This adds a new type 'none' as part of the WASM GC MVP.
The type can only be used in combination with a nullable reference, e.g.
'ref.null none'.
A 'nullref' is implicitly convertible to any nullable reference type.
Bug: v8:7748
Change-Id: Ic5ab6cc27094b3c9103ce3584452daa34633612f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3755136
Auto-Submit: Matthias Liedtke <mliedtke@google.com>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81670}
Suspender.{returnPromiseOnSuspend,suspendOnReturnedPromise}
are not tied to a specific suspender anymore, so move them to
WebAssembly.{returnPRomiseOnSuspend,suspendOnReturnedPromise}.
With this change, the suspender property is not needed anymore on the
function data. Convert it to a boolean flag that just indicates whether
a function uses the JS Promise Integration API.
R=ahaas@chromium.org
Bug: v8:12191
Change-Id: I1b6d8e3190ebf5049dbc7eedee448999cf077509
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3748660
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81639}
This field points to the start of an ArrayBuffer backing store, which
is guaranteed to be located inside the sandbox if it is enabled. As
such, this simply turns the field into a sandboxed pointer field.
Bug: chromium:1342548
Change-Id: I5a76e23cfc83b2a04cd461def1cd04337ccf5cf7
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/+/3749190
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Samuel Groß <saelo@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81579}
The streaming decoder did not properly check the ordering of sections
relative to the code section.
This CL fixes that for both empty and non-empty code sections.
The special path for empty code sections is not actually needed, so
remove it to simplify code paths.
Drive-by:
1. Refactor the existing code for checking section ordering to make it
more structured and readable.
2. Ensure that we either call {DecodeCodeSection} or {StartCodeSection},
but not both.
3. Remove {set_code_section}, merge it into {StartCodeSection}.
4. Simplify calls to {CalculateGlobalOffsets} (make them unconditional
and remove one redundant one).
R=ahaas@chromium.org
Bug: chromium:1336380
Change-Id: Ia2c5c115d43d2b5315e3b3c9e4a21175a36aa326
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3747860
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81554}
This opcode is being removed in favor of pre-declared non-defaultable
locals (details are still TBD).
Bug: v8:9495
Change-Id: I96ac053a1b5a852310c5dc0bbaeab0cbf5384663
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3738743
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81496}
Most often, the {ValueType::Ref} constructor was called with a
constant nullability. To make things more convenient, this CL renames
{Ref} to {RefMaybeNull}, and introduces {Ref} and {RefNull}
constructors with fixed nullability.
Bug: v8:7748
Change-Id: I664ff184ca936cc752e152c3c67546d79aa24390
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3732936
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81494}
This makes the internal V8 name consistent with the text-format name.
Bug: v8:7748
Change-Id: I44f7ac1eb5e634b4f829e596bf1f14caeb748d54
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3726291
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81491}
There were multiple bugs and no test coverage for br_on_cast and br_on_cast_fail, specifically for the paths in the decoder where those
checks get optimized away.
Bug: v8:7748
Change-Id: I6e5d6449152df0456b43938174f57055a4c63fdd
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3723503
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81349}
This merges the separate opcode name definitions from wasm-opcodes-inl.h
into the main opcode-defining macros in wasm-opcodes.h. This is simpler
(avoids a bunch of fairly complex macros) and easier to update when we
add new opcodes in the future.
The tests become obsolete because they would simply repeat the implementation.
Change-Id: Ib6421da5670079e7725659c1f4008251f8ff7aed
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3714244
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81280}
CreateJob() is similar to PostJob() but doesn't schedule anything
until Join() or Notify*() is called.
This allows
- CreateJob().Join() without too many worker.
- Early 'empty' CreateJob() for initialization
without causing spurious calls to GetMaxConcurrency()
Bug: chromium:1287665
Change-Id: I8fd8b139392ad30218f0cf8f580b2d76f1078777
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3668842
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81166}
Already after enabling Liftoff, the name did not match the semantics any
more. The callback was called after top-tier finished, not after initial
compilation of the module finished.
With dynamic tiering, the name is even less fitting.
This CL renames the "OnModuleCompiled" callback in the API to
"MoreFunctionsCanBeSerialized", which makes it more obvious what the
API should be used for. It also internally renames all related typedefs
and methods accordingly.
