a8e4b4bc7b
when we see a unaligned load from an insufficiently aligned global or
...
alloca, increase the alignment of the load, turning it into an aligned load.
This allows us to compile:
#include <xmmintrin.h>
__m128i foo(__m128i x){
static const unsigned int c_0[4] = { 0, 0, 0, 0 };
__m128i v_Zero = _mm_loadu_si128((__m128i*)c_0);
x = _mm_unpacklo_epi8(x, v_Zero);
return x;
}
into:
_foo:
punpcklbw _c_0.5944, %xmm0
ret
.data
.lcomm _c_0.5944,16,4 # c_0.5944
instead of:
_foo:
movdqu _c_0.5944, %xmm1
punpcklbw %xmm1, %xmm0
ret
.data
.lcomm _c_0.5944,16,2 # c_0.5944
llvm-svn: 40971
2007-08-09 19:05:49 +00:00
a05ec4dc52
GR16_ sub-register class should be GR8_, not GR8. That is, it should only be 8-bit registers in 32-bit mode. Ditto for GR32_.
...
llvm-svn: 40970
2007-08-09 18:05:17 +00:00
c339e45274
Update per review comments.
...
llvm-svn: 40965
2007-08-09 17:27:48 +00:00
9b1cc8cac0
Make NonLocal and None const in the right way. :-)
...
llvm-svn: 40961
2007-08-09 04:42:44 +00:00
42e3e5bec1
Traverse loop blocks' terminators to find split candidates.
...
llvm-svn: 40960
2007-08-09 01:39:01 +00:00
ba1a98a4e0
long double 9 of N. This finishes up the X86-32 bits
...
(constants are still not handled). Adds ConvertActions
to control fp-to-fp conversions (these are currently
defaulted for all other targets, so no changes there).
llvm-svn: 40958
2007-08-09 01:04:01 +00:00
f5124b36e4
Fix arguments for some Altivec instructions. From SWB.
...
llvm-svn: 40957
2007-08-09 00:49:19 +00:00
9d09c5ccda
If a target really needs to custom lower constants, it should be allowed
...
to do so.
llvm-svn: 40955
2007-08-08 23:23:31 +00:00
2b21c3c7a8
Add more comments to memdep.
...
llvm-svn: 40953
2007-08-08 22:26:03 +00:00
0183c797c4
Add cost analysis.
...
llvm-svn: 40952
2007-08-08 22:25:28 +00:00
fa788358d5
Make memdep fit in 80 cols.
...
llvm-svn: 40950
2007-08-08 22:01:54 +00:00
927f62a3a1
Add one more comment.
...
llvm-svn: 40949
2007-08-08 21:54:33 +00:00
064bad44d1
Cleanup and comment-ize the memdep header.
...
llvm-svn: 40948
2007-08-08 21:53:20 +00:00
0e34ee25ab
Preserve dom info while processing one iteration loop.
...
llvm-svn: 40947
2007-08-08 21:39:47 +00:00
b84d3b1c92
Change the None and NonLocal markers in memdep to be const.
...
llvm-svn: 40946
2007-08-08 21:39:39 +00:00
fd46badca5
Separate program name from error message with a :
...
llvm-svn: 40945
2007-08-08 21:19:01 +00:00
8abc5c82b7
Clear split info.
...
llvm-svn: 40944
2007-08-08 21:18:27 +00:00
f054db3242
Terminate an error message with a newline.
...
llvm-svn: 40943
2007-08-08 21:17:59 +00:00
593bf9ceb3
Handle multiple split conditions.
...
llvm-svn: 40941
2007-08-08 21:02:17 +00:00
8574ac95f0
Make it clear that getSTDIN returns null when stdin is empty.
...
llvm-svn: 40940
2007-08-08 20:02:20 +00:00
155533cf73
Make getSTDIN return null if the standard input is empty, as the header file
...
documentation implies and as its uses depend.
llvm-svn: 40939
2007-08-08 20:01:58 +00:00
8609c06956
Allow the filename "-" to be a place holder for stdin. This allows directing
...
stdin through llvm-ld and llvm-link.
llvm-svn: 40938
2007-08-08 19:52:29 +00:00
680862880d
Global values also don't undead-ify pointers in our dead alloca's set.
...
llvm-svn: 40936
2007-08-08 19:12:31 +00:00
ddf4aee543
Make handleEndBlock significantly faster with one trivial improvement,
...
and one hack to avoid hitting a bad case when the alias analysis is imprecise.
llvm-svn: 40935
2007-08-08 18:38:28 +00:00
50df9685b0
Small improvement: if a function doesn't access memory, we don't need to scan
...
it for potentially undeading pointers.
llvm-svn: 40933
2007-08-08 17:58:56 +00:00
52aaabf74d
Add some comments, remove a dead argument, and simplify some control flow.
