5df6d8de9b
When the user runs 'llc foo.bc -march=c', write the output to "foo.cbe.c", not
...
to "foo.s".
llvm-svn: 11485
2004-02-15 22:54:19 +00:00
7c437d3226
Break a line that's over 80cols into two.
...
llvm-svn: 11423
2004-02-13 23:46:47 +00:00
c5a4641c21
Do not advertise our -c option anymore
...
llvm-svn: 11413
2004-02-13 23:24:46 +00:00
92849b7dc9
Mercilessly rip the cbackend out of llvm-dis. Leave a helpful error message
...
for those who have not heard the news.
llvm-svn: 11412
2004-02-13 23:22:40 +00:00
3fba9dc096
Add support for -march=c
...
llvm-svn: 11410
2004-02-13 23:19:09 +00:00
9fae109121
Change how we create the cwriter, and add a buttload of libraries that it now
...
needs. This will be fixed shortly
llvm-svn: 11407
2004-02-13 23:01:14 +00:00
35145beb61
Add check-exit-code option, defaulting to true.
...
Add ProgramExitedNonzero argument to executeProgram(), and make it
tell its caller whether the program exited nonzero.
Move executeProgramWithCBE() out of line, to ExecutionDriver.cpp, and remove
its extra arguments which are always defaulted. Make it turn off
check-exit-code if the program exits nonzero while generating a reference
output.
Make diffProgram() assume that any nonzero exit code is a failure, if
check-exit-code is turned on.
llvm-svn: 11325
2004-02-11 18:37:32 +00:00
1dec4a1eda
Factor profile information loading OUT of llvm-prof into libanalysis
...
llvm-svn: 11316
2004-02-11 05:56:07 +00:00
d97d8faa17
Fix indentation of selectiondag.
...
I don't know why its indentation has been bugging me, but it has.
llvm-svn: 11180
2004-02-08 05:49:29 +00:00
8c146638b6
Minor change to breakpoint (lack of) support
...
llvm-svn: 11169
2004-02-08 00:06:20 +00:00
57b8ef2693
Add a -march=powerpc option. Automatically select it if this looks like a
...
big-endian, 32-bit module, or if __ppc__, __POWERPC__, or __APPLE__ are
defined.
llvm-svn: 11075
2004-02-02 19:06:12 +00:00
53f4c48393
Link in the PowerPC back-end.
...
llvm-svn: 11074
2004-02-02 19:05:52 +00:00
eddfe437c7
Now that tailduplication does not bork SSA form, run mem2reg earlier in gccas.
...
This tremendously improves the code generated by the LLVM optimizer, primarily
by making the inliner more aggressive. For example, it improves the stepanov
benchmark from 55.56 mega-additions/sec to 98.04 Ma/s. It also improves the
oopack/iterator benchmark from 338.3MFLOPS/s to 1103.4MFLOPS/s. Less noteworthy,
it improves oopack/matrix from 573 -> 641 MFLOPS/s.
llvm-svn: 11053
2004-02-01 07:24:53 +00:00
36eebb51f2
Fixed PR#197. The libcrtend library is removed from the library linking list
...
when creating native executables.
llvm-svn: 10979
2004-01-26 23:51:10 +00:00
e6a41815f3
Fixes for PR214. Use the SHLIBEXT variable instead of hardcoding .so into
...
every file.
llvm-svn: 10976
2004-01-26 20:59:41 +00:00
298d2f090f
Make sure to verify the result before writing out the bytecode file. Not doing
...
so can cause obscure errors downstream.
llvm-svn: 10840
2004-01-14 03:39:46 +00:00
2f1aa118a4
finegrainify namespacification
...
llvm-svn: 10839
2004-01-14 03:38:37 +00:00
03cbd1ab79
Remove standard C header wrappers
...
llvm-svn: 10748
2004-01-10 19:15:38 +00:00
56ca6b810a
Remove wrappers around std c files
...
llvm-svn: 10744
2004-01-10 19:12:09 +00:00
bd2977be1a
Ugh, fix some copy and pasted code that somehow never got compiled or something
...
:(
llvm-svn: 10706
2004-01-06 23:46:17 +00:00
fa13e8ab3e
* Implement set/show args
...
