Commit Graph

5454 Commits

Author SHA1 Message Date
Jim Grosbach
55dabaa73a ARM isel for vld1, opcode selection for register stride post-index pseudos.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143158 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-27 22:25:42 +00:00
Evan Cheng
824a70a384 Avoid partial CPSR dependency from loop backedges. rdar://10357570
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143145 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-27 21:21:05 +00:00
Jim Grosbach
88484c0030 Thumb2 t2LDMDB[_UPD] assembly parsing to recognize .w suffix.
rdar://10348844

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143110 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-27 17:33:59 +00:00
Jim Grosbach
036a67d670 Thumb2 t2MVNi assembly parsing to recognize ".w" suffix.
rdar://10348584


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143108 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-27 17:16:55 +00:00
Chad Rosier
6d64b3adab A branch predicated on a constant can just FastEmit an unconditional branch.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143086 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-27 00:21:16 +00:00
Chad Rosier
8ff2664f2f Add a TODO comment. FastISel works by parsing each basic block from the bottom
up.  Thus, improving the support for compares is goodness because it increases
the number of terminator instructions we can handle.  This creates many more 
opportunities for target specific fast-isel.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143079 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 23:34:37 +00:00
Chad Rosier
ade620065d Factor a little more code into EmitCmp, which should have been done in the first
place.  No functional change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143078 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 23:25:44 +00:00
Chad Rosier
75698f346f Use EmitCmp in SelectBranch. No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143076 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 23:17:28 +00:00
Chad Rosier
530f7cefd3 Factor out an EmitCmp function that can be used by both SelectCmp and
SelectBranch.  No functional change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143072 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 22:47:55 +00:00
Jim Grosbach
a581328ceb Thumb2 ldr pc-relative encoding fixes.
We were parsing label references to the i12 encoding, which isn't right.
They need to go to the pci variant instead.

More of rdar://10348687

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143068 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 22:22:01 +00:00
Jim Grosbach
758a519a22 ARM parse parenthesized expressions for label references.
Partial fix for rdar://10348687.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143063 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 21:14:08 +00:00
Lang Hames
75757f9fd1 Make sure short memsets on ARM lower to stores, even when optimizing for size.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143055 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 20:56:52 +00:00
Jim Grosbach
9c5edc00c4 Thumb2 remove redundant ".w" suffix from t2MVNCCi pattern.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143034 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 17:28:15 +00:00
James Molloy
015cca6c08 Revert r142530 at least temporarily while a discussion is had on llvm-commits regarding exactly how much optsize should optimize for size over performance.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143023 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 08:53:19 +00:00
Bill Wendling
de39d86f26 Use a worklist to prevent the iterator from becoming invalidated because of the 'removeSuccessor' call. Noticed in a Release+Asserts+Check buildbot.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143018 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 07:16:18 +00:00
Evan Cheng
f31151f34e Revert part of r142530. The patch potentially hurts performance especially
on Darwin platforms where -Os means optimize for size without hurting
performance.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143002 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-26 01:17:44 +00:00
Jim Grosbach
399cdca4d2 ARM assembly parsing and encoding for VLD1 with writeback.
Four entry register lists.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142882 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-25 00:14:01 +00:00
Jim Grosbach
b36e03d987 Nuke dead code. Nothing generates the VLD1d64QPseudo_UPD instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142877 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 23:40:46 +00:00
Jim Grosbach
5921675ff5 ARM assembly parsing and encoding for VLD1 w/ writeback.
Three entry register list variation.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142876 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 23:26:05 +00:00
Eli Friedman
5c89cb8cd6 Don't crash on variable insertelement on ARM. PR10258.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142871 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 23:08:52 +00:00
Evan Cheng
aed4952753 ARMConstantPoolMBB::print should print BB number.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142867 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 23:01:03 +00:00
Jim Grosbach
12431329d6 ARM assembly parsing and encoding for VLD1 w/ writeback.
One and two length register list variants.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142861 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 22:16:58 +00:00
Jim Grosbach
10b90a9bbf ARM refactor am6offset usage for VLD1.
Split am6offset into fixed and register offset variants so the instruction
encodings are explicit rather than relying an a magic reg0 marker.
Needed to being able to parse these.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142853 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 21:45:13 +00:00
Owen Anderson
a7c98f58ea Fix a NEON disassembly case that was broken in the recent refactorings. As more of this code gets refactored, a lot of these manual decoding hooks should get smaller and/or go away entirely.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142817 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 18:04:29 +00:00
Dan Gohman
692c1d8535 Change this overloaded use of Sched::Latency to be an overloaded
use of Sched::ILP instead, as Sched::Latency is going away.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142813 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 17:55:11 +00:00
Jim Grosbach
f8e74f816d Thumb2 LDM instructions can target PC. Make sure to encode it.
PR11220