One call of the callback in the streaming decoder was already wrong
before this CL and is being removed.
R=jkummerow@chromium.org, cbruni@chromium.org
Bug: v8:12899
Change-Id: I95c0fc9e32442383e47e4370e31277cc065bf0fe
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3687689
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81093}
The module's "prefix hash" is based on a prefix of the module bytes that
starts at the beginning of the module and stops at the code section.
In the case of the streaming decoder, if the code section is empty,
`AsyncStreamingProcessor::ProcessCodeSectionHeader()` is never called,
and we keep accumulating bytes in the hash after the code section. Fix
this by always calling into the streaming processor even if the code
section is empty.
R=ahaas@chromium.orgCC=clemensb@chromium.org
Bug: chromium:1334651
Change-Id: Id2a03468b355867868e589523c994c268c7b4eaf
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3695564
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81060}
Anyone using CopyablePersistentTraits should be using v8::Global, so
deprecate it and fix the uses in V8.
Bug: v8:12915
Change-Id: I25e6f2a03e070db9e9af9bbd9ea8cbc0f838c5ac
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3669254
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/main@{#81001}
When a 8x16 shuffle matches a packed byte to dword zero extension,
1. input1 is S128Zero after canonicalization,
2. the indices {0,4,8,16} are consecutive value in the range [0-15] and
other indices are in the range [16-31],
the shuffle can be matched to packed byte to dword zero extend. These
shuffles are commonly used in image processing.
Change-Id: I14d1e35401dbc5ecd91f67c46ea9762628835d01
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3547667
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Fanchen Kong <fanchen.kong@intel.com>
Cr-Commit-Position: refs/heads/main@{#80953}
Extend the effect of --freeze-flags-after-init to also protect updates
of individual flags instead of only the API.
For this, we wrap each flag in a {FlagValue} class which implicitly
converts to the value of the flag. Some cases still require the explicit
{value()} accessor though. That accessor is {constexpr}, in contrast to
the implicit conversion, because otherwise clang emits a lot of warnings
about dead code within "if (FLAG...)" scopes.
R=cbruni@chromium.org
Bug: v8:12887
Change-Id: I87d3457e49ceb317d34d6a21cf09c520d4171eb5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3683321
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Patrick Thier <pthier@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80938}
This fully removes the kFinishedTopTierCompilation event, and any
handling of it. In a dynamic tiering world, that event has no meaning
any more.
R=ahaas@chromium.org
Bug: v8:12899
Change-Id: I36484e36f7c36f2ac4fcb111e67a14509c2eefef
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3667081
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80930}
Prototype the instruction on the interpreter, and Arm64. Details of
instruction lowerings on all relevant architectures can be found at:
https://github.com/WebAssembly/relaxed-simd/issues/52
Bug: v8:12908
Change-Id: If8ffb82c38042191c67c9b5c23a231877d4f2159
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3679848
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Ilya Rezvov <irezvov@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80924}
We introduce a typing phase into the Turbofan compilation pipeline for
wasm-gc. It has two functionalities: (1) to type nodes that were not
typed during code generation (mainly phi nodes) and (2) to narrow types
as much as possible.
The following nodes are handled, which should be enough for our
purposes: TypeGuard, WasmTypeCast, AssertNotNull, Phi, LoadFromObject,
and LoadImmutableFromObject.
Loop phi types are computed by first assigning the type of the
non-recursive input, and updating once we have the type of the recursive
inputs, and repeating this process to a fixed point.
Drive-by: Remove the narrowing of function signatures during wasm
inlining, as it created some issues and should not be needed after this
series of changes.
Bug: v8:7748
Change-Id: I8a72488d5c221c4ae8257fc5abf6f0368cf10e96
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3678208
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80912}
Remove code size and compilation time sampling for the "top tier
finished" event. With dynamic tiering, this event will never be reached.
R=ahaas@chromium.org
Bug: v8:12899
Change-Id: I1b0d053e31fe8cd1f8ba3b23bfff4c5879569b45
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3647691
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80867}
This CL fixes all spots where wasm Turbofan code did not satisfy the
invariant that all nodes with effect outputs are connected to another
node. Also, it enables the related verification for wasm code.
Drive-by:
- Simplify how stack checks are removed during loop unrolling.
