Chris Lattner [Sun, 5 Sep 2010 02:18:34 +0000 (02:18 +0000)]
implement rdar://
6653118 - fastisel should fold loads where possible.
Since mem2reg isn't run at -O0, we get a ton of reloads from the stack,
for example, before, this code:
int foo(int x, int y, int z) {
return x+y+z;
}
used to compile into:
_foo: ## @foo
subq $12, %rsp
movl %edi, 8(%rsp)
movl %esi, 4(%rsp)
movl %edx, (%rsp)
movl 8(%rsp), %edx
movl 4(%rsp), %esi
addl %edx, %esi
movl (%rsp), %edx
addl %esi, %edx
movl %edx, %eax
addq $12, %rsp
ret
Now we produce:
_foo: ## @foo
subq $12, %rsp
movl %edi, 8(%rsp)
movl %esi, 4(%rsp)
movl %edx, (%rsp)
movl 8(%rsp), %edx
addl 4(%rsp), %edx ## Folded load
addl (%rsp), %edx ## Folded load
movl %edx, %eax
addq $12, %rsp
ret
Fewer instructions and less register use = faster compiles.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113102
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Sat, 4 Sep 2010 21:09:33 +0000 (21:09 +0000)]
Remove dead code.
Clobber ranges are no longer used when joining physical registers.
Instead, all aliases are checked for interference.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113084
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 18:45:02 +0000 (18:45 +0000)]
fix this to work with allocators that have reference type with compilers
that diagnose invalid references to references.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113078
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 18:19:16 +0000 (18:19 +0000)]
dead method.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113077
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 18:17:16 +0000 (18:17 +0000)]
delete dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113075
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 18:12:00 +0000 (18:12 +0000)]
zap dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113073
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 18:02:47 +0000 (18:02 +0000)]
remove dead code, mblaze uses SelectAddrRegImm/SelectAddrRegReg,
not SelectAddr
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113072
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 17:45:58 +0000 (17:45 +0000)]
zap dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113071
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 02:58:56 +0000 (02:58 +0000)]
Remove the last bit of isShuffleMaskLegal checks and improve the comment regarding mmx shuffles
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113059
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 02:50:13 +0000 (02:50 +0000)]
make explicit that we not handle several mmx shuffles
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113058
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Sat, 4 Sep 2010 02:42:48 +0000 (02:42 +0000)]
Fix LoopSimplify to notify ScalarEvolution when splitting a loop backedge
into an inner loop, as the new loop iteration may differ substantially.
This fixes PR8078.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113057
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 02:36:07 +0000 (02:36 +0000)]
Emit target specific nodes to handle palignr. Do not touch it for MMX versions yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113056
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 02:02:14 +0000 (02:02 +0000)]
Emit target specific nodes to handle splats starting at zero indicies
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113055
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 01:36:45 +0000 (01:36 +0000)]
Emit target specific nodes for isPSHUFHWMask and isPSHUFLWMask
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113050
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 01:22:57 +0000 (01:22 +0000)]
Emit target specific nodes for isSHUFPMask
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113048
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 00:50:08 +0000 (00:50 +0000)]
Previous isMOVLMask matching already emits targets nodes, remove check
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113047
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 00:46:16 +0000 (00:46 +0000)]
One more check from the original isShuffleMaskLegal goes away
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113045
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 00:43:12 +0000 (00:43 +0000)]
Remove a duplicated but useless check that i've inserted in the previous commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113044
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Sat, 4 Sep 2010 00:39:43 +0000 (00:39 +0000)]
Refactor some code and remove the extra checks for unpckl_undef and unpckh_undef
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113043
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 4 Sep 2010 00:12:30 +0000 (00:12 +0000)]
fix a bug in my licm rewrite when a load from the promoted memory
location is being re-stored to the memory location. We would get
a dangling pointer from the SSAUpdate data structure and miss a
use. This fixes PR8068
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113042
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 23:32:47 +0000 (23:32 +0000)]
Remove check for unpckh mask
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113035
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 23:31:50 +0000 (23:31 +0000)]
Remove check for unpckl mask
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113034
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 23:24:06 +0000 (23:24 +0000)]
Inline isShuffleMaskLegal into LowerVECTOR_SHUFFLE, so we can start
checking each standalone condition and decide whether emit target
specific nodes or remove the condition if it's already matched before.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113031
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Fri, 3 Sep 2010 22:47:08 +0000 (22:47 +0000)]
Propagate non-local comparisons. Fixes PR1757.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113025
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Fri, 3 Sep 2010 22:46:31 +0000 (22:46 +0000)]
Reapply this increase to the number of virtual registers. All of the
various breakages appear to be dealt with.
