4b98c4412f
Fix spelling.
...
llvm-svn: 12287
2004-03-11 10:14:21 +00:00
5935e0a002
In PhyRegAlloc::saveState(), dump Arguments' saved-state, and try to
...
make the output more compact.
Divorce state-saving from the doFinalization method; for some reason it's not
getting called when I want it to, at Reoptimizer time. Put the guts in
PhyRegAlloc::finishSavingState(). Put an abort() in it so that I can be really
really sure that it's getting called.
Update comments.
llvm-svn: 12286
2004-03-11 06:45:52 +00:00
44a540e869
Remove ghostly directory from the build
...
llvm-svn: 12285
2004-03-11 04:42:41 +00:00
9d08818de9
Embed a floating frame of the bugzilla query in the page. This way
...
people looking for open projects cannot miss the link :-)
llvm-svn: 12284
2004-03-11 02:50:35 +00:00
d2d7ccdb4b
Minor additions and cleanups
...
llvm-svn: 12279
2004-03-11 00:50:54 +00:00
0e09916d60
Move all the SaveState options and stuff inton one spot at the top of the file.
...
De-constify SaveStateToModule; we have to set both it and SaveRegAllocState
explicitly in the reoptimizer.
Make SaveRegAllocState an 'external location' option.
llvm-svn: 12278
2004-03-10 22:21:03 +00:00
f819263cc1
Only call verifySavedState if SaveRegAllocState is set AND debugging flag is on.
...
llvm-svn: 12277
2004-03-10 22:01:59 +00:00
d68b5893a6
Bugz fixed
...
llvm-svn: 12276
2004-03-10 21:43:47 +00:00
61f57617d7
Fix PR284: [indvars] Induction variable analysis violates LLVM invariants
...
llvm-svn: 12275
2004-03-10 21:42:19 +00:00
6fdbd60ed6
New testcase for PR284: [indvars] Induction variable analysis violates LLVM invariants
...
llvm-svn: 12274
2004-03-10 21:41:47 +00:00
1ccca74a65
Add link to bugzilla query of unassigned enhancements. Specific open
...
projects like 'port glibc to llvm' or 'improve nightly tester', should
have an unassigned enhancement bug opened for them so that they can be
tracked more easily. Open projects should only list generic projects
like 'compile programs with the LLVM compiler' or 'write a new backend
for target'.
llvm-svn: 12273
2004-03-10 19:38:33 +00:00
12f1eddc90
We need a logo.
...
llvm-svn: 12272
2004-03-10 19:22:29 +00:00
53d2513820
Add brainstorm for a random test vector generator
...
llvm-svn: 12271
2004-03-10 19:15:50 +00:00
73121d9395
Recommend using install-bytecode target
...
llvm-svn: 12270
2004-03-10 19:08:52 +00:00
1eec234f91
My fix for PR274 broke the build on Darwin/PPC. As I'm fairly certain this
...
bug only affects Linux systems that use GLIBC, I'm going to put ifdefs around
the array.
llvm-svn: 12269
2004-03-10 17:38:28 +00:00
0f31060c5a
Add support for 'install-bytecode' target, used for ONLY installing
...
bytecode-libs.
llvm-svn: 12268
2004-03-10 17:38:01 +00:00
8f1977ac07
Fix up a seriously outdated comment.
...
llvm-svn: 12267
2004-03-10 17:37:50 +00:00
6f61db7f8a
New testcase for PR275
...
llvm-svn: 12266
2004-03-10 03:07:45 +00:00
30f984671a
New testcase for PR276: llvm-g++ does not mangle method names that match stdlib function names
...
llvm-svn: 12265
2004-03-09 22:51:03 +00:00
b26b6fe9a7
implement new method
...
llvm-svn: 12264
2004-03-09 19:37:06 +00:00
f53977cd42
new method
...
llvm-svn: 12263
2004-03-09 19:36:59 +00:00
6623cd78f9
Spill explicit physical register defs as well.
...
llvm-svn: 12260
2004-03-09 08:35:13 +00:00
0766126e2e
Now that I read it again, this part in particular strikes me as kind of
...
pushy and contentious...
llvm-svn: 12259
2004-03-09 07:20:26 +00:00
f234284986
Check if printing of implicit uses is required for all types of shift
...
instructions.
llvm-svn: 12258
2004-03-09 06:10:15 +00:00
75a2efdd6c
b00g fixed
...
llvm-svn: 12257
2004-03-09 05:43:59 +00:00
96e450b3b2
Address PR274 - '[JIT] Programs cannot resolve the fstat function'
...
by trying to get the compiler to generate an undefined reference for it
and related functions which live in libc_nonshared.a on Linux.
Linkers... sigh.
llvm-svn: 12256
2004-03-09 05:22:10 +00:00
3a27bd889b
Hmm, who left this sitting around in my tree
...
llvm-svn: 12255
2004-03-09 04:49:13 +00:00
8a3f2f3600
Differentiate between extended precision floats (80-bit) and double precision floats (64-bit)
...
llvm-svn: 12254
2004-03-09 03:37:54 +00:00
8ac958bf43
Use newly added API to emit bytes for instructions that gas misassembles
...
llvm-svn: 12253
2004-03-09 03:35:34 +00:00
508b459d12
Add emitInstruction() API so that we can get the bytes of a simple instruction
...
llvm-svn: 12252
2004-03-09 03:34:53 +00:00
a7c2da3af5
Constify things a bit
...
llvm-svn: 12251
2004-03-09 03:30:12 +00:00
c567f402e5
Bug fixed
...
llvm-svn: 12250
2004-03-09 00:59:15 +00:00
349910567e
Checkin testcase for PR261
...
llvm-svn: 12249
2004-03-09 00:55:58 +00:00
0536f8df6e
John fixed this bug
...
llvm-svn: 12248
2004-03-09 00:40:49 +00:00
d734587436
Change PhyRegAlloc::saveStateForValue()'s arg type to deal with
...
