This is a CL in a string of CLs that aims to TNodify CSA. In particular,
there were some loads that were done in AnyTagged instead of
TaggedPointer. TNode-ifying them brings improvement in pointer
compression since we are able to decompress using the Pointer
decompression.
TNodified:
* LoadJSFunctionPrototype
* TryPrototypeChainLookup
* OrdinaryHasInstance
Also TNodified loads regarding:
* FeedbackCell::kValueOffset
* HeapObject::kMapOffset
* JSFunction::kSharedFunctionInfoOffset
* JSFunction::kFeedbackCellOffset
* Map::kInstanceTypeOffset
* Map::kInstanceDescriptorsOffset
* Map::kPrototypeOffset
Drive-by cleanup: StoreJSArrayLength and StoreElements were unused.
Bug: v8:6949, v8:9396
Change-Id: I89697b5c02490906be1eee63cf3d9e60a1094d48
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1755844
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63216}
This is the first in a series of changes to reduce the number of
bytecodes generated for the iteration protocol based operations.
The GetIterator bytecode introduced in this change currently loads the
@@iterator symbol from an object that was previously done using the
LdaNamedProperty bytecode. This change uses builtin-based mechanism
that would be extended to perform additional operations in the future
on absorbing the bytecodes associated with the GetIterator operation
from the iteration protocol.
Bug: v8:9489
Change-Id: I83b8b55c27bae8260bf227f355eeca1ba80cd8f0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1701852
Commit-Queue: Swapnil Gaikwad <swapnilgaikwad@google.com>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63139}
Cleans up a plethora of JumpIfUndefined().JumpIfNull()
occurances by introducing a new JumpIfUndefinedOrNull
bytecode.
Change-Id: I715e9dd82ca8309e0f3eb6514ddec19b4efe7dbe
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1743148
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63130}
This in in preparation for generic (off-heap/on-heap) bytecode
array accessor.
Bug: v8:7790
Change-Id: Ib419831ba1db95ab938179723ef5f130f01ae0d6
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1635895
Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62222}
- Lower LoadObjectField to LoadFromObject
- Mark LoadFromObject and StoreToObject as non-allocating
- Use optimizable BitcastTaggedSignedToWord in TaggedIsNotSmi check
R=jarin@chromium.org, tebbi@chromium.org
Change-Id: I42992d46597be795aee3702018f7efd93fcc6ebf
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1657926
Commit-Queue: Georg Schmid <gsps@google.com>
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62173}
This CL was generated by an automatic clang AST rewriter using this
matcher expression:
callExpr(
callee(
cxxMethodDecl(
hasName("operator->"),
ofClass(isSameOrDerivedFrom("v8::internal::Object"))
)
),
argumentCountIs(1)
)
The "->" at the expression location was then rewritten to ".".
R=jkummerow@chromium.orgTBR=mstarzinger@chromium.org,verwaest@chromium.org,yangguo@chromium.org
Bug: v8:9183, v8:3770
No-Try: true
No-Tree-Checks: true
Change-Id: I0a7ecabdeafe51d0cf427f5280af0c7cab96869e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1624209
Reviewed-by: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#61764}
We translate loads with TaggedXXX (XXX in {"", "Signed", "Pointer"})
representation in CSA into loads of CompressedXXX +
ChangeCompressedXXXToTaggedXXX in the raw-machine-assembler.
This way, CSA doesn't need to know about Compressed values since we
are introducing an explicit "decompress" node.
Also updating tests that were checking for the load nodes.
Cq-Include-Trybots: luci.v8.try:v8_linux64_pointer_compression_rel_ng
Cq-Include-Trybots: luci.v8.try:v8_linux64_arm64_pointer_compression_rel_ng
Bug: v8:8977, v8:7703
Change-Id: Ie22ca8123a25ef005c1ff7383776f9355020fa42
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1565897
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Santiago Aboy Solanes <solanes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60873}
We want to allocate feedback vectors lazily in lite mode. To do that,
we should create closures with the correct feedback cell. This cl
allocates feedback cell arrays to hold these feedback cells in lite mode.
This cl also modifies the compile lazy to builtin to expect these arrays
in the feedback cell.
Drive-by fix: InterpreterEntryTrampoline no longer has argument count in
a register. So updated comments and removed unnecessary push/pop of this
register.
