Commit Graph

13758 Commits

Author SHA1 Message Date
Chris Lattner cee3c6b1be This testcase causes the CBE to generate code that is not legal C.
llvm-svn: 15601
2004-08-09 23:56:17 +00:00
Misha Brukman 189229151f Stub definition of the PowerPC CodeEmitter class; this isn't functional (yet).
llvm-svn: 15600
2004-08-09 23:03:59 +00:00
Misha Brukman eca36a5aa0 CodePrinter -> AsmPrinter
llvm-svn: 15599
2004-08-09 22:27:45 +00:00
Chris Lattner 8e7260652b Fix InstCombine/2004-08-09-RemInfLoop.llx
This should go into the 1.3 branch

llvm-svn: 15593
2004-08-09 21:05:48 +00:00
Chris Lattner cf5dd04f8b New testcase
llvm-svn: 15592
2004-08-09 21:03:35 +00:00
Misha Brukman 976381e554 Do not mention Chris in public Makefiles
llvm-svn: 15591
2004-08-09 20:23:44 +00:00
Misha Brukman 265fd5dbbd Shorten comment header to 80 cols
llvm-svn: 15590
2004-08-09 20:13:14 +00:00
Misha Brukman f1c538e16e * Remove UIUC-specific comment
* Make header comment span 80 cols

llvm-svn: 15589
2004-08-09 20:07:44 +00:00
Misha Brukman 876d5f1bad Add note about AIX/PowerPC with disk space requirements. Wrap long line.
llvm-svn: 15588
2004-08-09 19:57:01 +00:00
Misha Brukman a71ab1cedd Remove Chris-related bytecode-dir comment
llvm-svn: 15587
2004-08-09 19:51:49 +00:00
Misha Brukman 8f951e579b Remove ClassPrefix variable as it's no longer used.
llvm-svn: 15586
2004-08-09 19:13:29 +00:00
Misha Brukman 920ae9524d Use the current target name instead of a ClassPrefix.
llvm-svn: 15585
2004-08-09 19:10:43 +00:00
Misha Brukman 65c7925816 * CFE builds on AIX/PowerPC
* Wrap long code lines

llvm-svn: 15584
2004-08-09 18:37:04 +00:00
Misha Brukman 0eb9cee698 The (future) SparcV8 JIT would do well to have a class prefix.
llvm-svn: 15583
2004-08-09 18:13:09 +00:00
Misha Brukman 18d27dc342 * Use Classname and ClassPrefix instead of hard-coded V9 values
* Simplify code and shorten lines by not recomputing values

llvm-svn: 15582
2004-08-09 17:47:45 +00:00
Misha Brukman 11bbbb9cdd Define the ClassPrefix for SparcV9.
llvm-svn: 15581
2004-08-09 17:46:49 +00:00
Misha Brukman d210ed1855 Define a ClassPrefix for PowerPC.
llvm-svn: 15580
2004-08-09 17:46:26 +00:00
Misha Brukman 8f7272c249 Classes need to have a prefix name, so that they can be tacked on to the pieces
that TableGen creates for them, e.g. CodeEmitter, AsmPrinter, etc.

llvm-svn: 15579
2004-08-09 17:46:06 +00:00
Misha Brukman b3401c4c50 Generate a code emitter for PowerPC as well, this will be used in the JIT.
llvm-svn: 15578
2004-08-09 17:24:32 +00:00
Misha Brukman 5295e1db10 Use instruction formats as defined in the PowerPC ISA manual
llvm-svn: 15577
2004-08-09 17:24:04 +00:00
Reid Spencer 4606127bf2 Move information on Configuration files to the ../CompilerDriver.html
file since it doesn't belong in the man page.

llvm-svn: 15575
2004-08-09 03:10:39 +00:00
Reid Spencer d5c0ccbe2c This is the initial draft of the Compiler Driver documentation. It is not
worthy of review at this point. There is much thought and content remaining
to be written.

llvm-svn: 15574
2004-08-09 03:08:29 +00:00
Reid Spencer 42817b7ea1 Fix stack size processing now that the return address isn't an implied
push onto the top of the stack like x86, which uses the local area
offset.  This will allow the removal of PowerPCPEI.cpp soon.

llvm-svn: 15573
2004-08-09 01:24:32 +00:00
Chris Lattner c0171089fb This isn't a bug, it's a missing feature
llvm-svn: 15571
2004-08-08 03:33:07 +00:00
Chris Lattner 4956a32c9e Fix another really nasty regression that Anshu pointed out. In cases where
dangling constant users were removed from a function, causing it to be dead,
we never removed the call graph edge from the external node to the function.

In most cases, this didn't cause a problem (by luck).  This should definitely
go into 1.3

llvm-svn: 15570
2004-08-08 03:29:50 +00:00
Chris Lattner 6b54110281 Add standard print/dump methods to CallGraph classes.
llvm-svn: 15569
2004-08-08 03:27:49 +00:00
Chris Lattner 92b9906199 Two fixes:
1. Fix a REALLY nasty cyclic replacement issue that Anshu discovered, causing
   nondeterminstic crashes and memory corruption.
2. For performance, don't go inserting constantexpr casts of GV pointers.