- Fix a test declaration in test-gc.cc.
Change-Id: Id32af8584ba0ec281f4bf7757bd2915e6d8bf443
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3676862
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80854}
Some tests disabled dynamic tiering in order to actually serialize
TurboFan code. This makes the tests a bit simpler, but does not reflect
real-world usage.
This CL enables dynamic tiering in those tests and when needed executes
the code until tier-up was observed.
R=ahaas@chromium.org
Bug: v8:12899
Change-Id: I34cb8cedbc5908d9e6ca09d56c51609d0c8b2d6a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3660262
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80849}
We introduce wasm types to Turbofan. They are represented by a new
subtype of {TypeBase}.
Types are attached to nodes during graph generation, as nodes are
assigned to wasm values. Therefore, phi nodes corresponding to merges
are not typed. Missing types will be computed, and assigned types will
be narrowed as much as possible, in a retyping phase that will be
introduced later.
Bug: v8:7748
Change-Id: I943559cf1d512edfab3bb2d22f8748c072cb1911
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3660248
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80778}
Bug: v8:12868
A slight modification to the existing DFA-based UTF-8 allocator to allow
decoding surrogates, for use in decoding WTF-8. We'll need to
additionally constrain the decoder to disallow surrogate pairs.
Change-Id: Ifddbf08d4eeeff8f270df52a68f01769ea790eec
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3652787
Commit-Queue: Andy Wingo <wingo@igalia.com>
Reviewed-by: Marja Hölttä <marja@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80654}
Triggering tier-up can happen very often, so the runtime function should
be as slim as possible.
This CL adds two DisallowGarbageCollection scopes and removes a
HandleScope which was unnecessarily created.
R=jkummerow@chromium.org
Bug: v8:12281
Change-Id: I43e7f2b449630856ac8dfb36d294fbd29191d0eb
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3652300
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80644}
If b is S128Zero, Shuffle(a,b,s) can be optimized to
Swizzle(a,s). By setting s[i] to 0x80, we can avoid access b.
If a is S128Zero, we can swap a and b first.
If one input of I8x16Shuffle is S128Zero, this patch can save
~60% instructions(7 of 12), and more than 30% improvement is
observed in local microbenchmarks.
Change-Id: I5953fa9064e01203cd4cf423c55dd5ed33cad57e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3544992
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Jie Pan <jie.pan@intel.com>
Cr-Commit-Position: refs/heads/main@{#80623}
Now that we require C++17 support, we can just use the standard
static_assert without message, instead of our STATIC_ASSERT macro.
R=leszeks@chromium.org
Bug: v8:12425
Change-Id: I1d4e39c310b533bcd3a4af33d027827e6c083afe
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3647353
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80524}
Prototype the instruction on the interpreter, and Arm64. Details of
instruction lowerings on all relevant architectures can be found at:
https://github.com/WebAssembly/relaxed-simd/issues/40
Bug: v8:12284
Change-Id: Id4cb3889d94cf0bb7169ea068efe5ca68cfcbad9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3636365
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80475}
This prevents ambiguity errors in C++20 due to ADL when casting types in
std::, which gains std::bit_cast<>().
Bug: chromium:1284275
Change-Id: I25046d1952a9304852e481ad8b84049c6769c289
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3625838
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80378}
This is a reland of commit 9d31f8663a
There were issues with --future flag implications on M1.
Original change's description:
> [rwx][mac] Support fast W^X permission switching on Apple Silicon (M1)
>
> ... for V8 code space. The feature is currently disabled.
>
> In order to use fast W^X permission switching we must allocate
> executable pages with readable writable executable permissions (RWX).
> However, MacOS on ARM64 ("Apple M1"/Apple Silicon) prohibits further
> permission changing of RWX memory pages. This means that the code page
> headers must be allocated with RWX permissions too because otherwise
> it wouldn't be possible to allocate a large code page over the freed
> regular code page and vice versa.
>
> When enabled, the new machinery works as follows:
>
> 1) when memory region is reserved for allocating executable pages, the
> whole region is committed with RWX permissions and then decommitted,
> 2) since reconfiguration of RWX page permissions is not allowed on
> MacOS on ARM64 ("Apple M1"/Apple Silicon), there must be no attempts
> to change them,
> 3) the request to set RWX permissions in the executable page region
> just recommits the pages without changing permissions (see (1), they
> were already allocated as RWX and then discarded),
> 4) in order to make executable pages inaccessible one must use
> OS::DiscardSystemPages() instead of OS::DecommitPages() or
> setting permissions to kNoAccess because the latter two are not
> allowed by the MacOS (see (2)).
