Commit Graph

821 Commits

Author SHA1 Message Date
Owen Anderson
080c092297 Add codegen and encoding support for the immediate form of vbic.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118291 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-05 19:27:46 +00:00
Evan Cheng
416941d50f Fix @llvm.prefetch isel. Selecting between pld / pldw using the first immediate rw. There is currently no intrinsic that matches to pli.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118237 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-04 05:19:35 +00:00
Owen Anderson
60f4870c22 Covert VORRIMM to be produced via early target-specific DAG combining, rather than legalization.
This is both the conceptually correct place for it, as well as allowing it to be more aggressive.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118204 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 23:15:26 +00:00
Owen Anderson
d966817f3c Add support for code generation of the one register with immediate form of vorr.
We could be more aggressive about making this work for a larger range of constants,
but this seems like a good start.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118201 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 22:44:51 +00:00
Evan Cheng
949c910fb3 Fix test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118187 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 18:21:33 +00:00
Bob Wilson
d168cef664 Add codegen patterns for VST1-lane instructions. Radar 8599955.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118176 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 16:24:53 +00:00
Bob Wilson
3468c2ee20 Check for extractelement with a variable operand for the element number.
For NEON we had been assuming this was always an immediate constant.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118175 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 16:24:50 +00:00
Evan Cheng
dfed19fe2c Fix preload instruction isel. Only v7 supports pli, and only v7 with mp extension supports pldw. Add subtarget attribute to denote mp extension support and legalize illegal ones to nothing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118160 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 06:34:55 +00:00
Evan Cheng
bc7deb0f75 Add support to match @llvm.prefetch to pld / pldw / pli. rdar://8601536.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118152 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 05:14:24 +00:00
Evan Cheng
8239daf7c8 Two sets of changes. Sorry they are intermingled.
1. Fix pre-ra scheduler so it doesn't try to push instructions above calls to
   "optimize for latency". Call instructions don't have the right latency and
   this is more likely to use introduce spills.
2. Fix if-converter cost function. For ARM, it should use instruction latencies,
   not # of micro-ops since multi-latency instructions is completely executed
   even when the predicate is false. Also, some instruction will be "slower"
   when they are predicated due to the register def becoming implicit input.
   rdar://8598427


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118135 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-03 00:45:17 +00:00
John Thompson
45c21ff044 Inline asm mult-alt constraint tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118107 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-02 23:01:44 +00:00
Jim Grosbach
ab3d00e535 Revert r114340 (improvements in Darwin function prologue/epilogue), as it broke
assumptions about stack layout. Specifically, LR must be saved next to FP.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@118026 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-02 17:35:25 +00:00
Bob Wilson
665814b6be Add support for alignment operands on VLD1-lane instructions.
This is another part of the fix for Radar 8599955.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117976 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 23:40:51 +00:00
Bob Wilson
baf0615426 Add VLD1-lane testcases for quad-register types.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117975 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 23:40:46 +00:00
Bob Wilson
b796bbb6de Add NEON VLD1-lane instructions. Partial fix for Radar 8599955.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117964 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 22:04:05 +00:00
Bill Wendling
40a5eb18b0 When we look at instructions to convert to setting the 's' flag, we need to look
at more than those which define CPSR. You can have this situation:

(1)  subs  ...
(2)  sub   r6, r5, r4
(3)  movge ...
(4)  cmp   r6, 0
(5)  movge ...

We cannot convert (2) to "subs" because (3) is using the CPSR set by
(1). There's an analogous situation here:

(1)  sub   r1, r2, r3
(2)  sub   r4, r5, r6
(3)  cmp   r4, ...
(5)  movge ...
(6)  cmp   r1, ...
(7)  movge ...

We cannot convert (1) to "subs" because of the intervening use of CPSR.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117950 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 20:41:43 +00:00
Bob Wilson
24645a1a6d NEON does not support truncating vector stores. Radar 8598391.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117940 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 18:31:39 +00:00
Bill Wendling
504fb4fb8e More tests to XFAIL. The arm-and-txt-peephole.ll test passes even when the
peephole optimizer is disabled. That's not good at all.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117905 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 05:59:43 +00:00
Bill Wendling
59707e8055 Disable because peephole is disabled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117903 91177308-0d34-0410-b5e6-96231b3b80d8
2010-11-01 05:48:44 +00:00
Evan Cheng
189c1ec4c1 Teach machine cse to eliminate instructions with multiple physreg uses and defs. rdar://8610857.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117745 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-29 23:36:03 +00:00
Bob Wilson
46455096be Remove DAG combiner patch to fold vector splats. Instcombiner does it now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117720 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-29 22:03:02 +00:00
Bob Wilson
0f1db1a6c6 Teach the DAG combiner to fold a splat of a splat. Radar 8597790.
Also do some minor refactoring to reduce indentation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117558 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28 17:06:14 +00:00
Evan Cheng
7e2fe9150f Re-commit 117518 and 117519 now that ARM MC test failures are out of the way.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117531 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28 06:47:08 +00:00
Evan Cheng
9e08ee5d16 Revert 117518 and 117519 for now. They changed scheduling and cause MC tests to fail. Ugh.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117520 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28 02:00:25 +00:00
Evan Cheng
0104d9de04 - Assign load / store with shifter op address modes the right itinerary classes.
- For now, loads of [r, r] addressing mode is the same as the
  [r, r lsl/lsr/asr #] variants. ARMBaseInstrInfo::getOperandLatency() should
  identify the former case and reduce the output latency by 1.
- Also identify [r, r << 2] case. This special form of shifter addressing mode
  is "free".


