 florian
|
c44b5d1043
* take care of tempinitcode when creating def information for tempcreatenodes
|
%!s(int64=11) %!d(string=hai) anos |
|
florian
|
cb8100bccf
* do not reuse registers in prec/succ
|
%!s(int64=11) %!d(string=hai) anos |
|
florian
|
1b355ce1e4
* apply cse to all non-regable variables if possible
|
%!s(int64=11) %!d(string=hai) anos |
 michael
|
03b859947f
* Fixed bug ID #25263
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
fa4aa67f35
* Fixed bug ID #24674
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
9bfe0b5e2a
* Fix bug ID #25262
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
1521541bf4
* Forgot to add this file prior to commit
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
d3ebc96dc8
* Fix bug ID #25412
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
7a3cd62e3a
* Patch from Dennis (Bug ID 25534) to add some APR functions to the interface
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
e164a16171
* Reworked TUploadedFile.Stream
|
%!s(int64=11) %!d(string=hai) anos |
 nickysn
|
be85998313
* fixes in optimize_op_const:
|
%!s(int64=11) %!d(string=hai) anos |
 Károly Balogh
|
2791e8ad15
use optimized cg loading functions in 64bit compares instead of hardcoded move.ls (move.l #X,dn is 48bit long, clr.l/moveq.l is 16bit long)
|
%!s(int64=11) %!d(string=hai) anos |
|
Károly Balogh
|
667ec17cae
tiny optimization for clearing address regs
|
%!s(int64=11) %!d(string=hai) anos |
|
Károly Balogh
|
047395df5f
implemented add/sub ops in a_op_reg_ref, since these are used often. also fixed up reference handling in a_op_const_ref a bit
|
%!s(int64=11) %!d(string=hai) anos |
|
Károly Balogh
|
cd4bb364e5
reenabled and fixed up tcg68k.op_const_ref a bit
|
%!s(int64=11) %!d(string=hai) anos |
 sergei
|
534ecbda9f
* SPARC: r26561 caused a_op_const_reg_reg used for zero-extending 8-bit values to be optimized away. Fixed by replacing it with an explicit instruction.
|
%!s(int64=11) %!d(string=hai) anos |
 masta
|
0736bf4ded
Try to change typical constant AND-masking operations into shifts for ARM
|
%!s(int64=11) %!d(string=hai) anos |
 svenbarth
|
a2f3f0d269
Explicitely add WinSock2 unit to rtl-extra so that it is installed as well.
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
72d43272ac
* More clear and flexible way to process content from stdin
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
3d892683a2
* Cleaner interface to set request content. Allow not reading content.
|
%!s(int64=11) %!d(string=hai) anos |
|
nickysn
|
85dd9e5789
+ added a size parameter to optimize_op_const and do a sign extension of the 'a' parameter up from the specified size, so that things like (i and $ffffffff) get optimized away the same way as (i and -1)
|
%!s(int64=11) %!d(string=hai) anos |
|
michael
|
bef2957e0b
* Use global class vars to create handler/request/response. Avoids having to do 2 overrides
|
%!s(int64=11) %!d(string=hai) anos |
|
Károly Balogh
|
e6db32e192
+ add missing "else" in r26558 to fix build
|
%!s(int64=11) %!d(string=hai) anos |
 Tomas Hajny
|
8242d060ce
* correct name for OS/2 and EMX import libraries
|
%!s(int64=11) %!d(string=hai) anos |
|
sergei
|
aadb407e31
* Convert int64 to double in more straightforward way.
|
%!s(int64=11) %!d(string=hai) anos |
|
sergei
|
6dd845a183
* Removed TTabCoef type and unused zero members in coefficient arrays (the actual number of coefficients is passed to polevl/p1evl anyways).
|
%!s(int64=11) %!d(string=hai) anos |
|
sergei
|
ab44638635
+ SPARC: initial peephole optimizer.
|
%!s(int64=11) %!d(string=hai) anos |
 pierre
|
42140bd11f
Add forgotten new file, should fix bug report 25571
|
%!s(int64=11) %!d(string=hai) anos |
|
sergei
|
cda81e1d18
* SPARC: emit "x shl 1" as "x+x", it yields shorter code.
|
%!s(int64=11) %!d(string=hai) anos |
|
pierre
|
408dc2d2c1
Regenerated after: Add -h output for new assemblers
|
%!s(int64=11) %!d(string=hai) anos |
