Commit Graph

7000 Commits

Author SHA1 Message Date
Michael J. Spencer
9a0bac41b3 Fix whitespace and tabs.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129517 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 14:33:36 +00:00
Bill Wendling
d336de318e As Dan pointed out, movzbl, movsbl, and friends are nicer than their alias
(movzx/movsx) because they give more information. Revert that part of the patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129498 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 01:46:37 +00:00
Bill Wendling
c6df9883da Have the X86 back-end emit the alias instead of what's being aliased. In most
cases, it's much nicer and more informative reading the alias.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129497 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-14 01:11:51 +00:00
Bill Wendling
eef965f04b Add an option to not print the alias of an instruction. It defaults to "print
the alias".


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129485 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 23:36:21 +00:00
Bill Wendling
f93f7b2446 Reapply r129401 with patch for clang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129419 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-13 00:36:11 +00:00
Bill Wendling
f9b2dc66c8 Revert r129401 for now. Clang is using the old way of doing things.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129403 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 22:59:27 +00:00
Bill Wendling
d5f323d70b Remove the unaligned load intrinsics in favor of using native unaligned loads.
Now that we have a first-class way to represent unaligned loads, the unaligned
load intrinsics are superfluous.

First part of <rdar://problem/8460511>.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129401 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-12 22:46:31 +00:00
Jay Foad
562b84b3ae Don't include Operator.h from InstrTypes.h.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129271 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-11 09:35:34 +00:00
Chris Lattner
15f8951799 fix rdar://8735979 - "int 3" doesn't match to "int3". Unfortunately,
InstAlias doesn't allow matching immediate operands, so we have to write
C++ code to do this.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129223 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-09 19:41:05 +00:00
Bill Wendling
44dcfd3625 Replace the old algorithm that emitted the "print the alias for an instruction"
with the newer, cleaner model. It uses the IAPrinter class to hold the
information that is needed to match an instruction with its alias. This also
takes into account the available features of the platform.

There is one bit of ugliness. The way the logic determines if a pattern is
unique is O(N**2), which is gross. But in reality, the number of items it's
checking against isn't large. So while it's N**2, it shouldn't be a massive time
sink.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129110 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-07 21:20:06 +00:00
Rafael Espindola
3d195faaad Add another case we are not optimizing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129012 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-06 17:35:32 +00:00
Rafael Espindola
8617897aaa The original issue has been fixed by not doing unnecessary sign extensions.
Change the test to force a sign extension and expose the problem again.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129011 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-06 17:19:35 +00:00
Joerg Sonnenberger
c15a91dfc8 Make OpcodeMask an unsigned long long literal to deal with overflow.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128847 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-04 21:38:17 +00:00
Joerg Sonnenberger
4a8ac8de1d Add support for the VIA PadLock instructions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128826 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-04 16:58:13 +00:00
Joerg Sonnenberger
229e4523f2 Expand Op0Mask by one bit in preparation for the PadLock prefixes.
Define most shift masks incrementally to reduce the redundant
hard-coding. Introduce new shift for the VEX flags to replace the
magic constant 32 in various places.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128822 91177308-0d34-0410-b5e6-96231b3b80d8
2011-04-04 15:58:30 +00:00
Evan Cheng
ee2e0e347e Don't try to create zero-sized stack objects.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128586 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-30 23:44:13 +00:00
Benjamin Kramer
050db52276 Make helper static.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128338 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-26 12:38:19 +00:00
NAKAMURA Takumi
a2e0762fae Target/X86: [PR8777][PR8778] Tweak alloca/chkstk for Windows targets.
FIXME: Some cleanups would be needed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128206 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-24 07:07:00 +00:00
Andrew Trick
f6c39412dd Revert r128175.
I'm backing this out for the second time. It was supposed to be fixed by r128164, but the mingw self-host must be defeating the fix.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128181 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-23 23:11:02 +00:00
Andrew Trick
d8fa01fbd7 Reapply Eli's r127852 now that the pre-RA scheduler can spill EFLAGS.
(target-specific branchless method for double-width relational comparisons on x86)


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128175 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-23 22:16:02 +00:00
Dan Gohman
b55d6b6a7e Fix fast-isel address mode folding to avoid folding instructions
outside of the current basic block. This fixes PR9500, rdar://9156159.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128041 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-22 00:04:35 +00:00
Bill Wendling
a5c177e70a We need to pass the TargetMachine object to the InstPrinter if we are printing
the alias of an InstAlias instead of the thing being aliased. Because we need to
know the features that are valid for an InstAlias.