Bug: v8:8394
Change-Id: I10d8ca67cebce61a284f0c80b200e1f0c24577a2
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1511274
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Commit-Queue: Mythri Alle <mythria@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60189}
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}
The original was reverted for breaking webkit layout tests:
https://ci.chromium.org/p/v8/builders/luci.v8.ci/V8-Blink%20Linux%2064/30270
It also caused the following clusterfuzz failures:
chromium:935832
This was a correctness bug due to not properly handling the case of arrays with prototypes other
than Array.prototype. Accesses that were TheHole were not being handled property, both in bounds
holes in holey arrays and out of bounds on either holey or packed arrays. Handling was incorrect
both in access-assembler and in Turbofan.
chromium:935932
This bug was that there was no handling for Has checks on the global object. Turbofan was emitting
code for a store (the 'else' condition on 'access_mode == AccessMode::kLoad'). It hit a DCHECK in
debug builds but in release could show up in different places. This is the bug that caused the
webkit layout test failure that led to the revert.
Both bugs are fixed by in CL, and tests are added for those cases.
Bug: v8:8733, chromium:935932, chromium:935832
Change-Id: Iba0dfcfce6e15d2c0815a7670ece67bc13ba1925
Reviewed-on: https://chromium-review.googlesource.com/c/1493132
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Matt Gardner <magardn@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#59958}
This is a reland of 35269f77f8
Switches on an expression that unconditionally throws would have all their
case statements dead, causing a DCHECK error in the SwitchBuilder. This
fixes up the DCHECK to allow dead labels.
Original change's description:
> [ignition] Skip binding dead labels
>
> BytecodeLabels for forward jumps may create a dead basic block if their
> corresponding jump was elided (due to it dead code elimination). We can
> avoid generating such dead basic blocks by skipping the label bind when
> no corresponding jump has been observed. This works because all jumps
> except JumpLoop are forward jumps, so we only have to special case one
> Bind for loop headers to bind unconditionally.
>
> Since Binds are now conditional on a jump existing, we can no longer rely
> on using Bind to get the current offset (e.g. at the beginning of a try
> block). Instead, we now expose the current offset in the bytecode array
> writer. Conveniently, this means that we can be a bit smarter about basic
> blocks around these statements.
>
> As a drive-by, remove the unused Bind(target,label) function.
>
> Bug: chromium:934166
> Change-Id: I532aa452fb083560d07b90da99caca0b1d082aa3
> Reviewed-on: https://chromium-review.googlesource.com/c/1488763
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#59942}
TBR=rmcilroy@chromium.org
Bug: chromium:934166
Change-Id: If6eab4162106717ce64a2dc477000c6a76354cb4
Reviewed-on: https://chromium-review.googlesource.com/c/1494535
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59948}
This reverts commit 35269f77f8.
Reason for revert: Fuzzer unhappy: https://ci.chromium.org/p/v8/builders/ci/V8%20Fuzzer/29792
Original change's description:
> [ignition] Skip binding dead labels
>
> BytecodeLabels for forward jumps may create a dead basic block if their
> corresponding jump was elided (due to it dead code elimination). We can
> avoid generating such dead basic blocks by skipping the label bind when
> no corresponding jump has been observed. This works because all jumps
> except JumpLoop are forward jumps, so we only have to special case one
> Bind for loop headers to bind unconditionally.
>
> Since Binds are now conditional on a jump existing, we can no longer rely
> on using Bind to get the current offset (e.g. at the beginning of a try
> block). Instead, we now expose the current offset in the bytecode array
> writer. Conveniently, this means that we can be a bit smarter about basic
> blocks around these statements.
>
> As a drive-by, remove the unused Bind(target,label) function.
>
> Bug: chromium:934166
> Change-Id: I532aa452fb083560d07b90da99caca0b1d082aa3
> Reviewed-on: https://chromium-review.googlesource.com/c/1488763
> Commit-Queue: Leszek Swirski <leszeks@chromium.org>
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#59942}
TBR=rmcilroy@chromium.org,leszeks@chromium.org
Change-Id: I8118e54e0afa5e08b0a0a874c952f8a01f1c3242
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: chromium:934166
Reviewed-on: https://chromium-review.googlesource.com/c/1494534
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59947}
BytecodeLabels for forward jumps may create a dead basic block if their
corresponding jump was elided (due to it dead code elimination). We can
avoid generating such dead basic blocks by skipping the label bind when
no corresponding jump has been observed. This works because all jumps
except JumpLoop are forward jumps, so we only have to special case one
Bind for loop headers to bind unconditionally.
