Commit Graph

15679 Commits

Author SHA1 Message Date
Chris Lattner 584a11ae22 Implement the getRegForInlineAsmConstraint method for PPC. With recent
sdisel changes, this eliminates a ton of copies around common inline asms.
For example:

int test2(int Y, int X) {
  asm("foo %0, %1" : "=r"(X): "r"(X));
  return X;
}

now compiles to:

_test2:
        foo r3, r4
        blr

instead of:

_test2:
        mr r2, r4
        foo r2, r2
        mr r3, r2
        blr

GCC produces:

_test2:
        foo r4, r4
        mr r3,r4
        blr

llvm-svn: 31367
2006-11-02 01:44:04 +00:00
Chris Lattner 55402d4403 Allow the getRegForInlineAsmConstraint method to return a register class with
no fixes physreg.  Treat this as permission to use any register in the register
class.  When this happens and it is safe, allow the llvm register allcoator to
allocate the register instead of doing it at isel time.  This eliminates a ton
of copies around common inline asms.  For example:

int test2(int Y, int X) {
  asm("foo %0, %1" : "=r"(X): "r"(X));
  return X;
}

now compiles to:

_test2:
        foo r3, r4
        blr

instead of:

_test2:
        mr r2, r4
        foo r2, r2
        mr r3, r2
        blr

GCC produces:

_test2:
        foo r4, r4
        mr r3,r4
        blr

llvm-svn: 31366
2006-11-02 01:41:49 +00:00
Evan Cheng 93cdd149f7 Rename
llvm-svn: 31364
2006-11-01 23:18:32 +00:00
Evan Cheng d8697deca3 Two-address instructions no longer have to be A := A op C. Now any pair of dest / src operands can be tied together.
llvm-svn: 31363
2006-11-01 23:06:55 +00:00
Devang Patel 2cb4f83b38 There can be more than one PHINode at the start of the block.
llvm-svn: 31362
2006-11-01 23:04:45 +00:00
Evan Cheng 1b2e60e970 Added getTiedToSrcOperand() to check for two-address'ness.
llvm-svn: 31360
2006-11-01 23:00:31 +00:00
Evan Cheng 1359196c4e Clean up.
llvm-svn: 31359
2006-11-01 22:39:30 +00:00
Devang Patel 44519a8feb Handle PHINode with only one incoming value.
This fixes http://llvm.org/bugs/show_bug.cgi?id=979

llvm-svn: 31358
2006-11-01 22:26:43 +00:00
Evan Cheng 47218fab42 CopyFromReg starts a live range so its use should not be considered a floater.
llvm-svn: 31356
2006-11-01 22:17:06 +00:00
Reid Spencer 20ffdbe7c1 Make ScalarEvolution actually use a ZeroExtend expression instead of
having SCZeroExtendExpr be equivalent to SCTruncate

llvm-svn: 31355
2006-11-01 21:53:12 +00:00
Chris Lattner 7cee6ddcd8 give branch folding a simple heuristic to decide which block to split so that
it inserts an uncond branch where it is less likely to cause a problem.  This
fixes some perf issues on ppc.

llvm-svn: 31354
2006-11-01 19:36:29 +00:00
Chris Lattner 5a0bd61c64 Fix GlobalOpt/2006-11-01-ShrinkGlobalPhiCrash.ll and McGill/chomp
llvm-svn: 31352
2006-11-01 18:03:33 +00:00
Rafael Espindola 3576405a26 print null values in bss
llvm-svn: 31349
2006-11-01 14:26:44 +00:00
Rafael Espindola 0cd8d14c7c implement zextload bool and truncstore bool
llvm-svn: 31348
2006-11-01 14:13:27 +00:00
Evan Cheng 91f120f4fa Add a printSetLabel that takes two id's.
llvm-svn: 31347
2006-11-01 09:23:08 +00:00
Chris Lattner eebea43b48 Factor gep instructions through phi nodes.
llvm-svn: 31346
2006-11-01 07:43:41 +00:00
Chris Lattner 14f82c7dcd Turn a phi of many loads into a phi of the address and a single load of the
result.  This can significantly shrink code and exposes identities more
aggressively.

