Changes:
- Remove TypeCheckBranchResult. Change TypeCheckBranch() to return bool.
Refactor call sites to reflect this (decouple current code
reachability check from type check).
- Unify TypeCheckBranch(), TypeCheckFallthrough(), and the type-checking
part of Return() into TypeCheckStackAgainstMerge().
- Make sure all TypeCheck* functions are only called within VALIDATE.
- In graph-builder-interface, rename end_env -> merge_env to reflect
its function for loops.
- Change expected error messages in some tests.
Change-Id: I857edc18db9c2454ad12d539ffe7a10e96367710
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2839560
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#74100}
The test fails because of the race condition between concurrent GC
threads and the destructor of the test.
Bug: v8:11413
Change-Id: I5a138a61c16ddf2398e7c54defe6cc4008ca3330
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2839552
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#74065}
The plan is to use VirtualMemoryCage both for the pointer compression
cage as well as the code range in a future CL. The PtrComprCage class is
removed in favor of using VirtualMemoryCage directly.
Bug: v8:11460
Change-Id: I4e34a3db1359319e3539ede587f6a73e0af03eec
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2824098
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Junliang Yan <junyan@redhat.com>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#74048}
This reverts commit d16eefe0f2.
It is not correct to check for node equality during the graph
construction phase, because we can have optimizations that will combine
same nodes. So it can happen that in wasm-compiler, the inputs to
shuffle are not the same, so we canonicalize using that knowledge that
it will not be the same, and allow indices > 15. But later we can have
optimizations that combine the 2 inputs (e.g. splat of the same
constants), and the instruction selector will see that the input nodes
are the same.
Bug: v8:11542,chromium:1199662
Change-Id: I21c175f4707708038710147f64d687d1b14c6ecc
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2829986
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Bill Budge <bbudge@chromium.org>
Cr-Commit-Position: refs/heads/master@{#74017}
This is a reland of ae0752df1b
Reland fixes:
* Remove UNREACHABLE() from constexpr switch, since we don't have a
CONSTEXPR_UNREACHABLE() (it's ok, the switch is exhaustive for the
enum anyway).
* Fix IsRegisterArray trait to use public inheritance and size_t for
std::array size.
Original change's description:
> [codegen] Add static interface descriptors
>
> Add a new CRTP StaticCallInterfaceDescriptor class, which provides
> static constexpr getters for a descriptor's registers, parameter counts,
> and so on. Each CallInterfaceDescriptor subclass is changed to extend
> StaticCallInterfaceDescriptor, with StaticCallInterfaceDescriptor itself
> extending CallInterfaceDescriptor to still provide a dynamic lookup
> where needed.
>
> StaticCallInterfaceDescriptor provides a couple of customisation points,
> where it reads its CRTP derived descriptor's static fields and
> functions, with default fallbacks where appropriate. With these
> customisation points, the definition of CallInterfaceDescriptor
> subclasses is simplified to:
>
> a) Providing parameter names (as before)
> b) Providing parameter types (as before)
> c) Optionally setting flags (like kNoContext or kAllowVarArgs) as
> static booleans on the class.
> d) Optionally providing a `registers()` method that returns a
> std::array<Register, N> of registers that may be used for
> parameters (if not provided, this defaults to the implementation
> specific default register set).
>
> Parameter registers (and register count) are automagically set based on
> the number of parameters and number of given registers, with extra magic
> to ignore no_reg registers (to reduce ia32 special casing). The
> CallInterfaceDescriptorData is initialized based on these static
> functions, rather than manual per-descriptor initializers.
>
> This allows us to skip loading descriptors dynamically for CallBuiltin
> in Sparkplug, and instead lets us use a bit of template magic to
> statically set up arguments for the calls. Any other users of statically
> known descriptors will also benefit, thanks to C++ picking the static
> methods over the dynamic methods on the base class when available.
>
> Because we can remove various virtual functions and trigger heavier
> inlining of constantly known values, binary size slightly decreases with
> this change.
>
> Note that torque-generated descriptors are changed to use the same magic,
> rather than having Torque-specific magic, for consistency.
>
> Bug: v8:11420
> Change-Id: Icc5e238b6313a08734feb564204a13226b450c22
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2814518
> Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Commit-Queue: Clemens Backes <clemensb@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#73996}
TBR=nicohartmann@chromium.org,clemensb@chromium.org,ishell@chromium.org,clemensb@chromium.org
Bug: v8:11420
Change-Id: Icd1f6cdb3c178e74460044b1e9623139929ceba8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2831872
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#74010}
This reverts commit ae0752df1b.