Since Binds are now conditional on a jump existing, we can no longer rely
on using Bind to get the current offset (e.g. at the beginning of a try
block). Instead, we now expose the current offset in the bytecode array
writer. Conveniently, this means that we can be a bit smarter about basic
blocks around these statements.
As a drive-by, remove the unused Bind(target,label) function.
Bug: chromium:934166
Change-Id: I532aa452fb083560d07b90da99caca0b1d082aa3
Reviewed-on: https://chromium-review.googlesource.com/c/1488763
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59942}
Includes various fixes and cleanups here and there.
Bug: v8:7703, v8:8852
Change-Id: I603eb0212cab3fecabfa15dceb70ee23b81cdb5a
Reviewed-on: https://chromium-review.googlesource.com/c/1491595
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59909}
This reverts commit 32fc0acfef.
Reason for revert:
https://ci.chromium.org/p/v8/builders/luci.v8.ci/V8-Blink%20Linux%2064/30270
layout test breakage:
https://test-results.appspot.com/data/layout_results/V8-Blink_Linux_64/30270/webkit_layout_tests%20%28with%20patch%29/layout-test-results/results.html
There is a dead node arriving in representation selection, which might indicate that the problem is not in this CL, but that this CL stirs up the node soup in such a way that dead code elimination gets confused.
Original change's description:
> Optimize `in` operator
>
> This change implements optimizations for the `in` operator for packed array
> elements and object properties. It adds a new feedback slot kind and an IC
> path similar to KeyedLoadIC for handling the lookups. TurboFan uses the
> feedback to optimize based on the maps and keys.
>
> For more details see:
> https://docs.google.com/document/d/1tIfzywY8AeNVcy_sen-5Xev21MeZwjcU8QhSdzHvXig
>
> This can provide 10x performance improvements of on loops of the form:
>
> for (let i = 0; i < ary.length; ++i) {
> if (i in ary) {
> ...
> }
> }
>
>
> Bug: v8:8733
> Change-Id: I766bf865a547a059e5bce5399bb6112e5d9a85c8
> Reviewed-on: https://chromium-review.googlesource.com/c/1432598
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Toon Verwaest <verwaest@chromium.org>
> Commit-Queue: Matt Gardner <magardn@microsoft.com>
> Cr-Commit-Position: refs/heads/master@{#59843}
TBR=ulan@chromium.org,rmcilroy@chromium.org,jkummerow@chromium.org,jarin@chromium.org,ishell@chromium.org,bmeurer@chromium.org,verwaest@chromium.org,magardn@microsoft.com
Change-Id: Ib2db974e5bed4c4a2b6b450f796bdc4b0b8fd562
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:8733
Reviewed-on: https://chromium-review.googlesource.com/c/1488761
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59857}
This change implements optimizations for the `in` operator for packed array
elements and object properties. It adds a new feedback slot kind and an IC
path similar to KeyedLoadIC for handling the lookups. TurboFan uses the
feedback to optimize based on the maps and keys.
For more details see:
https://docs.google.com/document/d/1tIfzywY8AeNVcy_sen-5Xev21MeZwjcU8QhSdzHvXig
This can provide 10x performance improvements of on loops of the form:
for (let i = 0; i < ary.length; ++i) {
if (i in ary) {
...
}
}
Bug: v8:8733
Change-Id: I766bf865a547a059e5bce5399bb6112e5d9a85c8
Reviewed-on: https://chromium-review.googlesource.com/c/1432598
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Commit-Queue: Matt Gardner <magardn@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#59843}
This takes heap-inl.h out of the "Giant Include Cluster".
Naturally, that means adding a bunch of explicit includes
in a bunch of places that relied on transitively including
them before.
As of this patch, no header file outside src/heap/ includes
heap-inl.h.
Bug: v8:8562,v8:8499
Change-Id: I65fa763f90e66afc30d105b9277792721f05a6d4
Reviewed-on: https://chromium-review.googlesource.com/c/1459659
Commit-Queue: Jakob Kummerow <jkummerow@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#59617}
As opposed to the register.
For subtle reasons, this fixes a deoptimizer bug with handling return
values in lazy deopt. Since the return values can now only overwrite
the accumulator, there is no danger of overwriting a captured object
that might be later used (since there is no "later").