llvm-svn: 31344
2006-11-01 07:13:54 +00:00
Evan Cheng 94e5bc9e83 Fix ldmxcsr JIT encoding.
llvm-svn: 31343
2006-11-01 06:53:52 +00:00
Chris Lattner dc826fc068 Fix a bug in the previous patch
llvm-svn: 31342
2006-11-01 04:55:47 +00:00
Chris Lattner cadac0c5c3 Fold things like "phi [add (a,b), add(c,d)]" into two phi's and one add.
This triggers thousands of times on multisource.

llvm-svn: 31341
2006-11-01 04:51:18 +00:00
Evan Cheng 415f365e5c Print jumptable index.
llvm-svn: 31340
2006-11-01 04:48:30 +00:00
Reid Spencer 5e2e0cb9aa Remove unnecessary sign conversions made possible by last patch.
llvm-svn: 31339
2006-11-01 03:45:43 +00:00
Reid Spencer f84930746e Fix a bug in the interpreter where divides of unmatched signed operands
would fail. E.g. udiv sint X, Y  or sdiv uint X, Y would fail to find a
type match in the switch statement and fail the operation.

llvm-svn: 31338
2006-11-01 03:41:05 +00:00
Chris Lattner f505a5a063 make tail merging more aggressive. If two blocks share a common tail, but the
tail is not an entire block for either of them, pick one, split it, then
merge the common part.

llvm-svn: 31336
2006-11-01 01:16:12 +00:00
Chris Lattner 3ac71b31f0 enable branch folding with an option
llvm-svn: 31335
2006-11-01 00:38:31 +00:00
Evan Cheng 7a6a5b9af5 Add constraints to Instruction class.
llvm-svn: 31332
2006-11-01 00:26:27 +00:00
Andrew Lenharth dfbf91e59d more shotenning
llvm-svn: 31331
2006-10-31 23:46:56 +00:00
Chris Lattner fe43befeda Compile CodeGen/PowerPC/fp-branch.ll to:
_intcoord_cond_next55:
LBB1_3: ;cond_next55
        lis r2, ha16(LCPI1_0)
        lfs f0, lo16(LCPI1_0)(r2)
        fcmpu cr0, f1, f0
        blt cr0, LBB1_2 ;cond_next62.exitStub
LBB1_1: ;bb72.exitStub
        li r3, 1
        blr
LBB1_2: ;cond_next62.exitStub
        li r3, 0
        blr

instead of:

_intcoord_cond_next55:
LBB1_3: ;cond_next55
        lis r2, ha16(LCPI1_0)
        lfs f0, lo16(LCPI1_0)(r2)
        fcmpu cr0, f1, f0
        bge cr0, LBB1_1 ;bb72.exitStub
LBB1_4: ;cond_next55
        lis r2, ha16(LCPI1_0)
        lfs f0, lo16(LCPI1_0)(r2)
        fcmpu cr0, f1, f0
        bnu cr0, LBB1_2 ;cond_next62.exitStub
LBB1_1: ;bb72.exitStub
        li r3, 1
        blr
LBB1_2: ;cond_next62.exitStub
        li r3, 0
        blr

llvm-svn: 31330
2006-10-31 23:06:00 +00:00
Chris Lattner 427301fdae look through isunordered to inline it into branch blocks.
llvm-svn: 31328
2006-10-31 22:37:42 +00:00
Evan Cheng a41c2d6432 Nuke dead code.
llvm-svn: 31327
2006-10-31 21:53:31 +00:00
Chris Lattner 44daa50bed allow the address of a global to be used with the "i" constraint when in
-static mode.  This implements PR882.

