 masta
|
073cab8d86
Optimize FoldShiftLdrStr in ARM Peephole optimizer
|
12 years ago |
 Jeppe Johansen
|
c4263ced51
Disable one peephole optimization for Thumb
|
12 years ago |
 florian
|
c0d4f189e9
* and takes three operands
|
12 years ago |
|
florian
|
be81b89147
* fix bugs in previous commit
|
12 years ago |
|
florian
|
086ae4b999
Merge r22905 and r22906
|
12 years ago |
|
florian
|
1eeeb309c7
* intial armv6m support, it is not working yet, constant pool insertation and conditional branch fixup is not working yet
|
12 years ago |
|
masta
|
96ed4e7932
Fixed access to freed memory in ShiftShiftShift2ShiftShift 1a on ARM
|
12 years ago |
|
florian
|
ef1c848198
* fix constant calculation for the AndAnd2And optimization if the first and is removed, noticed by Nico Erfurth
|
12 years ago |
|
florian
|
9d76168b57
* refactored Bl2B condition code to make it more readable
|
12 years ago |
|
florian
|
7184306a4c
+ option -Wx to generate thumb interworking safe code on arm
|
12 years ago |
|
florian
|
3fff969ced
* take care of instruction in between when doing ShiftShiftShift2ShiftShift optimizations
|
12 years ago |
|
florian
|
f1b8fad6dc
* make AndAnd2And working with other instructions in between
|
12 years ago |
|
florian
|
fcef2dc3df
* refactored some code and introduced usage of new method RegEndOfLife
|
12 years ago |
|
florian
|
82f3ec7922
* if cs_opt_level3 is not set, limit lookahead of GetNextInstructionUsingReg to one instruction:
|
12 years ago |
|
florian
|
a78af5b8fe
+ AndLslXsr2And and AndLsl2Lsl optimization
|
12 years ago |
|
florian
|
ff522d7e18
* improve ShiftShiftShift2ShiftShift to look further ahead
|
12 years ago |
|
florian
|
12d0c05ede
* remove bic instructions after lsr if possible
|
12 years ago |
|
florian
|
abfa6c1b43
* redo LsrAnd2Lsr optimization
|
12 years ago |
|
masta
|
fe520c215b
New ARM Peephole optimizer FoldShiftLdrStr
|
12 years ago |
|
florian
|
47d43750e4
* remove unused units from uses statements
|
12 years ago |
|
florian
|
903f18ea38
* get rid of calls which redirect the program flow only, Bl2B optimization
|
12 years ago |
|
florian
|
ff98d2567c
* don't crash on thumb instructions with only two operands when optimizing ADD, SUB, AND statements
|
12 years ago |
|
masta
|
1261d6617d
Properly handle MVN in RedundantMovProcess for ARM
|
12 years ago |
|
masta
|
3a017f76d0
Look ahead more than one instruction in FoldShiftProcess for ARM
|
12 years ago |
|
florian
|
8c73b0b17b
* disable broken MvnAnd2Bic optimization
|
12 years ago |
|
masta
|
e91b15b2a4
Disabled MulAdd2MLA and MulSub2MLS Peephole optimizers for thumb2
|
12 years ago |
|
florian
|
970405c0f3
o merging r22801 of Jeppe Johansen
|
12 years ago |
|
Jeppe Johansen
|
4e84431dde
Fix some optimizations which assume that there are 3 operands
|
12 years ago |
|
florian
|
04543b179f
o merge of the branch laksen/arm-embedded of Jeppe Johansen:
|
12 years ago |
|
Jeppe Johansen
|
5751bbecee
Properly do NR_DEFAULTFLAGS detection/allocation/deallocation
|
12 years ago |