Reason for revert: Predictably, constexpr issues on non-clang compilers.
Original change's description:
> [codegen] Add static interface descriptors
>
> Add a new CRTP StaticCallInterfaceDescriptor class, which provides
> static constexpr getters for a descriptor's registers, parameter counts,
> and so on. Each CallInterfaceDescriptor subclass is changed to extend
> StaticCallInterfaceDescriptor, with StaticCallInterfaceDescriptor itself
> extending CallInterfaceDescriptor to still provide a dynamic lookup
> where needed.
>
> StaticCallInterfaceDescriptor provides a couple of customisation points,
> where it reads its CRTP derived descriptor's static fields and
> functions, with default fallbacks where appropriate. With these
> customisation points, the definition of CallInterfaceDescriptor
> subclasses is simplified to:
>
> a) Providing parameter names (as before)
> b) Providing parameter types (as before)
> c) Optionally setting flags (like kNoContext or kAllowVarArgs) as
> static booleans on the class.
> d) Optionally providing a `registers()` method that returns a
> std::array<Register, N> of registers that may be used for
> parameters (if not provided, this defaults to the implementation
> specific default register set).
>
> Parameter registers (and register count) are automagically set based on
> the number of parameters and number of given registers, with extra magic
> to ignore no_reg registers (to reduce ia32 special casing). The
> CallInterfaceDescriptorData is initialized based on these static
> functions, rather than manual per-descriptor initializers.
>
> This allows us to skip loading descriptors dynamically for CallBuiltin
> in Sparkplug, and instead lets us use a bit of template magic to
> statically set up arguments for the calls. Any other users of statically
> known descriptors will also benefit, thanks to C++ picking the static
> methods over the dynamic methods on the base class when available.
>
> Because we can remove various virtual functions and trigger heavier
> inlining of constantly known values, binary size slightly decreases with
> this change.
>
> Note that torque-generated descriptors are changed to use the same magic,
> rather than having Torque-specific magic, for consistency.
>
> Bug: v8:11420
> Change-Id: Icc5e238b6313a08734feb564204a13226b450c22
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2814518
> Auto-Submit: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
> Reviewed-by: Clemens Backes <clemensb@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Commit-Queue: Clemens Backes <clemensb@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#73996}
Bug: v8:11420
Change-Id: Ie5469c9253fc140590ac30b72db6eb1d93f86806
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2831485
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Bot-Commit: Rubber Stamper <rubber-stamper@appspot.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#74000}
Add a new CRTP StaticCallInterfaceDescriptor class, which provides
static constexpr getters for a descriptor's registers, parameter counts,
and so on. Each CallInterfaceDescriptor subclass is changed to extend
StaticCallInterfaceDescriptor, with StaticCallInterfaceDescriptor itself
extending CallInterfaceDescriptor to still provide a dynamic lookup
where needed.
StaticCallInterfaceDescriptor provides a couple of customisation points,
where it reads its CRTP derived descriptor's static fields and
functions, with default fallbacks where appropriate. With these
customisation points, the definition of CallInterfaceDescriptor
subclasses is simplified to:
a) Providing parameter names (as before)
b) Providing parameter types (as before)
c) Optionally setting flags (like kNoContext or kAllowVarArgs) as
static booleans on the class.
d) Optionally providing a `registers()` method that returns a
std::array<Register, N> of registers that may be used for
parameters (if not provided, this defaults to the implementation
specific default register set).
Parameter registers (and register count) are automagically set based on
the number of parameters and number of given registers, with extra magic
to ignore no_reg registers (to reduce ia32 special casing). The
CallInterfaceDescriptorData is initialized based on these static
functions, rather than manual per-descriptor initializers.
This allows us to skip loading descriptors dynamically for CallBuiltin
in Sparkplug, and instead lets us use a bit of template magic to
statically set up arguments for the calls. Any other users of statically
known descriptors will also benefit, thanks to C++ picking the static
methods over the dynamic methods on the base class when available.
Because we can remove various virtual functions and trigger heavier
inlining of constantly known values, binary size slightly decreases with
this change.
