Commit Graph

79576 Commits

Author SHA1 Message Date
Craig Topper
a937633893 Fix a crash in AVX2 when trying to broadcast a double into a 128-bit vector. There is no vbroadcastsd xmm, but we do need to support 64-bit integers broadcasted into xmm. Also factor the AVX check into the isVectorBroadcast function. This makes more sense since the AVX2 check was already inside.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147844 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 08:23:59 +00:00
Craig Topper
1accb7ed98 Remove hasXMM/hasXMMInt functions. Move callers to hasSSE1/hasSSE2. This is the final piece to remove the AVX hack that disabled SSE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147843 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 06:54:16 +00:00
Craig Topper
d0a3117768 Remove hasSSE*orAVX functions and change all callers to use just hasSSE*. AVX is now an SSE level and no longer disables SSE checks.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147842 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 06:37:29 +00:00
Craig Topper
c6d59954d8 Instruction selection priority fixes to remove the XMM/XMMInt/orAVX predicates. Another commit will remove orAVX functions from X86SubTarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147841 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 06:30:56 +00:00
Evan Cheng
97b5beb7fe Allow machine-cse to look across MBB boundary when cse'ing instructions that
define physical registers. It's currently very restrictive, only catching
cases where the CE is in an immediate (and only) predecessor. But it catches
a surprising large number of cases.

rdar://10660865


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147827 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 02:02:58 +00:00
Andrew Trick
64925c55c6 Enable LSR IV Chains with sufficient heuristics.
These heuristics are sufficient for enabling IV chains by
default. Performance analysis has been done for i386, x86_64, and
thumbv7. The optimization is rarely important, but can significantly
speed up certain cases by eliminating spill code within the
loop. Unrolled loops are prime candidates for IV chains. In many
cases, the final code could still be improved with more target
specific optimization following LSR. The goal of this feature is for
LSR to make the best choice of induction variables.

Instruction selection may not completely take advantage of this
feature yet. As a result, there could be cases of slight code size
increase.

Code size can be worse on x86 because it doesn't support postincrement
addressing. In fact, when chains are formed, you may see redundant
address plus stride addition in the addressing mode. GenerateIVChains
tries to compensate for the common cases.

On ARM, code size increase can be mitigated by using postincrement
addressing, but downstream codegen currently misses some opportunities.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147826 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 01:45:08 +00:00
Jakob Stoklund Olesen
dae412bd32 Accurately model hardware alignment rounding.
On Thumb, the displacement computation hardware uses the address of the
current instruction rouned down to a multiple of 4.  Include this
rounding in the UserOffset we compute for each instruction.

When inline asm is present, the instruction alignment may not be known.
Constrain the maximum displacement instead in that case.

This makes it possible for CreateNewWater() and OffsetIsInRange() to
agree about the valid displacements.  When they disagree, infinite
looping happens.

As always, test cases for this stuff are insane.

<rdar://problem/10660175>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147825 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 01:34:59 +00:00
Rafael Espindola
9cce24a257 Remove the logging streamer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147820 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-10 00:40:39 +00:00
Jakob Stoklund Olesen
169db15717 Catch runaway ARMConstantIslandPass even in -Asserts builds.
The pass is prone to looping, and it is better to crash than loop
forever, even in a -Asserts build.

<rdar://problem/10660175>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147806 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 22:16:24 +00:00
Devang Patel
c16d96f6f4 Fix asm string wrt variants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147805 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 21:32:02 +00:00
Devang Patel
d06b01c038 Use descriptive variable name and remove incorrect operand number check.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147802 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 21:30:46 +00:00
Andrew Trick
22d20c218a Adding IV chain generation to LSR.
After collecting chains, check if any should be materialized. If so,
hide the chained IV users from the LSR solver. LSR will only solve for
the head of the chain. GenerateIVChains will then materialize the
chained IV users by computing the IV relative to its previous value in
the chain.

