Commit graph

2938 commits

Author SHA1 Message Date
Dale Weiler
86adb94d7d folding for lteqgt (less than equal to or greater than) operator a.k.a <=> which maps values to -1,0,1 depending on the result (think of the result as a troolan) .. it's a perl thing :P 2013-07-31 12:59:34 +00:00
Dale Weiler
9ed62eee58 Feed clang demon 2013-07-31 12:17:52 +00:00
Dale Weiler
c3da9b237b && and || op folding 2013-07-31 12:11:03 +00:00
Dale Weiler
b0460de935 Fold for div op 2013-07-31 11:48:43 +00:00
Dale Weiler
8dd125c8f3 Added constant folding support for '*' .. Clever ascii trick prevail :D 2013-07-31 10:40:17 +00:00
Dale Weiler
5e38c800f6 Some cleanups and more support for constant folding. 2013-07-31 09:56:45 +00:00
Dale Weiler
920dbaf1e0 Work in progress constant-folding rewrite. 2013-07-31 09:04:19 +00:00
Dale Weiler
d8b931fbcf Experimental/Initial try at in-ast constant folding. (for TYPE_FLOAT currently .. since comparisions on UTF8 strings need to be worked out yet ..) 2013-07-30 18:06:42 +00:00
Dale Weiler
2c975fe48f I'm in a state of confusion as to why constant folding was never done on comparisions, now it is ;) 2013-07-30 17:27:54 +00:00
Dale Weiler
d6ca5673dc Use the _t consistency naming scheme. Also various cleanups. 2013-07-30 16:00:51 +00:00
Dale Weiler
a9ab865add Move some things around 2013-07-30 15:32:24 +00:00
Dale Weiler
c8c25ef6f7 Some cleanups and smaller binaries! 2013-07-28 00:23:15 +00:00
Dale Weiler
6c0c7aac0f cherry-pick diagnostics testsuite into cooking. We can now create compiler diagnostic tests, i.e if the compiler errors on syntax, we can match the error messages. 2013-07-28 00:02:48 +00:00
Dale Weiler
9e2b17e715 Merge pull request #116 from ignatenkobrain/cooking
Add instruction for Fedora and other RPM distros
2013-07-27 11:29:14 -07:00
Igor Gnatenko
ab79dbd22a add instruction
Signed-off-by: Igor Gnatenko <i.gnatenko.brain@gmail.com>
2013-07-27 22:24:43 +04:00
Igor Gnatenko
23af0c0209 add updated spec and rename it
Signed-off-by: Igor Gnatenko <i.gnatenko.brain@gmail.com>
2013-07-27 22:09:39 +04:00
Igor Gnatenko
bee93f28b3 not needed yet
Signed-off-by: Igor Gnatenko <i.gnatenko.brain@gmail.com>
2013-07-27 22:08:43 +04:00
Dale Weiler
d684e99d86 Clear out for now. 2013-07-27 17:57:29 +00:00
Dale Weiler
557ad9da1a Make COMPILER_QCC default to please the stuborn Quake community. 2013-07-27 17:56:46 +00:00
Dale Weiler
ec0aaa72a3 Some typos 2013-07-27 16:39:17 +00:00
Dale Weiler
ce7a9d5a3d Not to be included 2013-07-27 16:35:15 +00:00
Dale Weiler
0291726c09 Fedora spec files (0.2.9 and 0.3.0) + INSTALL documentation, thanks Igor Gnatenko. 2013-07-27 16:33:33 +00:00
Dale Weiler
c7679722fb Please the whitespace gods 2013-07-27 11:48:55 +00:00
Dale Weiler
8db9724c5d Revert "Smaller memory footprint, 4/8 bytes vs 12/24 for individual token lex_ctx's. Use to be a 'shallow' copy of the details in lex_file, now it's a pointer to the contents in lex_file."
This reverts commit cddf70f46b.
2013-07-26 16:17:41 +00:00
Dale Weiler
cddf70f46b Smaller memory footprint, 4/8 bytes vs 12/24 for individual token lex_ctx's. Use to be a 'shallow' copy of the details in lex_file, now it's a pointer to the contents in lex_file. 2013-07-26 16:09:26 +00:00
Dale Weiler
4a3794ea2b Track strdup memory too 2013-07-26 14:57:21 +00:00
Dale Weiler
44d5481828 There was too many 16-sized buckets when compiling xonotic, increase the size: this is cheaper on memory and slightly faster. 2013-07-26 14:50:37 +00:00
Dale Weiler
629fe05083 Fix the ctype replacements to work with macro arguments that have side-effect, with exception to isspace which *might* need to become a function. 2013-07-26 12:28:02 +00:00
Dale Weiler
d411d60685 Eeasy fix for that 2013-07-21 22:07:22 +00:00
Dale Weiler
2bfea938b3 One last one 2013-07-21 22:00:07 +00:00
Dale Weiler
96ec279663 Another one 2013-07-21 21:56:03 +00:00
Dale Weiler
697ad9d3a7 Technically GCC was wrong here, since opts_ini_parse is guranteed to change error. 2013-07-21 21:51:34 +00:00
Dale Weiler
764b9abaf4 Cleanups 2013-07-19 11:43:22 +00:00
Dale Weiler
84ad8ec37a Update gmqcc.ini.example 2013-07-04 15:24:55 -04:00
Wolfgang Bumiller
17318af62f fix typo 2013-07-01 10:37:51 +02:00
Wolfgang Bumiller
b39a748984 actually no :) 2013-06-22 20:31:50 +02:00
Dale Weiler
b534aca263 Fix a case of the uninitialized blues 2013-06-22 02:25:19 +00:00
Dale Weiler
adc9e7bf22 Fix some more bugs (coverity you're a life saver) 2013-06-22 02:05:04 +00:00
Dale Weiler
560b45dd16 Buffer overflow! 2013-06-22 01:58:20 +00:00
Dale Weiler
bbffdde2dc Fix some bugs and a memleak in the testsuite. 2013-06-22 01:56:22 +00:00
Dale Weiler
063c50fce4 More bug fixes and dead code elimination 2013-06-22 01:23:59 +00:00
Dale Weiler
5429b6f189 Fix out of bound access 2013-06-22 01:16:24 +00:00
Dale Weiler
d39fb653aa Fix some more bugs, and use dup2 because dup leaks on some implementations. 2013-06-22 01:14:13 +00:00
Dale Weiler
e08f00bfcd Update clean rules in makefiles to remove coverity stuff 2013-06-22 00:16:44 +00:00
Dale Weiler
edf59e4f73 Fixed possible bug in VM. Added coverity rules to makefiles. 2013-06-22 00:15:25 +00:00
Dale Weiler
52e7394418 Fix uninitialized data 2013-06-22 00:00:17 +00:00
Dale Weiler
a982d4e524 Fix a sizeof bug 2013-06-21 23:55:47 +00:00
Dale Weiler
458cfcb48c I'm assuming this is a bug since all other read tests compare with itself's ir_value, and only this one doesn't. Otherwise why would you need to load to an ir_value for vec? 2013-06-21 23:54:09 +00:00
Dale Weiler
5bdf0aff81 Catch that one too 2013-06-21 23:51:49 +00:00
Dale Weiler
1929b129ee Improper use of a possible negative value. 2013-06-21 23:49:08 +00:00