This is part of a work-in-progress.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127986 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-21 04:13:46 +00:00
Evan Cheng
485fafc840 Re-apply r127953 with fixes: eliminate empty return block if it has no predecessors; update dominator tree if cfg is modified.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127981 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-21 01:19:09 +00:00
Daniel Dunbar
7a90e04fc7 Revert r127953, "SimplifyCFG has stopped duplicating returns into predecessors
to canonicalize IR", it broke a lot of things.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127954 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-19 21:47:14 +00:00
Evan Cheng
ae16d6b972 SimplifyCFG has stopped duplicating returns into predecessors to canonicalize IR
to have single return block (at least getting there) for optimizations. This
is general goodness but it would prevent some tailcall optimizations.
One specific case is code like this:
int f1(void);
int f2(void);
int f3(void);
int f4(void);
int f5(void);
int f6(void);
int foo(int x) {
  switch(x) {
  case 1: return f1();
  case 2: return f2();
  case 3: return f3();
  case 4: return f4();
  case 5: return f5();
  case 6: return f6();
  }
}

=>
LBB0_2:                                 ## %sw.bb
  callq   _f1
  popq    %rbp
  ret
LBB0_3:                                 ## %sw.bb1
  callq   _f2
  popq    %rbp
  ret
LBB0_4:                                 ## %sw.bb3
  callq   _f3
  popq    %rbp
  ret

This patch teaches codegenprep to duplicate returns when the return value
is a phi and where the phi operands are produced by tail calls followed by
an unconditional branch:

sw.bb7:                                           ; preds = %entry
  %call8 = tail call i32 @f5() nounwind
  br label %return
sw.bb9:                                           ; preds = %entry
  %call10 = tail call i32 @f6() nounwind
  br label %return
return:
  %retval.0 = phi i32 [ %call10, %sw.bb9 ], [ %call8, %sw.bb7 ], ... [ 0, %entry ]
  ret i32 %retval.0

This allows codegen to generate better code like this:

LBB0_2:                                 ## %sw.bb
        jmp     _f1                     ## TAILCALL
LBB0_3:                                 ## %sw.bb1
        jmp     _f2                     ## TAILCALL
LBB0_4:                                 ## %sw.bb3
        jmp     _f3                     ## TAILCALL

rdar://9147433


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127953 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-19 17:17:39 +00:00
Nadav Rotem
06cc324b9d Add support for legalizing UINT_TO_FP of vectors on platforms which do
not have native support for this operation (such as X86).
The legalized code uses two vector INT_TO_FP operations and is faster
than scalarizing.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127951 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-19 13:09:10 +00:00
Eli Friedman
b6192d2a9f Revert r127852; it's apparently causing an ICE on mingw.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127909 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-18 21:12:29 +00:00
Joerg Sonnenberger
96622aa063 Support explicit argument forms for the X86 string instructions.
For now, only the default segments are supported.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127875 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-18 11:59:40 +00:00
Eli Friedman
b4b8b0cc90 Add a target-specific branchless method for double-width relational
comparisons on x86.  Essentially, the way this works is that SUB+SBB sets
the relevant flags the same way a double-width CMP would.

This is a substantial improvement over the generic lowering in LLVM. The output
is also shorter than the gcc-generated output; I haven't done any detailed
benchmarking, though.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127852 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-18 02:34:11 +00:00
Cameron Zwarich
7bbf0ee97c Move more logic into getTypeForExtArgOrReturn.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127809 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-17 14:53:37 +00:00
Cameron Zwarich
4457968011 Rename getTypeForExtendedInteger() to getTypeForExtArgOrReturn().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127807 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-17 14:21:56 +00:00
Eli Friedman
3d831381a9 A couple new README entries.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127786 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-17 01:22:09 +00:00
Cameron Zwarich
ebe8173941 The x86-64 ABI says that a bool is only guaranteed to be sign-extended to a byte
rather than an int. Thankfully, this only causes LLVM to miss optimizations, not
generate incorrect code.

