Commit Graph

101614 Commits

Author SHA1 Message Date
Benjamin Kramer
c1c74fb2b4 ScalarEvolution: Compute exit counts for loops with a power-of-2 step.
If we have a loop of the form
for (unsigned n = 0; n != (k & -32); n += 32) {}
then we know that n is always divisible by 32 and the loop must
terminate. Even if we have a condition where the loop counter will
overflow it'll always hold this invariant.

PR19183. Our loop vectorizer creates this pattern and it's also
occasionally formed by loop counters derived from pointers.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204728 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 16:25:12 +00:00
Matt Arsenault
71a8d2baba Fix creating illegal setcc cond codes.
If GT/UGT or LT/ULT were set to expand, a comparison
with a constant would replace it with the illegal
cond code.

There are several more places later in this function that
will have the same basic problem.

Theoretically R600 should hit this problem for a test,
but for some reason it doesn't.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204727 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 16:09:21 +00:00
Evgeniy Stepanov
ffa7f428c5 [msan] Relax the test some more.
This may or may not fix the bots. R204720 did not.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204721 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 14:32:05 +00:00
Evgeniy Stepanov
0b9bcfa50f [msan] Make some tests less strict.
This may or may not fix the bots.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204720 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 14:15:14 +00:00
Rafael Espindola
2cf1cdfb4e Fix these tests on windows.
It is impossible to create a hard link to a non existing file, so create a
dummy file, create the link an delete the dummy file.

On windows one cannot remove the current directory, so chdir first.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204719 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 13:19:03 +00:00
Evgeniy Stepanov
800399636a [msan] More precise instrumentation of select IR.
Some bits of select result may be initialized even if select condition
is not.

https://code.google.com/p/memory-sanitizer/issues/detail?id=50


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204716 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 13:08:34 +00:00
Daniel Sanders
3527e5fcb3 [mips] '.set at=$0' should be equivalent to '.set noat'
Differential Revision: http://llvm-reviews.chandlerc.com/D3171

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204714 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 13:01:06 +00:00
Cameron McInally
3ec862b7ae Fix AVX2 Gather execution domains.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204713 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 12:36:38 +00:00
Daniel Sanders
c7e1663c24 [mips] Correct testcase for .set at=$reg and emit the new warnings for numeric registers too.
Summary:
Remove the XFAIL added in my previous commit and correct the test such that
it correctly tests the expansion of the assembler temporary.

Also added a test to check that $at is always $1 when written by the
user.

Corrected the new assembler temporary warnings so that they are emitted for
numeric registers too.

Differential Revision: http://llvm-reviews.chandlerc.com/D3169

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204711 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 11:16:03 +00:00
Daniel Sanders
c141b331b9 [mips] Fix assembler temporary expansion and add associated warnings about the use of $at.
Summary:
The assembler temporary is normally $at ($1) but can be reassigned using
'.set at=$reg'. Regardless of which register is nominated as the assembler
temporary, $at remains $1 when written by the user.

Adds warnings under the following conditions:
* The register nominated as the assembler temporary is used by the user.
* '.set noat' is in effect and $at is used by the user.
Both of these only work for named registers. I have a follow up commit that makes it work for numeric registers as well.

XFAIL set-at-directive.s since it incorrectly tests that $at is redefined by
'.set at=$reg'. Testcases will follow in a separate commit.

Patch by David Chisnall
His work was sponsored by: DARPA, AFRL

Differential Revision: http://llvm-reviews.chandlerc.com/D3167

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204710 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 10:57:07 +00:00
Yaron Keren
6e2aab69a6 Remove cmake module support for Visual C++ 2010 (MSVC10)
but keep the MSVC11 (Visual C++ 2012) support.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204706 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 09:34:20 +00:00
Erik Verbruggen
c5f5d0d234 Simplify loop that worked around bugs in old GCC/Xcode.
GCC 4.0.1 and Xcode 2 are no longer supported for building llvm/clang.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204705 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 09:06:18 +00:00
Yaron Keren
d8ceb275c2 Disable Visual C++ warning 4722 about aborting a destructor,
it has no value for us.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204704 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 08:42:49 +00:00
David Majnemer
4fd5cd06c8 WinCOFF: Add support for -fdata-sections
This is a pretty straight forward translation for COFF, we just need to
stick the data in a COMDAT section marked as
IMAGE_COMDAT_SELECT_NODUPLICATES.