AllocInfo.Instruction becoming an int.
llvm-svn: 12247
2004-03-08 23:22:03 +00:00
f0a73a3f42
Save argument list alloc state by recording it as the operands of Instruction
...
#-1. Other minor changes to deal with AllocInfo.Instruction becoming an int.
llvm-svn: 12246
2004-03-08 23:22:02 +00:00
0be7ea7289
Make AllocInfo's Instruction an int, so that we can overload it for arguments.
...
(Instruction #-1's operands = argument list).
llvm-svn: 12245
2004-03-08 23:22:01 +00:00
b22926616e
Fix some spelling and grammar.
...
llvm-svn: 12244
2004-03-08 23:06:46 +00:00
d6e1b28837
Update the profiling section
...
llvm-svn: 12243
2004-03-08 22:29:35 +00:00
9bde783c5c
Switch to using edge profiling information as the basic source of profile info
...
from using basic block counts.
llvm-svn: 12242
2004-03-08 22:04:08 +00:00
d6eda1a9ef
Switch over to using edge profile information as the basic profiling representation,
...
from basic block counts.
llvm-svn: 12241
2004-03-08 22:03:45 +00:00
bd481d588a
Refactor implementations
...
llvm-svn: 12240
2004-03-08 21:30:35 +00:00
755ffeb292
We don't want to make this a pure interface, as it makes all implementors
...
bear the burden of implementing what will be all exactly the same methods.
They just want to provide the information in differing ways.
llvm-svn: 12239
2004-03-08 21:30:18 +00:00
b08e4653cb
Rearrange some methods, implement the dominates method
...
llvm-svn: 12237
2004-03-08 21:07:12 +00:00
63b49d0574
Import the trace class from the reoptimizer
...
llvm-svn: 12236
2004-03-08 20:57:27 +00:00
9c222ed4b5
Default to using edge counts
...
llvm-svn: 12234
2004-03-08 20:04:46 +00:00
c71ea6e3d4
Annotate functions with edge counts as well, if they are available.
...
llvm-svn: 12233
2004-03-08 20:04:32 +00:00
5302943ff7
If we have edge counts, we can produce block counts. I've verified that
...
using an edge profile to produce block counts gives the exact same numbers
as using a block count directly.
llvm-svn: 12232
2004-03-08 20:03:52 +00:00
96cdd27b65
Adjust to new interface
...
llvm-svn: 12231
2004-03-08 18:51:45 +00:00
03745f3e80
Add the ability to put an annotation at the end of a basic block as well.
...
llvm-svn: 12230
2004-03-08 18:51:05 +00:00
8b8a0641a1
Add initial support for reading edge counts. This will be improved to enable
...
translation of edge counts into block/function counts when possible.
llvm-svn: 12229
2004-03-08 18:20:18 +00:00
178608ee2b
Add support for representing edge counts
...
llvm-svn: 12228
2004-03-08 18:19:37 +00:00
29ba1be907
Add edge profiling support to the runtime library
...
llvm-svn: 12227
2004-03-08 18:04:31 +00:00
3702b9d79b
Fix link to license: point to current version in CVS.
...
llvm-svn: 12226
2004-03-08 17:59:31 +00:00
721264aecc
Initial support for edge profiling
...
llvm-svn: 12225
2004-03-08 17:54:34 +00:00
dae48f93b0
Split utility functions out of BlockProfiling.cpp
...
llvm-svn: 12224
2004-03-08 17:06:13 +00:00
b16526108c
Remove the comment "Constants must always have an initial value.", which
...
is incorrect. Fix some formatting nastiness.
llvm-svn: 12223
2004-03-08 16:49:10 +00:00
d91e676700
finegrainify namespacification
...
llvm-svn: 12221
2004-03-08 16:45:53 +00:00
e2b27080e7
Insert functions into the module promptly, not lazily. This fixes a bug
...
I introduced last night. Note to self: test the *correct* tree...
llvm-svn: 12220
2004-03-08 16:14:19 +00:00
ea0789ca7b
Implement test/Regression/Assembler/2004-03-07-FunctionAddressAlignment.llx
...
llvm-svn: 12218
2004-03-08 06:17:35 +00:00
be3a169d18
New testcase for folding in some important situations. The first two
...
come up a lot in the code generated by the C++ front-end for pointers
to member functions. See PR166.
llvm-svn: 12217
2004-03-08 06:17:15 +00:00
c3ecaae16c
Remove Module::mutateConstantPointerRef, which is now thankfully dead!
...
This is one small step towards the complete obliteration of
ConstantPointerRef's entirely!! Woot!
llvm-svn: 12216
2004-03-08 06:16:10 +00:00
8301a593a9
Remove Module::mutateConstantPointerRef, which is now thankfully dead!
...
llvm-svn: 12215
2004-03-08 06:15:33 +00:00
9f4918ab93
Eliminate nightmarish API
...
llvm-svn: 12214
2004-03-08 06:11:10 +00:00
16e27f3af3
remove *THANKFULLY* dead method
...
llvm-svn: 12213
2004-03-08 06:10:32 +00:00
75998c0885
Eliminate a REALLY HORRIBLE API: mutateReferences, which is gross gross gross.
...
llvm-svn: 12212
2004-03-08 06:09:57 +00:00
7e201ecc31
New testcase for PR269
...
llvm-svn: 12211
2004-03-08 05:35:12 +00:00
f0cbbabea7
Bug fixed
...
llvm-svn: 12210
2004-03-08 04:55:45 +00:00
451e10e46a
Testcase that crashes the C front-end.
...
llvm-svn: 12209
2004-03-08 04:41:32 +00:00
84c697d520
Fix a bug handling globals that are constants, but are still external
...
llvm-svn: 12208
2004-03-08 03:52:24 +00:00
98502aae76
Avoid allocating special registers a bit more robustly
...
llvm-svn: 12207
2004-03-08 03:48:07 +00:00
0653e1167d
New code-quality testcase.
