Commit Graph

108499 Commits

Author SHA1 Message Date
Chandler Carruth
2ce5aac548 [x86] Cleanup and generate precise FileCheck assertions for a bunch of
SSE tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218947 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:37:58 +00:00
Chandler Carruth
9e7173e74d [x86] This is a terrible SSE1 test, but we should keep it. I've deleted
two functions that really didn't have any interesting assertions, and
generated more precise tests for one of the others.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218946 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:37:56 +00:00
Chandler Carruth
9a7125399b [x86] Merge two very similar tests and regenerate FileCheck lines for
them.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218945 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:37:53 +00:00
Lang Hames
07c5f89fa1 [BasicAA] Revert r218714 - Make better use of zext and sign information.
This patch broke 447.dealII on Darwin. I'm currently working on a reduced
test-case, but reverting for now to keep the bots happy.

<rdar://problem/18530107>



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218944 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:33:47 +00:00
Chandler Carruth
ea3d31f580 [x86] Regenerate a number of FileCheck assertions with my script for
test cases that will change with the new vector shuffle lowering. This
gives us a nice baseline for deltas against. I've checked and removed
the cases where there were weird register usage being pinned down, and
all of these are extremely pin-pointed tests so fully checking them
seems very appropriate.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218941 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:06:32 +00:00
Chandler Carruth
f71a17d0c6 [x86] Remove a couple of other overly isolated tests that are low-value
at this point. We have lots of tests of peephole optimizations with
insert and extract on vectors.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218940 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:06:30 +00:00
Chandler Carruth
95c5e556fc [x86] Remove a test that provides little value. There are plenty of
tests for zext of a vector.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218939 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:06:27 +00:00
Robin Morisset
6f6512cba2 Update Atomics.rst
Summary:
I changed various bits of the compilation of atomics recently, and forgot
updating the documentation. This patch just brings it up to date.

Test Plan: no change to the code

Reviewers: jfb

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D5590

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218937 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 01:04:20 +00:00
Chandler Carruth
b534439b8d [x86] Regenerate a bunch more avx512 test cases using my script to have
tighter, more strict FileCheck assertions. Some of these I really like
as they show case exactly what instruction sequences come out of these
microscopic functionality tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218936 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:50:03 +00:00
Chandler Carruth
77e97ffd61 [x86] Regenerate an avx512 test with my script to provide a nice
baseline for updates from the new vector shuffle lowering.

I've inspected the results here, and I couldn't find any register
allocation decisions where there should be any realistic way to register
allocate things differently. The closest was the imul test case. If you
see something here you'd like register number variables on, just shout
and I'll add them.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218935 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:44:46 +00:00
Eric Christopher
8f09464bc9 constify TargetMachine parameter.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218934 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:42:41 +00:00
Rui Ueyama
820664686a llvm-readobj: print COFF delay-load import table
This patch adds another iterator to access the delay-load import table
and use it from llvm-readobj.

http://reviews.llvm.org/D5594



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218933 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:41:58 +00:00
Chandler Carruth
5c10a029ad [x86] Remove some of the --show-mc-encoding flags from avx512 tests that
need to be updated for the new vector shuffle lowering.

After talking to Adam Nemet, Tim Northover, etc., it seems that testing
MC encodings in the same suite as the basic codegen isn't the right
approach. Instead, we're going to want dedicated MC tests for the
encodings. These encodings are starting to get in my way so I wanted to
cut them out early. The total set of instructions that should have
encoding tests added is:

  vpaddd
  vsqrtss
  vsqrtsd
  vmovlhps
  vmovhlps
  valignq
  vbroadcastss

Not too many parts of these tests were even using this. =]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218932 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:36:29 +00:00
Eric Christopher
59cacc9dec constify TargetMachine argument.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218930 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:17:59 +00:00
Eric Christopher
1340986490 We can grab the options struct from the TargetMachine, no need to
pass it down in the constructor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218929 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-03 00:10:03 +00:00
Adam Nemet
6955c9d1ac [AVX512] Pull pattern for subvector insert into the instruction definition
No functional change intended.