* Implement support for arguments to the 'run' command
* Minor cleanups and fixes
llvm-svn: 10703
2004-01-06 05:37:16 +00:00
5a62886928
implement the "null command" which repeats the last command
...
llvm-svn: 10702
2004-01-06 05:36:30 +00:00
dc24ced2ee
/me slaps head
...
llvm-svn: 10694
2004-01-05 05:47:19 +00:00
144992ea30
Add new directory
...
llvm-svn: 10688
2004-01-05 05:28:15 +00:00
0a903c38d7
Initial checkin of the LLVM Source-Level Debugger. This is incomplete, but
...
a good start. The status is documented in docs/SourceLevelDebugging.html
llvm-svn: 10687
2004-01-05 05:27:31 +00:00
238298acaf
Remove some pointless #includes
...
llvm-svn: 10653
2003-12-30 07:50:31 +00:00
c8cbec2cbb
Use getFileSize() instead of sys/stat.h directly
...
llvm-svn: 10652
2003-12-30 07:48:17 +00:00
ce8781cd67
Use new getFileSize function. Eliminate some using directives. Reorder #includes a bit
...
llvm-svn: 10651
2003-12-30 07:45:46 +00:00
cf1755a6f3
Add a -disable-opt flag to gccas, so that we can use -Wa,-disable-opt just
...
like -Wl,-disable-opt
llvm-svn: 10647
2003-12-30 03:24:27 +00:00
dabec3ba96
Pass extra arguments around n stuph
...
llvm-svn: 10631
2003-12-28 09:51:04 +00:00
99958ce7bb
Factor out code to ExecutionEngine
...
llvm-svn: 10614
2003-12-26 06:49:53 +00:00
e005b754f3
Simplify code
...
llvm-svn: 10613
2003-12-26 06:36:20 +00:00
079f4d5e71
* eliminate the -f argument to lli, as it was silly and never useful anyway
...
* Inline callMain function
* Remove hack from the ExecutionEngines where the 'run' method would automatically
run atExit functions. Fixing this requires explicitly calling exit if main returns
llvm-svn: 10611
2003-12-26 06:14:47 +00:00
141ea3a052
update comment
...
llvm-svn: 10607
2003-12-26 05:07:35 +00:00
80ce0e620e
Modified the logic so that library objects with main() are only linked in
...
if the program currently has main undefined.
llvm-svn: 10597
2003-12-23 20:27:14 +00:00
b6d3cb9449
Modified the linker so that it always links in an object from an archive
...
that defines the symbol "main." This is a hack that ensures that programs
that place their main function in a library and then link it in
(i.e. Apache 2.x) get their main function linked in.
There is probably a more correct way to do this, but this works for now.
llvm-svn: 10594
2003-12-23 17:37:06 +00:00
bd03ac8567
Rename method
...
llvm-svn: 10593
2003-12-23 09:41:45 +00:00
e5ac39d664
There is no reason to add -load support to LLC
...
llvm-svn: 10483
2003-12-15 23:10:25 +00:00
1f147379b7
Add an assertion to make sure we are at least getting argv[0] right.
...
Use a clearer error message when we fail to load a program.
llvm-svn: 10414
2003-12-12 00:47:19 +00:00
ed9fc4baad
Run tailcall elimination in a more logical place
...
llvm-svn: 10402
2003-12-11 17:50:32 +00:00
98cd4bf06f
Make 'opt -o -' work correctly instead of creating a file named './-'
...
llvm-svn: 10359
2003-12-10 14:41:33 +00:00
7dd715180d
Added the -h option for compatibility with other linkers.
...
llvm-svn: 10335
2003-12-09 15:39:11 +00:00
69d47e7314
Bugpoint had appalingly bad grammar. Fix some of it.
...
llvm-svn: 10308
2003-12-07 02:43:09 +00:00
233b069113
Do not leave a bunch of crud lying around
...
llvm-svn: 10307
2003-12-07 02:31:03 +00:00
b762ddbdf1
It is now after pldi. This issue has been fixed, so remove the hack
...
llvm-svn: 10306
2003-12-07 01:44:18 +00:00
c970a38824
New command line parsing. This isn't as perfect as I would have liked. The CommandLine Library needs to be extended, in order to parse the options and allow for optional dashes. In addition, the help option isn't correct since I do the parsing mostly myself. But this is in the ocorrect ar format.
...
llvm-svn: 10297
2003-12-06 23:01:25 +00:00
58efc9e0fe
be 3.4 happy
...
llvm-svn: 10265
2003-11-29 20:04:13 +00:00
982b285232
Fix compatibility with gcc 3.4
...
llvm-svn: 10262
2003-11-29 19:45:47 +00:00
8abd7dca76
The function resolving pass must be run, even if -disable-opt is specified
...
llvm-svn: 10250
2003-11-28 09:44:03 +00:00
fcaa6fbbd8
* The return value of LinkLibraries is ignored, so remove it.
...
* Finegrainify namespacification of Linker.cpp
* If linking a library in fails, do not STOP LINKING IN LIBRARIES AND
CONTINUE ANYWAY! Instead, just output the warning, and keep going. :)
llvm-svn: 10249
2003-11-28 07:44:09 +00:00
Chris Lattner