In theory, chained IV users could be exposed to LSR's solver. This
would be considerably complicated to implement and I'm not aware of a
case where we need it. In practice it's more important to
intelligently prune the search space of nontrivial loops before
running the solver, otherwise the solver is often forced to prune the
most optimal solutions. Hiding the chained users does this well, so
that LSR is more likely to find the best IV for the chain as a whole.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147801 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 21:18:52 +00:00
Andrew Trick
6c7d0ae8dc Adding collection of IV chains to LSR.
This collects a set of IV uses within the loop whose values can be
computed relative to each other in a sequence. Following checkins will
make use of this information.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147797 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 19:50:34 +00:00
Devang Patel
0dbcadaa2f Split AsmParser into two components - AsmParser and AsmParserVariant
AsmParser holds info specific to target parser.
AsmParserVariant holds info specific to asm variants supported by the target.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147787 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 19:13:28 +00:00
Andrew Trick
8bf295b1bf "Minor LSR debugging stuff"
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147785 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 18:58:16 +00:00
Devang Patel
bc44b9f8d2 Update language check. Do not ignore DW_LANG_Python.
Patch by  Joe Groff!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147781 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 17:49:47 +00:00
Benjamin Kramer
7e4ac32a4e Move assert to the right place.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147779 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 17:36:29 +00:00
Benjamin Kramer
47a8607721 InstCombine: Teach foldLogOpOfMaskedICmpsHelper that sign bit tests are bit tests.
This subsumes several other transforms while enabling us to catch more cases.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147777 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 17:23:27 +00:00
Chandler Carruth
4e3a40293f Don't rely on the fact that shift values are never very large, and thus
this substraction will result in small negative numbers at worst which
become very large positive numbers on assignment and are thus caught by
the <=4 check on the next line. The >0 check clearly intended to catch
these as negative numbers.

Spotted by inspection, and impossible to trigger given the shift widths
that can be used.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147773 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 09:47:25 +00:00
Chandler Carruth
a7cb699251 Cleanup and FileCheck-ize a test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147772 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 09:44:26 +00:00
Craig Topper
16de4632bb Remove AVX hack in X86Subtarget. AVX/AVX2 are now treated as an SSE level. Predicate functions have been altered to maintain previous names and behavior.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147770 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 09:02:13 +00:00
Craig Topper
8ffc964582 Add HasAVX predicate to some of the AVX patterns.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147769 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 08:34:00 +00:00
Craig Topper
47cf1003fa Reorder a bunch of patterns to put the AVX version first thus giving it priority over the SSE version. Another step towards trying to remove the AVX hack that disables SSE from X86Subtarget.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147768 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 08:10:38 +00:00
Craig Topper
5feb5dae93 Clean up patterns for MOVNT*. Not sure why there were floating point types on MOVNTPS and MOVNTDQ. And v4i64 was completely missing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147767 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 06:52:46 +00:00
Craig Topper
8974cd85cc Mark MOVNTI as being supported in SSE2 OR AVX mode. This instruction has no AVX equivalent so we should use the SSE version.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147766 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 06:38:55 +00:00
Craig Topper
dfa5f573e7 Move SSE2 logical operations PAND/POR/PXOR/PANDN above SSE1 logical operations ANDPS/ORPS/XORPS/ANDNPS. This fixes a pattern ordering issue that meant that the SSE2 instructions could never be directly selected since the SSE1 patterns would always match first. This is largely moot with the ExeDepsFix pass, but I'm trying to audit for all such ordering issues.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147765 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 05:07:01 +00:00
Craig Topper
6202e45d53 Change some places that were checking for AVX OR SSE1/2 to use hasXMM/hasXMMInt instead. Also fix one place that checked SSE3, but accidentally excluded AVX to use hasSSE3orAVX. This is a step towards removing the AVX hack from the X86Subtarget.h
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147764 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 02:28:15 +00:00
Rafael Espindola
1fe9737eb4 Don't print an unused label before .cfi_endproc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147763 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 00:17:29 +00:00
Craig Topper
39f227e4dd Don't disable MMX support when AVX is enabled. Fix predicates for MMX instructions that were added along with SSE instructions to check for AVX in addition to SSE level.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147762 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-09 00:11:29 +00:00
Craig Topper
a8224ddf80 Enable FISTTP* instructions when AVX is enabled.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147758 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 23:04:21 +00:00
Benjamin Kramer
d4242d8ab1 Tweak my last commit to be less conservative about uses.
We still save an instruction when just the "and" part is replaced.
Also change the code to match comments more closely.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147753 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 21:12:51 +00:00
Evan Cheng
79aa048d21 Don't forget to transfer implicit uses of return instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147752 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 20:41:16 +00:00
Evan Cheng
e811d0dd30 Avoid eraseing copies from a reserved register unless the definition can be
safely proven not to have been clobbered. No small test case possible.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147751 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 19:52:28 +00:00
Benjamin Kramer
dfb806f6ba InstCombine: If we have a bit test and a sign test anded/ored together, merge the sign bit into the bit test.
This is common in bit field code, e.g. checking if the first or the last bit of a bit field is set.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147749 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 18:32:24 +00:00
Victor Umansky
435d0bd09d Reverted commit #147601 upon Evan's request.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147748 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-08 17:20:33 +00:00
Rafael Espindola
f340a29e7a Remove MCELFStreamer.h.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147745 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 23:18:39 +00:00
Rafael Espindola
547be2699c Don't print a label before .cfi_startproc when we don't need to. This makes
the produce assembly when using CFI just a bit more readable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147743 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 22:42:19 +00:00
Benjamin Kramer
86132a721d Make clever use of alignment and padding to shrink GlobalValue.
-8 bytes on x86_64, no change on x86.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147742 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 21:17:16 +00:00
Jakob Stoklund Olesen
8f37a2422e Match SelectionDAG logic for enabling movt.
Darwin doesn't do static, and ELF targets only support static.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147740 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 20:49:15 +00:00
Craig Topper
eb3d460b25 Fix typo in the X86 backend readme. Patch from Jaeden Amero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147739 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 20:35:21 +00:00
Benjamin Kramer
f321e1075e Remove VectorExtras. This unused helper was written for a type of API that is discouraged now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147738 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 19:42:13 +00:00
Craig Topper
272cc580f8 Remove unnecessary check of hasAVX(). It's already included in hasXMM().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147734 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 18:48:43 +00:00
Craig Topper
704e1a0f04 Replace some uses of hasNUsesOfValue(0, X) with !hasAnyUseOfValue(X)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147733 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 18:31:09 +00:00
Benjamin Kramer
f0be7ca7e4 Port the trick to skip the check for empty buckets from StringMap to DenseMap.
This should fix the odd behavior that find() is slower than lookup().

