| Age | Commit message (Collapse) | Author |
|---|
|
Avoid three dependent loads to fetch the allocator on the
hot paths. Inline the `FixupUserRecordsAfter*UseInsertion()`
loop and use the fact that it's known to execute exactly one
or two iterations.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Bug: 181943478
Change-Id: I7fd4d48caebc6aeb13fb9a9f8146a06129c72b2e
|
|
It was added back in 2020 (r.android.com/1206763) but never used.
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Change-Id: Iace09f956bd520732588b3623722b74f6559a1fe
|
|
Based on a comment in r.android.com/3365219.
Bug: 368984521
Test: art/test/testrunner/testrunner.py -b --host --64 --optimizing
Change-Id: Ic8b945ba418df380158456bfb295edfd1dc0ad01
|
|
This resulted in a crash for debug builds, and mistakes in the
optimizations for release builds.
It requires a variable to be used in the loop we are unrolling
as well as in a catch phi that it is not part of the loop we
are unrolling.
We could potentially fully unroll that loop, but this requires
a refactor on how SuperblockCloner works. This CL makes it so
that we won't optimize such loops, which should be rare enough
to not cause any visible regressions.
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Bug: 368984521
Fixes: 368984521
Change-Id: I39f88586358573bbe76d9369d4ec67a942fd3eec
|
|
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Bug: 181943478
Change-Id: Ie05d4001e411a669e11b8edda375414e5da52ae2
|
|
ArenaBitVector creation guarantees it starts empty. Add a debug
check to make sure this assumption doesn't change.
Note that ArenaAllocator guarantees zero-initialized memory but
ScopedArenaAllocators do not. This is fine either way since the
BitVector constructor calls ClearAllBits.
Bug: 329037671
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Change-Id: Icbf5e5dd1869e80b5d5828ecca9f13de30c0242b
|
|
Bug: 328756212
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Test: m test-art-host-gtest
Change-Id: I9584e1b93e49265b84a9e45c8b283ebaf8ad3eb2
|
|
After the old implementation was renamed in
https://android-review.googlesource.com/2526708 ,
we introduce a new function with the old name but new
behavior, just `DCHECK()`-ing the instruction kind before
casting down the pointer. We change appropriate calls from
`As##type##OrNull()` to `As##type()` to avoid unncessary
run-time checks and reduce the size of libart-compiler.so.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Test: run-gtests.sh
Test: testrunner.py --target --optimizing
Bug: 181943478
Change-Id: I025681612a77ca2157fed4886ca47f2053975d4e
|
|
The null type check in the current implementation of
`HInstruction::As##type()` often cannot be optimized away
by clang++. It is therefore beneficial to have two functions
HInstruction::As##type()
HInstruction::As##type##OrNull()
where the first function never returns null but the second
one can return null. The additional text "OrNull" shall also
flag the possibility of yielding null to the developer which
may help avoid bugs similar to what we have seen previously.
This requires renaming the existing function that can return
null and introducing new function that cannot. However,
defining the new function `HInstruction::As##type()` in the
same change as renaming the old one would risk introducing
bugs by missing a rename. Therefore we simply rename the old
function here and the new function shall be introduced in a
separate change with all behavioral changes being explicit.
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Test: buildbot-build.sh --target
Bug: 181943478
Change-Id: I4defd85038e28fe3506903ba3f33f723682b3298
|
|
Follow-up to aosp/2442280. We haven't seen crashes with these ones,
but we can't guarantee that the RTI will be valid in these code paths.
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Change-Id: I80da85a6549ba0275a80027016363e0cf9fb8045
|
|
This reverts commit 0a51605ddd81635135463dab08b6f7c21b58ffb0.
Reason for revert: Reland after some of the required work
was merged in other CLs.
Also address a TODO from the original CL to mark required
symbols with EXPORT in `intrinsic_objects.h`.
Also mark symbols in new files as HIDDEN.
Bug: 186902856
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Change-Id: I936d448983928af23614ca82c2d0bf9a645e2c52
|
|
We can enable the optimization when the graph has try catch blocks,
but we do not perform the optimizations if the loops themselves
have try catch blocks.
Bug: 227283906
Change-Id: I8889d2ed7a5a260d5c2dcbbc5184cb7eaf3a8f9f
|
|
Follow-up to aosp/1988868 in which we added the (D)CHECK_IMPLIES
macro. This CL uses it on compiler/ occurrences found by a regex.