Bug: chromium:902608
Change-Id: I3a7a10bb1c7a6f4303a01d60f80680afcb7bc942
Reviewed-on: https://chromium-review.googlesource.com/c/1325901
Reviewed-by: Leszek Swirski <leszeks@chromium.org>
Commit-Queue: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#57349}
This is a reland of eccf186749
Original change's description:
> [interpreter] Separate bytecodes for one-shot property loads and stores
>
> Create LdaNamedPropertyNoFeedback and StaNamedPropertyNoFeedback
> for one-shot property loads and stores. This CL replaces the runtime
> calls with new bytecodes for named property load stores in one-shot code.
> the runtime calls needed extra set of consecutive registers and
> additional move instructions. This increased the size of
> bytecode-array and possibly extended the life time of objects.
> By replacing them with NoFeedback bytecodes we avoid these issues.
>
> Bug: v8:8072
> Change-Id: I20a38a5ce9940026171d870d354787fe0b7c5a6f
> Reviewed-on: https://chromium-review.googlesource.com/1196725
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Commit-Queue: Chandan Reddy <chandanreddy@google.com>
> Cr-Commit-Position: refs/heads/master@{#56211}
Bug: v8:8072
Change-Id: Ie8e52b37daf35c7bc08bb910d7b15a9b783354e4
Reviewed-on: https://chromium-review.googlesource.com/1245742
Commit-Queue: Chandan Reddy <chandanreddy@google.com>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#56266}
This reverts commit eccf186749.
Reason for revert: Speculative revert because it seems to introduce a pretty stable flake on gc stress tests, see https://bugs.chromium.org/p/v8/issues/detail?id=8229
Original change's description:
> [interpreter] Separate bytecodes for one-shot property loads and stores
>
> Create LdaNamedPropertyNoFeedback and StaNamedPropertyNoFeedback
> for one-shot property loads and stores. This CL replaces the runtime
> calls with new bytecodes for named property load stores in one-shot code.
> the runtime calls needed extra set of consecutive registers and
> additional move instructions. This increased the size of
> bytecode-array and possibly extended the life time of objects.
> By replacing them with NoFeedback bytecodes we avoid these issues.
>
> Bug: v8:8072
> Change-Id: I20a38a5ce9940026171d870d354787fe0b7c5a6f
> Reviewed-on: https://chromium-review.googlesource.com/1196725
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Commit-Queue: Chandan Reddy <chandanreddy@google.com>
> Cr-Commit-Position: refs/heads/master@{#56211}
TBR=rmcilroy@chromium.org,yangguo@chromium.org,jarin@chromium.org,neis@chromium.org,cbruni@chromium.org,chandanreddy@google.com
Change-Id: I445db58e6d4c275b434fabad5fad775bf259033f
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:8072
Reviewed-on: https://chromium-review.googlesource.com/1245421
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#56232}
Create LdaNamedPropertyNoFeedback and StaNamedPropertyNoFeedback
for one-shot property loads and stores. This CL replaces the runtime
calls with new bytecodes for named property load stores in one-shot code.
the runtime calls needed extra set of consecutive registers and
additional move instructions. This increased the size of
bytecode-array and possibly extended the life time of objects.
By replacing them with NoFeedback bytecodes we avoid these issues.
Bug: v8:8072
Change-Id: I20a38a5ce9940026171d870d354787fe0b7c5a6f
Reviewed-on: https://chromium-review.googlesource.com/1196725
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Chandan Reddy <chandanreddy@google.com>
Cr-Commit-Position: refs/heads/master@{#56211}
Remove %ToPrimitive, %ToPrimitive_Number, %SameValue and %SameValueZero,
as these runtime functions were only used from tests. For the %SameValue
we use Object.is() to test the internal algorithm (the actual one even),
and for %SameValueZero we use Set#has() - this was already the case for
most uses anyways.
Also drop %IsDate and %ValueOf, which didn't have uses at all.
Bug: v8:8015
Change-Id: Ice26d25e68aed4d5d8adac0547c56aedf9826b13
Reviewed-on: https://chromium-review.googlesource.com/1237677
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#56127}
We do not have to collect feedback for function calls in one-shot code.
This CL avoids allocating CallICslots for each function call by
emitting CallNoFeedback bytecodes. We save one CallICSlot (two entries
in feedback vector) per function call in One-shot.
Bug: v8:8072
Cq-Include-Trybots: luci.chromium.try:linux_chromium_headless_rel;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: Ic2580e5972acd5124c2e71d540985736ce797fe8
Reviewed-on: https://chromium-review.googlesource.com/1178051
Commit-Queue: Chandan Reddy <chandanreddy@google.com>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Camillo Bruni <cbruni@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55951}
This is a reland of 1c48d52bb1.
