Commit Graph

33 Commits

Author SHA1 Message Date
Andrew Trick
6ea2b9608a Allocate local registers in order for optimal coloring.
Also avoid locals evicting locals just because they want a cheaper register.

Problem: MI Sched knows exactly how many registers we have and assumes
they can be colored. In cases where we have large blocks, usually from
unrolled loops, greedy coloring fails. This is a source of
"regressions" from the MI Scheduler on x86. I noticed this issue on
x86 where we have long chains of two-address defs in the same live
range. It's easy to see this in matrix multiplication benchmarks like
IRSmk and even the unit test misched-matmul.ll.

A fundamental difference between the LLVM register allocator and
conventional graph coloring is that in our model a live range can't
discover its neighbors, it can only verify its neighbors. That's why
we initially went for greedy coloring and added eviction to deal with
the hard cases. However, for singly defined and two-address live
ranges, we can optimally color without visiting neighbors simply by
processing the live ranges in instruction order.

Other beneficial side effects:

It is much easier to understand and debug regalloc for large blocks
when the live ranges are allocated in order. Yes, global allocation is
still very confusing, but it's nice to be able to comprehend what
happened locally.

Heuristics could be added to bias register assignment based on
instruction locality (think late register pairing, banks...).

Intuituvely this will make some test cases that are on the threshold
of register pressure more stable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@187139 91177308-0d34-0410-b5e6-96231b3b80d8
2013-07-25 18:35:14 +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
Craig Topper
b82b5abf78 Simplify handling of v16i8 shuffles and fix a missed optimization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157043 91177308-0d34-0410-b5e6-96231b3b80d8
2012-05-18 06:42:06 +00:00
Evan Cheng
2a4410df44 Teach two-address pass to re-schedule two-address instructions (or the kill
instructions of the two-address operands) in order to avoid inserting copies.
This fixes the few regressions introduced when the two-address hack was
disabled (without regressing the improvements).
rdar://10422688


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144559 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-14 19:48:55 +00:00
Evan Cheng
623a7e146b Use a bigger hammer to fix PR11314 by disabling the "forcing two-address
instruction lower optimization" in the pre-RA scheduler.

The optimization, rather the hack, was done before MI use-list was available.
Now we should be able to implement it in a better way, perhaps in the
two-address pass until a MI scheduler is available.

Now that the scheduler has to backtrack to handle call sequences. Adding
artificial scheduling constraints is just not safe. Furthermore, the hack
is not taking all the other scheduling decisions into consideration so it's just
as likely to pessimize code. So I view disabling this optimization goodness
regardless of PR11314.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@144267 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-10 07:43:16 +00:00
Dan Gohman
65fd6564b8 Reapply r143206, with fixes. Disallow physical register lifetimes
across calls, and only check for nested dependences on the special
call-sequence-resource register.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143660 91177308-0d34-0410-b5e6-96231b3b80d8
2011-11-03 21:49:52 +00:00
Dan Gohman
6f3ddef7c5 Revert r143206, as there are still some failing tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143262 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-29 00:41:52 +00:00
Dan Gohman
bf923b815d Reapply r143177 and r143179 (reverting r143188), with scheduler
fixes: Use a separate register, instead of SP, as the
calling-convention resource, to avoid spurious conflicts with
actual uses of SP. Also, fix unscheduling of calling sequences,
which can be triggered by pseudo-two-address dependencies.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143206 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-28 17:55:38 +00:00
Duncan Sands
62c1d00dfd Speculatively disable Dan's commits 143177 and 143179 to see if
it fixes the dragonegg self-host (it looks like gcc is miscompiled).
Original commit messages:
Eliminate LegalizeOps' LegalizedNodes map and have it just call RAUW
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.

Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.

Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.

Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.

This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.

Delete #if 0 code accidentally left in.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143188 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-28 09:55:57 +00:00
Dan Gohman
2ba60e5930 Eliminate LegalizeOps' LegalizedNodes map and have it just call RAUW
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.

Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.

Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.

Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.

This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@143177 91177308-0d34-0410-b5e6-96231b3b80d8
2011-10-28 01:29:32 +00:00
Evan Cheng
f0b3c12919 Fix broken x86_64 tests which specify non-64-bit cpu's.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134756 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-08 22:29:33 +00:00
Jakob Stoklund Olesen
51458ed09e Be more aggressive about following hints.
RAGreedy::tryAssign will now evict interference from the preferred
register even when another register is free.

To support this, add the EvictionCost struct that counts how many hints
are broken by an eviction. We don't want to break one hint just to
satisfy another.

Rename canEvict to shouldEvict, and add the first bit of eviction policy
that doesn't depend on spill weights: Always make room in the preferred
register as long as the evictees can be split and aren't already
assigned to their preferred register.