N.B. We must be careful to avoid sticking entities with private linkage
in COMDAT groups.  COFF is pretty hostile to the renaming of entities so
we must be careful to disallow GlobalVariables with unstable names.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204703 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 06:14:26 +00:00
David Blaikie
9d5961d0d8 DebugInfo: Add GNU_addr_base and GNU_ranges_base only when there are addresses or ranges
Based on code review feedback from Eric in r204672.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204702 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 05:34:24 +00:00
Saleem Abdulrasool
f50e709043 test: fix CHECK lines
Thanks to gix for pointing out that the CHECK-LABEL lines were incorrect!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204700 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 03:39:39 +00:00
Andrew Trick
f5bf687cf7 SLP vectorizer: Don't hoist vector extracts of phis.
Extracts coming from phis were being hoisted, while all others were
sunk to their uses. This was inconsistent and didn't seem to serve a
purpose. Changing all extracts to be sunk to uses is a prerequisite
for adding block frequency to the SLP vectorizer's cost model.

I benchmarked the change in isolation (without block frequency). I
only saw noise on x86 and some potentially significant improvements on
ARM. No major regressions is good enough for me.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204699 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 02:18:47 +00:00
David Blaikie
b13d54f588 DebugInfo: Support debug_loc under fission
Implement debug_loc.dwo, as well as llvm-dwarfdump support for dumping
this section.

Outlined in the DWARF5 spec and http://gcc.gnu.org/wiki/DebugFission the
debug_loc.dwo section has more variation than the standard debug_loc,
allowing 3 different forms of entry (plus the end of list entry). GCC
seems to, and Clang certainly, only use one form, so I've just
implemented dumping support for that for now.

It wasn't immediately obvious that there was a good refactoring to share
the implementation of dumping support between debug_loc and
debug_loc.dwo, so they're separate for now - ideas welcome or I may come
back to it at some point.

As per a comment in the code, we could choose different forms that may
reduce the number of debug_addr entries we emit, but that will require
further study.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204697 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 01:44:02 +00:00
David Blaikie
6cfebbd64d DebugInfo: Remove unnecessary zero-size check
This seems excessive - switching section isn't expensive (or if it is
we're already being wasteful, since we emitted the debug_loc section
symbol earlier anyway) and otherwise there's no work that happens in
this function when the list is empty.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204696 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 01:43:56 +00:00
Justin Bogner
93065647b8 Support: Functions for consuming endian specific data from a buffer.
This adds a function to Endian.h that reads from and updates a pointer
into a buffer with endian specific data. This is more convenient for
stream-like reading of data than endian::read.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204693 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 01:04:44 +00:00
Manman Ren
692d1830f3 Register Allocator: check other options before using a CSR for the first time.
When register allocator's stage is RS_Spill, we choose spill over using the CSR
for the first time, if the spill cost is lower than CSRCost. 
When register allocator's stage is < RS_Split, we choose pre-splitting over
using the CSR for the first time, if the cost of splitting is lower than
CSRCost.

CSRCost is set with command-line option "regalloc-csr-first-time-cost". The
default value is 0 to generate the same codes as before this commit.

