- 11 Mar, 2019 25 commits
-
-
Clemens Hammacher authored
The counter is dead since January 2018 (https://crrev.com/c/860011). This CL reenables it. R=titzer@chromium.org Bug: v8:8975 Change-Id: I0245edd40bb972a89e223685f56ab02fbdf68aa0 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514677Reviewed-by:
Ben Titzer <titzer@chromium.org> Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#60166}
-
Clemens Hammacher authored
We currently see >20% of modules overflow on this counter, i.e. having more than 100k functions. Thus increase the histogram size to 1M, which is the maximum number of functions supported in v8. R=titzer@chromium.org Bug: v8:8975 Change-Id: Ie709ec37cdea90b78637da63b48aea0f4775cee5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514736Reviewed-by:
-
Clemens Hammacher authored
Currently the total allocated address space is collected twice per Wasm memory: Once on allocation, once on release. Recording it once provides the same data. R=titzer@chromium.org Bug: v8:8975 Change-Id: I99719ecb44b14ce5da1be4a74291377f91f4038a Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514738Reviewed-by:
-
Clemens Hammacher authored
The WasmCompileLazy runtime function used to return the start of the instructions of the newly compiled function. With garbage collection of wasm code, it would be tricky to keep this wasm code object alive until it is being called. Thus make the runtime function not return anything, and call via the jump table instead. The jump table was patched as part of lazy compilation to point to the compiled code. Drive-by: Merge {LazyCompileFunction} into single caller {CompileLazy}. R=mstarzinger@chromium.org Bug: v8:8217 Change-Id: Id3c02f4dfe2c536e335af2e7e1b9700203d48675 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1509472 Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Reviewed-by:Michael Starzinger <mstarzinger@chromium.org> Cr-Commit-Position: refs/heads/master@{#60163}
-
Simon Zünd authored
This CL changes the FieldAccessExpression to include a SourcePosition for the field, allowing a jump from such an access to the definition of a field inside a class or struct. Drive-by: Change the source position of a field from the position of its type to the position of its name. This will coincidentally improve the accuracy of error messages. R=danno@chromium.org Bug: v8:8880 Change-Id: I7996e8a57a2ff90e8bdca6f40c1b298d48071e23 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1512573Reviewed-by:
Daniel Clifford <danno@chromium.org> Commit-Queue: Simon Zünd <szuend@chromium.org> Cr-Commit-Position: refs/heads/master@{#60162}
-
Jakob Gruber authored
It will soon be possible to throw arbitrary exceptions from within interpreter execution (namely, in interrupts). We can thus no longer assume that an EXCEPTION return code means we need to throw a stack overflow exception. Bug: v8:8724 Change-Id: I10e24aba4305dc7b39248ced9a52735c59ab662c Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511474 Commit-Queue: Jakob Gruber <jgruber@chromium.org> Reviewed-by:
Peter Marshall <petermarshall@chromium.org> Cr-Commit-Position: refs/heads/master@{#60161}
-
Georg Neis authored
R=jarin@chromium.org Bug: chromium:940423 Change-Id: I9e357d8fe424b291bc25a1d389696df88deda6a3 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514735Reviewed-by:
Jaroslav Sevcik <jarin@chromium.org> Commit-Queue: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#60160}
-
Milad Farazmand authored
Fixing ThreadId according to the following changes: 656254b1 Change-Id: I1e1943ac7e3ed03799c213e566816bfe5c21967d Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1512718Reviewed-by:
Junliang Yan <jyan@ca.ibm.com> Commit-Queue: Milad Farazmand <miladfar@ca.ibm.com> Cr-Commit-Position: refs/heads/master@{#60159}
-
Clemens Hammacher authored