Very similar to the change I made for subvector extract in r218480.

test/CodeGen/X86/avx512-insert-extract.ll covers this.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218928 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 23:18:30 +00:00
Adam Nemet
d9e2cc7fa0 [AVX512] Refactor subvector inserts
No functional change.

Very similar to the extract refactoring I did in r218478.

Compared X86.td.expanded before and after.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218927 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 23:18:28 +00:00
Adam Nemet
a9014e5530 [AVX512] Fix i256mem->f256mem typo in VINSERTF64x4rm
Just like in the case of extracts, the refactoring is uncovering some typos in
the code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218926 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 23:18:26 +00:00
Rui Ueyama
793b1cb0a5 llvm-readobj: add a test for COFF import-by-ordinal symbols
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218924 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:40:55 +00:00
Hal Finkel
626236d9bc [PowerPC] Modern Book-E cores support sync
Older Book-E cores, such as the PPC 440, support only msync (which has the same
encoding as sync 0), but not any of the other sync forms. Newer Book-E cores,
however, do support sync, and for performance reasons we should allow the use
of the more-general form.

This refactors msync use into its own feature group so that it applies by
default only to older Book-E cores (of the relevant cores, we only have
definitions for the PPC440/450 currently).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218923 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:34:22 +00:00
Robin Morisset
2b1874cbd4 [Power] Improve the expansion of atomic loads/stores
Summary:
Atomic loads and store of up to the native size (32 bits, or 64 for PPC64)
can be lowered to a simple load or store instruction (as the synchronization
is already handled by AtomicExpand, and the atomicity is guaranteed thanks to
the alignment requirements of atomic accesses). This is exactly what this patch
does. Previously, these were implemented by complex
load-linked/store-conditional loops.. an obvious performance problem.

For example, this patch turns
```
define void @store_i8_unordered(i8* %mem) {
  store atomic i8 42, i8* %mem unordered, align 1
  ret void
}
```
from
```
_store_i8_unordered:                    ; @store_i8_unordered
; BB#0:
    rlwinm r2, r3, 3, 27, 28
    li r4, 42
    xori r5, r2, 24
    rlwinm r2, r3, 0, 0, 29
    li r3, 255
    slw r4, r4, r5
    slw r3, r3, r5
    and r4, r4, r3
LBB4_1:                                 ; =>This Inner Loop Header: Depth=1
    lwarx r5, 0, r2
    andc r5, r5, r3
    or r5, r4, r5
    stwcx. r5, 0, r2
    bne cr0, LBB4_1
; BB#2:
    blr
```
into
```
_store_i8_unordered:                    ; @store_i8_unordered
; BB#0:
    li r2, 42
    stb r2, 0(r3)
    blr

```
which looks like a pretty clear win to me.

Test Plan:
fixed the tests + new test for indexed accesses + make check-all

Reviewers: jfb, wschmidt, hfinkel

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D5587

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218922 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:27:07 +00:00
Chandler Carruth
bbb28e7e98 Fix the threshold added in r186434 (a re-apply of r185393) and updaated
to be a ManagedStatic in r218163 to not be a global variable written and
read to from within the innards of SpillPlacement.

This will fix a really scary race condition for anyone that has two
copies of LLVM running spill placement concurrently. Yikes!

This will also fix a really significant compile time hit that r218163
caused because the spill placement threshold read is actually in the
*very* hot path of this code. The memory fence on each read was showing
up as huge compile time regressions when spilling is responsible for
most of the compile time. For example, optimizing sanitized code showed
over 50% compile time regressions here. =/

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218921 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:23:14 +00:00
Juergen Ributzka
b3f91b0af7 [Stackmaps] Make ithe frame-pointer required for stackmaps.
Do not eliminate the frame pointer if there is a stackmap or patchpoint in the
function. All stackmap references should be FP relative.