This just fixes the zext at the return. We still insert an i32 ZextAssert when
reading a function's arguments, but it is followed by a truncate and another i8
ZextAssert so it is not optimized.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127766 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-16 22:20:18 +00:00
Sean Callanan
f88896b2a9 Enabled disassembler support for AVX instructions
in the instruction tables and fixed a few bugs that
were causing decode conflicts.  Rudimentary tests
are coming up in the next patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127646 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-15 01:28:15 +00:00
Sean Callanan
a21e2eae3d X86 table-generator and disassembler support for the AVX
instruction set.  This code adds support for the VEX prefix
and for the YMM registers accessible on AVX-enabled
architectures.  Instruction table support that enables AVX
instructions for the disassembler is in an upcoming patch.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127644 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-15 01:23:15 +00:00
Eric Christopher
de5e101b0d Change the x86 32-bit scheduler to register pressure and fix up the
corresponding testcases back to the previous versions.

Fixes some performance regressions only seen on 32-bit.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127441 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-11 01:05:58 +00:00
Stuart Hastings
03d5826164 Revert 127359; it broke lencod.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127382 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-10 00:25:53 +00:00
Evan Cheng
b0519e15f7 Re-commit 127368 and 127371. They are exonerated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127380 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-10 00:16:32 +00:00
Evan Cheng
02d7c92982 Revert 127368 and 127371 for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127376 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 23:53:17 +00:00
Evan Cheng
17adafc6c1 Change the definition of TargetRegisterInfo::getCrossCopyRegClass to be more
flexible.

If it returns a register class that's different from the input, then that's the
register class used for cross-register class copies.
If it returns a register class that's the same as the input, then no cross-
register class copies are needed (normal copies would do).
If it returns null, then it's not at all possible to copy registers of the
specified register class.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127368 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 22:47:38 +00:00
Benjamin Kramer
b64b497000 Fix a pasto that broke all x86_64-elf targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127365 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 22:07:13 +00:00
Stuart Hastings
2f26fa4838 X86 byval copies no longer always_inline. <rdar://problem/8706628>
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127359 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 21:10:30 +00:00
Jan Sjödin
d1cba8727a Add createELFObjectTargetWriter method to TargetAsmBackend, which enables construction of non-standard ELFObjectWriters that can be used in MCJIT.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127346 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 18:44:41 +00:00
NAKAMURA Takumi
3ca99435e9 Target/X86: Tweak va_arg for Win64 not to miss taking va_start when number of fixed args > 4.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127328 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-09 11:33:15 +00:00
Benjamin Kramer
c175a4bd7e X86: Fix the (saddo/ssub x, 1) -> incl/decl selection to check the right operand for 1.
Found by inspection.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127247 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-08 15:20:20 +00:00
Eric Christopher
7c2cdb1c05 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.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127208 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-08 02:42:25 +00:00
Cameron Zwarich
be2119e8e2 Move getRegPressureLimit() from TargetLoweringInfo to TargetRegisterInfo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127175 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-07 21:56:36 +00:00
Andrew Trick
e0ef509aeb Increased the register pressure limit on x86_64 from 8 to 12
regs. This is the only change in this checkin that may affects the
default scheduler. With better register tracking and heuristics, it
doesn't make sense to artificially lower the register limit so much.

Added -sched-high-latency-cycles and X86InstrInfo::isHighLatencyDef to
give the scheduler a way to account for div and sqrt on targets that
don't have an itinerary. It is currently defaults to 10 (the actual
number doesn't matter much), but only takes effect on non-default
schedulers: list-hybrid and list-ilp.

Added several heuristics that can be individually disabled for the
non-default sched=list-ilp mode. This helps us determine how much
better we can do on a given benchmark than the default
scheduler. Certain compute intensive loops run much faster in this
mode with the right set of heuristics, and it doesn't seem to have
much negative impact elsewhere. Not all of the heuristics are needed,
but we still need to experiment to decide which should be disabled by
default for sched=list-ilp.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127067 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-05 08:00:22 +00:00
Andrew Trick
8d4a422513 whitespace
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127065 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-05 06:31:54 +00:00
Eli Friedman
ac39bd534b PR9377: Handle x86 str with register operand in a way consistent with gas.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126970 91177308-0d34-0410-b5e6-96231b3b80d8
2011-03-04 00:10:17 +00:00