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117519 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-28 01:49:06 +00:00
Evan Cheng
f40deed62f Shifter ops are not always free. Do not fold them (especially to form
complex load / store addressing mode) when they have higher cost and
when they have more than one use.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117509 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-27 23:41:30 +00:00
Bob Wilson
f20700ca77 SelectionDAG shuffle nodes do not allow operands with different numbers of
elements than the result vector type.  So, when an instruction like:

%8 = shufflevector <2 x float> %4, <2 x float> %7, <4 x i32> <i32 1, i32 0, i32 3, i32 2>

is translated to a DAG, each operand is changed to a concat_vectors node that appends 2 undef elements.  That is:

shuffle [a,b], [c,d] is changed to:
shuffle [a,b,u,u], [c,d,u,u]

That's probably the right thing for x86 but for NEON, we'd much rather have:

shuffle [a,b,c,d], undef

Teach the DAG combiner how to do that transformation for ARM.  Radar 8597007.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117482 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-27 20:38:28 +00:00
Jim Grosbach
8ac98cb665 FileCheck'ize
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117401 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-26 21:26:47 +00:00
Bob Wilson
7c730e7790 When the "true" and "false" blocks of a diamond if-conversion are the same,
do not double-count the duplicate instructions by counting once from the
beginning and again from the end.  Keep track of where the duplicates from
the beginning ended and don't go past that point when counting duplicates
at the end.  Radar 8589805.

This change causes one of the MC/ARM/simple-fp-encoding tests to produce
different (better!) code without the vmovne instruction being tested.
I changed the test to produce vmovne and vmoveq instructions but moving
between register files in the opposite direction.  That's not quite the same
but predicated versions of those instructions weren't being tested before,
so at least the test coverage is not any worse, just different.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117333 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-26 00:02:24 +00:00
Rafael Espindola
cecbc3d282 Add support for emitting ARM file attributes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117275 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-25 17:50:35 +00:00
Jim Grosbach
b2331553f8 tidy up
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117185 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-22 23:46:04 +00:00
Jim Grosbach
871bf8752f Remove duplicate test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117158 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-22 22:04:28 +00:00
Jim Grosbach
d0ad5d52b8 tidy up.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117157 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-22 22:01:56 +00:00
Jim Grosbach
487119a08c FileCheck-ize a few tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@117156 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-22 21:55:03 +00:00
Andrew Trick
5b7a825ec5 putback r116983 and fix simple-fp-encoding.ll tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116992 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-21 03:40:16 +00:00
Owen Anderson
d9707e3d85 Revert r116983, which is breaking all the buildbots.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116987 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-21 03:11:16 +00:00
Evan Cheng
d6865de2d2 Add missing scheduling itineraries for transfers between core registers and VFP registers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116983 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-21 01:12:00 +00:00
Evan Cheng
2312842de0 Re-enable register pressure aware machine licm with fixes. Hoist() may have
erased the instruction during LICM so UpdateRegPressureAfter() should not
reference it afterwards.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116845 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-19 18:58:51 +00:00
Daniel Dunbar
9869413802 Revert r116781 "- Add a hook for target to determine whether an instruction def
is", which breaks some nightly tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116816 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-19 17:14:24 +00:00
Evan Cheng
11e8b74a7a - Add a hook for target to determine whether an instruction def is
"long latency" enough to hoist even if it may increase spilling. Reloading
  a value from spill slot is often cheaper than performing an expensive
  computation in the loop. For X86, that means machine LICM will hoist
  SQRT, DIV, etc. ARM will be somewhat aggressive with VFP and NEON
  instructions.
- Enable register pressure aware machine LICM by default.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116781 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-19 00:55:07 +00:00
Bob Wilson
3454ed9545 Support alignment for NEON vld-lane and vst-lane instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116776 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-19 00:16:32 +00:00
Eric Christopher
6e5367d6a3 Revert r116220 - thus turning arm fast isel back on by default.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116762 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-18 22:53:53 +00:00
Bob Wilson
cfbece50f6 ARM instructions that are both predicated and set the condition codes
have been printed with the "S" modifier after the predicate.  With ARM's
unified syntax, they are supposed to go in the other order.  We fixed this
for Thumb when we switched to unified syntax but missed changing it for
ARM.  Apparently we don't generate these instructions often because no one
noticed until now.  Thanks to Bill Wendling for the testcase!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116563 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-15 03:23:44 +00:00
Jim Grosbach
7032f922b1 Refactor the MOVsr[al]_flag and RRX pseudo-instructions to really be pseudos
and let the ARMExpandPseudoInsts pass fix them up into the real (MOVs)
instruction form.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116534 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-14 22:57:13 +00:00
Jim Grosbach
792e9796b3 Tweak the ARM backend to use the RRX mnemonic instead of the 'mov a, b, rrx'
pseudonym.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116512 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-14 20:43:44 +00:00
Eric Christopher
8ff9a9da0a Found a bug turning this on by default. Disable again for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116220 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-11 20:26:21 +00:00
Eric Christopher
a3d210733a Remove now non-existent option.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116219 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-11 20:21:21 +00:00
Evan Cheng
d2ca813549 Correct some load / store instruction itinerary mistakes:
1. Cortex-A8 load / store multiplies can only issue on ALU0.
2. Eliminate A8_Issue, A8_LSPipe will correctly limit the load / store issues.
3. Correctly model all vld1 and vld2 variants.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116134 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-09 01:03:04 +00:00
Bill Wendling
5ed5c38423 Simplify test and move into a generic "crash" ll file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116130 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-09 00:29:04 +00:00
Bill Wendling
0aa38b9381 Check to make sure that the iterator isn't at the beginning of the basic block
before decrementing. <rdar://problem/8529919>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116126 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-09 00:03:48 +00:00