Gitiles
Code Review
Sign In
gerrit-public.fairphone.software
/
platform
/
art
/
1f47b678db70d57aa5194266651a8f4336fc851e
/
compiler
/
optimizing
0580d96
Fix a crash with unresolved classes.
by Nicolas Geoffray
· 9 years ago
744a1c6
ART: Don't set initial RTI for BoundType if input untyped
by David Brazdil
· 9 years ago
15693bf
ART: Resolve ambiguous ArraySets
by David Brazdil
· 9 years ago
f555258
ART: Create BoundType for CheckCast early
by David Brazdil
· 9 years ago
fd2140f
ART: Make opt inliner a little bit cleaner/faster
by Andreas Gampe
· 9 years ago
d87f3ea
ART: Use Primitive::Is64BitType in SsaBuilder::TypePhiFromInputs
by David Brazdil
· 9 years ago
f196a43
Merge "X86: templatize GenerateTestAndBranch and friends"
by David Brazdil
· 9 years ago
06856d3
Merge "Detect phi cycles."
by Nicolas Geoffray
· 9 years ago
a3f0bf3
Merge "Revert "Revert "Tweak inlining heuristics."""
by Nicolas Geoffray
· 9 years ago
5949fa0
Revert "Revert "Tweak inlining heuristics.""
by Nicolas Geoffray
· 9 years ago
5f332cb
Merge "MIPS32: improvements in code generation (mostly 64-bit ALU ops)"
by Nicolas Geoffray
· 9 years ago
b7371a5
Merge "Remove bogus DCHECK in induction analysis."
by Nicolas Geoffray
· 9 years ago
152408f
X86: templatize GenerateTestAndBranch and friends
by Mark Mendell
· 9 years ago
b35302b
Remove bogus DCHECK in induction analysis.
by Nicolas Geoffray
· 9 years ago
295abc1
ART: Set RTI of HArm64IntermediateAddress
by David Brazdil
· 9 years ago
4833f5a
ART: Refactor SsaBuilder for more precise typing info
by David Brazdil
· 9 years ago
5d75afe
Improved side-effects/can-throw information on intrinsics.
by Aart Bik
· 9 years ago
fa0dc72
Merge "On x64, cmpl can never take a int64 immediate."
by Nicolas Geoffray
· 9 years ago
6ce0173
On x64, cmpl can never take a int64 immediate.
by Nicolas Geoffray
· 9 years ago
1c421aa
Merge "Fix code generation for String.<init> on x64."
by Nicolas Geoffray
· 9 years ago
7f59d59
Fix code generation for String.<init> on x64.
by Nicolas Geoffray
· 9 years ago
e6d0d8d
ART: Disable Math.round intrinsics
by Andreas Gampe
· 9 years ago
095b1df
Revert "Make Math.round consistent on arm64."
by Andreas Gampe
· 9 years ago
40041c9
Make Math.round consistent on arm64.
by Nicolas Geoffray
· 9 years ago
dcdc85b
Dex2oat support for multiple oat file and image file outputs.
by Jeff Hao
· 9 years ago
0cf4493
Generate more stack maps during native debugging.
by David Srbecky
· 9 years ago
5f7b58e
Rewrite HInstruction::Is/As<type>().
by Vladimir Marko
· 9 years ago
9865bde
Rename NullHandle to ScopedNullHandle
by Mathieu Chartier
· 9 years ago
803cbb9
For LSE, further optimize stores for singleton references.
by Mingyao Yang
· 9 years ago
280a65b
Merge "MIPS64: Fuse long and FP compare & condition in Optimizing."
by Roland Levillain
· 9 years ago
ecf52df
ART: Fix bug in LSE
by David Brazdil
· 9 years ago
2739411
Merge "Disable the UnsafeCASObject intrinsic with read barriers."
by Roland Levillain
· 9 years ago
391b866
Disable the UnsafeCASObject intrinsic with read barriers.