Our UMA data shows a lot of small modules, and I have the suspicion we are loosing some numbers about the bigger ones. Thus sample the module code size after baseline compilation finished. At that point the majority of the code was generated. Sampling after top-tier finished is not that easy since we do not spawn a foreground task at that point. R=mstarzinger@chromium.org Bug: v8:8217 Change-Id: Icaa4a2efb201d24cbc8d2e1b8da516ae26574f01 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1508675 Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Reviewed-by:
-
Maya Lekova authored
Bug: v8:7790 R=neis@chromium.org Change-Id: I10085cff40e14ea63074e29649af55fa2c0ea462 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514494 Commit-Queue: Maya Lekova <mslekova@chromium.org> Commit-Queue: Georg Neis <neis@chromium.org> Reviewed-by:
Georg Neis <neis@chromium.org> Auto-Submit: Maya Lekova <mslekova@chromium.org> Cr-Commit-Position: refs/heads/master@{#60157}
-
Georg Neis authored
It wasn't always guaranteed that they were serialized before taking the dependency. Bug: chromium:940361 Change-Id: Id5e5e14532809e7496546c2011176e33848506ce Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514495Reviewed-by:
Maya Lekova <mslekova@chromium.org> Commit-Queue: Georg Neis <neis@chromium.org> Cr-Commit-Position: refs/heads/master@{#60156}
-
Simon Zünd authored
R=tmrts@chromium.org Change-Id: I652a142c815c7268700de3b743f09495748ec399 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514634Reviewed-by:
Jakob Gruber <jgruber@chromium.org> Reviewed-by:
Tamer Tas <tmrts@chromium.org> Commit-Queue: Simon Zünd <szuend@chromium.org> Cr-Commit-Position: refs/heads/master@{#60155}
-
Jakob Gruber authored
Similar to NativeRegExpMacroAssembler::Result, the regexp interpreter will need a RETRY return code in case the subject string representation changes during an interrupt. This CL adds a new IrregexpInterpreter::Result type to decouple from RegExpImpl::Result. Bug: v8:8724 Change-Id: I946fc0cbc4d7d8631312b72f13a45abeb9986905 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511472Reviewed-by:
-
Maya Lekova authored
This reverts commit beaca8cf. Reason for revert: Broke presubmit bot - https://ci.chromium.org/p/v8/builders/ci/V8%20Presubmit/2938 Note that the problem is not with this CL itself, but it uncovers some presubmit issue in Torque code. Until the latter is fixed, I'm reverting to unblock the tree. Original change's description: > [presubmit] use the correct path for third party libraries > > This CL ensures that presubmit script checks Torque files in third_party > dependencies. > > R=szuend@chromium.org > TBR=machenbach@chromium.org,sergiyb@chromium.org > CC=yangguo@chromium.org > > No-Try: true > Change-Id: I9e2b193defbebe7ae85cfc5d14ce50c2ac367e9b > Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513674 > Reviewed-by: Tamer Tas <tmrts@chromium.org> > Reviewed-by: Simon Zünd <szuend@chromium.org> > Commit-Queue: Tamer Tas <tmrts@chromium.org> > Cr-Commit-Position: refs/heads/master@{#60152} TBR=tmrts@chromium.org,szuend@chromium.org Change-Id: If8e2db0801f51ef737243ccfcc909d05fb42e3e6 No-Presubmit: true No-Tree-Checks: true No-Try: true Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514633Reviewed-by:
-
Tamer Tas authored
This CL ensures that presubmit script checks Torque files in third_party dependencies. R=szuend@chromium.org TBR=machenbach@chromium.org,sergiyb@chromium.org CC=yangguo@chromium.org No-Try: true Change-Id: I9e2b193defbebe7ae85cfc5d14ce50c2ac367e9b Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513674Reviewed-by:
Simon Zünd <szuend@chromium.org> Commit-Queue: Tamer Tas <tmrts@chromium.org> Cr-Commit-Position: refs/heads/master@{#60152}
-
Simon Zünd authored
