 > Change-Id: Icdc7aeac77819b113985b424feda814a072d5406 > Reviewed-on: https://chromium-review.googlesource.com/684295 > Reviewed-by: Michael Achenbach <machenbach@chromium.org> > Commit-Queue: Yang Guo <yangguo@chromium.org> > Cr-Commit-Position: refs/heads/master@{#48161} TBR=machenbach@chromium.org,yangguo@chromium.org Change-Id: I35a9b575e4f7fe5c45c9dc6f9e774c3e6d30049c No-Presubmit: true No-Tree-Checks: true No-Try: true Bug: chromium:764327 Reviewed-on: https://chromium-review.googlesource.com/684315Reviewed-by:
-
Yang Guo authored
This is to easier diagnose build issues involving the snapshot. Sample error message for mismatching snapshot: # # Fatal error in ../../src/snapshot/snapshot-common.cc, line 286 # Version mismatch between V8 binary and snapshot. # V8 binary version: 6.3.1 (candidate) # Snapshot version: 6.3.0 (candidate) # The snapshot consists of 2820444 bytes and contains 1 contexts. # R=machenbach@chromium.org Bug: chromium:764327 Change-Id: Icdc7aeac77819b113985b424feda814a072d5406 Reviewed-on: https://chromium-review.googlesource.com/684295Reviewed-by:
-
Yang Guo authored
- Add kProduceExhaustiveCodeCache to v8::ScriptCompiler::CompileOptions to request eager compilation to add as much as possible to the code cache for the script. - Repurpose ParseInfo::kLazy flag. - Remove ParseInfo::kDebug flag. - Remove --serialize-toplevel as it has become obsolete. R=marja@chromium.org Bug: chromium:768705 Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng Change-Id: Ife14f7a1d1c02e525f0b9dbfd2452013d67c7167 Reviewed-on: https://chromium-review.googlesource.com/684019 Commit-Queue: Yang Guo <yangguo@chromium.org> Reviewed-by:
-
Andreas Haas authored
NOTRY=true R=clemensh@chromium.org Change-Id: I1d4d2eee771103a91f0b4878056b8cd72e06337c Reviewed-on: https://chromium-review.googlesource.com/684077Reviewed-by:
-
