Commit Graph

19998 Commits

Author SHA1 Message Date
Rafael Espindola
7240ae3cac Align i64 and f64 at 8 byte on x86-64.
This is mandated table 3.1 at
http://www.x86-64.org/documentation/abi.pdf



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41642 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-31 12:23:58 +00:00
Evan Cheng
d4e4937b79 Remove an unnecessary element, saving 4 bytes per LiveInterval.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41641 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-31 08:26:44 +00:00
Evan Cheng
c14b144ec3 Use std::map instead of a (potentially very sparse) array to track val# defined by copy from the other live range. Minor compile time win when number of val# is large.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41640 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-31 08:04:17 +00:00
Dale Johannesen
eaf089430e Enhance APFloat to retain bits of NaNs (fixes oggenc).
Use APFloat interfaces for more references, mostly
of ConstantFPSDNode.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41632 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-31 04:03:46 +00:00
Raul Herbster
8c132633c8 Instruction formats added used to generate multiply instructions of V5TE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41629 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 23:34:14 +00:00
Raul Herbster
b94e60872e Unused relocation type reloc_arm_absolute removed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41628 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 23:31:35 +00:00
Raul Herbster
9c1a3827ce Comments added. It now generates V5TE multiply instructions. However, it is still necessary to model PUWLSH bits more clearly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41627 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 23:29:26 +00:00
Raul Herbster
37fb5b154c ARM instruction table was modified by adding information to generate multiply instruction of V5TE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41626 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 23:25:47 +00:00
Raul Herbster
d05c04c169 JITInfo now resolves function addrs and also relocations. It always emits a stub.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41625 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 23:21:27 +00:00
Evan Cheng
f4c3a59dba Added support to fold X86 load / store instructions. This allow rematerialized loads to be folded into their uses.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41599 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 05:54:07 +00:00
Evan Cheng
34c2a9f57c Try fold re-materialized load instructions into its uses.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41598 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 05:53:02 +00:00
Evan Cheng
35b35c5c32 Add a variant of foldMemoryOperand to fold any load / store, not just load / store from / to stack slots.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41597 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 05:52:20 +00:00
Evan Cheng
2f39426ec9 Mark load instructions with isLoad = 1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41595 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 05:49:43 +00:00
Bill Wendling
b8a80f03bf Use i64 on a PPC64 machine
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41590 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 00:59:19 +00:00
Dale Johannesen
f04afdbb48 Change LegalFPImmediates to use APFloat.
Add APFloat interfaces to ConstantFP, SelectionDAG.
Fix integer bit in double->APFloat conversion.
Convert LegalizeDAG to use APFloat interface in
ConstantFPSDNode uses.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41587 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-30 00:23:21 +00:00
Anton Korobeynikov
055c5449a4 Fix use of declaration inside case block
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41584 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-29 23:18:48 +00:00
Evan Cheng
7ecb38be0a Change LiveRange so it keeps a pointer to the VNInfo rather than an index.
Changes related modules so VNInfo's are not copied. This decrease
copy coalescing time by 45% and overall compilation time by 10% on siod.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41579 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-29 20:45:00 +00:00
Anton Korobeynikov
066f7b40f8 Lower FRAME_TO_ADDR_OFFSET to zero by default (if not custom lowered)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41578 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-29 19:28:29 +00:00
Duncan Sands
ee46574931 Move getX86RegNum into X86RegisterInfo and use it
in the trampoline lowering.  Lookup the jump and
mov opcodes for the trampoline rather than hard
coding them.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41577 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-29 19:01:20 +00:00
Chris Lattner
d80fb8b366 update these
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41565 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-29 16:15:23 +00:00
Dan Gohman
3e1a7aef17 Add an option, -view-sunit-dags, for viewing the actual SUnit DAGs used by
scheduling.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41556 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 20:32:58 +00:00
Evan Cheng
1a66f0a4f2 Recover most of the compile time regression due to recent live interval changes.
1. Eliminate the costly live interval "swapping".
2. Change ValueNumberInfo container from SmallVector to std::vector. The former
   performs slowly when the vector size is very large.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41536 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 08:28:51 +00:00
Bruno Cardoso Lopes
51195af45f Added method to get Mips register numbers
Changed the stack frame layout, StackGrowsUp fits better to Mips strange stack.
Stack offset calculation bug fixed!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41529 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 05:13:42 +00:00
Bruno Cardoso Lopes
a2b1bb5296 Changed stack allocation On LowerFORMAL_ARGUMENTS.
Added comments about new stack allocation.
Expand SelectCC for i32 results