by Roland Levillain
· 9 years ago
570a920
Merge "Revert "Revert "X86: Use locked add rather than mfence"""
by Aart Bik
· 9 years ago
299a939
MIPS64: Fuse long and FP compare & condition in Optimizing.
by Alexey Frunze
· 9 years ago
14c4e90
Merge "Revert "Revert "ART: Reduce the instructions generated by packed switch."""
by Vladimir Marko
· 9 years ago
f3e0ee2
Revert "Revert "ART: Reduce the instructions generated by packed switch.""
by Vladimir Marko
· 9 years ago
3e3e4a7
Fix braino in parallel move resolver.
by Nicolas Geoffray
· 9 years ago
17077d8
Revert "Revert "X86: Use locked add rather than mfence""
by Mark P Mendell
· 9 years ago
5c7aed3
MIPS32: improvements in code generation (mostly 64-bit ALU ops)
by Alexey Frunze
· 9 years ago
1c70f18
Merge "Revert "X86: Use locked add rather than mfence""
by Aart Bik
· 9 years ago
0da3b91
Revert "X86: Use locked add rather than mfence"
by Aart Bik
· 9 years ago
c3ca1e6
Merge "X86: Use locked add rather than mfence"
by Aart Bik
· 9 years ago
698fa97
Remove spurious references to kEmitCompilerReadBarrier in MIPS.
by Roland Levillain
· 9 years ago
cbf8af8
Merge "MIPS32: Fuse long and FP compare & condition in Optimizing."
by Roland Levillain
· 9 years ago
4741516
Merge "Revert "Revert "Introduce support for hardware simulators, starting with ARM64"""
by Roland Levillain
· 9 years ago
d7d3538
Merge "Revert "ART: Reduce the instructions generated by packed switch.""
by Nicolas Geoffray
· 9 years ago
f5f64ef
Detect phi cycles.
by Nicolas Geoffray
· 9 years ago
b4c1376
Revert "ART: Reduce the instructions generated by packed switch."
by Nicolas Geoffray
· 9 years ago
96f721d
Merge "Revert "ART: Set RTI of Arm64IntermediateAddress""
by Nicolas Geoffray
· 9 years ago
dce90b9
Revert "ART: Set RTI of Arm64IntermediateAddress"
by Nicolas Geoffray
· 9 years ago
68289a5
Revert "ART: Refactor SsaBuilder for more precise typing info"
by Alex Light
· 9 years ago
cd7b0ee
MIPS32: Fuse long and FP compare & condition in Optimizing.
by Alexey Frunze
· 9 years ago
7b3e4f9
X86: Use locked add rather than mfence
by Mark Mendell
· 9 years ago
7d57d7f
Various induction/range analysis improvements.
by Aart Bik
· 9 years ago
1e7f8db
x86-64 Baker's read barrier fast path implementation.
by Roland Levillain
· 9 years ago
7c1559a
x86 Baker's read barrier fast path implementation.
by Roland Levillain
· 9 years ago
e36ae94
ART: Set RTI of Arm64IntermediateAddress
by David Brazdil
· 9 years ago
d9510df
ART: Refactor SsaBuilder for more precise typing info
by David Brazdil
· 9 years ago
351dddf
Optimizing: Clean up after HRor.
by Vladimir Marko
· 9 years ago
58dcb02
Merge "Replace rotate patterns and invokes with HRor IR."
by Vladimir Marko
· 9 years ago
40a04bf
Replace rotate patterns and invokes with HRor IR.
by Scott Wakeling
· 9 years ago
bf479be
Merge "Don't generate a slow path for strings in the dex cache."
by Nicolas Geoffray
· 9 years ago
376cbcc
Merge "Optimizing: Add direct calls to math intrinsics"
by Nicolas Geoffray
· 9 years ago
8ab1d64
For LSE, add a few non-aliasing cases due to pre-existence.
by Mingyao Yang
· 9 years ago
a4f1220
Optimizing: Add direct calls to math intrinsics
by Mark Mendell
· 9 years ago
917d016
Don't generate a slow path for strings in the dex cache.
by Nicolas Geoffray
· 9 years ago
f71b3ad
Get source mapping table from stack maps.
