 michael
|
00bac4d8e7
* Add Fail with format
|
11 years ago |
|
michael
|
c2a9dccc14
* Serial unit from Mark Morgan LLoyd
|
11 years ago |
|
michael
|
2a048692a5
* Patch from Mark Morgan LLoyd, adding some functions (part 2)
|
11 years ago |
|
michael
|
0ecee7cd3f
* Patch from Mark Morgan LLoyd, adding some functions
|
11 years ago |
 svenbarth
|
e734ca7a9b
M68k: Fix passing of environment variables. Fixes the compiler not being able to execute the external assembler and linker.
|
11 years ago |
 sergei
|
04f32c2a01
* Do not attempt to retrieve caller context if initial frame is not between StackTop and frame pointer of current procedure (aka stack bottom). Reverts part of r21843, fixing regression of tests/test/cg/traise6.pp after r26843.
|
11 years ago |
|
svenbarth
|
2a65c05a8c
Clean up dead code.
|
11 years ago |
|
svenbarth
|
908c0f7293
M68k: Correctly handle the case of assigning a Boolean expression to a 64-bit Boolean. Fixes tw20889.
|
11 years ago |
|
sergei
|
ae81e0889b
* Updated internal declarations of RTTI structures to match their counterparts in typinfo unit and actual data generated by compiler. It went unnoticed this far because fields whose location differ with and without 'packed' record attribute are not used by internal routines, and tests use declarations from typinfo unit (the declarations in question are not in interface, so tests cannot be done against them).
|
11 years ago |
|
sergei
|
b26e36431a
* Instead of three individual try..except blocks in fpc_pushexceptobject, guard the entire procedure at the caller side (it is called only from fpc_raiseexception). This additionally guards against possible crashes in getmem() due to corrupted heap. Furthermore, try..except block is not necessary at all for such one-time purposes.
|
11 years ago |
 florian
|
6697939903
* replaced some aren't, weren't, can't by are not, were not, cannot
|
11 years ago |
|
florian
|
76b2074e59
* updated year in logo from 2013 to 2014
|
11 years ago |
|
florian
|
13cb468a8e
* fixed modification information for some avx instructions
|
11 years ago |
|
svenbarth
|
0321a77d72
Fix the generation of alignment instructions for m68k. This not only fixes quite some tests, but most importantly the native m68k compiler!
|
11 years ago |
|
florian
|
c00d752813
+ create avx code in abs as well
|
11 years ago |
|
sergei
|
d004607876
* fpc_raiseexception does not return a result.
|
11 years ago |
|
sergei
|
749ff6b19d
* fpc_popobjectstack: clean up.
|
11 years ago |
|
svenbarth
|
435dec656f
* m68k/n68kcnv.pas, tm68ktypeconvnode.second_int_to_bool:
|
11 years ago |
|
sergei
|
1f22c2de22
- Text i/o routines are not expected to return a value.
|
11 years ago |
|
michael
|
ff2b6a90d9
* Added tests for various strutils functions by Karl-Michael Schindler (bug ID 25756)
|
11 years ago |
 Károly Balogh
|
1c930b5d24
sysutils/GetEnvironmentVariable() now handles PATH lengths above 255 chars
|
11 years ago |
|
Károly Balogh
|
e286a7462c
a slightly less broken getpathstring() for dos unit. yes, sadly this code is still in use on MorphOS, sysutils wraps to it
|
11 years ago |
|
florian
|
ab3f5744e0
* skip dec/inc type checking in unspecialized generic methods, resolves #25603
|
11 years ago |
|
Károly Balogh
|
31ec76422a
redirect .s files to the RAM: drive on Amiga while compiling
|
11 years ago |
|
Károly Balogh
|
c6d60979d2
set errno after syscalls. this fixes tiorte test at least. this code ideally should be in assembly, but it seems the assembler reader needs some fixing first...
|
11 years ago |
|
Károly Balogh
|
50c59a1cc2
don't try to fillchar/move by a negative amount, fixes tmem test
|
11 years ago |
|
Károly Balogh
|
74533a7c90
avoid generating addressing like <label>+8(aX), since this might fail during linking if label is more than 32k away, fixes linking with lineinfo unit here
|
11 years ago |
|
svenbarth
|
60a07770f1
Fix cycling after addition of the new case simplification.
|
11 years ago |
|
svenbarth
|
500920030b
Implement simplification of ordinal case nodes if the case expression is constant (might be useful when inlining).
|
11 years ago |
 nickysn
|
83dd2aed4a
* refactored the pass_typecheck code for and/or/xor/comparisons, so that the logic is simpler, but still yields exactly the same result
|
11 years ago |