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142801 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-24 17:16:24 +00:00
Benjamin Kramer
1a2f9886a2 Move various generated tables into read-only memory, fixing up const correctness along the way.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142726 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-22 16:50:00 +00:00
Bill Wendling
b8dcb314f7 The different flavors of ARM have different valid subsets of registers. Check
that the set of callee-saved registers is correct for the specific platform.
<rdar://problem/10313708> & ctor_dtor_count & ctor_dtor_count-2


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142706 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-22 00:29:28 +00:00
Jim Grosbach
224180e81b Assembly parsing for 4-register sequential variant of VLD2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142704 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 23:58:57 +00:00
Jim Grosbach
4661d4cac3 Assembly parsing for 2-register sequential variant of VLD2.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142691 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 22:21:10 +00:00
Jim Grosbach
b6310316db Assembly parsing for 4-register variant of VLD1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142682 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 20:35:01 +00:00
Jim Grosbach
cdcfa28056 Assembly parsing for 3-register variant of VLD1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142675 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 20:02:19 +00:00
Jim Grosbach
280dfad489 ARM VLD parsing and encoding.
Next step in the ongoing saga of NEON load/store assmebly parsing. Handle
VLD1 instructions that take a two-register register list.

Adjust the instruction definitions to only have the single encoded register
as an operand. The super-register from the pseudo is kept as an implicit def,
so passes which come after pseudo-expansion still know that the instruction
defines the other subregs.




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142670 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 18:54:25 +00:00
Owen Anderson
7784f1d2d8 Don't automatically set the "fc" bits on MSR instructions if the user didn't ask for them. This is a divergence from gas' behavior, but it is correct per the documentation and allows us to forge ahead with roundtrip testing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142669 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 18:43:28 +00:00
Jim Grosbach
7926db8268 Nuke an #if0 that got accidentally left in.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142658 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 16:59:08 +00:00
Jim Grosbach
293a5f69fa whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142657 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 16:56:40 +00:00
Jim Grosbach
fe7b4998c6 Remove some outdated comments.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142653 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-21 16:14:12 +00:00
Owen Anderson
cd20c58e98 Revert r142618, r142622, and r142624, which were based on an incorrect reading of the ARMv7 docs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142626 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 22:23:58 +00:00
Owen Anderson
a3a6f215d0 Separate out ARM MSR instructions into M-class versions and AR-class versions. This fixes some roundtripping failures.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142618 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 21:24:38 +00:00
Bill Wendling
767f8be9ee Add missing operand. <rdar://problem/10313323>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142615 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 20:37:11 +00:00
Jim Grosbach
01817c39a9 Tidy up. Trailing whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142591 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 17:28:20 +00:00
Jim Grosbach
6b09c77b7a ARM VLD1/VST1 (one register, no writeback) assembly parsing and encoding.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142583 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 15:04:25 +00:00
Jim Grosbach
d0b614754e ARM VTBX (one register) assembly parsing and encoding.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142581 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 14:48:50 +00:00
Chad Rosier
6690bca623 Revert 142337. Thumb1 still doesn't support dynamic stack realignment. :(
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142557 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-20 00:07:12 +00:00
James Molloy
cdd8e46bec Use literal pool loads instead of MOVW/MOVT for materializing global addresses when optimizing for size.
On spec/gcc, this caused a codesize improvement of ~1.9% for ARM mode and ~4.9% for Thumb(2) mode. This is
codesize including literal pools.

The pools themselves doubled in size for ARM mode and quintupled for Thumb mode, leaving suggestion that there
is still perhaps redundancy in LLVM's use of constant pools that could be decreased by sharing entries.

Fixes PR11087.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142530 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-19 14:11:07 +00:00
Bill Wendling
922ad78776 Make sure we emit the 'movw' and 'movt' only if it's supported. Otherwise, use a constant pool.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142485 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-19 09:24:02 +00:00
Bill Wendling
b677a135ac Remove some dead code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142484 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-19 09:04:11 +00:00
Bill Wendling
15a1a226be Emit the MOVT instruction only if the # LPads is > 64K.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142460 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-18 23:19:55 +00:00
Bill Wendling
a5871dc104 For Thumb mode, we need to use a constant pool if the value is too large to be
used with the CMP instruction.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142458 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-18 23:11:05 +00:00
Jim Grosbach
862019c37f ARM VTBL (one register) assembly parsing and encoding.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@142441 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-18 23:02:30 +00:00