Test: art/test/testrunner/testrunner.py --host --64 --optimizing -b
Change-Id: If63aed969bfb8b31d6fbbcb3bca2b04314c894b7
|
|
Support Loop Versioning in SuberblockCloner as a tool to
enable further optimization (e.g. Dynamic Loop Unrolling).
The patch brings the feature in without enabling it.
Replace std::cout with LOG(INFO) for debug dumps.
Test: superblock_cloner_test.
Test: test-art-target.
Change-Id: I303cabfb752b8c3c8597abfc0ac261e8616e8cee
|
|
ART vectorizer assumes that there is single size of SIMD
register used for the whole program. Make this assumption explicit
and refactor the code.
Note: This is a base for the future introduction of SIMD slots of
size other than 8 or 16 bytes.
Test: test-art-target, test-art-host.
Change-Id: Id699d5e3590ca8c655ecd9f9ed4e63f49e3c4f9c
|
|
This reverts commit e2727154f25e0db9a5bb92af494d8e47b181dfcf.
Reason for revert: Breaks ASAN tests (ODR violation).
Bug: 142365358
Change-Id: I38103d74a1297256c81d90872b6902ff1e9ef7a4
|
|
Make symbols in compiler/optimizing hidden by a namespace
attribute. The unit intrinsic_objects.{h,cc} is excluded as
it is needed by dex2oat.
As the symbols are no longer exported, gtests are now linked
with the static version of the libartd-compiler library.
libart-compiler.so size:
- before:
arm: 2396152
arm64: 3345280
- after:
arm: 2016176 (-371KiB, -15.9%)
arm64: 2874480 (-460KiB, -14.1%)
Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing --jit
Bug: 142365358
Change-Id: I1fb04a33351f53f00b389a1642e81a68e40912a8
|
|
Because loop unrolling is part of a general loop optimization pass,
it needs to update induction ranges as it will invalidate its
instruction cache with new instructions.
Bug: 131174583
Test: 696-loop
Change-Id: Id3628efe316b58f69abbd9ebd43e891a8e42529f
|
|
Change the default parameters for HashSet<std::string> to
allow passing StringPiece as a key, avoiding an unnecessary
allocation. Use the HashSet<std::string> instead of
std::unordered_set<std::string>. Rename HashSet<> functions
that mirror std::unordered_multiset<> to lower-case.
Fix CompilerDriver::LoadImageClasses() to avoid using
invalidated iterator.
Test: m test-art-host-gtest
Test: testrunner.py --host
Change-Id: I7f8b82ee0b07befc5a0ee1c420b08a2068ad931e
|
|
Support copying subgraphs with a single exit and live-outs -
values which are defined inside the subgraph and have uses outside.
Test: test-art-target, test-art-host.
Test: 530-checker-peel-unroll.
Change-Id: Id06a6f08d14c6e86f6437d662e24489f955e9edf
|
|
Implement scalar loop peeling for invariant exits elimination
(on arm64). If the loop exit condition is loop invariant then
loop peeling + GVN + DCE can eliminate this exit in the loop
body. Note: GVN and DCE aren't applied during loop optimizations.
Note: this functionality is turned off by default now.
Test: test-art-host, test-art-target, boot-to-gui.
Change-Id: I98d20054a431838b452dc06bd25c075eb445960c
|
|
Implement scalar loop unrolling for small loops
(on arm64) with known trip count to reduce loop check
and branch penalty and to provide more opportunities
for instruction scheduling.
Note: this functionality is turned off by default now.
Test: cloner_test.cc
Test: test-art-target, test-art-host
Change-Id: Ic27fd8fb0bc0d7b69251252da37b8b510bc30acc
|
|
Implement loop peeling/unrolling routines using SuperblockCloner.
Fixes bug b/74198030 and provides tests for it.
Bug: b/74198030
Test: superblock_cloner_test.cc, loop_optimization_test.cc.
Change-Id: Id0d9a91575c88f8e45186441b14e903b89b007dd
|
|
SuperblockCloner provides a feature of cloning subgraphs in a
smart, high level way without fine grain manipulation with IR;
data flow and graph properties are resolved/adjusted automatically.
The clone transformation is defined by specifying a set of basic
blocks to copy and a set of rules how to treat edges, remap their
successors. By using this approach such optimizations as Branch
Target Expansion, Loop Peeling, Loop Unrolling can be implemented.
Test: superblock_cloner_test.cc.
Change-Id: Ibeede38195376ca35f44ba9015491e50b3a5b87e
|