...
llvm-svn: 12205
2004-03-08 02:12:36 +00:00
653e662a93
Implement folding explicit load instructions into binary operations. For a
...
testcase like this:
int %test(int* %P, int %A) {
%Pv = load int* %P
%B = add int %A, %Pv
ret int %B
}
We now generate:
test:
mov %ECX, DWORD PTR [%ESP + 4]
mov %EAX, DWORD PTR [%ESP + 8]
add %EAX, DWORD PTR [%ECX]
ret
Instead of:
test:
mov %EAX, DWORD PTR [%ESP + 4]
mov %ECX, DWORD PTR [%ESP + 8]
mov %EAX, DWORD PTR [%EAX]
add %EAX, %ECX
ret
... saving one instruction, and often a register. Note that there are a lot
of other instructions that could use this, but they aren't handled. I'm not
really interested in adding them, but mul/div and all of the FP instructions
could be supported as well if someone wanted to add them.
llvm-svn: 12204
2004-03-08 01:58:35 +00:00
1dd6afe6a2
Rearrange and refactor some code. No functionality changes.
...
llvm-svn: 12203
2004-03-08 01:18:36 +00:00
fe6f2e3e80
Implement ArgumentPromotion/aggregate-promote.ll
...
This allows pointers to aggregate objects, whose elements are only read, to
be promoted and passed in by element instead of by reference. This can
enable a LOT of subsequent optimizations in the caller function.
It's worth pointing out that this stuff happens a LOT of C++ programs, because
objects in templates are generally passed around by reference. When these
templates are instantiated on small aggregate or scalar types, however, it is
more efficient to pass them in by value than by reference.
This transformation triggers most on C++ codes (e.g. 334 times on eon), but
does happen on C codes as well. For example, on mesa it triggers 72 times,
and on gcc it triggers 35 times. this is amazingly good considering that
we are using 'basicaa' so far.
llvm-svn: 12202
2004-03-08 01:04:36 +00:00
1f7f22c42c
New testcase
...
llvm-svn: 12201
2004-03-08 00:58:19 +00:00
cc544e57f3
Implement: ArgumentPromotion/chained.ll
...
llvm-svn: 12200
2004-03-07 22:52:53 +00:00
332a8bdf31
New testcase for missed case
...
llvm-svn: 12199
2004-03-07 22:52:41 +00:00
64b8d697ad
Fix another minor bug, exposed by perlbmk
...
llvm-svn: 12198
2004-03-07 22:43:27 +00:00
538fee7aa2
Since 'load null' is undefined, we can make it do whatever we want. Returning
...
a zero value is the most likely way to cause further simplification, so we do it.
llvm-svn: 12197
2004-03-07 22:16:24 +00:00
ac2e2d3bdd
Run the new pass in gccld now that it passes all tests
...
llvm-svn: 12196
2004-03-07 22:12:40 +00:00
6770842b67
Fix a minor bug and turn debug output into, well, debug output.
...
llvm-svn: 12195
2004-03-07 21:54:50 +00:00
1f607f2f2e
Add prototype
...
llvm-svn: 12194
2004-03-07 21:30:08 +00:00
483ae01c9c
New LLVM pass: argument promotion. This version only handles simple scalar
...
variables.
llvm-svn: 12193
2004-03-07 21:29:54 +00:00
f2b8692b73
New testcase
...
llvm-svn: 12192
2004-03-07 21:28:54 +00:00
d6f62ba55b
Add memory operand version of conditional move.
...
llvm-svn: 12190
2004-03-07 03:19:11 +00:00
cb98644e9b
As I wrote in the docs, simple is the default spiller :-)
...
llvm-svn: 12189
2004-03-06 23:08:44 +00:00
79850121ad
Add simple spiller.
...
llvm-svn: 12188
2004-03-06 22:38:29 +00:00
36fd868f08
The default register allocator is local not simple.
...
llvm-svn: 12187
2004-03-06 22:27:41 +00:00
4e0669cef0
Support return values of basic integer types.
...
Emit RETL instruction to return instead of funny JMPL.
Fix indentation.
llvm-svn: 12186
2004-03-06 05:32:28 +00:00
c65b97d4f6
Sort stanzas into Sparc V8 book page number order.
...
Add RET, RETL. Rename SAVE, RESTORE & JMPL for consistency.
llvm-svn: 12185
2004-03-06 05:32:13 +00:00
84b4c9208d
Hack it so we do not try to allocate values to G0.
...
llvm-svn: 12184
2004-03-06 05:31:32 +00:00
901cfa9e2e
Make prolog align stack properly. Make epilog not touch any registers.
...
llvm-svn: 12183
2004-03-06 05:31:21 +00:00
5a1c7a8be3
Emit register names in lowercase, as required by the assembler.
...
llvm-svn: 12182
2004-03-06 05:30:21 +00:00
639b04aca3
Teach getRegClassForType where to find FP registers
...
llvm-svn: 12180
2004-03-06 03:54:13 +00:00
2cb2905c96
Fix a minor bug
...
llvm-svn: 12169
2004-03-05 22:04:07 +00:00
d2e88a21b7
Unbreak the build on Sparc.
...
llvm-svn: 12161
2004-03-05 20:04:40 +00:00
0c76d1146f
Asm output is looking a lot better; not correct for all operands yet though.
...
llvm-svn: 12143
2004-03-05 08:39:09 +00:00
6c271edcd7
Fix a bug in a previous checkin that broke 175.vpr
...
llvm-svn: 12128
2004-03-04 21:36:57 +00:00
bc3381f267
Add support for strto* and v*printf
...
llvm-svn: 12127
2004-03-04 21:03:54 +00:00
0ed8162c1a
Add non-crappy support for varargs
...
llvm-svn: 12126
2004-03-04 20:33:47 +00:00
3562ea84ab
Implement a FIXME, improving the efficiency of DSA on povray.
...