Note that torque-generated descriptors are changed to use the same magic,
rather than having Torque-specific magic, for consistency.
Bug: v8:11420
Change-Id: Icc5e238b6313a08734feb564204a13226b450c22
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2814518
Auto-Submit: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Nico Hartmann <nicohartmann@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73996}
The typing of br_table was relaxed in
https://github.com/WebAssembly/spec/pull/1305. Before, we had to compute
the greatest lower bound of all branch types and make sure that stack
values are subtypes of that type. Now, we have to check that the stack
values are subtypes of each individual branch. This makes a difference
only in polymorphic stacks, but greatly simplifies the code, especially
with the upcoming introduction of a much more complex type system in
wasm-gc.
Change-Id: I6e3b410cfe0e71a97623b3030b3575ef707c4900
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2827897
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73982}
Changes:
- Implement function subtyping in wasm-subtyping.cc.
- Add Signature::Build(), which takes initializer lists for the return
and parameter types.
- Only throw kTrapFuncSigMismatch in call_indirect, change that trap's
message.
- Add a missing "return 0" in function-body-decoder-impl.h
- Fix a faulty check in wasm-objects.cc.
- Improve some comments.
- Write tests. Improve readability of subtyping-unittest.
Bug: v8:7748
Change-Id: I1caba09d5bd01cfd4d6125f300cd9c16af7aba99
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2822633
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73972}
Migrating unittests from Blink that were not already covered by cppgc.
Bug: chromium:1056170
Change-Id: If31591c3f1e99562028087c2b818f5ceb8122ec9
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2821542
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73960}
Multivalue has been shipped for a while now, so it is time to remove
its experimental feature flag.
Additional change: Set kV8MaxWasmFunctionReturns to the old
kV8MaxWasmFunctionMultiReturns value.
Change-Id: I5c4d33b036e64a7221de17f0e97119bb0a036838
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2817790
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73927}
The pointer compression cage is the virtual memory reservation
that all compressed pointers fall within. This CL splits pointer
compression into two modes: a per-Isolate cage and a shared cage
among multiple Isolates.
When multiple Isolates are sharing a cage, they can decompress
each others' pointers and share the same virtual memory range.
Bug: v8:11460
Change-Id: I7b89b7413b8e7ca6b8b6faafd083dc387542a8b4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2783674
Reviewed-by: Dan Elphick <delphick@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Shu-yu Guo <syg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73916}
Make runtime-call-stats a compile-time flag. Disabling RCS saves roughly
1MB binary size on 64bit systems and yields minor performance
improvements.
Bug: v8:11299
Change-Id: Ia1db75e330a665db5251b685c164b96857e38d2d
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2799766
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Camillo Bruni <cbruni@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73910}
This is a reland of 62ff82e44b
Original change's description:
> cppgc: Add tests for in-construction during ctor
>
> Adds explicit tests that check that an object is marked as in
> construction while running the constructor.
>
> Bug: chromium:1056170
> Change-Id: I7f7340832e1bc31cec98784c261ed86deb402e72
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2811238
> Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
> Reviewed-by: Omer Katz <omerkatz@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#73869}
Bug: chromium:1056170
Change-Id: I38c8579dc2ed437f2ad530bd552b5ef037ba8621
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2817603
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Auto-Submit: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73880}
Adds explicit tests that check that an object is marked as in
construction while running the constructor.
Bug: chromium:1056170
Change-Id: I7f7340832e1bc31cec98784c261ed86deb402e72
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2811238
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73869}
The test doesn't fail anymore after it was updated to check platforms
with page size > 4096:
402806e87f
("[cppgc][unittests] Only expect guard pages support on 4k platforms.")
R=jkummerow@chromium.org
Bug: v8:11587
Change-Id: I6be93e9561b9db0d0f948c5e12fea6a067eb0a76
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2813538
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73857}
The only valid way to define a GCed type T is by inheriting from
GarbageCollected<T>. Since this is prone to typos (see tests), add a
simple check that covers most interesting use cases.
The static assert covers
A -> B -> GarbageCollected<C>
The static assert does not cover
A -> B -> C -> GarbageCollected<B>
(In order to do so, we would need __direct_bases() support which is
not yet available for C++.)
Bug: pdfium:1670, chromium:1056170
Change-Id: I494de48992f8ba9a1f0f9daad60584d828717403
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2810415
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73854}
Now that all users are migrated to Jobs API.