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41527 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 05:08:16 +00:00
Bruno Cardoso Lopes
dc0c04c0ed Mask directive completed with CalleeSave info
Comments for Mips directives added.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41526 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 05:06:17 +00:00
Bruno Cardoso Lopes
2d4575ea5a Added methods to record SPOffsets from LowerFORMAL_ARGUMENTS
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41525 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 05:04:41 +00:00
Chris Lattner
92103de8ce Cut off crazy computation. This helps PR1622 slightly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41522 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-28 04:23:55 +00:00
Devang Patel
1cd39a4f6e Use simpler test to filter loops.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41516 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 21:34:31 +00:00
David Greene
f1355a55f8 Update InvokeInst to work like CallInst
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41506 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 19:04:21 +00:00
Rafael Espindola
44c8265cf8 Add a comment about using libc memset/memcpy or generating inline code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41502 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 17:48:26 +00:00
Dan Gohman
e9c8fa095e Make DAGCombiner's global alias analysis query more precise in the case
where both pointers have non-zero offsets.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41491 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 16:32:11 +00:00
Dan Gohman
5f43f92c69 If the source and destination pointers in an llvm.memmove are known
to not alias each other, it can be translated as an llvm.memcpy.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41489 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 16:26:13 +00:00
Dan Gohman
e3784951af Change comments to refer to @malloc and @free instead of %malloc and %free.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41488 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 16:11:48 +00:00
Duncan Sands
6590b0457c There is an impedance matching problem between LLVM and
gcc exception handling: if an exception unwinds through
an invoke, then execution must branch to the invoke's
unwind target.  We previously tried to enforce this by
appending a cleanup action to every selector, however
this does not always work correctly due to an optimization
in the C++ unwinding runtime: if only cleanups would be
run while unwinding an exception, then the program just
terminates without actually executing the cleanups, as
invoke semantics would require.  I was hoping this
wouldn't be a problem, but in fact it turns out to be the
cause of all the remaining failures in the LLVM testsuite
(these also fail with -enable-correct-eh-support, so turning
on -enable-eh didn't make things worse!).  Instead we need
to append a full-blown catch-all to the end of each
selector.  The correct way of doing this depends on the
personality function, i.e. it is language dependent, so
can only be done by gcc.  Thus this patch which generalizes
the eh.selector intrinsic so that it can handle all possible
kinds of action table entries (before it didn't accomodate
cleanups): now 0 indicates a cleanup, and filters have to be
specified using the number of type infos plus one rather than
the number of type infos.  Related gcc patches will cause
Ada to pass a cleanup (0) to force the selector to always
fire, while C++ will use a C++ catch-all (null).


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41484 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 15:47:50 +00:00
Dan Gohman
81975f6dfd Add explicit keywords and remove spurious trailing semicolons.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41482 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 14:50:10 +00:00
Rafael Espindola
6b83b5d1ae call libc memcpy/memset if array size is bigger then threshold.
Coping 100MB array (after a warmup) shows that glibc 2.6.1 implementation on
x86-64 (core 2) is 30% faster (from 0.270917s to 0.188079s)



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41479 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-27 10:18:20 +00:00
Anton Korobeynikov
9f528e6280 Don't promote volatile loads/stores. This is needed (for example) to handle setjmp/longjmp properly.
This fixes PR1520.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41461 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-26 21:43:30 +00:00
Owen Anderson
5dfcf4318a Don't DSe volatile stores.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41456 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-26 21:14:47 +00:00
Dale Johannesen
e6c1742914 Revise per review comments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41409 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-26 01:18:27 +00:00
Dale Johannesen
87503a63d5 Add APFloat interface to ConstantFPSDNode. Change
over uses in DAGCombiner.  Fix interfaces to work
with APFloats.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41407 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 22:10:57 +00:00
Devang Patel
d79faee69a Move exit condition and exit branch from exiting block into loop header and dominator info. This avoid execution of dead iteration. Loop is already filter in the beginning such that this change is safe.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41394 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 02:39:24 +00:00
Devang Patel
be97c986e1 Constant split values needs upper bound and lower bound check, just like any other split value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41389 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 01:09:14 +00:00
Chris Lattner
09ec1b0583 Allow target constants to be illegal types. The target should
know how to handle them.  This fixes 
test/CodeGen/Generic/asm-large-immediate.ll


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41388 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 01:00:22 +00:00
Devang Patel
4a69da9cb0 While calculating upper loop bound for first loop and lower loop bound for second loop, take care of edge cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41387 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 00:56:38 +00:00
Chris Lattner
7df31dc89b Teach the dag scheduler to handle inline asm nodes with multi-value immediate operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41386 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 00:53:07 +00:00
Bill Wendling
cf4bb317ef The personality function might need to be declared as:
.set  Lset0,___gxx_personality_v0-.
    .long Lset0

on some targets. Make it so!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41385 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 00:51:55 +00:00
Chris Lattner
48884cd80b rename isOperandValidForConstraint to LowerAsmOperandForConstraint,
changing the interface to allow for future changes.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41384 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-25 00:47:38 +00:00
Dale Johannesen
12595d7b16 Poison APFloat::operator==. Replace existing uses with bitwiseIsEqual.
This means backing out the preceding change to Constants.cpp, alas.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41378 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-24 22:09:56 +00:00
Dale Johannesen
8bb369b807 Use APFloat internally for ConstantFPSDNode.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41372 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-24 20:59:15 +00:00
Chris Lattner
cd406fe123 sink clone() down the class hierarchy from CmpInst into ICmpInst/FCmpInst.
This eliminates a conditional on that path, and ensures ICmpInst/FCmpInst
both have an out-of-line virtual method to home the class.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@41371 91177308-0d34-0410-b5e6-96231b3b80d8
2007-08-24 20:48:18 +00:00