diff options
author | buzbee <buzbee@google.com> | 2013-09-14 16:15:25 -0700 |
---|---|---|
committer | buzbee <buzbee@google.com> | 2013-10-02 00:49:08 -0700 |
commit | b48819db07f9a0992a72173380c24249d7fc648a (patch) | |
tree | 2b02366b8e105c025b64d9b27c76dc6c05b88f53 /compiler/dex/compiler_enums.h | |
parent | 65d1b22d0b02fb0111f69013163c8170e68392f1 (diff) | |
download | android_art-b48819db07f9a0992a72173380c24249d7fc648a.tar.gz android_art-b48819db07f9a0992a72173380c24249d7fc648a.tar.bz2 android_art-b48819db07f9a0992a72173380c24249d7fc648a.zip |
Compile-time tuning: assembly phase
Not as much compile-time gain from reworking the assembly phase as I'd
hoped, but still worthwhile. Should see ~2% improvement thanks to
the assembly rework. On the other hand, expect some huge gains for some
application thanks to better detection of large machine-generated init
methods. Thinkfree shows a 25% improvement.
The major assembly change was to establish thread the LIR nodes that
require fixup into a fixup chain. Only those are processed during the
final assembly pass(es). This doesn't help for methods which only
require a single pass to assemble, but does speed up the larger methods
which required multiple assembly passes.
Also replaced the block_map_ basic block lookup table (which contained
space for a BasicBlock* for each dex instruction unit) with a block id
map - cutting its space requirements by half in a 32-bit pointer
environment.
Changes:
o Reduce size of LIR struct by 12.5% (one of the big memory users)
o Repurpose the use/def portion of the LIR after optimization complete.
o Encode instruction bits to LIR
o Thread LIR nodes requiring pc fixup
o Change follow-on assembly passes to only consider fixup LIRs
o Switch on pc-rel fixup kind
o Fast-path for small methods - single pass assembly
o Avoid using cb[n]z for null checks (almost always exceed displacement)
o Improve detection of large initialization methods.
o Rework def/use flag setup.
o Remove a sequential search from FindBlock using lookup table of 16-bit
block ids rather than full block pointers.
o Eliminate pcRelFixup and use fixup kind instead.
o Add check for 16-bit overflow on dex offset.
Change-Id: I4c6615f83fed46f84629ad6cfe4237205a9562b4
Diffstat (limited to 'compiler/dex/compiler_enums.h')
-rw-r--r-- | compiler/dex/compiler_enums.h | 21 |
1 files changed, 21 insertions, 0 deletions
diff --git a/compiler/dex/compiler_enums.h b/compiler/dex/compiler_enums.h index 97a682f2aa..17b5bb5b19 100644 --- a/compiler/dex/compiler_enums.h +++ b/compiler/dex/compiler_enums.h @@ -412,6 +412,27 @@ enum OatBitMapKind { std::ostream& operator<<(std::ostream& os, const OatBitMapKind& kind); +// LIR fixup kinds for Arm +enum FixupKind { + kFixupNone, + kFixupLabel, // For labels we just adjust the offset. + kFixupLoad, // Mostly for imediates. + kFixupVLoad, // FP load which *may* be pc-relative. + kFixupCBxZ, // Cbz, Cbnz. + kFixupPushPop, // Not really pc relative, but changes size based on args. + kFixupCondBranch, // Conditional branch + kFixupT1Branch, // Thumb1 Unconditional branch + kFixupT2Branch, // Thumb2 Unconditional branch + kFixupBlx1, // Blx1 (start of Blx1/Blx2 pair). + kFixupBl1, // Bl1 (start of Bl1/Bl2 pair). + kFixupAdr, // Adr. + kFixupMovImmLST, // kThumb2MovImm16LST. + kFixupMovImmHST, // kThumb2MovImm16HST. + kFixupAlign4, // Align to 4-byte boundary. +}; + +std::ostream& operator<<(std::ostream& os, const FixupKind& kind); + } // namespace art #endif // ART_COMPILER_DEX_COMPILER_ENUMS_H_ |