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147731 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 13:12:07 +00:00
Craig Topper
cc274526ef Add some DAG combines for SUBC/SUBE. If nothing uses the carry/borrow out of subc, turn it into a sub. Turn (subc x, x) into 0 with no borrow. Turn (subc x, 0) into x with no borrow. Turn (subc -1, x) into (xor x, -1) with no borrow. Turn sube with no borrow in into subc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147728 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 09:06:39 +00:00
Cameron Zwarich
82f000266a Fix TableGen so that it will emit the correct signature for FastEmit_f:
/// FastEmit_f - This method is called by target-independent code
  /// to request that an instruction with the given type, opcode, and
  /// floating-point immediate operand be emitted.
  virtual unsigned FastEmit_f(MVT VT,
                              MVT RetVT,
                              unsigned Opcode,
                              const ConstantFP *FPImm);

Currently, it emits an accidentally overloaded version without the const on the
ConstantFP*. This doesn't affect anything in the tree, since nothing causes that
method to be autogenerated, but I have been playing with some ARM TableGen
refactorings that hit this problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147727 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 08:18:37 +00:00
Jakob Stoklund Olesen
34786a3ad7 Optimize reserved register coalescing.
Reserved registers don't have proper live ranges, their LiveInterval
simply has a snippet of liveness for each def.  Virtual registers with a
single value that is a copy of a reserved register (typically %esp) can
be coalesced with the reserved register if the live range doesn't
overlap any reserved register defs.

When coalescing with a reserved register, don't modify the reserved
register live range.  Just leave it as a bunch of dead defs.  This
eliminates quadratic coalescer behavior in i386 functions with many
function calls.

PR11699

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147726 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 07:39:50 +00:00
Jakob Stoklund Olesen
4281e20aab Use the 'regalloc' debug tag for most register allocator tracing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147725 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 07:39:47 +00:00
Andrew Trick
24f670f1ba Enable redundant phi elimination after LSR.
This will be more important as we extend the LSR pass in ways that don't rely on the formula solver. In particular, we need it for constructing IV chains.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147724 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 07:08:17 +00:00
Eli Bendersky
0ff100dec0 Fix dead link
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@147721 91177308-0d34-0410-b5e6-96231b3b80d8
2012-01-07 04:11:27 +00:00