On termination of the worker pool in the main process, a SIGTERM is
sent from pool to worker. It was meant to terminate long-running
tests in the worker process. The signal handler on the worker side,
however, was only registered during test execution. During the
remaining logic (<1% of the time probably) the default system
behavior for SIGTERM would be used (which will likely just kill
the process). The ungracefully killed process might be killed while
writing to the results queue, which then remains with corrupted data.
Later when the main process cleans up the queue, it hangs.
We now register a default handler in the worker process that catches
the SIGTERM and also gracefully stops the processing loop. Like
that, the SIGTERM signal will always be handled in workers and never
fall back to SIGKILL.
However, a small time window exists when the SIGTERM was caught
right when starting a test process, but when the test-abort handler
was not registered yet. We keep fixing this as a TODO. Worst case,
the main process will block until the last test run is done.
Bug: v8:13113
Change-Id: Ib60f82c6a1569da042c9f44f7b516e2f40a46f93
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3882972
Reviewed-by: Alexander Schulze <alexschulze@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/main@{#83101}
Now that all external pointers have been sandboxed,
V8_SANDBOXED_EXTERNAL_POINTERS is no longer needed. This change also
shrinks external pointer slots to 32 bits when the sandbox is enabled.
Bug: v8:10391
Change-Id: Iccbef27ac107b988cb23fe9ef66da6fe0bae087a
Cq-Include-Trybots: luci.v8.try:v8_linux64_heap_sandbox_dbg_ng,v8_linux_arm64_sim_heap_sandbox_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3869269
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Samuel Groß <saelo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#83083}
.. which sets the --interrupt-budget-for-maglev to a very low value
s.t. that tiering to Maglev happens very early. Note this affects
both normal tierup and OSR.
Also add flag handling to fuzzer.py, both as added globally with
probability 0.1, and added to InterruptBudgetFuzzer (which I also
updated with other tiering-related flags).
Bug: v8:7700
Change-Id: I844cf53a6a2da459565d0ad0ccae02b04853cd26
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3878165
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Jakob Linke <jgruber@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#83033}
This is a reland of commit d00c040547
Changes since revert: Use AsAtomicTagged instead of
base::AsAtomicPointer to store a hash value in the forwarding table.
Original change's description:
> [strings] Support shared external strings>
>
> With this CL shared strings can be externalized and external strings can
> be shared.
> The StringForwardingTable is used to delay the real transition to the
> next full GC. On the API side strings marked for externalization will
> look like externalized strings.
>
> Bug: v8:12957
> Change-Id: I53b6509129bc5679c06bdf99421bdb41ea5d9082
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3849643
> Reviewed-by: Shu-yu Guo <syg@chromium.org>
> Reviewed-by: Camillo Bruni <cbruni@chromium.org>
> Commit-Queue: Patrick Thier <pthier@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Reviewed-by: Victor Gomes <victorgomes@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#82966}
Bug: v8:12957
Change-Id: I17715e927e4339240a6aa12a3c4a3c2ea50eb567
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3871211
Commit-Queue: Patrick Thier <pthier@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Victor Gomes <victorgomes@chromium.org>
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/main@{#83018}
This adds Python logging to several code locations that previously
used to just print. The locations aren't yet complete. The changed
code locations should help for investigating hanging test runs.
The default level is WARNING for running tests locally, and INFO
when called from bots that pass the --swarming parameter.
Bug: v8:13113
Change-Id: If3a336703e7c346a5c718f2359b1a80e37e1ca6d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3876183
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Alexander Schulze <alexschulze@chromium.org>
Cr-Commit-Position: refs/heads/main@{#83006}
This reverts commit d00c040547.
Reason for revert: Failing CI tests on Mac arm64
Original change's description:
> [strings] Support shared external strings
>
> With this CL shared strings can be externalized and external strings can
> be shared.
> The StringForwardingTable is used to delay the real transition to the
> next full GC. On the API side strings marked for externalization will
> look like externalized strings.