> 5) since code space page headers are allocated as RWX pages it's also
> necessary to switch between W^X modes when updating the data in the
> page headers (i.e. when marking, updating stats, wiring pages in
> lists, etc.). The new CodePageHeaderModificationScope class is used
> in the respective places. On unrelated configurations it's a no-op.
>
> The fast permission switching can't be used for V8 configuration with
> enabled pointer compression and disabled external code space because
> a) the pointer compression cage has to be reserved with MAP_JIT flag
> which is too expensive,
> b) in case of shared pointer compression cage if the code range will
> be deleted while the cage is still alive then attempt to configure
> permissions of pages that were previously set to RWX will fail.
>
> This also CL extends the unmapper unit tests with permissions tracking
> for discarded pages.
>
> Bug: v8:12797
> Change-Id: Idb28cbc481306477589eee9962d2e75167d87c61
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3579303
> Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Commit-Queue: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#80238}
Bug: v8:12797
Change-Id: I0fe86666f31bad37d7074e217555c95900d2afba
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3610433
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80259}
This reverts commit 9d31f8663a.
Reason for revert: crashes on Mac/arm64 bots:
https://ci.chromium.org/ui/p/v8/builders/ci/V8%20Mac%20-%20arm64%20-%20debug/5923/overview
Original change's description:
> [rwx][mac] Support fast W^X permission switching on Apple Silicon (M1)
>
> ... for V8 code space. The feature is currently disabled.
>
> In order to use fast W^X permission switching we must allocate
> executable pages with readable writable executable permissions (RWX).
> However, MacOS on ARM64 ("Apple M1"/Apple Silicon) prohibits further
> permission changing of RWX memory pages. This means that the code page
> headers must be allocated with RWX permissions too because otherwise
> it wouldn't be possible to allocate a large code page over the freed
> regular code page and vice versa.
>
> When enabled, the new machinery works as follows:
>
> 1) when memory region is reserved for allocating executable pages, the
> whole region is committed with RWX permissions and then decommitted,
> 2) since reconfiguration of RWX page permissions is not allowed on
> MacOS on ARM64 ("Apple M1"/Apple Silicon), there must be no attempts
> to change them,
> 3) the request to set RWX permissions in the executable page region
> just recommits the pages without changing permissions (see (1), they
> were already allocated as RWX and then discarded),
> 4) in order to make executable pages inaccessible one must use
> OS::DiscardSystemPages() instead of OS::DecommitPages() or
> setting permissions to kNoAccess because the latter two are not
> allowed by the MacOS (see (2)).
> 5) since code space page headers are allocated as RWX pages it's also
> necessary to switch between W^X modes when updating the data in the
> page headers (i.e. when marking, updating stats, wiring pages in
> lists, etc.). The new CodePageHeaderModificationScope class is used
> in the respective places. On unrelated configurations it's a no-op.
>
> The fast permission switching can't be used for V8 configuration with
> enabled pointer compression and disabled external code space because
> a) the pointer compression cage has to be reserved with MAP_JIT flag
> which is too expensive,
> b) in case of shared pointer compression cage if the code range will
> be deleted while the cage is still alive then attempt to configure
> permissions of pages that were previously set to RWX will fail.
>
> This also CL extends the unmapper unit tests with permissions tracking
> for discarded pages.
>
> Bug: v8:12797
> Change-Id: Idb28cbc481306477589eee9962d2e75167d87c61
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3579303
> Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
> Commit-Queue: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#80238}
Bug: v8:12797
Change-Id: Ic07948e036db36326d464a2a901d052aa060a406
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3611665
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Auto-Submit: Adam Klein <adamk@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80239}
... for V8 code space. The feature is currently disabled.
In order to use fast W^X permission switching we must allocate
executable pages with readable writable executable permissions (RWX).
However, MacOS on ARM64 ("Apple M1"/Apple Silicon) prohibits further
permission changing of RWX memory pages. This means that the code page
headers must be allocated with RWX permissions too because otherwise
it wouldn't be possible to allocate a large code page over the freed
regular code page and vice versa.
