Chris Lattner
|
4cdf86adaf
|
Add support for variable argument functions!
llvm-svn: 6044
|
2003-05-08 18:41:45 +00:00 |
|
Chris Lattner
|
f5a5ed1a59
|
Implement varargs support for LLI!
llvm-svn: 6043
|
2003-05-08 16:52:43 +00:00 |
|
Chris Lattner
|
ba9142280a
|
Add a pointersize/endianness safe load routine to match the store routine
llvm-svn: 6042
|
2003-05-08 16:52:16 +00:00 |
|
Chris Lattner
|
818643d3e9
|
A large number of simple changes:
* s/Method/Function
* Kill some obsolete (external) functions that used to be to support tracing
llvm-svn: 6041
|
2003-05-08 16:18:31 +00:00 |
|
Chris Lattner
|
6aa2e81ec4
|
Add support for recording arguments passed through the ... of a varargs function
llvm-svn: 6040
|
2003-05-08 16:06:52 +00:00 |
|
Chris Lattner
|
75a5272ec6
|
va_start only takes on argument now.
llvm-svn: 6039
|
2003-05-08 15:55:44 +00:00 |
|
Chris Lattner
|
746e408dd9
|
llvm.va_start only takes on argument now
Verify that va_start doesn't occur in functions with no varargs
llvm-svn: 6038
|
2003-05-08 15:55:31 +00:00 |
|
Chris Lattner
|
9487ecd232
|
Update information about the switch instruction
llvm-svn: 6037
|
2003-05-08 05:08:48 +00:00 |
|
Chris Lattner
|
16e530dfbf
|
* Add information about LLVM intrinsics
* Add information about the va_arg instruction
* Add information about the llvm.va_* intrinsics
llvm-svn: 6036
|
2003-05-08 04:57:36 +00:00 |
|
Chris Lattner
|
5dd2913d57
|
Add more support for intrinsic functions and for varargs stuff
llvm-svn: 6035
|
2003-05-08 03:47:33 +00:00 |
|
Chris Lattner
|
157c888810
|
Test the whole range of varargs handling stuff
llvm-svn: 6034
|
2003-05-08 03:46:51 +00:00 |
|
Chris Lattner
|
6b02f9dcc3
|
Initial support for intrinsic functions
llvm-svn: 6033
|
2003-05-08 03:34:12 +00:00 |
|
Chris Lattner
|
1ea6fe252b
|
Initial checkin for LLVM intrinsic functions
llvm-svn: 6032
|
2003-05-08 03:33:54 +00:00 |
|
Chris Lattner
|
a7ee27e0fd
|
Handle va_arg instruction correctly
llvm-svn: 6030
|
2003-05-08 02:50:13 +00:00 |
|
Chris Lattner
|
e1faff42f4
|
Add support for the new va_arg instruction
llvm-svn: 6029
|
2003-05-08 02:44:12 +00:00 |
|
Chris Lattner
|
1c7dcb91d4
|
Add support for the new VarArgInst instruction, reindent some stuff
llvm-svn: 6028
|
2003-05-08 02:43:06 +00:00 |
|
Chris Lattner
|
bb2a0102ae
|
Add new VarArgInst class for the va_arg instruction
llvm-svn: 6027
|
2003-05-08 02:42:50 +00:00 |
|
Chris Lattner
|
8fa300dc19
|
Add support for new va_arg instruction
llvm-svn: 6026
|
2003-05-08 02:41:11 +00:00 |
|
Chris Lattner
|
a36927b4e4
|
New testcase
llvm-svn: 6025
|
2003-05-08 02:39:37 +00:00 |
|
Chris Lattner
|
f1f55e385c
|
Don't inline functions that take variable numbers of arguments!
llvm-svn: 6024
|
2003-05-08 02:36:43 +00:00 |
|
Chris Lattner
|
d381dc9316
|
Remove using decl
llvm-svn: 6023
|
2003-05-08 02:34:52 +00:00 |
|
Chris Lattner
|
c412b950d2
|
Remove using declarations
llvm-svn: 6022
|
2003-05-08 02:08:14 +00:00 |
|
Misha Brukman
|
3e328da21d
|
Added the initial version of the TableGen description for the Sparc backend.
llvm-svn: 6021
|
2003-05-07 21:52:39 +00:00 |
|
Misha Brukman
|
8cff58b409
|
Fixed syntax error (in documentation!): missing parenthesis.
llvm-svn: 6020
|
2003-05-07 21:47:39 +00:00 |
|
Misha Brukman
|
4e7a30e565
|
Grammar fix: `of course' is separated from the sentences by commas on both
sides.