This reduces CBU time from 145s -> 122s (debug build), reduces # allocated nodes
from 129420 to 116477.
llvm-svn: 12125
2004-03-04 19:47:04 +00:00
9c609df3c9
Support -print-machineinstrs
...
llvm-svn: 12124
2004-03-04 19:22:16 +00:00
30f94a72f4
Speed up the cbu pass from taking somewhere near the age of the universe to about 90s on povray
...
llvm-svn: 12123
2004-03-04 19:16:35 +00:00
8351d8c1a8
make -print-machineinstrs work for both SparcV9 and X86
...
llvm-svn: 12122
2004-03-04 19:16:23 +00:00
b9501c1f8c
Add assertion for scale verification.
...
llvm-svn: 12120
2004-03-04 18:05:02 +00:00
8a4159708b
Fix bug introduced by yesterday's changes.
...
llvm-svn: 12119
2004-03-04 18:02:07 +00:00
33cb4260a5
Hide variable from other functions.
...
llvm-svn: 12118
2004-03-04 17:50:44 +00:00
536b131c9c
Fix BU datastructures with povray!
...
The problem was that we were merging a field of a node with a value that was
deleted. Thanks to bugpoint for reducing povray to a nice small 3 function
example. :)
llvm-svn: 12116
2004-03-04 17:06:53 +00:00
8db52980b7
Minor changes, remove some debugging code that got checked in somehow.
...
Make sure to scope the NodeMap passed into cloneInto so that it doesn't point
to nodes that are deleted. Add some FIXME's for future performance enhancements.
llvm-svn: 12115
2004-03-04 17:05:28 +00:00
82ec4a561b
Asm printer support, based on x86 - only prints mnemonics for now
...
llvm-svn: 12113
2004-03-04 06:00:41 +00:00
8ace44fe0d
Double-FP pseudo-registers.
...
llvm-svn: 12112
2004-03-04 05:15:03 +00:00
92c95f812d
Subtract instructions; minor cleanups
...
llvm-svn: 12111
2004-03-04 04:37:45 +00:00
d2e89190d9
Floating point regs
...
llvm-svn: 12110
2004-03-04 04:37:22 +00:00
f48eb7519a
Only clone nodes that are needed in the caller, don't clone ALL aux calls. This improves
...
povray from having ~600K nodes and 300K call nodes to 65K nodes and 25K call nodes.
llvm-svn: 12109
2004-03-04 03:57:53 +00:00
956fb06a4a
Simple copyConstantToReg support, SETHIi and ORri
...
llvm-svn: 12107
2004-03-04 00:56:25 +00:00
165ff998e0
Support add - note, still missing important copyConstantToRegister stuff
...
llvm-svn: 12106
2004-03-03 23:03:14 +00:00
559487301f
Fix a minor bug handling incomplete programs
...
llvm-svn: 12105
2004-03-03 23:00:19 +00:00
d9da2678fc
Fix a DSA bug that caused DSA to generate incredibly huge graphs and take forever to
...
do it on povray. The problem is that we were not copying globals from callees to
callers unless the existed in both graphs. We should have copied them in the case
where the global pointed to a node that was copied as well.
llvm-svn: 12104
2004-03-03 22:01:09 +00:00
91be5332b2
Rename method, add new method
...
llvm-svn: 12103
2004-03-03 22:00:20 +00:00
5fc189157c
Deinline methods, add fast exit
...
llvm-svn: 12102
2004-03-03 20:55:27 +00:00
06bd20676d
De-inline methods
...
llvm-svn: 12101
2004-03-03 20:55:15 +00:00
015972103d
Make MachineOperand's value named 'contents'. Make really, really sure
...
it is always completely initialized and copied.
Also, fix up many comments and asserts.
llvm-svn: 12100
2004-03-03 19:07:27 +00:00
3001c6e264
Squash /usr/dcs path; use a 'require <version>' instead.
...
llvm-svn: 12099
2004-03-03 17:38:51 +00:00
c0c260dd37
Add my regression finding by binary searching cvs script.
...
llvm-svn: 12095
2004-03-03 08:14:25 +00:00
140b2cbd28
Fix a node mapping problem that was causing the pool allocator to locally allocate
...
nodes that were globally live, thus breaking programs.
llvm-svn: 12094
2004-03-03 05:34:31 +00:00
93e9878c2c
SPECIFY a target data to initialize the CBE target with. Until now we have
...
been using the default target data layout object to lower malloc instructions,
causing us to allocate more memory than we needed! This could improve the
performance of the CBE generated code substantially!
llvm-svn: 12088
2004-03-03 02:14:15 +00:00
87ed2a4096
Add a new constructor
...
llvm-svn: 12087
2004-03-03 02:12:47 +00:00
7abcc387de
Don't emit things like malloc(16*1). Allocation instructions are fixed arity now.
...
llvm-svn: 12086
2004-03-03 01:40:53 +00:00
2e92b469c8
FINALLY be able to get symbolic type names in the globals graph!
...
llvm-svn: 12082
2004-03-02 21:39:43 +00:00
9e100fc3f7
Really, only if reopen
...
llvm-svn: 12080
2004-03-02 20:46:18 +00:00
626f47f946
Added EH support.
...
llvm-svn: 12075
2004-03-02 15:54:25 +00:00
cb8937197e
Make <di> into <dt> (definition list term).
...
llvm-svn: 12072
2004-03-02 03:36:10 +00:00
c566ca36a1
Doxygenify comments.
...
llvm-svn: 12071
2004-03-02 00:22:19 +00:00
f44acae31e
Implement ExtractCodeRegion()
...
llvm-svn: 12070
2004-03-02 00:20:57 +00:00
1998b73492
Add prototype for ExtractCodeRegion()
...
llvm-svn: 12069
2004-03-02 00:20:32 +00:00
f272f9b3d5
Make a note that this is usually used via bugpoint.
...
llvm-svn: 12068
2004-03-02 00:19:09 +00:00
a6025e6480
Doxygenify some comments.