>
> Bug: v8:12957
> Change-Id: I53b6509129bc5679c06bdf99421bdb41ea5d9082
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3849643
> Reviewed-by: Shu-yu Guo <syg@chromium.org>
> Reviewed-by: Camillo Bruni <cbruni@chromium.org>
> Commit-Queue: Patrick Thier <pthier@chromium.org>
> Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
> Reviewed-by: Victor Gomes <victorgomes@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#82966}
Bug: v8:12957
Change-Id: I13155fcc788d217db56cbfd1c9e4457a81a9dbd7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3870486
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Matthias Liedtke <mliedtke@chromium.org>
Owners-Override: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Matthias Liedtke <mliedtke@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82970}
With this CL shared strings can be externalized and external strings can
be shared.
The StringForwardingTable is used to delay the real transition to the
next full GC. On the API side strings marked for externalization will
look like externalized strings.
Bug: v8:12957
Change-Id: I53b6509129bc5679c06bdf99421bdb41ea5d9082
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3849643
Reviewed-by: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Patrick Thier <pthier@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Victor Gomes <victorgomes@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82966}
This ignores one of --assert-types and --stress-concurrent-inlining if
used together. We already filter those from trials, but they also might
get picked from // Flags lines in test cases.
No-Try: true
Bug: chromium:1359829
Change-Id: I7e46afb53bdbb0a871c7443a5a66a339046eb3de
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3871195
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Auto-Submit: Michael Achenbach <machenbach@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82961}
Add logic to drop cyclic contradictory flags from
correctness-fuzzing command lines. Add the currently known
biggest offenders.
Without this, the correctness fuzzing harness runs into a CHECK
failure during smoke testing, when attempting to pass cyclic flags
to d8. It fails fast, but uselessly burns fuzzing time.
This change drops one of the known cyclic flags instead to make the
test run still useful. The precedence is right to left like in the
V8 test framework.
Additionally on Clusterfuzz, all crashes during smoke testing are
deduped as one crash report. We don't know if there are other
problems before this one is fixed/hidden.
No-Try: true
Bug: chromium:1330303
Change-Id: I06cbb4655cd3cf467f5cce6f84dba653834ca72e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3865562
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Alexander Schulze <alexschulze@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82939}
.. and update old entries:
- Remove --no-untrusted-code-mitigations (flag no longer exists)
- Make --stress-concurrent-inlining-attach-code less likely since it
mostly overlaps with --stress-concurrent-inlining and there haven't
been many reports in this are lately.
Bug: v8:7700
Change-Id: Ic88ea74ca88a7c14edfc39bc0a55b846bb01e465
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3867506
Reviewed-by: Victor Gomes <victorgomes@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Jakob Linke <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82882}
If we see a default ctor, walk up the constructors until we find a non-
default one.
Default ctors can only be skipped if there are no class fields / private
brands.
This CL implements the Ignition parts; Sparkplug, Maglev and TF will
be implemented as follow ups. (This is fine, since this feature is
behind a flag.)
Bug: v8:13091
Change-Id: Ie8ca8aedb01bd4b13adf1063332a5cdf41ab358a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3804601
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Commit-Queue: Marja Hölttä <marja@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82872}
Due to https://crbug.com/1195263, large typed arrays can have an
observable difference on the same architecture, depending on
optimization behavior.
For differential fuzzing comparisons, we previously used a proxy
for typed arrays that capped the maximum size only when comparing
different architectures - there an observable difference is WAI.
We move the capping code and make it arch-independent for now until
the bug above gets fixed, since it caused too many duplicates, which
degrade fuzzing performance.
Bug: chromium:1195263
Change-Id: Ic81c383e547413378cbe037de3c38eb900a9e5dc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3866173
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82860}
There are 2 kinds of WasmToJs code: one uses a WasmFrame and the other
a simple TypedFrame TF frame. This CL introduces WasmToJsFunctionFrame
(as a simple TypedFrame) to dfferentiate from the WasmFrame counterpart.
This was not an issue before
https://crrev.com/c9c490891ab3d63fc790770cb1c76f8013ba963f
since we used to use a single master function for pointer iteration.