by David Srbecky
· 9 years ago
d480156
Merge "Fix braino when resolving an invoke-super."
by Nicolas Geoffray
· 9 years ago
41844e5
Fix braino when resolving an invoke-super.
by Nicolas Geoffray
· 9 years ago
4a34064
Merge "ART: Fix bug in DCE not removing phis from catch phi uses"
by David Brazdil
· 9 years ago
04ff4e8
ART: Fix bug in DCE not removing phis from catch phi uses
by David Brazdil
· 9 years ago
6d0aefd
Merge "MIPS32: Improve integer division by constants"
by Nicolas Geoffray
· 9 years ago
f1975e4
Merge "Simplify and rename IsLoopInvariant() test."
by Mingyao Yang
· 9 years ago
2c1164c
Merge "ART: Reduce the instructions generated by packed switch."
by Vladimir Marko
· 9 years ago
4b467ed
Simplify and rename IsLoopInvariant() test.
by Mingyao Yang
· 9 years ago
eb5ddd7
Merge "ART: Check invoke-interface earlier in verifier"
by Andreas Gampe
· 9 years ago
59f054d
ART: Reduce the instructions generated by packed switch.
by Zheng Xu
· 9 years ago
dae2414
ART: Check invoke-interface earlier in verifier
by Andreas Gampe
· 9 years ago
8993caf
ART: Enable all optimizations except LSE under try/catch
by David Brazdil
· 9 years ago
bf84a3d
Annotate Boolean literals more uniformly in Optimizing's intrinsics.
by Roland Levillain
· 9 years ago
038f4f2
Merge "Fix art::x86::IntrinsicLocationsBuilderX86::VisitUnsafeGetLong."
by Roland Levillain
· 9 years ago
5bde68f
Merge "MIPS32: Bit rotation intrinsics"
by Andreas Gampe
· 9 years ago
fdacff8
Merge "MIPS32: int java.lang.*.numberOfTrailingZeros"
by Andreas Gampe
· 9 years ago
e15f6e2
Merge "MIPS32: int java.lang.*.numberOfLeadingZeros"
by Andreas Gampe
· 9 years ago
d382994
Merge "Step-wise improvement of range analysis with outer loop induction."
by Aart Bik
· 9 years ago
eddbfb7
Merge "MIPS32: java.lang.*.reverse"
by Andreas Gampe
· 9 years ago
73be1e8
Inline monomorphic calls.
by Nicolas Geoffray
· 9 years ago
b9f8119
Fix art::x86::IntrinsicLocationsBuilderX86::VisitUnsafeGetLong.
by Roland Levillain
· 9 years ago
b738d4f
Step-wise improvement of range analysis with outer loop induction.
by Aart Bik
· 9 years ago
47229aa
Merge "Revert "Revert "Don't use the compiler driver for method resolution."""
by Nicolas Geoffray
· 9 years ago
e523423
Revert "Revert "Don't use the compiler driver for method resolution.""
by Nicolas Geoffray
· 9 years ago
b5cdc5d
Merge "Clean up optimizing compiler stats"
by Calin Juravle
· 9 years ago
d585b2c
Merge "Revert "Revert "ART: Do not eagerly type LoopPhi [null, ...]"""
by David Brazdil
· 9 years ago
fe86070
Revert "Revert "ART: Do not eagerly type LoopPhi [null, ...]""
by David Brazdil
· 9 years ago
874121e
Merge "Optimizing: Add checker tests for sharpening."
by Vladimir Marko
· 9 years ago
f64242a
Optimizing: Add checker tests for sharpening.
by Vladimir Marko
· 9 years ago
08a84ac
Merge "Revert "ART: Do not eagerly type LoopPhi [null, ...]""
by David Brazdil
· 9 years ago
5f3fb18
Revert "ART: Do not eagerly type LoopPhi [null, ...]"
by David Brazdil
· 9 years ago
4a34277
Dynamic BCE (based on induction range analysis)
by Aart Bik
· 9 years ago
Next »