...
llvm-svn: 12064
2004-03-01 23:53:11 +00:00
45da18d2a4
Fix <p> missing <p> tag
...
llvm-svn: 12063
2004-03-01 23:25:38 +00:00
31953c7a10
Add a spiller option to llc. A simple spiller will come soon. When we get CFG in the machine code represenation a global spiller will also be possible. Also document the linear scan register allocator but mark it as experimental for now.
...
llvm-svn: 12062
2004-03-01 23:18:15 +00:00
b76d234ee9
Add the long awaited memory operand folding support for linear scan
...
llvm-svn: 12058
2004-03-01 20:05:10 +00:00
b22d09cc5e
* If a badref has a name, print it out for ease of debugging
...
* Doxygenify (some) comments
llvm-svn: 12057
2004-03-01 19:48:13 +00:00
d86982547f
Add new method
...
llvm-svn: 12056
2004-03-01 19:36:50 +00:00
c8d23b19fb
Correctly add an array marker on a node when appropriate!
...
llvm-svn: 12055
2004-03-01 19:02:54 +00:00
88cc178e3b
Fix grammar and doxygenify comments.
...
llvm-svn: 12054
2004-03-01 18:31:19 +00:00
5af2be7d09
* Add implementation of ExtractBasicBlock()
...
* Add comments to ExtractLoop()
llvm-svn: 12053
2004-03-01 18:28:34 +00:00
f68f121dd3
Add ability to extract a single basic block into a new function.
...
llvm-svn: 12052
2004-03-01 18:27:13 +00:00
773d6f66db
* HTML 4.01 Strict compliance
...
* Removed unnecessary tabs in the entire file
llvm-svn: 12051
2004-03-01 18:21:04 +00:00
c501f5530b
HTML 4.01 Strict compliance.
...
llvm-svn: 12050
2004-03-01 17:47:27 +00:00
f96405a1ce
Add this back, as its absence introduces assertions, and it seems to work now
...
that Instructions are annotable again
llvm-svn: 12045
2004-03-01 15:28:27 +00:00
e63e8ea884
fix bug in previous checkin
...
llvm-svn: 12044
2004-03-01 15:05:17 +00:00
427cec1395
TargetCacheInfo has been removed; its only uses were to propagate a constant
...
(16) into certain areas of the SPARC V9 back-end. I'm fairly sure the US IIIi's
dcache has 32-byte lines, so I'm not sure where the 16 came from. However, in
the interest of not breaking things any more than they already are, I'm going
to leave the constant alone.
llvm-svn: 12043
2004-03-01 06:43:29 +00:00
48a503be04
Adding new Modulo Scheduling graph files.
...
llvm-svn: 12031
2004-03-01 02:50:57 +00:00
dd10fbe775
Removing old graph files with new graph files that I wrote. Updated ModuloScheduling pass, but still in progress.
...
llvm-svn: 12030
2004-03-01 02:50:01 +00:00
6f6e0f29ad
Expand on my note-to-self
...
llvm-svn: 12029
2004-03-01 02:44:44 +00:00
1f4642c47c
Handle passing constant integers to functions much more efficiently. Instead
...
of generating this code:
mov %EAX, 4
mov DWORD PTR [%ESP], %EAX
mov %AX, 123
movsx %EAX, %AX
mov DWORD PTR [%ESP + 4], %EAX
call Y
we now generate:
mov DWORD PTR [%ESP], 4
mov DWORD PTR [%ESP + 4], 123
call Y
Which hurts the eyes less. :)
Considering that register pressure around call sites is already high (with all
of the callee clobber registers n stuff), this may help a lot.
llvm-svn: 12028
2004-03-01 02:42:43 +00:00
b19dcc6651
Don't look in the .libs directories that libtool makes
...
llvm-svn: 12027
2004-03-01 02:41:22 +00:00
5c7d3cda78
Fix a minor code-quality issue. When passing 8 and 16-bit integer constants
...
to function calls, we would emit dead code, like this:
int Y(int, short, double);
int X() {
Y(4, 123, 4);
}
--- Old
X:
sub %ESP, 20
mov %EAX, 4
mov DWORD PTR [%ESP], %EAX
*** mov %AX, 123
mov %AX, 123
movsx %EAX, %AX
mov DWORD PTR [%ESP + 4], %EAX
fld QWORD PTR [.CPIX_0]
fstp QWORD PTR [%ESP + 8]
call Y
mov %EAX, 0
# IMPLICIT_USE %EAX %ESP
add %ESP, 20
ret
Now we emit:
X:
sub %ESP, 20
mov %EAX, 4
mov DWORD PTR [%ESP], %EAX
mov %AX, 123
movsx %EAX, %AX
mov DWORD PTR [%ESP + 4], %EAX
fld QWORD PTR [.CPIX_0]
fstp QWORD PTR [%ESP + 8]
call Y
mov %EAX, 0
# IMPLICIT_USE %EAX %ESP
add %ESP, 20
ret
Next up, eliminate the mov AX and movsx entirely!
llvm-svn: 12026
2004-03-01 02:34:08 +00:00
482cf01a1e
Fix the "partial pool allocator" on em3d and others. The problem is that
...
DSNodes, unlike other GraphTraits nodes, can have null outgoing edges, and
df_iterator doesn't take this into consideration. As a workaround, the
successor iterator now handles null nodes and 'indicates' that null has
no successors.
llvm-svn: 12025
2004-03-01 01:42:26 +00:00
b4c203ce67
Make Module annotable. Reid has a bunch of code that depends on this, and
...
we really don't win that much by eliminating this (not many Modules are
allocated), so it's not worth it. When we can, we should revisit this in
the future.
llvm-svn: 12023
2004-03-01 01:25:37 +00:00
5cf39339d1
Disable tail duplication in a case that breaks on Olden/tsp
...
llvm-svn: 12021
2004-03-01 01:12:13 +00:00
2a4c5b050e
Doxygenify comments.