Patch by Pekka Jääskeläinen.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113024
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 3 Sep 2010 22:12:56 +0000 (22:12 +0000)]
Disable the asserts that check that normalization is perfectly
invertible. ScalarEvolution's folding routines don't always succeed
in canonicalizing equal expressions to a single canonical form, and
this can cause these asserts to fail, even though there's no actual
correctness problem. This fixes PR8066.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113021
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 22:09:41 +0000 (22:09 +0000)]
Reapply considered harmfull part of rr112934 and r112942.
"Use target specific nodes instead of relying in unpckl and
unpckh pattern fragments during isel time. Also place a
depth limit in getShuffleScalarElt.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113020
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Fri, 3 Sep 2010 21:45:15 +0000 (21:45 +0000)]
previous patch was a little too tricky for its own good. Don't try to
overload UserInInstr. Explicitly check Allocatable. The early exit in the
condition will mean the performance impact of the extra test should be
minimal.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113016
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Fri, 3 Sep 2010 21:23:00 +0000 (21:23 +0000)]
Remove the rest of the nonexistent 64-bit AVX instructions.
Bruno, please review.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113014
91177308-0d34-0410-b5e6-
96231b3b80d8
David Greene [Fri, 3 Sep 2010 21:00:49 +0000 (21:00 +0000)]
Generalize getFieldType to work on all TypedInits. Add a couple of testcases from
Amaury Pouly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113010
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 20:44:26 +0000 (20:44 +0000)]
Reapply last harmless part of r112934, the pattern fragment to match X86Unpcklpd
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113009
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 20:20:02 +0000 (20:20 +0000)]
Reintroduce a simple function refactoring done in r112934, also without any functionality changes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113008
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 20:10:35 +0000 (20:10 +0000)]
Reapply piecies of r112942 and r112934 which don't do
functional changes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113007
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 19:55:05 +0000 (19:55 +0000)]
Reapply Fix comment
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113006
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Fri, 3 Sep 2010 19:38:11 +0000 (19:38 +0000)]
Revert r112934, "- Use specific nodes to match unpckl masks.", which introduced
some infinite loop and select failures.
- Apologies for eager reverting, but its branch day.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@113000
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Fri, 3 Sep 2010 19:38:08 +0000 (19:38 +0000)]
Revert r112938 "Fix comment", which depends on r112934, which introduced some
infinite loop and select failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112999
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Fri, 3 Sep 2010 19:38:05 +0000 (19:38 +0000)]
Revert r112942, "Use punpckh and unpckh family of nodes instead of using unpckh
mask pattern fragment", which depends on r112934, which introduced some infinite
loop and select failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112998
91177308-0d34-0410-b5e6-
96231b3b80d8
Bob Wilson [Fri, 3 Sep 2010 19:20:37 +0000 (19:20 +0000)]
Add a missing check when legalizing a vector extending load. This doesn't
solve the root problem, but it corrects the bug in the code I added to
support legalizing in the case where the non-extended type is also legal.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112997
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 3 Sep 2010 19:09:46 +0000 (19:09 +0000)]
Get rid of "passing signed into unsigned parameter" warning on PPC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112995
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Fri, 3 Sep 2010 19:08:37 +0000 (19:08 +0000)]
Add support for simplifying a load from a computed value to a load from a global when it
is provable that they're equivalent. This fixes PR4855.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112994
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 3 Sep 2010 18:41:20 +0000 (18:41 +0000)]
Quiesce warning about non-virtual d'tor in virtual class.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112991
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Fri, 3 Sep 2010 18:37:12 +0000 (18:37 +0000)]
Re-apply r112883:
"For ARM stack frames that utilize variable sized objects and have either
large local stack areas or require dynamic stack realignment, allocate a
base register via which to access the local frame. This allows efficient
access to frame indices not accessible via the FP (either due to being out
of range or due to dynamic realignment) or the SP (due to variable sized
object allocation). In particular, this greatly improves efficiency of access
to spill slots in Thumb functions which contain VLAs."
r112986 fixed a latent bug exposed by the above.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112989
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Fri, 3 Sep 2010 18:36:56 +0000 (18:36 +0000)]
VirtRegRewriter checks for early clobbers before it reuses an available stack
slot.
Teach it to also check for early clobbered aliases, and early clobber operands
following the current operand.