With a value of 15 (1 << 14 is the entry frequency), I measured performance
gain of 3% on 253.perlbmk and 1.7% on 197.parser, with instrumented PGO,
on an arm device.

rdar://16162005


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204690 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 00:16:25 +00:00
Kevin Enderby
4a88cd08da Fix crashes when assembler directives are used that are not
for Mach-O object files by generating an error instead.

rdar://16335232


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204687 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-25 00:05:50 +00:00
Manman Ren
66124f9e89 Register Allocator: refactoring (no functionality change).
Factor out two functions calculateRegionSplitCost and doRegionSplit
from tryRegionSplit. These two functions will be used in coming patches.

rdar://16162005


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204684 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 23:23:42 +00:00
David Blaikie
0f53443d82 DebugInfo: Simplify debug loc list handling by keeping separate lists
Rather than using a flat list with "empty" entries (ala the actual
on-disk format), keep separate lists for each variable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204680 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 22:38:38 +00:00
David Blaikie
7385c3207c DwarfDebug: Simplify debug_loc merging
No functional change intended.

Merging up-front rather than delaying this task until later. This just
seems simpler and more efficient (avoiding growing the debug loc list
only to have to skip over those post-merged entries, etc).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204679 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 22:27:06 +00:00
Adrian Prantl
1b14452fe4 Get rid of an unnecessary use of the * and & operators.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204673 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 21:33:01 +00:00
David Blaikie
882d89ead8 DebugInfo: Add DW_AT_GNU_ranges_base to skeleton CUs
This is used to avoid relocations in the dwo file by allowing
DW_AT_ranges specified in debug_info.dwo to be relative to this base
address. (r204667 implements the base-relative DW_AT_ranges side of
this)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204672 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 21:31:35 +00:00
Justin Bogner
2563e201c1 Support: Document Endian.h functions
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204671 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 21:30:55 +00:00
David Blaikie
f4ec1889db DebugInfo: Implement relative addressing for DW_AT_ranges under fission
This removes the debug_ranges relocations from debug_info.dwo (but
doesn't implement the DW_AT_GNU_ranges_base which is also necessary for
correct functioning)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204668 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 21:07:27 +00:00
David Blaikie
05e62b507a DebugInfo: Don't emit relocations to abbreviations in debug_info.dwo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204667 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:53:02 +00:00
David Blaikie
f2c696ac66 DwarfDebug: Remove an unused parameter
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204665 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:31:01 +00:00
Matt Arsenault
e130844e41 R600: Don't viewCFG() under DEBUG() except on failure.
Having these popping up every time you use -debug is really
irritating.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204664 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:29:02 +00:00
David Blaikie
fcd547ce03 Remove unused parameter
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204663 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:28:10 +00:00
Matt Arsenault
add2e2ec8f R600/SI: Fix extra mov from legalizing 64-bit SALU ops.
Check the register class of each operand individually
to avoid an extra copy to a vgpr.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204662 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:08:13 +00:00
Matt Arsenault
3a96e61469 R600/SI: Sub-optimial fix for 64-bit immediates with SALU ops.
No longer asserts, but now you get moves loading legal immediates
into the split 32-bit operations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204661 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:08:09 +00:00
Matt Arsenault
db1807144a R600/SI: Fix 64-bit bit ops that require the VALU.
Try to match scalar and first like the other instructions.
Expand 64-bit ands to a pair of 32-bit ands since that is not
available on the VALU.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204660 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 20:08:05 +00:00
Yaron Keren
d816eda3c7 In Release modes, Visual Studio complains that the Operator destructor in User.cpp
never returns, which is true by design. 

Initially assumed that the reason is llvm_unreachable being dependent on NDEBUG.

However, even if llvm_unreachable is replaced by __assume(false), VC still warns in
Release modes but not in Debug modes...

The real reason turned out to be optimization flags.
With /Od in Debug modes the warning is not issued whereas with /O1 it is.