...
llvm-svn: 12015
2004-02-29 23:55:11 +00:00
c91e1ff50d
* Remove function to find "main" in a Module, there's a method for that
...
* Removing extraneous empty space and empty comment lines
llvm-svn: 12014
2004-02-29 23:09:10 +00:00
43640d7138
Fix -debug-pass=Executions, which relied on Function, Module, and BasicBlock
...
being annotable
llvm-svn: 12013
2004-02-29 22:37:04 +00:00
2de229f31b
Fix bug: test/Regression/Transforms/LowerInvoke/2004-02-29-PHICrash.llx
...
... which tickled the lowerinvoke pass because it used the BCE routines.
llvm-svn: 12012
2004-02-29 22:24:41 +00:00
9c9a4c4abd
Testcase for a bug that caused a whole bunch of testcases to die in the nightly
...
tester last night.
llvm-svn: 12011
2004-02-29 22:15:51 +00:00
604cc83933
Add an assert
...
llvm-svn: 12010
2004-02-29 22:01:51 +00:00
e0c3c18802
Add back #include I messed up
...
llvm-svn: 12009
2004-02-29 21:40:53 +00:00
90ecb92838
Urg, forgot to check this in.
...
llvm-svn: 12007
2004-02-29 21:03:08 +00:00
f2bb31c5fd
Module does not need to be annotatable aka annotable
...
llvm-svn: 12005
2004-02-29 19:27:55 +00:00
fcb4f9d00b
Remove public header. It's been moved to lib/Target/SparcV9
...
llvm-svn: 12004
2004-02-29 19:13:20 +00:00
3859031b9b
Move the private MachineInstrAnnot.h into a private directory.
...
llvm-svn: 12003
2004-02-29 19:12:51 +00:00
d5a49c477d
Remove use of an ugly header
...
llvm-svn: 12002
2004-02-29 19:04:31 +00:00
6194ee84ba
Move methods out of .h file
...
llvm-svn: 12001
2004-02-29 19:02:39 +00:00
23840c7db5
Do not use explicit casts that hide the dependence on Instruction being
...
annotable
llvm-svn: 12000
2004-02-29 19:02:26 +00:00
79b8d467da
Ugh, the old sparc backend attaches MachineCodeForInstruction annotations on
...
LLVM instructions. Because it contains an explicit cast, we didn't catch it.
I guess instruction's will be annotable for the duration of the sparcv9's
existence.
llvm-svn: 11999
2004-02-29 18:54:23 +00:00
9a4653edfa
Add instruction name description.
...
llvm-svn: 11998
2004-02-29 18:44:03 +00:00
0824ffc697
Use correct template for SHLD and SHRD instructions so that the memory
...
operand size is correctly specified.
llvm-svn: 11997
2004-02-29 09:19:40 +00:00
c7fd0770a0
Improve allocation order:
...
1) For 8-bit registers try to use first the ones that are parts of the
same register (AL then AH). This way we only alias 2 16/32-bit
registers after allocating 4 8-bit variables.
2) Move EBX as the last register to allocate. This will cause less
spills to happen since we will have 8-bit registers available up to
register excaustion (assuming we use the allocation order). It
would be nice if we could push all of the 8-bit aliased registers
towards the end but we much prefer to keep callee saved register to
the end to avoid saving them on entry and exit of the function.
For example this gives a slight reduction of spills with linear scan
on 164.gzip.
Before:
11221 asm-printer - Number of machine instrs printed
975 spiller - Number of loads added
675 spiller - Number of stores added
398 spiller - Number of register spills
After:
11182 asm-printer - Number of machine instrs printed
952 spiller - Number of loads added
652 spiller - Number of stores added
386 spiller - Number of register spills
llvm-svn: 11996
2004-02-29 09:17:01 +00:00
ea81b79a97
A big X86 instruction rename. The instructions are renamed to make
...
their names more decriptive. A name consists of the base name, a
default operand size followed by a character per operand with an
optional special size. For example:
ADD8rr -> add, 8-bit register, 8-bit register
IMUL16rmi -> imul, 16-bit register, 16-bit memory, 16-bit immediate
IMUL16rmi8 -> imul, 16-bit register, 16-bit memory, 8-bit immediate
MOVSX32rm16 -> movsx, 32-bit register, 16-bit memory
llvm-svn: 11995
2004-02-29 08:50:03 +00:00
2f80d586b8
Remove dead member variables of SparcV9SchedInfo and TargetSchedInfo
...
llvm-svn: 11994
2004-02-29 08:40:03 +00:00
1e36fb030c
Eliminate the X86-specific BMI functions, using BuildMI instead.
...
Replace uses of addZImm with addImm.
llvm-svn: 11992
2004-02-29 07:22:16 +00:00
9a97573267
Fix a miscompilation of 197.parser that occurs when you have single basic
...
block loops.
llvm-svn: 11990
2004-02-29 07:10:16 +00:00
bf2963ef91
Fix PR255: [tailduplication] Single basic block loops are very rare
...
Note that this is a band-aid put over a band-aid. This just undisables
tail duplication in on very specific case that it seems to work in.
llvm-svn: 11989
2004-02-29 06:41:20 +00:00
d470a0f042
Update comment at head of file. Also fix C 'typedef struct' nonsense I
...
inadvertently left in here.
llvm-svn: 11988
2004-02-29 06:33:28 +00:00
ed01da8f0b
Adjust to change in TII ctor arguments
...
llvm-svn: 11987
2004-02-29 06:31:44 +00:00
f20abac9bc
Eliminate the distinction between "real" and "unreal" instructions
...
llvm-svn: 11986
2004-02-29 06:31:16 +00:00
e65125dcdf
Add more architectures, and ELF64 stuff.
...
llvm-svn: 11985
2004-02-29 06:30:25 +00:00
ca89812db7
These two virtual methods are never called.