This fixes PR21107.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218920 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:21:49 +00:00
Duncan P. N. Exon Smith
32e192aeb3 Revert "DI: Fold constant arguments into a single MDString"
This reverts commit r218914 while I investigate some bots.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218918 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:15:31 +00:00
Rui Ueyama
7bd81ba05b Rename data -> Data
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218916 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:13:44 +00:00
Rui Ueyama
3d49ad0cbc llvm-readobj: print COFF imported symbols
This patch defines a new iterator for the imported symbols.
Make a change to COFFDumper to use that iterator to print
out imported symbols and its ordinals.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218915 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 22:05:29 +00:00
Duncan P. N. Exon Smith
0917b70630 DI: Fold constant arguments into a single MDString
This patch addresses the first stage of PR17891 by folding constant
arguments together into a single MDString.  Integers are stringified and
a `\0` character is used as a separator.

Part of PR17891.

Note: I've attached my testcases upgrade scripts to the PR.  If I've
just broken your out-of-tree testcases, they might help.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218914 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 21:56:57 +00:00
Chandler Carruth
bf21d40070 [x86] Teach the new vector shuffle lowering to widen floating point
elements as well as integer elements in order to form simpler shuffle
patterns.

This is the primary reason why we were failing to match some of the
2-and-2 floating point shuffles such as PR21140. Even after fixing this
we need to support some extra patterns in the backend in order to match
the resulting X86ISD::UNPCKL nodes into the correct instructions. This
commit should fix PR21140 and includes more comprehensive testing of
insertion patterns in v4 shuffles.

Not all of the added tests are beautiful. For example, we don't have
clever instructions to insert-via-load in the integer domain. There are
also some places where we aren't sufficiently cunning with our use of
movq and movd, but that's future work.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218911 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 21:37:14 +00:00
Sanjay Patel
b417a35b5b Remove unused function attribute params.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218909 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 21:12:04 +00:00
Duncan P. N. Exon Smith
49048ecf56 LTO: Document the Boolean argument from r218784
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218907 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 21:11:04 +00:00
Sanjay Patel
0ff4eb684f Optimize square root squared (PR21126).
When unsafe-fp-math is enabled, we can turn sqrt(X) * sqrt(X) into X.

This can happen in the real world when calculating x ** 3/2. This occurs
in test-suite/SingleSource/Benchmarks/BenchmarkGame/n-body.c.

Differential Revision: http://reviews.llvm.org/D5584



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218906 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 21:10:54 +00:00
Chandler Carruth
130d072eb7 [x86] Move the vperm2f128 test to be vperm2x128 and test both the
floating point and integer domains.

Merge the AVX2 test into it and add an extra RUN line. Generate clean
FileCheck statements with my script. Remove the now merged AVX2 tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218903 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 20:11:11 +00:00
Justin Bogner
d76512681e InstrProf: Avoid linear search in a hot loop
Every time we were adding or removing an expression when generating a
coverage mapping we were doing a linear search to try and deduplicate
the list. The indices in the list are important, so we can't just
replace it by a DenseMap entirely, but an auxilliary DenseMap for fast
lookup massively improves the performance issues I was seeing here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218892 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 17:14:18 +00:00
Rui Ueyama
f3cd10bdc3 This patch adds a new flag "-coff-imports" to llvm-readobj.
When the flag is given, the command prints out the COFF import table.

Currently only the import table directory will be printed.
I'm going to make another patch to print out the imported symbols.

The implementation of import directory entry iterator in
COFFObjectFile.cpp was buggy. This patch fixes that too.

http://reviews.llvm.org/D5569



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218891 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 17:02:18 +00:00
Justin Bogner
ddc725b9b8 Reapply "InstrProf: Don't keep a large sparse list around just to zero it"
When I was preparing r218879 for commit, I removed an early return
that I decided was just noise. It wasn't. This is r218879 no-crash
edition.

This reverts commit r218881, reapplying r218879.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218887 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 16:43:31 +00:00
Adrian Prantl
f9e44c8bf8 Remove an extra whitespace.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218886 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 16:42:15 +00:00
Adrian Prantl
8ceff5c2b2 Pretty-printer: Paper over an ambiguity between line table entries
and tagged mdnodes.

fixes http://llvm.org/bugs/show_bug.cgi?id=21131

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218885 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 16:42:13 +00:00
Justin Bogner
a740e5d8d6 Revert "InstrProf: Don't keep a large sparse list around just to zero it"
This seems to be crashing on some buildbots. Reverting to investigate.