It turned out that IterableToList doesn't always behave according to
the ES operation with the same name. Specifically, it allows holey arrays
to take its fast path, which produces an output array with holes where
actually "undefined" elements should appear.
This CL changes the version of IterableToList that is used for spreads
(IterableToListWithSymbolLookup) such that holey arrays take the slow path.
It also includes tests for such situations.
Original change's description:
> [interpreter] Add bytecode for leading array spreads.
>
> This CL improves the performance of creating [...a, b] or [...a].
> If the array literal has a leading spread, this CL emits the bytecode
> [CreateArrayFromIterable] to create the literal. CreateArrayFromIterable
> is implemented by [IterableToListDefault] builtin to create the initial
> array for the leading spread. IterableToListDefault has a fast path to
> clone efficiently if the spread is an actual array.
>
> The bytecode generated is now shorter. Bytecode generation is refactored
> into to BuildCreateArrayLiteral, which allows VisitCallSuper to benefit
> from this optimization also.
> For now, turbofan also lowers the bytecode to the builtin.
>
> The idiomatic use of [...a] to clone the array a now performs better
> than a simple for-loop, but still does not match the performance of slice.
>
> Bug: v8:7980
>
> Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: Ibde659c82d3c7aa1b1777a3d2f6426ac8cc15e35
> Reviewed-on: https://chromium-review.googlesource.com/1181024
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Commit-Queue: Georg Neis <neis@chromium.org>
> Commit-Queue: Hai Dang <dhai@google.com>
> Cr-Commit-Position: refs/heads/master@{#55520}
Bug: v8:7980
Change-Id: I0b5603a12d2b588327658bf0a9b214bd0f22e237
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1201882
Commit-Queue: Hai Dang <dhai@google.com>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55639}
This reverts commit 1c48d52bb1.
Reason for revert: Clusterfuzz found something.
Original change's description:
> [interpreter] Add bytecode for leading array spreads.
>
> This CL improves the performance of creating [...a, b] or [...a].
> If the array literal has a leading spread, this CL emits the bytecode
> [CreateArrayFromIterable] to create the literal. CreateArrayFromIterable
> is implemented by [IterableToListDefault] builtin to create the initial
> array for the leading spread. IterableToListDefault has a fast path to
> clone efficiently if the spread is an actual array.
>
> The bytecode generated is now shorter. Bytecode generation is refactored
> into to BuildCreateArrayLiteral, which allows VisitCallSuper to benefit
> from this optimization also.
> For now, turbofan also lowers the bytecode to the builtin.
>
> The idiomatic use of [...a] to clone the array a now performs better
> than a simple for-loop, but still does not match the performance of slice.
>
> Bug: v8:7980
>
> Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: Ibde659c82d3c7aa1b1777a3d2f6426ac8cc15e35
> Reviewed-on: https://chromium-review.googlesource.com/1181024
> Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
> Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
> Reviewed-by: Jakob Gruber <jgruber@chromium.org>
> Reviewed-by: Georg Neis <neis@chromium.org>
> Commit-Queue: Georg Neis <neis@chromium.org>
> Commit-Queue: Hai Dang <dhai@google.com>
> Cr-Commit-Position: refs/heads/master@{#55520}
TBR=rmcilroy@chromium.org,neis@chromium.org,sigurds@chromium.org,gsathya@chromium.org,jgruber@chromium.org,dhai@google.com
Change-Id: I1c86ddcc24274da9f5a8dd3d8bf8d869cbb55cb6
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:7980
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1199303
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55544}
This CL improves the performance of creating [...a, b] or [...a].
If the array literal has a leading spread, this CL emits the bytecode
[CreateArrayFromIterable] to create the literal. CreateArrayFromIterable
is implemented by [IterableToListDefault] builtin to create the initial
array for the leading spread. IterableToListDefault has a fast path to
clone efficiently if the spread is an actual array.
The bytecode generated is now shorter. Bytecode generation is refactored
into to BuildCreateArrayLiteral, which allows VisitCallSuper to benefit
from this optimization also.
For now, turbofan also lowers the bytecode to the builtin.
The idiomatic use of [...a] to clone the array a now performs better
than a simple for-loop, but still does not match the performance of slice.
