- 11 Jun, 2018 25 commits
-
-
Dominik Inführ authored
Bug: chromium:844008 Change-Id: If248ca90213c5cb3dfb0e3936a9ac824ed3c3119 Reviewed-on: https://chromium-review.googlesource.com/1095276Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Commit-Queue: Dominik Inführ <dinfuehr@google.com> Cr-Commit-Position: refs/heads/master@{#53646}
-
Clemens Hammacher authored
The current implementation allocates a new {NativeModule} if trap handlers are disabled. Afterwards, all export wrappers need to be patched to call to the code objects in this new native module (this patching is done in {CodeSpecialization}). With the jump table, we want to get rid of this patching, hence we need to reuse the same jump table, hence we cannot allocate a new {NativeModule}. Instead, we should update the existing one with the new code. R=mstarzinger@chromium.org CC=eholk@chromium.org Bug: v8:7143, v8:7758 Change-Id: If2f395d462752b9084ed1e5f0a81f71d400ccfca Reviewed-on: https://chromium-review.googlesource.com/1095262Reviewed-by:Eric Holk <eholk@chromium.org> Reviewed-by:
Michael Starzinger <mstarzinger@chromium.org> Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#53645}
-
Michael Starzinger authored
This switches Liftoff code to use WebAssembly runtime stubs instead of builtin calls for handling traps, similar to what we use in TurboFan generated code as well. R=clemensh@chromium.org BUG=v8:7424 Change-Id: If2554067b3e294220306c67861bb1fb14db7b492 Reviewed-on: https://chromium-review.googlesource.com/1087275 Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Reviewed-by:
Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#53644}
-
Dominik Inführ authored
Now that V8 uses EphemeronHashTable as backing store for JSWeakCollections one pass over the heap is enough. In the old implementation a second pass was necessary to find the owner of the FixedArray to determine if its entries are strong or weak references. Bug: chromium:844008 Change-Id: I04bdf7d480c9be301831698571be5b226d20ac20 Reviewed-on: https://chromium-review.googlesource.com/1090910Reviewed-by:
Alexei Filippov <alph@chromium.org> Commit-Queue: Dominik Inführ <dinfuehr@google.com> Cr-Commit-Position: refs/heads/master@{#53643}
-
Ivica Bogosavljevic authored
Fix afd6351d After `[csa][builtins] Port CompileLazy* to CSA` landed, there is an implicit expectation that the address of the JS code is stored in the target register. Due to optimizations on MIPSR6, this address was not properly loaded. We fix this by loading target address explicitly before performing the jump. TEST=mjsunit/json2,mjsunit/packed-elements Change-Id: Ida9cb062910415bf15aa40dd20a17fc42fdba719 Reviewed-on: https://chromium-review.googlesource.com/1095155Reviewed-by:
Sreten Kovacevic <sreten.kovacevic@mips.com> Commit-Queue: Sreten Kovacevic <sreten.kovacevic@mips.com> Cr-Commit-Position: refs/heads/master@{#53642}
-
Dominik Inführ authored
Mark JSWeakCollections and EphemeronHashTables concurrently to the application to reduce pause times. Bug: chromium:844008 Change-Id: I0a8fe592e651b368c5c12846d02d55bbf4058ff2 Reviewed-on: https://chromium-review.googlesource.com/1095035Reviewed-by:
-
Michael Achenbach authored
This adds a new gn arg v8_custom_deps to add a dependency that's dynamically added under v8/custom_deps. E.g. for depending on a target in v8/custom_deps/test_root/BUILD.gn set v8_custom_deps = "custom_deps/test_root:target_name". Bug: chromium:846711 Change-Id: If03d48dba07cb36dc1b981602e5dc267deec086b Reviewed-on: https://chromium-review.googlesource.com/1095218Reviewed-by:
Sergiy Byelozyorov <sergiyb@chromium.org> Commit-Queue: Michael Achenbach <machenbach@chromium.org> Cr-Commit-Position: refs/heads/master@{#53640}
-
Michael Starzinger authored
R=sigurds@chromium.org Change-Id: I544b804fd7ccb90e64a945e0da9e43b01b5afdf3 Reviewed-on: https://chromium-review.googlesource.com/1095294Reviewed-by:
Sigurd Schneider <sigurds@chromium.org> Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#53639}
-
Stephan Herhut authored