Bug: chromium:1196703
Change-Id: Ic48cce441c1793b1b33f0fc3d6a60847f2eefb2f
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2810156
Commit-Queue: Etienne Pierre-Doray <etiennep@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73838}
These are used by v8_wrapper/heap_test_utilities.* in Blink.
See crrev.com/c/2787126 for usage.
Bug: chromium:1056170
Change-Id: I329b1823f2ac21181a3536577ed72bee3d591347
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2786842
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73805}
Full GCs on non-main-frame context disposals show up on real-world web
workloads and often cause missed frames. Remove and let the regular
scheduler take over these workloads.
Bug: chromium:1191325
Change-Id: Ib58419e4623c096321860db05c36ddf9c8e9f4e4
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2773347
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73799}
Introduce two machine nodes for FABD and fold Float32/64 Abs,Sub
during instruction selection.
This gives ~1% speed improvement of the Bullet physics engine
compiled as wasm.
Change-Id: Ifd985538e6ebb280bc0eaf11b0ebfc687891cf91
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2786854
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Commit-Queue: Martyn Capewell <martyn.capewell@arm.com>
Cr-Commit-Position: refs/heads/master@{#73765}
The delegate instruction is only allowed inside the try section of a try
block.
R=ahaas@chromium.org
Bug: chromium:1193169
Change-Id: I71756b18903dc39e487721b401d18bc4e213cbff
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2791565
Commit-Queue: Thibaud Michaud <thibaudm@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73720}
It could happen (e.g. with --stress-background-compile) that the main
thread blocks for a background thread but the background thread requests
a GC from the main thread. This would result in a deadlock. Avoid this
by parking the main thread for potentially blocking operations and allow
allocations while the main thread is parked.
This CL introduces new states for the main thread: CollectionRequested
and ParkedCollectionRequested. These states will force Safepoint(),
Park() and Unpark() on the main thread into slow paths. The slow path
can then perform a GC on the main thread - right before parking or after
the main thread got unparked.
Bug: v8:10315
Change-Id: If7ef31622d27320613139a0b7f79086fe3200f99
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2731528
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73707}
AddImmediate ends up pushing repeated immediates very often
unecessarily. Add support for Int64 immediates being inlined into
InstructionOperand if they fit into the payload (which is almost always
the case). Also add a seperate rpo_immediate vector for RPO numbers to
avoid having to add them to the immediates_ vector multiple times.
Ideally the RPO values would also be inlined, however JumpThreading
needs to patch RPO targets throughout the instruction stream, so we
need an indirection.
Change-Id: I75b1cdb05917f85d4f740a34c3720dd9cf0ee29c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2782282
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73687}
Add Float32Select and Float64Select as OptionalOperators and insert
these, if supported, when handling a Select expression in the wasm
graph builder. FlagsContinuation have been modified to support the
select operation and code generation support has been added for arm64.
This improves the 'Bullet' physics benchmark by ~2-3%.
Change-Id: I928c3085c9136ad8baeeb34c71c47c1c8338844c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2763871
Commit-Queue: Martyn Capewell <martyn.capewell@arm.com>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73657}
We now canonicalize earlier in the pipeline, and don't need to worry
about non-canonicalized shuffles.
Bug: v8:11542,v8:11257
Change-Id: If9f5c44061465be339c98e479fd8c5a437bbd74b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2778673
Reviewed-by: Bill Budge <bbudge@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73645}
LoadImmutable represents a load from a position in memory that is known
to be immutable, e.g. an immutable IsolateRoot or an immutable field of
a WasmInstanceObject. Because the returned value cannot change through
the execution of a function, LoadImmutable is a pure operator and does
not have effect or control edges.
This will allow more aggressive optimizations of loads of fields of
the Isolate and Instance that are known to be immutable.
Requires that the memory in question has been initialized at function
start even through inlining.
Note: We may reconsider this approach once we have escape analysis for
wasm, and replace it with immutable load/initialize operators that live
inside the effect chain and are less restriced.
Bug: v8:11510
Change-Id: I5e8e4f27d7008f39f01175ffa95a9c531ba63e66
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2775568
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73594}
When swizzle is called with a v128.const node, we can check that the
indices are either all in bounds, or if they are out of bounds the top
bit of each byte is set. This will match exactly pshufb behavior, and so
we can omit the paddusb (and getting external reference).