...
llvm-svn: 11984
2004-02-29 05:59:33 +00:00
71a899dea6
Remove a TON of flags that noone cares about
...
llvm-svn: 11983
2004-02-29 05:58:30 +00:00
59d9e9126d
Noone calls these virtual methods
...
llvm-svn: 11982
2004-02-29 05:58:16 +00:00
79d9a93a71
This is the only file in the system that uses this enum. eliminate it.
...
llvm-svn: 11981
2004-02-29 05:57:59 +00:00
f611ca055c
Scrap a huge layer of cruft out of this interface.
...
llvm-svn: 11980
2004-02-29 05:57:21 +00:00
e5f08917ae
Implement initial prolog/epilog code insertion methods.
...
llvm-svn: 11979
2004-02-29 05:18:30 +00:00
5e7b3180e5
Remove unneeded #include
...
llvm-svn: 11978
2004-02-29 05:15:56 +00:00
44b1b22a2d
int64_t -> int
...
llvm-svn: 11977
2004-02-29 05:07:02 +00:00
a33f2b3731
Continue Alkis's int64_t cleanup. This makes all of the immediate related
...
methods take an int or unsigned value instead of int64_t.
Also, add an 'addImm' method to the MachineInstrBuilder class, because the
fact that the hardware sign or zero extends it does not/should not matter
to the code generator. Once the old sparc backend is removed the difference
can be eliminated.
llvm-svn: 11976
2004-02-29 05:06:49 +00:00
349b8c928d
Add BuildMI variants that take a MBB::iterator
...
llvm-svn: 11975
2004-02-29 04:55:28 +00:00
876f6f96d0
Use correct template for ADC instruction with memory operands.
...
llvm-svn: 11974
2004-02-29 02:18:17 +00:00
b1862e4d3b
Add an instruction selector capable of selecting 'ret void'
...
llvm-svn: 11973
2004-02-29 00:27:00 +00:00
fa63580517
SHLD and SHRD take 32-bit operands but an 8-bit immediate. Rename them
...
to denote this fact.
llvm-svn: 11972
2004-02-28 23:46:44 +00:00
4953ae085a
Floating point loads/stores act on memory operands. Rename them to
...
denote this fact.
llvm-svn: 11971
2004-02-28 23:42:35 +00:00
c6948fa762
Rename instruction templates to be easier to the human eye to
...
parse. The name is now I (operand size)*. For example:
Im32 -> instruction with 32-bit memory operands.
Im16i8 -> instruction with 16-bit memory operands and 8 bit immediate
operands.
llvm-svn: 11970
2004-02-28 23:09:03 +00:00
5b5dee4afb
Uncomment instructions that take both an immediate and a memory
...
operand but their sizes differ.
llvm-svn: 11969
2004-02-28 22:06:59 +00:00
c0f3741a16
Fix my sloppiness
...
llvm-svn: 11968
2004-02-28 22:06:03 +00:00
194939086d
Each instruction now has both an ImmType and a MemType. This describes
...
the size of the immediate and the memory operand on instructions that
use them. This resolves problems with instructions that take both a
memory and an immediate operand but their sizes differ (i.e. ADDmi32b).
llvm-svn: 11967
2004-02-28 22:02:05 +00:00
465a5cc291
Fix typo in comment
...
llvm-svn: 11966
2004-02-28 21:55:18 +00:00
c116ede471
Add a testcase for switch instruction induction
...
llvm-svn: 11965
2004-02-28 21:30:39 +00:00
d3e6ae263c
Implement switch->br and br->switch folding by ripping out the switch->switch
...
and br->br code and generalizing it. This allows us to compile code like this:
int test(Instruction *I) {
if (isa<CastInst>(I))
return foo(7);
else if (isa<BranchInst>(I))
return foo(123);
else if (isa<UnwindInst>(I))
return foo(1241);
else if (isa<SetCondInst>(I))
return foo(1);
else if (isa<VAArgInst>(I))
return foo(42);
return foo(-1);
}
into:
int %_Z4testPN4llvm11InstructionE("struct.llvm::Instruction"* %I) {
entry:
%tmp.1.i.i.i.i.i.i.i = getelementptr "struct.llvm::Instruction"* %I, long 0, ubyte 4 ; <uint*> [#uses=1]
%tmp.2.i.i.i.i.i.i.i = load uint* %tmp.1.i.i.i.i.i.i.i ; <uint> [#uses=2]
%tmp.2.i.i.i.i.i.i = seteq uint %tmp.2.i.i.i.i.i.i.i, 27 ; <bool> [#uses=0]
switch uint %tmp.2.i.i.i.i.i.i.i, label %endif.0 [
uint 27, label %then.0
uint 2, label %then.1
uint 5, label %then.2
uint 14, label %then.3
uint 15, label %then.3
uint 16, label %then.3
uint 17, label %then.3
uint 18, label %then.3
uint 19, label %then.3
uint 32, label %then.4
]
...
As well as handling the cases in 176.gcc and many other programs more effectively.
llvm-svn: 11964
2004-02-28 21:28:10 +00:00
72bb8fcb15
Change this so that LLC actually tries to run the code generator, though it will
...
immediately abort due to lack of an instruction selector. :)
llvm-svn: 11963
2004-02-28 20:21:45 +00:00
72170c4a2e
Add a hook to run with the V8 target, though it doesn't currently work. Also
...
mark the PPC backend as experimental
llvm-svn: 11962
2004-02-28 19:55:16 +00:00
fc5232ba3a
Add hook for V8 target
...
llvm-svn: 11961
2004-02-28 19:54:16 +00:00
990c279f5c
SparcV8 now builds.
...
llvm-svn: 11960
2004-02-28 19:54:00 +00:00
daeeaa72cd
fine grainify namespacification
...
llvm-svn: 11959
2004-02-28 19:53:18 +00:00
1a3ee3e961
Finegrainify namespacification
...
llvm-svn: 11958
2004-02-28 19:52:49 +00:00
e40fd90b0b
Tab completion is our friend.