With the recent changes to Array#sort, the main algorithm does not need to bail out anymore. Only the initial copying into the workarray, as well as the final copying back into the original backing store might cause a switch from fast-path to the slow-path. This CL changes the slow-path so sorting itself is not restarted and the slow-path will continue copying where the fast-path left off. R=jgruber@chromium.org Bug: v8:7382 Change-Id: I4ab61daa62bb816f4f6e16e60bde1f948ad1e7db Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507717 Commit-Queue: Simon Zünd <szuend@chromium.org> Reviewed-by:
-
Simon Zünd authored
With the recent changes to Array#sort, some bailout labels and accessor checks became superfluous. This CL removes them along with some other minor cleanup work. R=jgruber@chromium.org Bug: v8:8834 Change-Id: I7429482ceaccbe743e2b8190d83bfa2c34875b11 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1507678Reviewed-by:
-
Andreas Haas authored
The table_object instance field is not needed anymore because its purpose is fulfilled now by the tables field I introduced to support multiple tables. In addition I removed {table_instances_} from the {InstanceBuilder}. This field existed because tables could exist without a WasmTableObject. With recent changes, WasmTableObjects always exist. R=mstarzinger@chromium.org Bug: v8:7581 Change-Id: I5e8e3d2910f7ed7ae74d61eff660f9451b3493ac Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1466641 Commit-Queue: Andreas Haas <ahaas@chromium.org> Reviewed-by: -
Andreas Haas authored
I forgot to add the test file to https://crrev.com/c/1463519. R=mstarzinger@chromium.org Bug: v8:7581 Change-Id: I88d50fe83ae60a8170110504ce7f765aa28db517 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511480Reviewed-by:
-
Mythri authored
This is a pre-work for allocating feedback vectors lazily. Feedback cells are required to share the feedback vectors across the different closures of the same function. Currently, they are held in the CreateClosureSlot in the feedback vector. With lazy feedback vector allocation, we may not have a feedback vector. However, we still need a place to store the feedback cells, so if feedback vector is allocated in future it can still be shared across closures. Here is the detailed design doc: https://docs.google.com/document/d/1m2PTNChrlJqw9MiwK_xEJfqbFHAgEHmgGqmIN49PaBY/edit BUG=v8:8394 Change-Id: Ib406d862b2809b1293bfecdcfcf8dea3127cb1c7 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1503753 Commit-Queue: Mythri Alle <mythria@chromium.org> Reviewed-by:
Toon Verwaest <verwaest@chromium.org> Reviewed-by:
Ross McIlroy <rmcilroy@chromium.org> Reviewed-by:
Ulan Degenbaev <ulan@chromium.org> Reviewed-by:
Benedikt Meurer <bmeurer@chromium.org> Cr-Commit-Position: refs/heads/master@{#60147}
-
Clemens Hammacher authored
The {id_} stored in {ThreadId} should not be atomic. Only getting a new id for the current thread needs to be atomic. If any user of {ThreadId} needs atomicity, that user should wrap {ThreadId} in a {std::atomic} instead. Drive-by: Remove {Equals} method, use {operator==} instead. Drive-by: Move static methods after member methods. R=ishell@chromium.org Cq-Include-Trybots: luci.v8.try:v8_linux64_tsan_rel Bug: v8:8834 Change-Id: Id0470eb2fa907948843ac1153e2dc5dcd9a8fbc8 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1494006Reviewed-by:Leszek Swirski <leszeks@chromium.org> Reviewed-by:
Igor Sheludko <ishell@chromium.org> Commit-Queue: Clemens Hammacher <clemensh@chromium.org> Cr-Commit-Position: refs/heads/master@{#60146}
-
Michael Lippautz authored
v8::EmbedderHeapTracer::TracePrologue may call back into V8 during StartMarking. In this case we expect that the write barriers are set up and consistent, i.e., global flag matches page flag. Blink calls back into V8 in a corner case where sweeping is finalized on incremental marking start which may trigger resettting a V8 Value which may trigger DescriptorArray re-shuffling. Bug: chromium:940003 Change-Id: Ia15c798d0faaab802df1c3b569b5b6a323a4fe59 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1514492Reviewed-by:
Hannes Payer <hpayer@chromium.org> Commit-Queue: Michael Lippautz <mlippautz@chromium.org> Cr-Commit-Position: refs/heads/master@{#60145}
-
Shiyu Zhang authored
Currently, if input types are not string or number, ToString builtin will fall into runtime and a loop of ToPrimitive and type-checks is done in runtime, which is slow. This CL reimplements ToString to add support for that ToPrimitive and type-checks loop in CSA instead of runtime to improve performance. This will benefit Array.prototype.toString/join a lot when the array elements are objects. This Cl improves the performance of Speedometer2.0 EmberJS-Debug case by ~14% on Atom. Change-Id: I27c2669097be1e542e30119cdffcf79c0d16a0eb Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1498698 Commit-Queue: Shiyu Zhang <shiyu.zhang@intel.com> Reviewed-by:
-
Jaroslav Sevcik authored
If the branch associated with the condition is kDead, the current node will be killed anyway, so let us just survive the lowering. Bug: chromium:935092 Change-Id: If7b39e3b5452d6c9bc5199080eb38725e6c4eab5 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1488769Reviewed-by:
Tobias Tebbi <tebbi@chromium.org> Commit-Queue: Jaroslav Sevcik <jarin@chromium.org> Cr-Commit-Position: refs/heads/master@{#60143}
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/8da0480..61c98e6 Rolling v8/test/wasm-js/data: https://chromium.googlesource.com/external/github.com/WebAssembly/spec/+log/b552862..299e687 TBR=machenbach@chromium.org,hablich@chromium.org,sergiyb@chromium.org Change-Id: If1c6e07dac6b0cc58a03307e737353ebb9fceb57 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513938Reviewed-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@{#60142}
-
- 10 Mar, 2019 1 commit
-
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/0ee4cbc..8da0480 Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/26a1944..b88eedc TBR=machenbach@chromium.org,hablich@chromium.org,sergiyb@chromium.org Change-Id: I67809fa36e3df8e65e2bb59a62d902c91675be23 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513319Reviewed-by:
-
- 09 Mar, 2019 6 commits
-
-
Sergiy Belozorov authored
TBR=sergiyb@chromium.org Bug: chromium:934964 Change-Id: I95ffd94f31c60b0e82cac7653ad4e5978bd3b9d9 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1503759Reviewed-by:
Sergiy Belozorov <sergiyb@chromium.org> Commit-Queue: Sergiy Belozorov <sergiyb@chromium.org> Cr-Commit-Position: refs/heads/master@{#60140}
-
Anna Henningsen authored
This should not be used anymore (and it definitely is not by Node.js or Chromium). Change-Id: I4a1ce1fda98efd197a64ce0969dae5c8b18f6e97 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511484Reviewed-by:
Yang Guo <yangguo@chromium.org> Commit-Queue: Yang Guo <yangguo@chromium.org> Cr-Commit-Position: refs/heads/master@{#60139}
-
Deepti Gandluri authored
Integer splats should use an operand when a register is not allocated. Bug: V8:8927 Change-Id: I14c80b7b073fae3754ec32f4fa8605af399ef341 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513102Reviewed-by:
Bill Budge <bbudge@chromium.org> Commit-Queue: Deepti Gandluri <gdeepti@chromium.org> Cr-Commit-Position: refs/heads/master@{#60138}
-
v8-ci-autoroll-builder authored
Rolling v8/build: https://chromium.googlesource.com/chromium/src/build/+log/ba214bf..0ee4cbc Rolling v8/third_party/catapult: https://chromium.googlesource.com/catapult/+log/a07cd8f..26a1944 Rolling v8/third_party/depot_tools: https://chromium.googlesource.com/chromium/tools/depot_tools/+log/2f3c820..3485a26 TBR=machenbach@chromium.org,hablich@chromium.org,sergiyb@chromium.org Change-Id: I8b2d600c00d4a84db60c36ae87fd6e8ecede8b56 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1513165Reviewed-by:
-
Yu Yin authored
We want to compare two inputs so need to perform the same operation(ExtractBits) on them. Change-Id: I6c81884fdd34dfa125b842f010cd40f8a6816a0f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511132Reviewed-by:
-
Deepti Gandluri authored
Change-Id: Ie080683af1d990e5205c75a2a199f0581d826811 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511630 Commit-Queue: Deepti Gandluri <gdeepti@chromium.org> Reviewed-by:
Ben Smith <binji@chromium.org> Cr-Commit-Position: refs/heads/master@{#60135}
-
- 08 Mar, 2019 8 commits
-
-
Sathya Gunasekaran authored
This allows the devtools to preview the private fields that are installed on an object. Change-Id: I6d8aad7ad0e51cdf18f6139b4bb8665e4b606aa5 Bug: v8:8773, v8:8337 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1487914 Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org> Reviewed-by:
Alexei Filippov <alph@chromium.org> Reviewed-by:
Dmitry Gozman <dgozman@chromium.org> Cr-Commit-Position: refs/heads/master@{#60134}
-
Z Duong Nguyen-Huu authored
Script, PrototypeInfo, JSCollectionIterator, JSWeakCollection, JSSloppyArgumentsObject, JSArgumentsObject Bug: v8:8952 Change-Id: Id0d7049f3d5971a0fd4af56dd4850ad3b44c22e9 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1506198 Commit-Queue: Z Nguyen-Huu <duongn@microsoft.com> Reviewed-by:
-
Matt Gardner authored
Call to ReduceKeyedLoadFromHeapConstant got lost in rebasing, as did the kHas check in ReduceElementAccessOnString. Added some tests to ensure both cases are covered. Change-Id: I8d6992c33315436b6228471b9bc57e3b267ad09f Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1508837Reviewed-by:
-
Z Duong Nguyen-Huu authored
This is the reland of https://chromium-review.googlesource.com/c/v8/v8/+/1495898 builtin_function_id corresponded to BuiltinFunctionId (a manually maintained list of 'interesting' functionsmainly used during optimization). With this change, we nuke builtin-function-id in favor of builtin-id and 8 bits is freed up in SFI. Bug: v8:6993 Change-Id: I7e1681cc2a95864c71ce8bdda075481310607166 Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1506445Reviewed-by:
-
Z Duong Nguyen-Huu authored
TemplateInfo, FunctionTemplateRareData, FunctionTemplateInfo, ObjectTemplateInfo Bug: v8:8952 Change-Id: Ib355cf825ca8c57be45d3a55ff595d356c8737ce Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1504774Reviewed-by:
-
Junliang Yan authored
the problem is that we call irregexp code in two ways: 1. CallCFunction9 from CSA builtins 2. Through GeneratedCode::Call from the runtime. 1 is a standard C call and expects the target to be a FD, 2 is our own implementation where we dynamically generate a FD. So there's a mismatch between the two. Change-Id: I8391db30fa7586d296b5d1880a7f44dafad21a2a Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1487341 Commit-Queue: Junliang Yan <jyan@ca.ibm.com> Auto-Submit: Junliang Yan <jyan@ca.ibm.com> Reviewed-by:
-
Mike Stanton authored
This CL contains find, findIndex, every and some. Now that we've established the pattern on the torque side for iterating array builtins, it's a very easy port, which nonetheless decreases code size in the snapshot, w00t! Bug: v8:8906 Change-Id: I3082d8e3e298e55733a42d6b441e5812b7f12f3d Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1496976 Commit-Queue: Michael Stanton <mvstanton@chromium.org> Reviewed-by:
-
Clemens Hammacher authored
Reusing the same {Binary} object (with the same {ArrayBuffer} underneath) speeds up the limits test with 1M functions by a factor of 11x in an optdebug build. R=titzer@chromium.org Change-Id: I36d032d652c66f5b7f5a80399588652d7e3946ec Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511475Reviewed-by:
-