This fixes the miscompilation in PR8044 where EC registers eax and ecx were
being used for inputs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112988
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Fri, 3 Sep 2010 18:33:08 +0000 (18:33 +0000)]
Add a test for PR4413, which was apparently fixed at some point in the past.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112987
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Fri, 3 Sep 2010 18:28:19 +0000 (18:28 +0000)]
Check the local frame alignment for determining whether dynamic stack
alignment should be performed. Otherwise dynamic realignment may trigger
when the register allocator has already used the frame pointer as a general
purpose register. That is, we need to make sure that the list of reserved
registers doesn't change after register allocation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112986
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Fri, 3 Sep 2010 18:17:16 +0000 (18:17 +0000)]
Add 'const' to getter function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112984
91177308-0d34-0410-b5e6-
96231b3b80d8
Bob Wilson [Fri, 3 Sep 2010 18:16:02 +0000 (18:16 +0000)]
Finish converting the rest of the NEON VLD instructions to use pseudo-
instructions prior to regalloc. Since it's getting a little close to
the 2.8 branch deadline, I'll have to leave the rest of the instructions
handled by the NEONPreAllocPass for now, but I didn't want to leave half
of the VLD instructions converted and the other half not.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112983
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 3 Sep 2010 18:06:11 +0000 (18:06 +0000)]
Remove a TODO comment; this is now filed in bugzilla (PR8005).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112982
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 3 Sep 2010 17:29:33 +0000 (17:29 +0000)]
Fix Windows stuff to follow the existing source organization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112976
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Fri, 3 Sep 2010 16:58:25 +0000 (16:58 +0000)]
Add PR number to test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112971
91177308-0d34-0410-b5e6-
96231b3b80d8
Mikhail Glushenkov [Fri, 3 Sep 2010 15:42:45 +0000 (15:42 +0000)]
Rename 'libprofile/exported_symbols.lst' to 'libprofile/libprofile.exports'.
Makefile.rules assumes that all EXPORTED_SYMBOL_FILEs have the .exports
extension.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112964
91177308-0d34-0410-b5e6-
96231b3b80d8
Mikhail Glushenkov [Fri, 3 Sep 2010 15:42:38 +0000 (15:42 +0000)]
Revert "Rules for win32 ld shouldn't fire when cross-compiling."
This reverts commit
012b374e5b46fe93a8fdf605f5929fa8239e0960.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112963
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Fri, 3 Sep 2010 15:26:42 +0000 (15:26 +0000)]
Revert "For ARM stack frames that utilize variable sized objects and have either", it is breaking oggenc with Clang for ARMv6.
This reverts commit
8d6e29cfda270be483abf638850311670829ee65.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112962
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Fri, 3 Sep 2010 12:13:18 +0000 (12:13 +0000)]
Zap dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112955
91177308-0d34-0410-b5e6-
96231b3b80d8
Mikhail Glushenkov [Fri, 3 Sep 2010 09:57:52 +0000 (09:57 +0000)]
Rules for win32 ld shouldn't fire when cross-compiling.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112954
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Fri, 3 Sep 2010 08:31:48 +0000 (08:31 +0000)]
Reapply commit 112702 which was speculatively reverted by echristo.
Original commit message:
Use the SSAUpdator to turn calls to eh.exception that are not in a
landing pad into uses of registers rather than loads from a stack
slot. Doesn't touch the 'orrible hack code - Bill needs to persuade
me harder :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112952
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Fri, 3 Sep 2010 04:16:28 +0000 (04:16 +0000)]
Re-enable CorrelatedValuePropagation. I've tested nightly tests, llvm-gcc
bootstrap, and clang self-host.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112948
91177308-0d34-0410-b5e6-
96231b3b80d8
NAKAMURA Takumi [Fri, 3 Sep 2010 03:24:08 +0000 (03:24 +0000)]
test/CodeGen/X86: Add explicit -mtriple=(i686|x86_64)-linux for Win32 host.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112947
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Fri, 3 Sep 2010 02:22:23 +0000 (02:22 +0000)]
Fix cmake build without native target selected.
The cmake (+ MSVC) build is broken if you don't select your native
target.
e.g. 'cmake -D LLVM_TARGETS_TO_BUILD="MyNonNativeTarget" .'
This is because cmake currently sets the LLVM_NATIVE_* definitions
regardless of whether the native target is selected (causing build
errors).
Patch by Mike Gist!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112946
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 02:08:45 +0000 (02:08 +0000)]
AVX doesn't support mm operations neither its instrinsics.
