Commit Graph

14 Commits

Author SHA1 Message Date
Andrew Trick 8485257d6d 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.

llvm-svn: 187139
2013-07-25 18:35:14 +00:00
Stephen Lin d24ab20e9b 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

llvm-svn: 186280
2013-07-14 06:24:09 +00:00
Andrew Trick 121124acf8 Revert "Temporarily enable MI-Sched on X86."
This reverts commit 98a9b72e8c56dc13a2617de84503a3d78352789c.

llvm-svn: 184823
2013-06-25 02:48:58 +00:00
Andrew Trick 5a1e0af838 Temporarily enable MI-Sched on X86.
Sorry for the unit test churn. I'll try to make the change permanently
next time.

llvm-svn: 184705
2013-06-24 09:13:20 +00:00
Benjamin Kramer ef6494f24d PR13578: Teach MachineCSE that instructions that use a constant register can be CSE'd safely.
This is common e.g. when doing rip-relative addressing on x86_64.

llvm-svn: 161728
2012-08-11 19:05:13 +00:00
Preston Gurd f2ea70ae4a Fix remaining lit tests which were failing when run on an Atom
processor.

Patches by Tyler Nowicki, Andy Zhang, and Preston Gurd!

llvm-svn: 160520
2012-07-19 18:53:21 +00:00
Jakob Stoklund Olesen 37492eac8c Don't break the IV update in TLI::SimplifySetCC().
LSR always tries to make the ICmp in the loop latch use the incremented
induction variable. This allows the induction variable to be kept in a
single register.

When the induction variable limit is equal to the stride,
SimplifySetCC() would break LSR's hard work by transforming:

   (icmp (add iv, stride), stride) --> (cmp iv, 0)

This forced us to use lea for the IC update, preventing the simpler
incl+cmp.

<rdar://problem/7643606>
<rdar://problem/11184260>

llvm-svn: 154119
2012-04-05 20:30:20 +00:00
NAKAMURA Takumi 270354100a test/CodeGen/X86/lsr-loop-exit-cond.ll: Try to appease linux and freebsd bots to specify explicit -mtriple=x86_64-darwin.
I guess it expects -relocation-model=pic.

llvm-svn: 144290
2011-11-10 14:18:59 +00:00
Evan Cheng d33b2d6b7a 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.

llvm-svn: 144267
2011-11-10 07:43:16 +00:00
Andrew Trick bfbd972b1f In the pre-RA scheduler, maintain cmp+br proximity.
This is done by pushing physical register definitions close to their
use, which happens to handle flag definitions if they're not glued to
the branch. This seems to be generally a good thing though, so I
didn't need to add a target hook yet.

The primary motivation is to generate code closer to what people
expect and rule out missed opportunity from enabling macro-op
fusion. As a side benefit, we get several 2-5% gains on x86
benchmarks. There is one regression:
SingleSource/Benchmarks/Shootout/lists slows down be -10%. But this is
an independent scheduler bug that will be tracked separately.
See rdar://problem/9283108.

Incidentally, pre-RA scheduling is only half the solution. Fixing the
later passes is tracked by:
<rdar://problem/8932804> [pre-RA-sched] on x86, attempt to schedule CMP/TEST adjacent with condition jump

Fixes:
<rdar://problem/9262453> Scheduler unnecessary break of cmp/jump fusion

llvm-svn: 129508
2011-04-14 05:15:06 +00:00
Eric Christopher eb19e9e9fc Turn on list-ilp scheduling by default on x86 and x86-64, fix up
testcases accordingly. Some are currently xfailed and will be filed
as bugs to be fixed or understood.

Performance results:

roughly neutral on SPEC
some micro benchmarks in the llvm suite are up between 100 and 150%, only
a pair of regressions that are due to be investigated

john-the-ripper saw:
10% improvement in traditional DES
8% improvement in BSDI DES
59% improvement in FreeBSD MD5
67% improvement in OpenBSD Blowfish
14% improvement in LM DES

Small compile time impact.

llvm-svn: 127208
2011-03-08 02:42:25 +00:00
Dan Gohman 8fdda8a655 This test doesn't need the ssp attribute.
llvm-svn: 105440
2010-06-04 00:14:48 +00:00
Daniel Dunbar b9ea94c990 Eliminate uses of %prcontext.
- I'd appreciate it if someone else eyeballs my changes to make sure I captured
   the intent of the test.

llvm-svn: 81083
2009-09-05 11:35:16 +00:00
Evan Cheng 78a4eb844b Teach LSR to optimize more loop exit compares, i.e. change them to use postinc iv value. Previously LSR would only optimize those which are in the loop latch block. However, if LSR can prove it is safe (and profitable), it's now possible to change those not in the latch blocks to use postinc values.
Also, if the compare is the only use, LSR would place the iv increment instruction before the compare instead in the latch.

llvm-svn: 71485
2009-05-11 22:33:01 +00:00