llvm-svn: 6019
|
2003-05-07 21:47:16 +00:00 |
|
Chris Lattner
|
ba6d6034d5
|
Fix bug: Jello/2003-05-07-ArgumentTest.llx
Also fixes yacr2/ks benchmarks
llvm-svn: 6017
|
2003-05-07 20:31:37 +00:00 |
|
Chris Lattner
|
9e94685ca2
|
New testcase for problem that is actually generic to all 32 bit LLI targets
llvm-svn: 6016
|
2003-05-07 20:23:32 +00:00 |
|
Chris Lattner
|
d3d53723a2
|
Add a vector to keep track of which registers are allocatable. Remove FIXMEs
llvm-svn: 6015
|
2003-05-07 20:08:36 +00:00 |
|
Chris Lattner
|
d2fae9f27c
|
Add comments, add a vector to keep track of which registers are allocatable
llvm-svn: 6014
|
2003-05-07 20:07:58 +00:00 |
|
Chris Lattner
|
59c46ee655
|
IMPLICIT_DEF was not the beautiful elegant solution I thought it was going
to be. In fact, it made stuff worse. :(
llvm-svn: 6013
|
2003-05-07 19:21:28 +00:00 |
|
Chris Lattner
|
15d2a1b0b9
|
Re-add gross hack, it's still necessary. :(
llvm-svn: 6012
|
2003-05-06 21:44:54 +00:00 |
|
Chris Lattner
|
bfeffcec97
|
Remove hideously nasty hack
llvm-svn: 6011
|
2003-05-06 21:32:39 +00:00 |
|
Chris Lattner
|
54c4b56baf
|
Emit pseudo instructions to indicate that some registers are live on entrance
and exit of the function. This fixes bug: Jello/2003-05-06-LivenessClobber.llx
and the Fhourstones benchmark
llvm-svn: 6010
|
2003-05-06 21:32:22 +00:00 |
|
Chris Lattner
|
b6fab2a875
|
Don't have a cow with new pseudo instructions
llvm-svn: 6009
|
2003-05-06 21:31:47 +00:00 |
|
Chris Lattner
|
e4c33a0ca0
|
Add two new pseudo instructions
llvm-svn: 6008
|
2003-05-06 21:31:31 +00:00 |
|
Chris Lattner
|
1269cdf9de
|
Add a new open project
llvm-svn: 6007
|
2003-05-06 20:47:11 +00:00 |
|
Chris Lattner
|
792fe48132
|
New testcase distilled from the fhourstones benchmark
llvm-svn: 6006
|
2003-05-06 19:41:25 +00:00 |
|
Chris Lattner
|
c29d6c71e2
|
Make sure to compiel the LowerSwitch tests
llvm-svn: 6005
|
2003-05-06 19:04:02 +00:00 |
|
Chris Lattner
|
ff78e3f3e4
|
Fix memory corruption problem
llvm-svn: 6003
|
2003-05-06 18:45:02 +00:00 |
|
Misha Brukman
|
cf277f9143
|
Debug output should go to cerr, not cout, because that's where bytecode goes.
llvm-svn: 6002
|
2003-05-04 22:51:30 +00:00 |
|
Chris Lattner
|
9c0a8fe1d0
|
New testcase
llvm-svn: 6001
|
2003-05-03 18:55:29 +00:00 |
|
Chris Lattner
|
5636fe923e
|
Fix bug: UnitTests/2003-05-02-DependantPHI.c
Fix testcase MultiSource/Ptrdist-ks
llvm-svn: 6000
|
2003-05-03 07:11:00 +00:00 |
|
Chris Lattner
|
0ec2b1188a
|
Fix miscompilation in Fhourstones and bug: LICM/2003-05-02-LoadHoist.ll
llvm-svn: 5994
|
2003-05-03 03:42:08 +00:00 |
|
Chris Lattner
|
ea8aec9b82
|
Fix miscompilation found in Fhourstones by bugpoint
llvm-svn: 5993
|
2003-05-03 03:39:35 +00:00 |
|
Misha Brukman
|
0cd9e6371c
|
It's 'separate'. The 'i.e.' is kinda nit-picking, but think of it this way:
'separate' is the one causing you to recompile all your passes anyway.
llvm-svn: 5992
|
2003-05-03 03:31:06 +00:00 |
|
Chris Lattner
|
63f5ef84be
|
Make a comment more accurate
llvm-svn: 5991
|
2003-05-03 03:20:36 +00:00 |
|
Chris Lattner
|
198cc8f556
|
Add support for debugging miscompilations with the use of the JIT or CBE!
llvm-svn: 5990
|
2003-05-03 03:19:41 +00:00 |
|
Chris Lattner
|
b65c30446d
|
Bugpoint is always verbose, eliminate option
llvm-svn: 5989
|
2003-05-03 03:18:41 +00:00 |
|
Chris Lattner
|
21b4596a3a
|
* Eliminate using decls
* Fix some styleistic things
* Fix bug: CWriter/2002-05-16-NameCollide.ll
llvm-svn: 5988
|
2003-05-03 03:14:35 +00:00 |
|
Chris Lattner
|
504a7cd060
|
Fix buggy tests
llvm-svn: 5987
|
2003-05-03 03:08:54 +00:00 |
|