The graph_zone_ field in PipelineData was initialized to nullptr instead of the correct zone when running the wasm compilation pipeline. Now it is set to the actual zone of the graph that is being compiled. Bug: chromium:844591 Change-Id: I8d9a688a292471a45dedcd01b03244baed60e378 Reviewed-on: https://chromium-review.googlesource.com/1095254Reviewed-by:
-
Sigurd Schneider authored
Bug: v8:7327 Change-Id: Id354b039977a82af8b2c6bba416cf5a1f804ca71 Reviewed-on: https://chromium-review.googlesource.com/1095257 Commit-Queue: Sigurd Schneider <sigurds@chromium.org> Reviewed-by:
Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#53637}
-
Sigurd Schneider authored
This CL allows selection/highlighting of wasm source when a graph node is clicked. Bug: v8:7327 Change-Id: I4a3347a83c8a38804feabffefaefd761596005c3 Reviewed-on: https://chromium-review.googlesource.com/1092712 Commit-Queue: Sigurd Schneider <sigurds@chromium.org> Reviewed-by:
-
Michael Starzinger authored
This new scope allows to avoid emitting calls to {Builtin::kAbort} in debug code when requested, but still trap when the debug code fails. It can be used to keep generated code independent of builtins. R=jgruber@chromium.org Change-Id: I77864a1a10ec5b52bccfd76981ab5f4ff33bc727 Reviewed-on: https://chromium-review.googlesource.com/1095179 Commit-Queue: Michael Starzinger <mstarzinger@chromium.org> Reviewed-by:Jakob Gruber <jgruber@chromium.org> Cr-Commit-Position: refs/heads/master@{#53635}
-
Michael Achenbach authored
TBR=gsathya@chromium.org NOTRY=true Bug: v8:7841 Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng Change-Id: Idb77549e6ede9ea903bfd706db10f6d7ce701a97 Reviewed-on: https://chromium-review.googlesource.com/1095038 Commit-Queue: Michael Achenbach <machenbach@chromium.org> Reviewed-by:
Michael Achenbach <machenbach@chromium.org> Cr-Commit-Position: refs/heads/master@{#53634}
-
Sigurd Schneider authored
Bug: v8:7327 Change-Id: I445a6589612f12c5fffa2bd23831b961c94d960e Reviewed-on: https://chromium-review.googlesource.com/1092535Reviewed-by:
Tobias Tebbi <tebbi@chromium.org> Commit-Queue: Sigurd Schneider <sigurds@chromium.org> Cr-Commit-Position: refs/heads/master@{#53633}
-
Sigurd Schneider authored
D3.v5 needs a merge before applying combined update and enter actions. Bug: v8:7327 Change-Id: If5a044c18e4229c5eadb18c18aea191bcbacef32 Reviewed-on: https://chromium-review.googlesource.com/1095188Reviewed-by:
-
Sigurd Schneider authored
- Improve typing by introducing PhaseView interface. - Recalculate scale extent after resizing. - Fix null sentinel which should have been undefined. Bug: v8:7327 Change-Id: I06881ac3f5681cb419b5da9c6b8aa3a6b2652088 Reviewed-on: https://chromium-review.googlesource.com/1090914Reviewed-by:
-
Sigurd Schneider authored
This CL adds a TFS stub for RegExp#test and moves several checks to the JSCallReducer. In particular, the JSCallReducer checks that - property {exec} on the regexp is still the original exec - property {lastIndex} on the regexp is a non-negative smi The stub does not repeat these checks in release mode. This effectively means that if the regexp is known, we can perform these checks at compile time, and get away with a map dependency. Bug: v8:7779, v8:7200 Change-Id: I0c6d711d4f1d2f6f325a1c02855b0e1b62e014c8 Reviewed-on: https://chromium-review.googlesource.com/1074654 Commit-Queue: Sigurd Schneider <sigurds@chromium.org> Reviewed-by: -
Leszek Swirski authored
Whenever an Isolate is available on a variable, field, or method parameter, use that instead of GetIsolate(). Also convert simple cases of the one-argument handle constructor to either use an available Isolate, or use GetIsolate() if their first parameter is a variable. Bug: v8:7786 Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng Change-Id: I52805905a9ca8729615ead78859f43d5e8f605f1 Reviewed-on: https://chromium-review.googlesource.com/1092853 Commit-Queue: Leszek Swirski <leszeks@chromium.org> Reviewed-by:
Toon Verwaest <verwaest@chromium.org> Reviewed-by:
-
Dominik Inführ authored
Emphasize this code handles ephemerons (key-value pairs) and not ephemeral objects (short-lived objects). Bug: chromium:844008 Change-Id: I1dbf97f69ea5d17b1eb446d92dffef0cc3d5d08f Reviewed-on: https://chromium-review.googlesource.com/1095101Reviewed-by:
-
Sigurd Schneider authored
...and add a js-perf-test instead. Bug: v8:7726 Change-Id: I0cd17642c76887a5a24f54792625d62b9cee3667 Reviewed-on: https://chromium-review.googlesource.com/1094637 Commit-Queue: Sigurd Schneider <sigurds@chromium.org> Reviewed-by:
Daniel Clifford <danno@chromium.org> Reviewed-by:
-
Michael Starzinger authored
R=jgruber@chromium.org BUG=v8:7424 Change-Id: I15fa004b52229ab808401e45fb9d8885f99d8b5e Reviewed-on: https://chromium-review.googlesource.com/1092735Reviewed-by:
-
Clemens Hammacher authored
For errors, it just printed "Failure: expected <Error()> found <Error()>" and completely omitted the specific error type and the message. The new output is: Failure: expected: Error(Error: my explicit error) found: Error(ReferenceError: ffi is not defined) R=mstarzinger@chromium.org Change-Id: Ie17a97e4413c4585b9560fd1c408018ee8c06701 Reviewed-on: https://chromium-review.googlesource.com/1092746Reviewed-by:
-
Matheus Marchini authored
ExistingCodeLogger was behaving incorrectly when the CodeEventHandler API was used in combination with --interpreted-frames-native-stack. Instead of collecting copied trampolines as InterpretedFunction:functionName, they were being collected as Builtin:IntepreterEntryTrampolines. This patch adds special handling for copied trampolines when using ExistingCodeLogger. R=yangguo@google.com Change-Id: I3ee4be03800122d28d53b51b20c60dcf6263e4c1 Reviewed-on: https://chromium-review.googlesource.com/1087813Reviewed-by:
Yang Guo <yangguo@chromium.org> Commit-Queue: Yang Guo <yangguo@chromium.org> Cr-Commit-Position: refs/heads/master@{#53624}
-
jgruber authored
Internal references create absolute pointers within the code and must therefore be disallowed for embedded builtins to remain position-independent. Drive-by: remove related cctest. This test used to be relevant before embedding was fully implemented, but by now it is useless and rather misleading since it gives a false sense of safety. Bug: v8:6666 Change-Id: I57a62274b57c3ef1303d5114c68e2a9b1f92bda4 Reviewed-on: https://chromium-review.googlesource.com/1092732 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by:
-
Sathya Gunasekaran authored
OrderedHashTableHandler (to be renamed to OrderedHashTable) is the interface that abstracts away the two different ordered hash tables. All operations on the two ordered hash tables must be performed through this new interface so that we can seamlessly migrate from one table to another behind the scenes. Bug: v8:6443, v8:7569 Change-Id: Ifc0a38974605b63e0a2a36b4aafb8dc68a081f4a Reviewed-on: https://chromium-review.googlesource.com/1059865 Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by:
Camillo Bruni <cbruni@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Cr-Commit-Position: refs/heads/master@{#53622}
-
- 08 Jun, 2018 15 commits
-
-
Michael Achenbach authored
NOTRY=true Bug: chromium:846711 Change-Id: I5d7f366db2e80fbe539b34a5dcb7475e4e38bd29 Reviewed-on: https://chromium-review.googlesource.com/1088694Reviewed-by:
-
Dominik Inführ authored
The GC now stores EphemeronHashTables in a Worklist. Bug: chromium:844008 Change-Id: I6ad95b3b49484901d75c6fd92b59065c438fcd03 Reviewed-on: https://chromium-review.googlesource.com/1093092 Commit-Queue: Dominik Inführ <dinfuehr@google.com> Reviewed-by:
-
Dominik Inführ authored
JSWeakCollections used to be chained in a singly-linked list. The GC now stores the JSWeakCollection's backing store EphemeronHashTable in a Worklist. This should allow easier parallelization in the future. Bug: chromium:844008 Change-Id: Icc0c1b537c23e8d24dc5b27452fb8106562c9a09 Reviewed-on: https://chromium-review.googlesource.com/1092496 Commit-Queue: Dominik Inführ <dinfuehr@google.com> Reviewed-by:
-
Sathya Gunasekaran authored
Bug: v8:7834 Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng Change-Id: I1fcbcf0b019a0fbf0f17a676bd31a686a6d5377d Reviewed-on: https://chromium-review.googlesource.com/1091880 Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by:
Mathias Bynens <mathias@chromium.org> Cr-Commit-Position: refs/heads/master@{#53618}