The AVX versions of PALIGN and PABS* should only exist for
128-bit. Remove the unnecessary stuff.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112944
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 01:39:08 +0000 (01:39 +0000)]
Use punpckh and unpckh family of nodes instead of using unpckh mask pattern fragment
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112942
91177308-0d34-0410-b5e6-
96231b3b80d8
Bob Wilson [Fri, 3 Sep 2010 01:35:08 +0000 (01:35 +0000)]
Replace NEON vabdl, vaba, and vabal intrinsics with combinations of the
vabd intrinsic and add and/or zext operations. In the case of vaba, this
also avoids the need for a DAG combine pattern to combine vabd with add.
Update tests. Auto-upgrade the old intrinsics.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112941
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 01:28:51 +0000 (01:28 +0000)]
Fix comment
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112938
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Fri, 3 Sep 2010 01:24:00 +0000 (01:24 +0000)]
- Use specific nodes to match unpckl masks.
- Teach getShuffleScalarElt how to handle more target
specific nodes, so the DAGCombine can make use of it.
- Add another hack to avoid the node update problem
during legalization. More description on the comments
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112934
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Fri, 3 Sep 2010 00:39:50 +0000 (00:39 +0000)]
Stop emitting predicate functions. They are no longer used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112924
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Fri, 3 Sep 2010 00:35:47 +0000 (00:35 +0000)]
Simple branch instruction support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112923
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Fri, 3 Sep 2010 00:35:18 +0000 (00:35 +0000)]
Don't call Predicate_* from X86 target.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112921
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Fri, 3 Sep 2010 00:35:16 +0000 (00:35 +0000)]
Remove Predicate_* calls from MBlaze and XCore
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112920
91177308-0d34-0410-b5e6-
96231b3b80d8
Jakob Stoklund Olesen [Fri, 3 Sep 2010 00:35:13 +0000 (00:35 +0000)]
Remove Predicate_* calls from Mips
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112919
91177308-0d34-0410-b5e6-
96231b3b80d8
Eric Christopher [Thu, 2 Sep 2010 23:43:26 +0000 (23:43 +0000)]
Add basic support for materializing constants (including fp) and
stores.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112912
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:32:55 +0000 (23:32 +0000)]
update one more test
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112910
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:22:50 +0000 (23:22 +0000)]
this is a release note worthy change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112907
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:21:44 +0000 (23:21 +0000)]
add a new "llvm-dis -show-annotations" option, which causes it to print
#uses comments, with a testcase.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112906
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:09:42 +0000 (23:09 +0000)]
lets get crazy and name the header file the exact class name,
not a scrunched version of it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112904
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:07:12 +0000 (23:07 +0000)]
AsmPrinter has a formatted stream, pass it down through AsmAnnotationWriter
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112903
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 2 Sep 2010 23:03:46 +0000 (23:03 +0000)]
Properly emit __chkstk call instead of __alloca on non-mingw windows targets.
Patch by Cameron Esfahani!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112902
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 23:03:10 +0000 (23:03 +0000)]
stop forcing a noop AssemblyAnnotationWriter to silence #uses
comments, these don't happen anymore.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112901
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Thu, 2 Sep 2010 23:01:10 +0000 (23:01 +0000)]
There is no need to use .set here.
Thanks Chris!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112900
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:52:10 +0000 (22:52 +0000)]
reapply 112894:
Remove #uses comments from functions: they we're padded out to column 50
and were potentially confusing for externally visible functions.
going further, remove the "<i8**> [#uses=3]" comments entirely. They
add a lot of noise, confuse people about what the IR is, and don't add
any particular value. When the types are long it makes it really really
hard to read IR.
If someone is interested in this sort of thing, the right way to do this
is to implement an AsmAnnotationWriter that produces the same output, and
add a flag to llvm-dis (only) to produce this output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112899
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Thu, 2 Sep 2010 22:43:39 +0000 (22:43 +0000)]
Move insertps mask decoding to header file
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112896
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:42:07 +0000 (22:42 +0000)]
revert patch, need to update clang tests too
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112895
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:41:24 +0000 (22:41 +0000)]
Remove #uses comments from functions: they we're padded out to column 50
and were potentially confusing for externally visible functions.
going further, remove the "<i8**> [#uses=3]" comments entirely. They
add a lot of noise, confuse people about what the IR is, and don't add
any particular value. When the types are long it makes it really really
hard to read IR.