This reverts commit r218879.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218881 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 16:15:27 +00:00
Justin Bogner
1476756523 InstrProf: Don't keep a large sparse list around just to zero it
The Terms vector here represented a polynomial of of all possible
counters, and is used to simplify expressions when generating coverage
mapping. There are a few problems with this:

1. Keeping the vector as a member is wasteful, since we clear it every
   time we use it.
2. Most expressions refer to a subset of the counters, so we end up
   iterating over a large number of zeros doing nothing a lot of the
   time.

This updates the user of the vector to store the terms locally, and
uses a sort and combine approach so that we only operate on counters
that are actually used in a given expression. For small cases this
makes very little difference, but in cases with a very large number of
counted regions this is a significant performance fix.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218879 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 16:04:03 +00:00
Sanjay Patel
dafb357bb5 Use the local variable that other clauses around here are already using.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218876 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 15:20:45 +00:00
Sanjay Patel
e165693c7a Remove duplicate function names from comments. NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218875 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 15:13:22 +00:00
Tilmann Scheller
ad7783df73 [NVPTX] Remove dead code.
Found by the Clang static analyzer.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218874 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 15:12:48 +00:00
Joerg Sonnenberger
92583e0712 Support padding unaligned data in .text.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218870 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 13:41:42 +00:00
Aaron Ballman
1af4bec1e2 Silence a -Wsign-compare warning. NFC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218868 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 13:17:11 +00:00
Zinovy Nis
31713943aa [BUG][INDVAR] Fix for PR21014: wrong SCEV operands commuting for non-commutative instructions
My commit rL216160 introduced a bug PR21014: IndVars widens code 'for (i = ; i < ...; i++) arr[ CONST - i]' into 'for (i = ; i < ...; i++) arr[ i - CONST]'
thus inverting index expression. This patch fixes it. 
Thanks to Jörg Sonnenberger for pointing.

Differential Revision: http://reviews.llvm.org/D5576



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218867 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 13:01:15 +00:00
Chandler Carruth
fd94d1bd76 [x86] Just delete the last combine test file.
This file isn't really doing anything useful. Many of the tests that
seem to be combined are also repeats from other test files. Many of the
other tests, despite the comment that they should be combined into
a single shuffle... well... aren't combined into a single shuffle.
=/

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218862 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 08:05:57 +00:00
Chandler Carruth
7d663050b4 [x86] Merge still more combine tests into the common file. These at
least seem *slightly* more interesting test wise, although given how
spotily we actually combine anything, I remain somewhat suspicious.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218861 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 08:02:34 +00:00
Chandler Carruth
cd912001b4 [x86] Merge the third combining test into the generic one and add proper
checks for all the ISA variants.

If the SSE2 checks here terrify you, good. This is (in large part) the
kind of amazingly bad code that is holding LLVM back when vectorizing on
older ISAs.

At the same time, these tests seem increasingly dubious to me. There are
a very large number of tests and it isn't clear that they are
systematically covering a specific set of functionality. Anyways,
I don't want to reduce testing during the transition, I just want to
consolidate it to where it is easier to manage.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218860 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 07:56:47 +00:00
Chandler Carruth
84b3f53bcb [x86] Merge the second set of vector combining tests into a common test
file.

Some of these really don't make sense to test -- we're testing for the
*lack* of combining two shuffles into one, presumably because the two
would generate better shuffles in the end. But if you look at the
generated code shown here, in many cases the generated code is, frankly,
terrible. Or we combine any two generated shuffles back into a single
instruction! I've left a FIXME to revisit these decisions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218859 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 07:42:58 +00:00
Chandler Carruth
84c7078ddc [x86] Merge the bitwise operation shuffle combining into the common test
file, adding assertions across the ISA variants for it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@218858 91177308-0d34-0410-b5e6-96231b3b80d8
2014-10-02 07:30:24 +00:00