...
llvm-svn: 11957
2004-02-28 19:45:39 +00:00
0179fb5881
Clean up rules
...
llvm-svn: 11956
2004-02-28 19:43:40 +00:00
74598091a1
Bring this directory into "it actually compiles" land
...
llvm-svn: 11955
2004-02-28 19:37:18 +00:00
3290952f8b
Fix multiple inclusion problem
...
llvm-svn: 11954
2004-02-28 19:31:32 +00:00
772eafa332
if there is already a prototype for malloc/free, use it, even if it's incorrect.
...
Do not just inject a new prototype.
llvm-svn: 11951
2004-02-28 18:51:45 +00:00
36326de312
Do not remove an active template argument even if the superclass had one of the
...
same name
llvm-svn: 11950
2004-02-28 17:41:48 +00:00
178f0f1405
fix a bug in the testcase
...
llvm-svn: 11949
2004-02-28 17:33:21 +00:00
994f2d1406
Ignore X = X assignments that was causing Alkis's rewrite of X86.td to crash
...
tblgen.
llvm-svn: 11948
2004-02-28 17:31:28 +00:00
1ddddf536a
new testcase for a tblgen bug that alkis ran into
...
llvm-svn: 11947
2004-02-28 16:43:44 +00:00
d959ab9208
Assert instead of going into an infinite loop!
...
llvm-svn: 11946
2004-02-28 16:31:53 +00:00
18577345aa
ELF constants and data structures.
...
llvm-svn: 11945
2004-02-28 06:26:20 +00:00
2debead504
Do not generate instructions with mismatched memory/immediate sized
...
operands. The X86 backend doesn't handle them properly right now.
llvm-svn: 11944
2004-02-28 06:01:43 +00:00
eff5f51807
new testcase for intrinsic folding
...
llvm-svn: 11943
2004-02-28 05:28:42 +00:00
b66f42427a
The instruction combining pass removes dead instructions, there is no need
...
to run the die pass after it.
llvm-svn: 11942
2004-02-28 05:26:06 +00:00
51ea127bf3
Rename AddUsesToWorkList -> AddUsersToWorkList, as that is what it does.
...
Create a new AddUsesToWorkList method
optimize memmove/set/cpy of zero bytes to a noop.
llvm-svn: 11941
2004-02-28 05:22:00 +00:00
f3a366062c
Turn 'free null' into nothing
...
llvm-svn: 11940
2004-02-28 04:57:37 +00:00
8a2c28fdda
Right, it's really Extractor, not Extraction.
...
llvm-svn: 11939
2004-02-28 03:37:58 +00:00
370c58a44a
New Function-level transformation utils.
...
llvm-svn: 11938
2004-02-28 03:33:30 +00:00
3151628a0d
Add the prototype for the LoopExtractor Pass.
...
llvm-svn: 11937
2004-02-28 03:33:17 +00:00
03a11340ff
A pass that uses the generic CodeExtractor to rip out *every* loop in every
...
function, as long as the loop isn't the only one in that function. This should
help debugging passes easier with BugPoint.
llvm-svn: 11936
2004-02-28 03:33:01 +00:00
caa1a5abeb
A generic code extractor: given a list of BasicBlocks, it will rip them out into
...
a new function, taking care of inputs and outputs.
llvm-svn: 11935
2004-02-28 03:26:20 +00:00
be74d5f73a
Two testcases for loops: one with outputs, one without.
...
llvm-svn: 11934
2004-02-28 03:20:41 +00:00
24b3d0bdae
Further comment updates.
...
llvm-svn: 11933
2004-02-28 03:20:31 +00:00
f87966b8c4
Update comments.
...
llvm-svn: 11932
2004-02-28 03:12:31 +00:00
2dbc79df84
My previous commit broke the jit. The shift instructions always take
...
an 8-bit immediate. So mark the shifts that take immediates as taking
an 8-bit argument. The rest with the implicit use of CL are marked
appropriately.
A bug still exists:
def SHLDmri32 : I2A8 <"shld", 0xA4, MRMDestMem>, TB; // [mem32] <<= [mem32],R32 imm8
The immediate in the above instruction is 8-bit but the memory
reference is 32-bit. The printer prints this as an 8-bit reference
which confuses the assembler. Same with SHRDmri32.
llvm-svn: 11931
2004-02-28 02:56:26 +00:00
ffeef64177
Turn off the SparcV9MachineCodeDestructionPass for now, because it's buggy
...
llvm-svn: 11930
2004-02-27 21:15:40 +00:00
50d19bf1de
Correct DestroyMachineFunction's getPassName
...
llvm-svn: 11929
2004-02-27 21:01:14 +00:00
06a573f63d
Only clone global nodes between graphs if both graphs have the global.
...
llvm-svn: 11928
2004-02-27 20:05:15 +00:00
98f8ca4017
ADD MORE FUNCTIONS!
...
llvm-svn: 11927
2004-02-27 20:04:48 +00:00
b10b04c5ec
Fix argument size for SHL, SHR, SAR, SHLD and SHRD families of
...
instructions.
llvm-svn: 11923
2004-02-27 19:46:30 +00:00
75ed0f67bf
Fix encoding of ADD and SUB family of instructions. Also rearrange
...
them so that they are consistent with AND, XOR, etc...
llvm-svn: 11922
2004-02-27 18:57:00 +00:00
58270fcf1f
Rename MRMS[0-7]{r,m} to MRM[0-7]{r,m}.
...
llvm-svn: 11921
2004-02-27 18:55:12 +00:00
e82c217b2f
setcond instructions don't have aliasing implications.
...
llvm-svn: 11919
2004-02-27 18:09:25 +00:00
a3b47cae4c
Added d and hbd MultiSource applications.
...
llvm-svn: 11915
2004-02-27 18:02:06 +00:00
3705370df1
Fix Regression/Assembler/2004-02-27-SelfUseAssertError.ll
...
llvm-svn: 11913
2004-02-27 17:28:25 +00:00
Alkis Evlogimenos