No-Tree-Checks: true
Bug: v8:13243
Change-Id: I97f7e8c897159ca3cafa65ff6ddf836c5ef7b76e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3865969
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Victor Gomes <victorgomes@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82856}
This CL:
- Separates Wasm frame pointer iteration from TypedFrame iteration
- Introduces the frame StubWithContext that correspond to a TF-optimized
code without JS linkage
- Fixes WasmToJsFrame to be a WasmFrame
- Fixes the hack with JsToWasmFrame that calls TypedFrame when not
the right builtin
- Cleans up TypedFrame::Iterate
Change-Id: Ie238df5188f2e2f081f5353b4538b75b6e79f64a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3859787
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Victor Gomes <victorgomes@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82809}
Bugs that are older than the switch of v8_foozzie.py to Python3
bisect to the switch commit unfortunately. This change attempts to
let bisect run longer if a python2 executable still exists.
No-Try: true
Bug: chromium:1355824
Change-Id: I457a50af21704ddd2985793861eee8be5601a673
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3856574
Reviewed-by: Liviu Rau <liviurau@google.com>
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82728}
--stress-opt never did what we wanted it to; it ran its runs in
different contexts (therefore not able to share feedback across runs),
and even if it didn't, each run would create new closures for any
defined closures, so we'd still more than likely end up poly- or
mega-morphic.
Fuzzers cover this use case better than --stress-opt ever did, so now
it's just using precious bot time. We can get rid of it.
Bug: v8:10386
Change-Id: Ibbb9207d887b4b1dc4ec9093858d477c0f95eb37
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3803228
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82722}
Now roll_merge_gerrit.py waits for a +1, and immediately submits. With
auto-submit and rubber-stamper bot, this makes the script fully
fire-and-forget.
This also fixes the commit message update to include the change id.
Bug: v8:12849
Change-Id: I63784bfc1b2a16dfcd308b11e67d9da9c2ff3f8a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3804249
Reviewed-by: Liviu Rau <liviurau@google.com>
Commit-Queue: Liviu Rau <liviurau@google.com>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82664}
Namely:
- AccessorInfo::getter and AccessorInfo::js_getter,
- CallHandlerInfo::callback and CallHandlerInfo::js_callback.
The redirected/non-redirected callback distinction is required only
for simulated builds but we wasted memory also for all native builds.
Now we store these fields in "redirected" form which allows us to call
them directly from builtins or generated code. In case it's necessary
to call a callback from C++ code the C function address is read from
the redirection. This additional indirection makes the callback calls
from C++ code in simulated builds slower but saves memory for native
builds.
This CL should recover a part of memory regression caused by inlining
Foreign fields into AccessorInfo and CallHandlerInfo.
Bug: v8:12949, chromium:1336105, chromium:1335930
Change-Id: I38470ed21ee23b281247c11a9531542c7e4acca1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3835686
Reviewed-by: Jakob Linke <jgruber@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82631}
This CL changes the WasmTypeInfo class to have a direct ExternalPointer
to the native type structure instead of using a Foreign. This in turn
makes it possible to use a unique pointer tag for that external pointer
when the sandbox is enabled.
Bug: v8:10391, v8:12949
Change-Id: Ifee4d2103cabfa6a7299d0d09e06d387034e5f8f
Cq-Include-Trybots: luci.v8.try:v8_linux64_heap_sandbox_dbg_ng,v8_linux_arm64_sim_heap_sandbox_dbg_ng
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3829085
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Samuel Groß <saelo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82504}
Once lazy compilation lands we want to continue testing the eager
compilation configuration. For that we add --no-wasm-lazy-compilation to
the "stress" variant.
Bug: v8:12852
Change-Id: I1777aaeb1c8cec0359128ed4b3d34a21aa3ce37e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3810249
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82470}
When trying to understand why a given module fails to validate, it
can be helpful to disassemble it as far as possible until reaching
the erroneous byte(s).
Change-Id: I0056ba1a81b85a486c0446d15bbf54ccb2e8332e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3827866
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#82433}
- Unify old Pool interface with the new context related interface
- Add single threaded execution pool
- Defer task killing back to OS context
- Defer process listing in indicators back to OS context
Bug: v8:12785
Cq-Include-Trybots: luci.v8.try:v8_numfuzz_dbg_ng,v8_numfuzz_ng,v8_numfuzz_tsan_ng,v8_android_arm64_n5x_rel_ng
Change-Id: I8ffe01c5d567411203f69ecc451c718ff35d81c9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3781347
Reviewed-by: Alexander Schulze <alexschulze@chromium.org>
Commit-Queue: Liviu Rau <liviurau@google.com>
Cr-Commit-Position: refs/heads/main@{#82371}