Also make the CSR avoidance more accurate. When looking for a cheaper
register it is OK to use a new volatile register. Only CSR aliases that
have never been used before should be avoided.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134735 91177308-0d34-0410-b5e6-96231b3b80d8
2011-07-08 20:46:18 +00:00
Jakob Stoklund Olesen
61396aebee Fix a batch of x86 tests to be coalescer independent.
Most of these tests require a single mov instruction that can come either before
or after a 2-addr instruction. -join-physregs changes the behavior, but the
results are equivalent.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@130891 91177308-0d34-0410-b5e6-96231b3b80d8
2011-05-04 23:54:51 +00:00
Jakob Stoklund Olesen
b0e47cdf3d Fix register-dependent X86 tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128867 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-05 00:32:44 +00:00
Evan Cheng
835580fc3a Canonicalize X86ISD::MOVDDUP nodes to v2f64 to make sure all cases match. Also eliminate unneeded isel patterns. rdar://8520311
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@115977 91177308-0d34-0410-b5e6-96231b3b80d8
2010-10-07 20:50:20 +00:00
Dan Gohman
fe5e4dabbf Teach EmitLiveInCopies to omit copies for unused virtual registers,
and to clean up unused incoming physregs from the live-in list.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@106805 91177308-0d34-0410-b5e6-96231b3b80d8
2010-06-24 22:23:02 +00:00
Jakob Stoklund Olesen
bfcd61b907 Enable -sse-domain-fix by default. Now with tests!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99954 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-30 22:47:00 +00:00
Evan Cheng
bdc652bab8 Turning off post-ra scheduling for x86. It isn't a consistent win.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98810 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-18 06:55:42 +00:00
Chris Lattner
d1b7382983 Fix some issues in WalkChainUsers dealing with
CopyToReg/CopyFromReg/INLINEASM.  These are annoying because
they have the same opcode before an after isel.  Fix this by
setting their NodeID to -1 to indicate that they are selected,
just like what automatically happens when selecting things that
end up being machine nodes.

With that done, give IsLegalToFold a new flag that causes it to
ignore chains.  This lets the HandleMergeInputChains routine be
the one place that validates chains after a match is successful,
enabling the new hotness in chain processing.  This smarter
chain processing eliminates the need for "PreprocessRMW" in the
X86 and MSP430 backends and enables MSP to start matching it's
multiple mem operand instructions more aggressively.

I currently #if out the dead code in the X86 backend and MSP 
backend, I'll remove it for real in a follow-on patch.

The testcase changes are:
  test/CodeGen/X86/sse3.ll: we generate better code
  test/CodeGen/X86/store_op_load_fold2.ll: PreprocessRMW was 
      miscompiling this before, we now generate correct code
      Convert it to filecheck while I'm at it.
  test/CodeGen/MSP430/Inst16mm.ll: Add a testcase for mem/mem
      folding to make anton happy. :)




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97596 91177308-0d34-0410-b5e6-96231b3b80d8
2010-03-02 22:20:06 +00:00
Evan Cheng
00a99a3584 Run codegen dce pass for all targets at all optimization levels. Previously it's
only run for x86 with fastisel. I've found it being very effective in
eliminating some obvious dead code as result of formal parameter lowering
especially when tail call optimization eliminated the need for some of the loads
from fixed frame objects. It also shrinks a number of the tests. A couple of
tests no longer make sense and are now eliminated.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@95493 91177308-0d34-0410-b5e6-96231b3b80d8
2010-02-06 09:07:11 +00:00
Dan Gohman
d84ea47b89 Don't enable the post-RA scheduler on x86 except at -O3. In its
current form, it is too expensive in compile time.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@90781 91177308-0d34-0410-b5e6-96231b3b80d8
2009-12-07 19:04:31 +00:00
Eric Christopher
761411c21b Fix a couple of shuffle patterns to use movhlps instead
of movhps as the constraint.  Changes optimizations so
update testcases as appropriate as well.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@86360 91177308-0d34-0410-b5e6-96231b3b80d8
2009-11-07 08:45:53 +00:00
Evan Cheng
b6c215b63f Update tests for 84931.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@84932 91177308-0d34-0410-b5e6-96231b3b80d8
2009-10-23 05:58:34 +00:00
David Goodwin
4c3715c2e5 Allow the target to select the level of anti-dependence breaking that should be performed by the post-RA scheduler. The default is none.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@84911 91177308-0d34-0410-b5e6-96231b3b80d8
2009-10-22 23:19:17 +00:00
Dan Gohman
c1ae8c9b8f Revert r84658 and r84691. They were causing llvm-gcc bootstrap to fail.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@84727 91177308-0d34-0410-b5e6-96231b3b80d8
2009-10-21 01:44:44 +00:00
David Goodwin
480c529e02 Checkpoint more aggressive anti-dependency breaking for post-ra scheduler.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@84658 91177308-0d34-0410-b5e6-96231b3b80d8
2009-10-20 19:54:44 +00:00
Evan Cheng
d36076e4a3 Turn on post-alloc scheduling for x86.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@84431 91177308-0d34-0410-b5e6-96231b3b80d8
2009-10-18 19:57:27 +00:00
Dan Gohman
36a0947820 Eliminate more uses of llvm-as and llvm-dis.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@81290 91177308-0d34-0410-b5e6-96231b3b80d8
2009-09-08 23:54:48 +00:00
Chris Lattner
d842962e27 change selectiondag to add the sign extended versions of immediate operands
to instructions instead of zero extended ones.  This makes the asmprinter
print signed values more consistently.  This apparently only really affects
the X86 backend.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@81265 91177308-0d34-0410-b5e6-96231b3b80d8
2009-09-08 23:05:44 +00:00
Chris Lattner
d04fee1e2b specify a target triple so global variable manglings are consistent etc.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@79118 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-15 17:35:05 +00:00
Chris Lattner
8e1fad4bb7 merge a bunch more sse3 tests into sse3.ll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@79115 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-15 17:21:44 +00:00
Chris Lattner
075ee999d3 convert test to filecheck format.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@79114 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-15 17:05:03 +00:00
Chris Lattner
bd0463098b rename test
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@79113 91177308-0d34-0410-b5e6-96231b3b80d8
2009-08-15 17:01:44 +00:00