Bug: v8:7980
Cq-Include-Trybots: luci.v8.try:v8_linux_noi18n_rel_ng
Change-Id: Ibde659c82d3c7aa1b1777a3d2f6426ac8cc15e35
Reviewed-on: https://chromium-review.googlesource.com/1181024
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Georg Neis <neis@chromium.org>
Commit-Queue: Hai Dang <dhai@google.com>
Cr-Commit-Position: refs/heads/master@{#55520}
As discussed in
https://docs.google.com/document/d/1sBdGe8RHgeYP850cKSSgGABTyfMdvaEWLy-vertuTCo/edit?ts=5b3ba5cc#,
this CL introduces a new bytecode (CloneObject), and a new IC type.
In this prototype implementation, the type feedback looks like the
following:
Uninitialized case:
{ uninitialized_sentinel, uninitialized_sentinel }
Monomorphic case:
{ weak 'source' map, strong 'result' map }
Polymorphic case:
{ WeakFixedArray with { weak 'source' map, strong 'result' map }, cleared value }
Megamorphic case:
{ megamorphic_sentinel, cleared_Value }
In the fast case, Object cloning is done by allocating an object with
the saved result map, and a shallow clone of the fast properties from
the source object, as well as cloned fast elements from the source object.
If at any point the fast case can't be taken, the IC transitions to the
slow case and remains there.
This prototype CL does not include any TurboFan optimization, and the
CloneObject operation is merely reduced to a stub call.
It may still be possible to get some further improvements by somehow
incorporating compile-time boilerplate elements into the cloned object,
or simplifying how the boilerplate elements are inserted into the
object.
In terms of performance, we improve the ObjectSpread score in JSTests/ObjectLiteralSpread/
by about 8x, with substantial improvements over the Babel and ObjectAssign scores.
R=gsathya@chromium.org, mvstanton@chromium.org, rmcilroy@chromium.org, neis@chromium.org, bmeurer@chromium.org
BUG=v8:7611
Change-Id: I79e1796eb77016fb4feba0e1d3bb9abb348c183e
Reviewed-on: https://chromium-review.googlesource.com/1127472
Commit-Queue: Caitlin Potter <caitp@igalia.com>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Michael Stanton <mvstanton@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#54595}
Currently each isolate stores its own array of
{CallInterfaceDescriptorData}. This array has size 173, and each entry
has 40 bytes. That's already 7kB per isolate.
Additionally, each {CallInterfaceDescriptorData} allocates two
heap-allocated arrays, which probably add up to more than the static
size of the {CallInterfaceDescriptorData}. Note that all the
{CallInterfaceDescriptorData} instances are initialized eagerly on
isolate creation.
Since {CallInterfaceDescriptor} is totally isolate independent itself,
this CL refactors the current design to avoid a copy of them per
isolate, and instead shares them process-wide. Still, we need to free
the allocated heap arrays when the last isolate dies to avoid leaks.
This can probably be refactored later by statically initializing more
and avoiding the heap allocations all together.
This refactoring will also allow us to use {CallInterfaceDescriptor}s
from wasm background compilation threads, which are not bound to any
isolate.
R=mstarzinger@chromium.org, titzer@chromium.org
Bug: v8:6600
Change-Id: If8625b89951eec8fa8986b49a5c166e874a72494
Reviewed-on: https://chromium-review.googlesource.com/1100879
Commit-Queue: Clemens Hammacher <clemensh@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#53803}
Return the raw Object* when accessing the constant pool of bytecode
with the bytecode array accessor, to avoid needing an isolate there.
If the returned value needs to be a handle, we create the handle
later.
Bug: v8:7786
Change-Id: Ifeac2a06f0383230bf7e9bfc1b751d9750ecfb51
Reviewed-on: https://chromium-review.googlesource.com/1102334
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#53784}
Instead rely on the scope info containing the name as well.
Change-Id: Ie1f96ea023a793b11209510566f6831b1dfd40ab
Reviewed-on: https://chromium-review.googlesource.com/1042567
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#52983}
There are likely cleanups that can be done after this CL:
- context-related functions in the interpreter and compiler take ScopeInfo as
well as ScopeType and slot-count as input. The latter 2 should be directly
derived from the former. We should be able to drop FunctionContextParameters.
- ContextExtension is probably not needed anymore, since we now always have the
correct scope_info directly in the SCOPE_INFO_INDEX slot.
Bug: v8:7066
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: Ie1f6134c686a9f2183e54730d9cdd598a9e5ab67
Reviewed-on: https://chromium-review.googlesource.com/785151
Commit-Queue: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Adam Klein <adamk@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#52952}