...
No functionality change.
llvm-svn: 40932
2007-08-08 17:50:09 +00:00
d1fd0db2ae
eliminate redundant conditions from the signless types conversion.
...
llvm-svn: 40927
2007-08-08 16:19:57 +00:00
7574ef3ac4
Handle functions with no name better.
...
llvm-svn: 40926
2007-08-08 16:07:23 +00:00
d771b793fe
Adding kill info to val#.
...
llvm-svn: 40925
2007-08-08 07:03:29 +00:00
785f9986bd
significantly speed up constant folding of calls (and thus all clients that use
...
ConstantFoldInstruction on calls) by avoiding Value::getName(). getName() constructs
and returns an std::string, which does heap allocation stuff. This slightly speeds up
instcombine.
llvm-svn: 40924
2007-08-08 06:55:43 +00:00
415750e132
Speed up updateDFSNumbers with two observations:
...
1. domtree is a tree, not a graph. There is no need to avoid revisiting nodes with a set.
2. the worklist can contain the child iterator pointers so we don't get N^2 rescanning of children.
This speeds up updateDFSNumbers significantly, making it basically free. On the testcase in PR1432,
this speeds up loopsimplify by another 3x, dropping it from the 12th most expensive pass to the to
the 30th. :) It used to be #1 .
llvm-svn: 40923
2007-08-08 06:24:20 +00:00
b17ab03081
A few more small cleanups.
...
llvm-svn: 40922
2007-08-08 06:06:02 +00:00
103947125c
Clean up and bug fix.
...
llvm-svn: 40921
2007-08-08 05:56:18 +00:00
c63d4c2e4e
reimplement dfs number computation to be significantly faster. This speeds up
...
natural loop canonicalization (which does many cfg xforms) by 4.3x, for
example. This also fixes a bug in postdom dfnumber computation.
llvm-svn: 40920
2007-08-08 05:51:24 +00:00
0aecf0ebef
First round of cleanups from Chris' feedback.
...
llvm-svn: 40919
2007-08-08 04:52:29 +00:00
a8c2f38617
- Each val# can have multiple kills.
...
- Fix some minor bugs related to special markers on val# def. ~0U means
undefined, ~1U means dead val#.
llvm-svn: 40916
2007-08-08 03:00:28 +00:00
68de1ae816
Embrace patch review feedback.
...
llvm-svn: 40915
2007-08-08 01:51:27 +00:00
c236617ea0
Remove a dead assertion.
...
llvm-svn: 40914
2007-08-08 01:00:21 +00:00
0d0fee269a
- LiveInterval value#'s now have 3 components: def instruction #,
...
kill instruction #, and source register number (iff the value# is defined by a
copy).
- Now def instruction # is set for every value#, not just for copy defined ones.
- Update some outdated code related inactive live ranges.
- Kill info not yet set. That's next patch.
llvm-svn: 40913
2007-08-07 23:49:57 +00:00
c7e53bdcfd
Fix new compare instruction's signness. Caught by Chris during review.
...
llvm-svn: 40912
2007-08-07 23:17:52 +00:00
549c0d5175
Fix comment.
...
llvm-svn: 40911
2007-08-07 23:16:03 +00:00
0cc1a76283
Don't insert nearly as many redundant phi nodes.
...
llvm-svn: 40909
2007-08-07 23:12:31 +00:00
4e7ff3593c
Fix spelling of mtvscr and mfvscr.
...
llvm-svn: 40908
2007-08-07 23:08:00 +00:00
a47f7d7cfd
Long double patch 8 of N: make it partially work in
...
SSE mode (all but conversions <-> other FP types, I think):
>>Do not mark all-80-bit operations as "Requires[FPStack]"
(which really means "not SSE").
>>Refactor load-and-extend to facilitate this.
>>Update comments.
>>Handle long double in SSE when computing FP_REG_KILL.
llvm-svn: 40906
2007-08-07 20:29:26 +00:00
8feb9d67e8
Who thought up this crazy formatting scheme?
...
llvm-svn: 40905
2007-08-07 17:57:36 +00:00
4391aa042c
Okay, over/back tags don't next.
...
llvm-svn: 40904
2007-08-07 17:48:56 +00:00
19211b6528
Use eraseFromParent().
...
llvm-svn: 40903
2007-08-07 17:45:35 +00:00
5f5bb46e95
Try an indent level for better formatting.
...
Add the -version option.
llvm-svn: 40902
2007-08-07 17:43:48 +00:00
d64584f64b
Add the code generation options.
...
llvm-svn: 40900
2007-08-07 17:12:43 +00:00
c600d65e2b
Make this code more efficient
...
llvm-svn: 40899
2007-08-07 16:57:55 +00:00
Chris Lattner