When enabled, the new machinery works as follows:
1) when memory region is reserved for allocating executable pages, the
whole region is committed with RWX permissions and then decommitted,
2) since reconfiguration of RWX page permissions is not allowed on
MacOS on ARM64 ("Apple M1"/Apple Silicon), there must be no attempts
to change them,
3) the request to set RWX permissions in the executable page region
just recommits the pages without changing permissions (see (1), they
were already allocated as RWX and then discarded),
4) in order to make executable pages inaccessible one must use
OS::DiscardSystemPages() instead of OS::DecommitPages() or
setting permissions to kNoAccess because the latter two are not
allowed by the MacOS (see (2)).
5) since code space page headers are allocated as RWX pages it's also
necessary to switch between W^X modes when updating the data in the
page headers (i.e. when marking, updating stats, wiring pages in
lists, etc.). The new CodePageHeaderModificationScope class is used
in the respective places. On unrelated configurations it's a no-op.
The fast permission switching can't be used for V8 configuration with
enabled pointer compression and disabled external code space because
a) the pointer compression cage has to be reserved with MAP_JIT flag
which is too expensive,
b) in case of shared pointer compression cage if the code range will
be deleted while the cage is still alive then attempt to configure
permissions of pages that were previously set to RWX will fail.
This also CL extends the unmapper unit tests with permissions tracking
for discarded pages.
Bug: v8:12797
Change-Id: Idb28cbc481306477589eee9962d2e75167d87c61
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3579303
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80238}
This introduces an AssemblerBufferCache class which will (later) cache
the backing store of AssemblerBuffers. This is needed for PKU-protected
assembler buffers, which are expensive to allocate and deallocate.
For now, the AssemblerBufferCache does not do any caching, this will be
added in a follow-up CL.
R=thibaudm@chromium.org
Bug: v8:12809
Change-Id: I4a7ccff49c9930584a9fcda8899cfe38cfc61419
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3593136
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80145}
These were originally proposed as a part of the fixed-width SIMD
proposal, and were then migrated to the relaxed-simd proposal
which also deems these operations out of scope.
Github issue: https://github.com/WebAssembly/relaxed-simd/issues/4
Bug: v8:12284
Change-Id: I65ceb6dfd25c43cf49bd7ec5b5ecd6b32cc3516a
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3595970
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Deepti Gandluri <gdeepti@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80125}
{LiftoffOptions} already contains many (optional) parameters for Liftoff
compilation, but not all of them.
This CL moves the function index and the {for_debugging} field also into
that struct, to further reduce the number of parameters to
{ExecuteLiftoffCompilation} and to improve readability by having a
factory-like initialization of the {LiftoffOptions} struct.
That struct is now also passed down to the LiftoffCompiler directly
instead of unpacking the fields again.
R=thibaudm@chromium.org
Bug: v8:12809
Change-Id: I8824a1908f214cbf4c21f113934fef3ece1bf88b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3513894
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80032}
Immediate version of the Bitclear instruction can be used for logical
And with some immediates. It can also be used to implement
And(x, Not(imm)) in a single instruction. This patch gives ~0.5% runtime
improvement in one benchmark on Neoverse N1.
Change-Id: Ia926c6746f0c252f81626c6fca21c4dfb41679d9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3160667
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Martyn Capewell <martyn.capewell@arm.com>
Cr-Commit-Position: refs/heads/main@{#80015}
This is a reland of commit 4d8e1846a7
One of the Mac arm64 bots failed to link an exported thread_local
static variable (crbug/1316800).
Original change's description:
> [rwx][mac] Introduce RwxMemoryWriteScope
>
> ... as a single bottleneck that encapsulates the semantics and
> implementation of fast per-thread W^X permission switching supported
> by Apple Silicon (arm64 M1).
> On other architectures this class is a no-op.
>
> Bug: v8:12797
> Change-Id: Ica842ff9f843e20b7f61fd7e80591e7a1fd29771
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3586986
> Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
> Commit-Queue: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/main@{#79994}
Bug: v8:12797
Change-Id: Ifbd15c233bb343f11daa89b1328b5bf65c4806f4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3591332
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/main@{#80011}