llvm-svn: 31326
2006-10-31 20:13:11 +00:00
Chris Lattner 60ce558aaa implement the 'c' inline asm modifier character
llvm-svn: 31325
2006-10-31 20:12:30 +00:00
Chris Lattner 1fd360e13a handle global address constant sdnodes
llvm-svn: 31323
2006-10-31 20:01:56 +00:00
Andrew Lenharth 8b20fa42da Let us play simplify the td file (and fix a few missed sub and mul patterns).
llvm-svn: 31322
2006-10-31 19:52:12 +00:00
Chris Lattner f6a6966cd2 handle "st" as "st(0)"
llvm-svn: 31320
2006-10-31 19:42:44 +00:00
Chris Lattner 6f043b90ea TargetLowering::isOperandValidForConstraint
llvm-svn: 31319
2006-10-31 19:41:18 +00:00
Chris Lattner 8c6949e5b2 Change the prototype for TargetLowering::isOperandValidForConstraint
llvm-svn: 31318
2006-10-31 19:40:43 +00:00
Chris Lattner 984d6e1669 generalize the fix for PR977 to also fix
Transforms/LCSSA/2006-10-31-UnreachableBlock-2.ll

llvm-svn: 31317
2006-10-31 18:56:48 +00:00
Chris Lattner eb68f080ef Fix PR977 and Transforms/LCSSA/2006-10-31-UnreachableBlock.ll
llvm-svn: 31315
2006-10-31 17:52:18 +00:00
Andrew Lenharth 692e4155aa Add all that branch mangling niftiness
llvm-svn: 31313
2006-10-31 16:49:55 +00:00
Rafael Espindola 4e825336a0 add support for calling functions when the caller has variable sized objects
llvm-svn: 31312
2006-10-31 13:03:26 +00:00
Anton Korobeynikov aa4c0f9374 1. Clean up code due to changes in SwitchTo*Section(2)
2. Added partial debug support for mingw\cygwin targets (the same as
   Linux\ELF). Please note, that currently mingw\cygwin uses 'stabs' format
   for storing debug info by default, thus many (runtime) libraries has
   this information included. These formats shouldn't be mixed in one binary
   ('stabs' & 'DWARF'), otherwise binutils tools will be confused.

llvm-svn: 31311
2006-10-31 08:31:24 +00:00
Chris Lattner 968f803928 Turn an assert into an error message. This is commonly triggered when
we don't support a specific constraint yet.  When this happens, print the
unsupported constraint.

llvm-svn: 31310
2006-10-31 07:33:13 +00:00
Anton Korobeynikov 49993529eb Unbreaking static ctors patch.
Defaulting second arguments of SwitchTo*Section, this should make things
somehow clearer.

llvm-svn: 31306
2006-10-31 06:11:06 +00:00
Evan Cheng e6d584765f Fix a typo which can break jumptables.
llvm-svn: 31305
2006-10-31 02:31:00 +00:00
Reid Spencer 2545fe6d93 Make this compile again.
llvm-svn: 31304
2006-10-31 01:45:56 +00:00
Evan Cheng de5ecd6bc1 Apply Aton's LLVM patch for PR973: Linux ctors / dtors support.
llvm-svn: 31303
2006-10-31 01:26:55 +00:00
Reid Spencer 7eda51d4cf Fix a problem introduced by a last-minute change (logic negation).
llvm-svn: 31302
2006-10-30 23:34:32 +00:00
Chris Lattner 3e36e07db2 fix miscompilation of llvm.isunordered, where we branched on the opposite
condition.  This fixes miscompilation of Olden/bh and many others.

llvm-svn: 31301
2006-10-30 23:02:25 +00:00
Reid Spencer b51b5c0b1f Add debug support for X86/ELF targets (Linux). This allows llvm-gcc4
generated object modules to be debugged with gdb. Hopefully this helps
pre-release debugging.

llvm-svn: 31299
2006-10-30 22:32:30 +00:00