I could not find any documentation to this effect, but it is reproducable:

Try compiling http://msdn.microsoft.com/en-us/library/khwfyc5d(v=vs.90).aspx
with /O1 and then with /Od.




git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204659 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 19:48:13 +00:00
Matt Arsenault
6c199d8212 R600: Implement isNarrowingProfitable.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204658 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 19:43:31 +00:00
Matt Arsenault
03cd663eb1 R600/SI: Move splitting 64-bit immediates to separate function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204651 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 18:26:52 +00:00
Aaron Ballman
9f00cd9602 Adding some very nascent information about the clang tablegen backends, with a promise to add more information later.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204635 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 18:18:31 +00:00
Ulrich Weigand
47eac58333 [PowerPC] Generate little-endian object files
As a first step towards real little-endian code generation, this patch
changes the PowerPC MC layer to actually generate little-endian object
files.  This involves passing the little-endian flag through the various
layers, including down to createELFObjectWriter so we actually get basic
little-endian ELF objects, emitting instructions in little-endian order,
and handling fixups and relocations as appropriate for little-endian.

The bulk of the patch is to update most test cases in test/MC/PowerPC
to verify both big- and little-endian encodings.  (The only test cases
*not* updated are those that create actual big-endian ABI code, like
the TLS tests.)

Note that while the object files are now little-endian, the generated
code itself is not yet updated, in particular, it still does not adhere
to the ELFv2 ABI.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204634 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 18:16:09 +00:00
Quentin Colombet
4768df00c4 [X86][ISelDAG] Add missing fallback patterns for avx2 broadcast instructions.
Those patterns are used when the load cannot be folded into the related broadcast
during the select phase.
This happens when the load gets additional uses that were not anticipated during
the previous lowering phases (constant vector to constant load, then constant
load reused) or when selection DAG is not able to prove that folding the load
will not create a cycle in the DAG.

<rdar://problem/16074331>


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204631 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 17:54:19 +00:00
Matt Arsenault
875870fdb4 R600/SI: Fix 64-bit private loads.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204630 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 17:50:46 +00:00
Hans Wennborg
657c472bb0 VS integration installer: set SUCCESS=1 if we find VS 2013
Previously we would print an error message on machines where the only VS
version we find is 2013, even though we successfully install the integration
files for it.

Also, we shouldn't have two END labels.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204629 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 17:33:22 +00:00
Eli Bendersky
b6f02c13da Add test to test/CodeGen/NVPTX for "alloca buffer" arguments.
Make sure such IR gets properly lowered to PTX.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204624 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:52:30 +00:00
Adam Nemet
a1b54dd1ff [X86] Fix non-determinism in LowerVectorAllZeroTest
This can be observed with the old testcase of CodeGen/X86/pr12312.ll:

47c47
<       vorps   %ymm0, %ymm1, %ymm0
---
>       vorps   %ymm1, %ymm0, %ymm0
97c97
<       vorps   %ymm1, %ymm0, %ymm0
---
>       vorps   %ymm0, %ymm1, %ymm0

The vector VecIns is populated with all the values from VecInMap. This is done
while iterating VecInMap.  VecInMap uses a hash of pointer values so the
resulting order can vary depending on the memory layout.

The fix is to populate the vector VecIns earlier as VecInMap is populated.
This is done in DAG traversal order.

Fixes <rdar://problem/16398806>

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204623 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:52:08 +00:00
Daniel Sanders
8ce101ed10 [mips] Add error message when trying to use $at in '.set noat' mode.
Summary:
Patch by David Chisnall
His work was sponsored by: DARPA, AFRL

Differential Revision: http://llvm-reviews.chandlerc.com/D3158

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204621 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:48:01 +00:00
Eli Bendersky
2685aa8713 Removes the NVPTXSplitBBatBar pass.
This pass is a historic remnant and actually causes less efficient code to be
generated in some cases.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204620 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:36:39 +00:00
Tom Stellard
4ddee6a5da R600/SI: Fix warning with gcc 4.8.2
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204618 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:12:34 +00:00
Tom Stellard
65b5e9b4ef R600/SI: Promote fp64 SELECT to i64
This type promotion is replacing a Tablegen pattern and it is already
covered by existing tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204617 91177308-0d34-0410-b5e6-96231b3b80d8
2014-03-24 16:07:30 +00:00