This should definitely go into 1.3

llvm-svn: 15568
2004-08-08 01:30:07 +00:00
Chris Lattner 6a93462144 This DEBUG is buggy. comment it out because it's not worth fixing. This
should go into 1.3

llvm-svn: 15567
2004-08-08 01:27:56 +00:00
Reid Spencer 6c70fdd7b8 Added information about the configuration file. This is a temporary home
for this information as it belongs in a larger document on the subject of
llvmc.

llvm-svn: 15566
2004-08-07 16:30:14 +00:00
Reid Spencer 2bce8006d5 Get rid of a warning when compiling optimized. Uninitialized variable has
been initialized.

llvm-svn: 15565
2004-08-07 15:19:31 +00:00
Chris Lattner 57cafee576 Ok get rid of the REST of the tabs
llvm-svn: 15564
2004-08-07 07:18:41 +00:00
Chris Lattner 0ee8279979 Death to tabs
llvm-svn: 15563
2004-08-07 07:07:57 +00:00
Reid Spencer e132906258 Correct spelling mistakes. Wrap lines at 80 columns.
llvm-svn: 15562
2004-08-06 22:56:49 +00:00
Reid Spencer fd5ea75c3e Various updates suggested by Misha and Chris. Moved material that is
aimed more at the comiler writer than the llvmc user later in the document.

llvm-svn: 15561
2004-08-06 22:28:47 +00:00
Brian Gaeke 8d38dee7b4 dumpSavedState has outlived its usefulness.
llvm-svn: 15555
2004-08-06 19:11:43 +00:00
Chris Lattner 9ad9ab8952 New testcase for critical 1.3 feature
llvm-svn: 15553
2004-08-06 18:09:02 +00:00
Reid Spencer 9a908daf38 First version of a manual page for the llvmc compiler driver. This tool is
being documented before being written to foster discussion about what the
tool should do. Eventually, we'll achieve a decision point at which time
this document forms the requirements for the tool from which it will be
coded.

llvm-svn: 15552
2004-08-06 16:58:48 +00:00
Alkis Evlogimenos 079fbde482 Split assertion to two in order to give better assertion messages.
llvm-svn: 15543
2004-08-06 14:33:37 +00:00
Chris Lattner 5064097632 Now without stinky dos newlines!
llvm-svn: 15537
2004-08-06 07:45:37 +00:00
Chris Lattner a8dcf2423e Changes commited for Nate Begeman:
Use a PowerPC specific prolog epilog inserter to control where spilled
callee save regs are placed on the stack.
Get rid of implicit return address stack slot, save return address reg
(LR) in appropriate slot
Improve code generated for functions that don't have calls or access
globals


Note from Chris: PowerPCPEI will eventually be eliminated, once the
functionality is merged into CodeGen/PrologEpilogInserter.cpp

llvm-svn: 15536
2004-08-06 06:58:50 +00:00
Brian Gaeke a1a0643e38 Update the To-Do list according to my notes + assertions
llvm-svn: 15535
2004-08-06 04:38:13 +00:00
John Criswell 2fc39a9c2f Added Louis Gerbarg. Louis is given credit in the CREDITS.TXT file, so I
assume Louis also holds copyright.

llvm-svn: 15534
2004-08-05 23:46:27 +00:00
Chris Lattner d21e3d020e More prominantly mention ppc support, more accurately describe its status
llvm-svn: 15533
2004-08-05 21:53:13 +00:00
Misha Brukman 467ea6be23 * Mention PowerPC as new feature, and experimental feature
* Wrap long lines to 80 cols

llvm-svn: 15532
2004-08-05 21:04:03 +00:00
John Criswell e7ccf6445e Add additional copyright notice for the PowerPC backend.
Thanks Nate!

llvm-svn: 15531
2004-08-05 20:36:00 +00:00
Brian Gaeke c51892528b Split out -disable-codegen into -disable-llc and -disable-jit.
llvm-svn: 15530
2004-08-05 19:54:59 +00:00
Misha Brukman 72b2a27617 We now have the PowerPC backend, thanks to Nate Begeman and Luis Gerbarg!
llvm-svn: 15529
2004-08-05 19:09:47 +00:00
Misha Brukman 7d41ed5271 Simplify makefile by combining all TableGen dependencies into one variable
llvm-svn: 15527
2004-08-05 18:34:15 +00:00
Chris Lattner c76471842a while we're at it, make the libraries be on separate lines
llvm-svn: 15526
2004-08-05 18:32:57 +00:00
Misha Brukman eabe770156 Lines need to end with \ to make sure they're actually continued
llvm-svn: 15525
2004-08-05 18:31:33 +00:00