 florian
|
637976e83f
* patch by Marģers to unify internal error numbers, resolves #37888
|
4 years ago |
|
florian
|
8c5606b41d
+ support mmx shifting
|
7 years ago |
|
florian
|
c1de51454c
ti386shlshrnode.second_64bit:
|
7 years ago |
 nickysn
|
af235cae86
* use TEST CL,32 instead of TEST ECX,32 in the beginning of a 64-bit shl/shr
|
8 years ago |
|
nickysn
|
d7b8d8dd54
* don't emit the "SUB ECX,32" instruction on i386, when doing a 64-bit shift by
|
8 years ago |
|
nickysn
|
7e8c89435f
* avoid the AND instruction in the i386 shr64/shl64 code, by using TEST+JZ,
|
8 years ago |
|
nickysn
|
6580dfee39
* generate better i386 code for 64-bit shl/shr, by masking the shift count by
|
8 years ago |
 sergei
|
5356f17fa5
* i386: switch the div/mod node to shared code, leaving in place the specific optimization for division by power of 2.
|
11 years ago |
|
sergei
|
ed46a07f62
* Using x86-specific capability of shifting with carry flag helps to reduce amount of instructions.
|
11 years ago |
|
sergei
|
92cf25b9a5
* Reworked i386 division by constant optimization to reuse code from powerpc64. The algorithm is slightly different, signed version is one instruction shorter, unsigned one is one instruction longer (typically). The new algorithm is easily scalable for x86_64 target, unlike the old one.
|
11 years ago |
 Jonas Maebe
|
edd42aa42a
* moved subsetref/reg and bit_set/test support from cgobj to hlcgobj for
|
13 years ago |
|
Jonas Maebe
|
9ec3cc2022
* fixed sign of location.size for divmodn, 64 shlshrn and for muln
|
17 years ago |
 daniel
|
9adb202a92
* Rework the constexprint to allow operations from low(int64) to high(qword).
|
18 years ago |
 peter
|
658c46b903
* remove tdictionary and tindexarray
|
19 years ago |
|
florian
|
85d63d9fa9
* settings refactored
|
19 years ago |
|
peter
|
3078a1927f
* remove ttype
|
19 years ago |
|
peter
|
b7fe6797bf
Merged revisions 2921-2922,2925 via svnmerge from
|
19 years ago |
|
peter
|
588cccb3ca
* remove some obsolete i386 specific code
|
19 years ago |
|
peter
|
d8dc5679e3
* aktcputype and aktoptimizecputype was missing in the previous optimization commits
|
19 years ago |
|
peter
|
b6e35a200e
* rewrite of optimizer options
|
19 years ago |
|
peter
|
4bee83c736
* range check errors
|
19 years ago |
|
florian
|
bec0daebfa
+ optimized <dword> div <const dword>
|
19 years ago |
|
peter
|
5942d23b1b
* fix range error
|
19 years ago |
|
florian
|
3582108be1
* signed divisions with constants are always done without idiv
|
19 years ago |
|
peter
|
a3ab2053c9
* support multiple asmlabel types, renamed getlabel to
|
20 years ago |
 fpc
|
790a4fe2d3
* log and id tags removed
|
20 years ago |
|
fpc
|
50778076c3
initial import
|
20 years ago |
|
peter
|
e417e34496
* truncate log
|
20 years ago |
|
peter
|
c95a859f0a
* generic tlocation
|
21 years ago |
|
peter
|
65c3ba277c
* ungetregister is now only used for cpuregisters, renamed to
|
21 years ago |