If someone is interested in this sort of thing, the right way to do this
is to implement an AsmAnnotationWriter that produces the same output, and
add a flag to llvm-dis (only) to produce this output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112894
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:38:56 +0000 (22:38 +0000)]
more test cleanup
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112892
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:35:33 +0000 (22:35 +0000)]
remove some noise from tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112889
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 2 Sep 2010 22:32:38 +0000 (22:32 +0000)]
Use the FindProgramByName fallback only with known absolute paths.
I wasn't able to convince myself that all GetMainExecutable
implementations always return absolute paths; this prevents
unexpected behavior in case they ever don't.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112888
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:32:02 +0000 (22:32 +0000)]
we are past the point where these tests are useful.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112887
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Thu, 2 Sep 2010 22:31:53 +0000 (22:31 +0000)]
Revert int_x86_mmx_palignr_b intrinsic to match llvm-gcc's version.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112886
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 2 Sep 2010 22:31:32 +0000 (22:31 +0000)]
Revert win64 changes. They seem to be incomplete
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112885
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Thu, 2 Sep 2010 22:29:01 +0000 (22:29 +0000)]
For ARM stack frames that utilize variable sized objects and have either
large local stack areas or require dynamic stack realignment, allocate a
base register via which to access the local frame. This allows efficient
access to frame indices not accessible via the FP (either due to being out
of range or due to dynamic realignment) or the SP (due to variable sized
object allocation). In particular, this greatly improves efficiency of access
to spill slots in Thumb functions which contain VLAs.
rdar://
7352504
rdar://
8374540
rdar://
8355680
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112883
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Thu, 2 Sep 2010 22:26:35 +0000 (22:26 +0000)]
- Change __builtin_ia32_palignr intrinsic type to match the pattern in clang.
- Add patterns to match the following MMX builtins:
* __builtin_ia32_vec_init_v8qi
* __builtin_ia32_vec_init_v4hi
* __builtin_ia32_vec_init_v2si
* __builtin_ia32_vec_ext_v2si
These builtins do not correspond to a single MMX instruction. They will have
to be lowered -- most likely in the back-end.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112881
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Thu, 2 Sep 2010 22:19:10 +0000 (22:19 +0000)]
fix more AST updating bugs, correcting miscompilation in PR8041
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112878
91177308-0d34-0410-b5e6-
96231b3b80d8
Owen Anderson [Thu, 2 Sep 2010 22:16:52 +0000 (22:16 +0000)]
Remove incorrect and poorly tested code for trying to reason about values on default edges of
switches. Just return the conservatively correct answer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112876
91177308-0d34-0410-b5e6-
96231b3b80d8
Anton Korobeynikov [Thu, 2 Sep 2010 22:16:28 +0000 (22:16 +0000)]
Properly allocate win64 shadow reg area.
Patch by Jan Sjodin!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112875
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 2 Sep 2010 22:14:51 +0000 (22:14 +0000)]
Remove obsolete comments about throwing exceptions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112874
91177308-0d34-0410-b5e6-
96231b3b80d8
John McCall [Thu, 2 Sep 2010 21:55:03 +0000 (21:55 +0000)]
After some discussion with djg, teach SmallVector to grow from a zero
capacity and remove the workaround in SmallVector<T,0>. There are some
theoretical benefits to a N->2N+1 growth policy anyway.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112870
91177308-0d34-0410-b5e6-
96231b3b80d8
Bruno Cardoso Lopes [Thu, 2 Sep 2010 21:51:11 +0000 (21:51 +0000)]
Move decoding of insertps back to avoid unused warnings in x86 isel lowering, and fix movlhps/movhlps to decode 4 elements shuffles
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112869
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Thu, 2 Sep 2010 21:29:42 +0000 (21:29 +0000)]
Detect undef value early and save unnecessary NodeMap query.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112864
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Thu, 2 Sep 2010 21:18:42 +0000 (21:18 +0000)]
Don't narrow the load and store in a load+twiddle+store sequence unless
there are clearly no stores between the load and the store. This fixes
this miscompile reported as PR7833.
This breaks the test/CodeGen/X86/narrow_op-2.ll optimization, which is
safe, but awkward to prove safe. Move it to X86's README.txt.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112861
91177308-0d34-0410-b5e6-
96231b3b80d8
Devang Patel [Thu, 2 Sep 2010 21:02:27 +0000 (21:02 +0000)]
Tidy up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112858
91177308-0d34-0410-b5e6-
96231b3b80d8
Sandeep Patel [Thu, 2 Sep 2010 20:19:24 +0000 (20:19 +0000)]
Fix an unnecessary XFAIL
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112853
91177308-0d34-0410-b5e6-
96231b3b80d8