Bug: v8:10992
Change-Id: I5479a9eb92ebcfc12bedff5efd3e72bb4a43ff40
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2766222
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73583}
This is a more canonical type name, and is in line with {kVoidCode}.
Change-Id: Iaae9524b6fb6ecaafd63ce81cf30e3d01ca3e525
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2775565
Commit-Queue: Manos Koukoutos <manoskouk@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Simon Zünd <szuend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73557}
This allows to hold a constexpr (empty) "builder" object instead of
creating it for every use.
R=ahaas@chromium.org
Bug: v8:11384
Change-Id: Ib5e13c58e81a950bb5dd0e8eefe4021bc77d8b64
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2773801
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Andreas Haas <ahaas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73548}
If memory64 is enabled, memory.grow should consume and return an i64
instead of i32.
This CL implements this for both TurboFan and Liftoff, and adds
validation and execution tests at different layers.
R=manoskouk@chromium.org
Bug: v8:10949
Change-Id: I0b725dbd0d5767bda4609747c1f4aad163c35304
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2773800
Reviewed-by: Manos Koukoutos <manoskouk@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73542}
The test was extended in this CL: https://crrev.com/c/2762420
It now uses wasm::SimdShuffle, which is only available if webassembly is
enabled.
Thus, #if out the test if webassembly is disabled.
Drive-by: Add a missing include.
R=jkummerow@chromium.orgCC=zhin@chromium.org
Bug: v8:11238
Change-Id: I1b53d0145467b58616a161944fb88d2ca256fd58
Cq-Include-Trybots: luci.v8.try:v8_linux64_no_wasm_compile_rel
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2772978
Reviewed-by: Zhi An Ng <zhin@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73517}
We do not emit this opcode in the asm.js to wasm translator, hence we do
not need to handle this case in the function body decoder.
Replace it by a DCHECK to catch bugs where we might accidentally emit
the opcode.
R=thibaudm@chromium.org
Change-Id: Ibaa07099f30b871c28e3d3cebae96277f46e3164
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2768619
Reviewed-by: Thibaud Michaud <thibaudm@chromium.org>
Commit-Queue: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73513}
We currently canonicalize shuffles in the architecture specific
instruction selector. This has the drawback that if we want to pattern
match on nodes that have a shuffle as input, they need to individually
canonicalize the shuffle. There can also be a subtle bug if we
canonicalize the same shuffle node twice (see bug for details).
This moves the canonicalization to "construction time", in
wasm-compiler, when building the graph. As such, any pattern matches in
instruction-selector will only need to deal with canonicalized shuffles.
We introduce a new kind of parameter for shuffle nodes,
ShuffleParameter, to store the 16 bytes plus a bool indicating if this
is a swizzle. A swizzle essentially: inputs to the shuffle are the same
or all indices only touch 1 input. We calculate this when
canonicalizing, so store this bit of information inside of the node's
parameter.
We update the tests in x64 to handle special cases where, even though
the node's inputs are not swapped (due to canonicalization), they need
to be swapped for the specific instruction selected (e.g. palignr). The
test data also contains canonicalized shuffles, so we have to manually
canonicalize them.
Bug: v8:11542
Change-Id: I4e78082267bd03d6caedf43d68d81ef3f5f364a8
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2762420
Reviewed-by: Bill Budge <bbudge@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73495}
This is a reland of d76064df4f
Original change's description:
> cppgc: Rework GC info creation
>
> Previously, GCInfoTrait relied on the non-trivial constructor of a
> static object for registering a new GCInfo object. The generated code
> is required to be thread-safe which is achieved by introducing guard
> variables in the compiler.
>
> The new version is similar to Blink in that it relies on zero
> initialization of a trivially constructible atomic.
>
> Compared to guard variables that are created per GCInfo registration,
> the atomic creates less bloat (~20bytes/type) and also results in a
> better fast path.
>
> Minimum example: https://godbolt.org/z/qrdTf8
>
> Bug: chromium:1056170
> Change-Id: I95efbbf035b655d0440c9477f5391e310e2b71fa
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2764750
> Reviewed-by: Omer Katz <omerkatz@chromium.org>
> Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#73463}
Bug: chromium:1056170
Change-Id: I01e60beabc1d279d352361657f408f113aac768e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2767021
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Auto-Submit: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73471}
