Commit Graph

37 Commits

Author SHA1 Message Date
Tim Northover
9ed30bb230 ARM: use LLVM IR to represent the vshrn operation
vshrn is just the combination of a right shift and a truncate (and the limits
on the immediate value actually mean the signedness of the shift doesn't
matter). Using that representation allows us to get rid of an ARM-specific
intrinsic, share more code with AArch64 and hopefully get better code out of
the mid-end optimisers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@201085 91177308-0d34-0410-b5e6-96231b3b80d8
2014-02-10 14:04:07 +00:00
Matthias Braun
1dfe206062 Revert "Tests: Be less dependent on a specific schedule/regalloc"
This reverts r192454

Apparently FileCheck isn't as smart as I though and does not enforce a
topological order between variable defs+uses.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@192472 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-11 18:09:19 +00:00
Matthias Braun
5b51fd5b55 Tests: Be less dependent on a specific schedule/regalloc
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@192454 91177308-0d34-0410-b5e6-96231b3b80d8
2013-10-11 15:40:12 +00:00
Tim Northover
32c2bfda77 ARM: implement some simple f64 materializations.
Previously we used a const-pool load for virtually all 64-bit floating values.
Actually, we can get quite a few common values (including 0.0, 1.0) via "vmov"
instructions of one stripe or another.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188773 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-20 08:57:11 +00:00
Stephen Lin
8b2b8a1835 Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally.
This update was done with the following bash script:

  find test/CodeGen -name "*.ll" | \
  while read NAME; do
    echo "$NAME"
    if ! grep -q "^; *RUN: *llc.*debug" $NAME; then
      TEMP=`mktemp -t temp`
      cp $NAME $TEMP
      sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \
      while read FUNC; do
        sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP
      done
      sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP
      sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP
      sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP
      sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP
      mv $TEMP $NAME
    fi
  done


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186280 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-14 06:24:09 +00:00
Kristof Beyls
29e05fe7a8 Make ARMAsmPrinter generate the correct alignment specifier syntax in instructions.
The Printer will now print instructions with the correct alignment specifier syntax, like
    vld1.8  {d16}, [r0:64]



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@175884 91177308-0d34-0410-b5e6-96231b3b80d8
2013-02-22 10:01:33 +00:00
Jakob Stoklund Olesen
5cf178f281 Enable the new coalescer algorithm by default.
The new coalescer is better at merging values into unused vector lanes,
improving NEON code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164794 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-27 21:06:02 +00:00
Evan Cheng
2dad6b501b Try to make these tests more portable.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164320 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-20 21:35:21 +00:00
Evan Cheng
d10eab0a95 Use vld1 / vst2 for unaligned v2f64 load / store. e.g. Use vld1.16 for 2-byte
aligned address. Based on patch by David Peixotto.

Also use vld1.64 / vst1.64 with 128-bit alignment to take advantage of alignment
hints. rdar://12090772, rdar://12238782


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164089 91177308-0d34-0410-b5e6-96231b3b80d8
2012-09-18 01:42:45 +00:00
Nadav Rotem
4ac9081c71 This commit contains a few changes that had to go in together.
1. Simplify xor/and/or (bitcast(A), bitcast(B)) -> bitcast(op (A,B))
   (and also scalar_to_vector).

2. Xor/and/or are indifferent to the swizzle operation (shuffle of one src).
   Simplify xor/and/or (shuff(A), shuff(B)) -> shuff(op (A, B))

3. Optimize swizzles of shuffles:  shuff(shuff(x, y), undef) -> shuff(x, y).

4. Fix an X86ISelLowering optimization which was very bitcast-sensitive.

Code which was previously compiled to this:

movd    (%rsi), %xmm0
movdqa  .LCPI0_0(%rip), %xmm2
pshufb  %xmm2, %xmm0
movd    (%rdi), %xmm1
pshufb  %xmm2, %xmm1
pxor    %xmm0, %xmm1
pshufb  .LCPI0_1(%rip), %xmm1
movd    %xmm1, (%rdi)
ret

Now compiles to this:

movl    (%rsi), %eax
xorl    %eax, (%rdi)
ret




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@153848 91177308-0d34-0410-b5e6-96231b3b80d8
2012-04-01 19:31:22 +00:00
Jim Grosbach
ffc658b056 ARM VLDR/VSTR instructions don't need a size suffix.
Canonicallize on the non-suffixed form, but continue to accept assembly that
has any correctly sized type suffix.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144583 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14 23:03:21 +00:00
Benjamin Kramer
70be28a5ad Simplify some uses of utohexstr.
As a side effect hex is printed lowercase instead of uppercase now.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144013 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-07 21:00:59 +00:00
Owen Anderson
43967a97cf Remove VMOVDneon and VMOVQ, which are just aliases for VORR. This continues to simplify the path towards an auto-generated disassembler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135290 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-15 18:46:47 +00:00
Jakob Stoklund Olesen
ca6fd009ad Fix ARM tests to be register allocator independent.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128680 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-31 22:14:03 +00:00
Evan Cheng
48575f6ea7 Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
   of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
   additional pipeline stall. So it's frequently better to single codegen
   vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
   stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
   vmla + vmla is very bad. But this isn't ideal either:
     vmul
     vadd
     vmla
   Instead, we want to expand the second vmla:
     vmla
     vmul
     vadd
   Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
   faster.

Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.

A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
   compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
   fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
   vmla / vmls will trigger one of the special hazards.

Work in progress, only A+B are enabled.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@120960 91177308-0d34-0410-b5e6-96231b3b80d8
2010-12-05 22:04:16 +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
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
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
Bob Wilson
7d24705f65 Change register allocation order for ARM VFP and NEON registers to put the
callee-saved registers at the end of the lists.  Also prefer to avoid using
the low registers that are in register subclasses required by certain
instructions, so that those registers will more likely be available when needed.
This change makes a huge improvement in spilling in some cases.  Thanks to
Jakob for helping me realize the problem.

Most of this patch is fixing the testsuite.  There are quite a few places
where we're checking for specific registers.  I changed those to wildcards
in places where that doesn't weaken the tests.  The spill-q.ll and
thumb2-spill-q.ll tests stopped spilling with this change, so I added a bunch
of live values to force spills on those tests.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@116055 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-08 06:15:13 +00:00
Bob Wilson
ffde080ae6 Convert VLD1 and VLD2 instructions to use pseudo-instructions until
after regalloc.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112825 91177308-0d34-0410-b5e6-96231b3b80d8
2010-09-02 16:00:54 +00:00
Bob Wilson
7a9ef44b3b Add alignment arguments to all the NEON load/store intrinsics.
Update all the tests using those intrinsics and add support for
auto-upgrading bitcode files with the old versions of the intrinsics.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@112271 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-27 17:13:24 +00:00
Bob Wilson
d58b51e67c Replace some NEON vmovl intrinsic that I missed earlier.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@111696 91177308-0d34-0410-b5e6-96231b3b80d8
2010-08-20 23:22:43 +00:00
Bob Wilson
cba270d042 Use a target-specific VMOVIMM DAG node instead of BUILD_VECTOR to represent
NEON VMOV-immediate instructions.  This simplifies some things.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@108275 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-13 21:16:48 +00:00
Bob Wilson
a0148c360e Print "dregpair" NEON operands with a space between them, for readability and
consistency with other instructions that have lists of register operands.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107944 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-09 00:47:20 +00:00
Bob Wilson
f1d93ca920 Reenable DAG combining for vector shuffles. It looks like it was temporarily
disabled and then never turned back on again.  Adjust some tests, one because
this change avoids an unnecessary instruction, and the other to make it
continue testing what it was intended to test.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@107941 91177308-0d34-0410-b5e6-96231b3b80d8
2010-07-09 00:38:12 +00:00
Dan Gohman
ee1c870778 Eliminate the other half of the BRCOND optimization, and update
as many tests as possible.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@106749 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-24 15:24:03 +00:00
Rafael Espindola
1e81966626 Remove arm_apcscc from the test files. It is the default and doing this
matches what llvm-gcc and clang now produce.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@106221 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-17 15:18:27 +00:00
Evan Cheng
046fa3f90a Fix some latency computation bugs: if the use is not a machine opcode do not just return zero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@105061 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-28 23:26:21 +00:00
Evan Cheng
f7d87ee158 Change ARM scheduling default to list-hybrid if the target supports floating point instructions (and is not using soft float).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104307 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-21 00:43:17 +00:00
Jakob Stoklund Olesen
3437352887 TwoAddressInstructionPass doesn't really know how to merge live intervals when
lowering REG_SEQUENCE instructions.

Insert copies for REG_SEQUENCE sources not killed to avoid breaking later passes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104146 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-19 20:08:00 +00:00
Evan Cheng
27e4840e03 Fix PR7162: Use source register classes and sub-indices to determine the correct register class of the definitions of REG_SEQUENCE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104050 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-18 20:03:28 +00:00
Evan Cheng
c6dcce3ba5 Fix PR7175. Insert copies of a REG_SEQUENCE source if it is used by other REG_SEQUENCE instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103994 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-17 23:24:12 +00:00
Evan Cheng
44bfdd3d78 Fix PR7156. If the sources of a REG_SEQUENCE are all IMPLICIT_DEF's. Replace it with an IMPLICIT_DEF rather than deleting it or else it would be left without a def.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103984 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-17 22:09:49 +00:00
Evan Cheng
53c779bb3a Careful with reg_sequence coalescing to not to overwrite sub-register indices.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103971 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-17 20:57:12 +00:00
Evan Cheng
6206124250 Turn on -neon-reg-sequence by default.
Using NEON load / store multiple instructions will no longer create gobs of vmov of D registers!


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103960 91177308-0d34-0410-b5e6-96231b3b80d8
2010-05-17 19:51:20 +00:00