-
Rodrigo Bruno authored
Bug: chromium:845409 Change-Id: Id8b89e6dac48bba9065ac9f04ce48f951aa2186b Reviewed-on: https://chromium-review.googlesource.com/1092860 Commit-Queue: Rodrigo Bruno <rfbpb@google.com> Reviewed-by:
-
Ulan Degenbaev authored
The mock histogram functions cannot be cleared and can be called on isolate tear down if incremental marking is in progress. Bug: chromium:850508 Tbr: mlippautz@chromium.org No-Presubmit: true No-Tree-Checks: true No-Try: true Change-Id: I99e52aaa81c863f71e195aeed691b37da9e71da6 Reviewed-on: https://chromium-review.googlesource.com/1093073Reviewed-by:
-
Ulan Degenbaev authored
Bug: chromium:850508 Tbr: mlippautz@chromium.org No-Presubmit: true No-Tree-Checks: true No-Try: true Change-Id: I99adf68c4b3e26decf3b68c8062ce74cbf27874d Reviewed-on: https://chromium-review.googlesource.com/1093071Reviewed-by:
-
Ulan Degenbaev authored
This adds the following histograms recorded corresponding to V8.GCFinalizeMC phases: - V8.GCFinalizeMC.Clear - V8.GCFinalizeMC.Epilogue - V8.GCFinalizeMC.Evacuate - V8.GCFinalizeMC.Finish - V8.GCFinalizeMC.Mark - V8.GCFinalizeMC.Prologue - V8.GCFinalizeMC.Sweep Bug: chromium:850508 Change-Id: I47adc125a9a28436d09e35db68a8e8198cbee2dd Reviewed-on: https://chromium-review.googlesource.com/1091311 Commit-Queue: Ulan Degenbaev <ulan@chromium.org> Reviewed-by:
Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#53614}
-
Dominik Inführ authored
Emit full header and key/value-pairs. Bug: chromium:844008 Change-Id: Ia94b841ff97d024d6ba27b1bcc2f993f95fc11fa Reviewed-on: https://chromium-review.googlesource.com/1092698Reviewed-by:
-
Tobias Tebbi authored
Reland "[turbofan] enable switch jump tables with --no-untrusted-code-mitigations, also for stubs and Wasm" But for builtins, jump tables are disabled to be compatible with embedded builtins. This is a reland of 884bec9f Original change's description: > [turbofan] enable switch jump tables with --no-untrusted-code-mitigations, > also for stubs and Wasm > > Bug: chromium:845851 > Change-Id: I9b860dc26f8b35d629235b82fc5fffe04bf10493 > Reviewed-on: https://chromium-review.googlesource.com/1076151 > Commit-Queue: Tobias Tebbi <tebbi@chromium.org> > Reviewed-by: Michael Starzinger <mstarzinger@chromium.org> > Reviewed-by: Jaroslav Sevcik <jarin@chromium.org> > Cr-Commit-Position: refs/heads/master@{#53525} Bug: chromium:845851 Change-Id: I66c300f875a46a3f2a68730fda94b8196f38aa97 Reviewed-on: https://chromium-review.googlesource.com/1087468 Commit-Queue: Tobias Tebbi <tebbi@chromium.org> Reviewed-by:
-
Simon Zünd authored
This CL implements NumberIsNaN in Torque and removes it from the CodeStubAssembler. It is currently used in TypedArray.p.sort and Array.p.sort. R=jgruber@chromium.org Change-Id: Iaaa81901f89c0df68997d4501607b302a0449012 Reviewed-on: https://chromium-review.googlesource.com/1092532 Commit-Queue: Simon Zünd <szuend@google.com> Reviewed-by:
-
Clemens Hammacher authored
Since we never extract pools from a {DisjointAllocationPool}, the {Allocate} method can just return an {AddressRange}, and also {Merge} just needs to merge a single {AddressRange}. Drive-by: Make {AddressRange} a proper struct, for DCHECKs and better accessors. R=mstarzinger@chromium.org Bug: v8:7754 Change-Id: I19fd02b2c6d8eb5316a5e994835b89be9cfa792b Reviewed-on: https://chromium-review.googlesource.com/1090723 Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Reviewed-by: -
jgruber authored
This class can contain members and functions common across all platforms. Bug: v8:6666 Change-Id: I8f232f806457164a2401f74c7140fd035ad05096 Reviewed-on: https://chromium-review.googlesource.com/1086940 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by:
-
Rodrigo Bruno authored
Bug: chromium:845409 Change-Id: If1fbb03258fdfae528069ba37bbef0395ddb230d Reviewed-on: https://chromium-review.googlesource.com/1090920Reviewed-by:
-
Igor Sheludko authored
Bug: v8:5269, v8:7703 Change-Id: I0b766ab4ca55edc05d1849745fd3604be6c726d6 Reviewed-on: https://chromium-review.googlesource.com/1087460 Commit-Queue: Igor Sheludko <ishell@chromium.org> Reviewed-by:
-
