083bb46370
regenerate.
...
llvm-svn: 45085
2007-12-17 01:17:35 +00:00
edf0788758
Change the PointerType api for creating pointer types. The old functionality of PointerType::get() has become PointerType::getUnqual(), which returns a pointer in the generic address space. The new prototype of PointerType::get() requires both a type and an address space.
...
llvm-svn: 45082
2007-12-17 01:12:55 +00:00
55c6d4f22f
Make it clear in the LangRef that allocation instructions only operated on the generic address space. Implement support in the verifier for ensuring this is true.
...
llvm-svn: 45080
2007-12-17 01:00:21 +00:00
56ed48036b
Revert this part of r45073 until the verifier is
...
changed not to reject invoke of inline asm.
llvm-svn: 45077
2007-12-16 21:01:21 +00:00
2af27c202c
don't violate C TBAA rules, use FloatToBits instead.
...
llvm-svn: 45076
2007-12-16 20:41:33 +00:00
e3b05fe31b
fix a questionable cast, thanks to Mike Stump for pointing this out.
...
llvm-svn: 45075
2007-12-16 20:26:54 +00:00
dab6bd902e
Fix the JIT encoding of cmp*ss, which aborts with this assertion currently:
...
X86CodeEmitter.cpp:378: failed assertion `0 && "Immediate size not set!"'
I *think* this is right, but Evan, please verify. It also looks like
CMPSDrr and maybe others are missing this info. Evan, plz investigate.
llvm-svn: 45074
2007-12-16 20:12:41 +00:00
8e4847ee95
Make instcombine promote inline asm calls to 'nounwind'
...
calls. Remove special casing of inline asm from the
inliner. There is a potential problem: the verifier
rejects invokes of inline asm (not sure why). If an
asm call is not marked "nounwind" in some .ll, and
instcombine is not run, but the inliner is run, then
an illegal module will be created. This is bad but
I'm not sure what the best approach is. I'm tempted
to remove the check in the verifier...
llvm-svn: 45073
2007-12-16 15:51:49 +00:00
fcf2085731
Remove spurious warnings from GCC:
...
warning: suggest a space before ';' or explicit braces around empty
body in 'for' statement
Patch by Mike Stump (modified slightly by yours truly).
llvm-svn: 45071
2007-12-16 09:16:12 +00:00
7b8a741189
Break local interferences in StrongPHIElimination. One step closer...
...
llvm-svn: 45070
2007-12-16 05:44:27 +00:00
ccb3981256
A few more comments.
...
llvm-svn: 45069
2007-12-16 04:07:23 +00:00
ce1489e575
Define addString() and lookup() out-of-line to dissuade the C++ compiler from inlining it.
...
llvm-svn: 45068
2007-12-16 01:36:16 +00:00
fbb9e418b5
Provide GraphTraits and DOTGraphTraits interface for Trie.
...
Retoss private/public stuff.
Make copy ctor and operator= private.
llvm-svn: 45067
2007-12-16 01:27:04 +00:00
06e60b714b
Constify graph arguments
...
llvm-svn: 45066
2007-12-16 00:42:19 +00:00
659c3f8dd1
Use references in DF iterators. This eliminates copy-ctor calls on huge objects (graphs)
...
llvm-svn: 45063
2007-12-15 22:23:24 +00:00
26d6539e70
These are more correctly called signaling NaNs.
...
llvm-svn: 45059
2007-12-15 17:37:40 +00:00
23d2d4dc6c
Make better use of instructions that clear high bits; fix various 2-wide shuffle bugs.
...
llvm-svn: 45058
2007-12-15 03:00:47 +00:00
cbeec14a6f
__builtin_ia32_movqv4si is now expanded to a shuffle.
...
llvm-svn: 45057
2007-12-15 02:54:12 +00:00
0aa7133f82
Start committing working test cases for CellSPU.
...
llvm-svn: 45050
2007-12-15 00:38:50 +00:00
9556729128
Actually, MOVPQIto64mr is a dup of MOVPQI2QImr, MOV64toPQIrm is a dup of MOVQI2PQIrm.
...
llvm-svn: 45041
2007-12-14 20:08:14 +00:00
e28372c0d6
Fix (mem) <-> low 64-bits of xmm bugs pointed out by David Greene. Mac OS X Leopard assembler recognizes movq.
...
llvm-svn: 45040
2007-12-14 19:54:07 +00:00
1202d1b1c4
Teach the interpreter to read and write memory in the
...
endianness of the target not of the host. Done by the
simple expedient of reversing bytes for primitive types
if the host and target endianness don't match. This is
correct for integer and pointer types. I don't know if
it is correct for floating point types.
llvm-svn: 45039
2007-12-14 19:38:31 +00:00
f7cefdd5f0
x86-32 long doubles are 4-byte aligned on the stack
...
for parameter passing (only for that, on Darwin).
llvm-svn: 45038
2007-12-14 19:25:34 +00:00
a56e6ff9a7
Fix bsf / bsr jit encoding.
...
llvm-svn: 45037
2007-12-14 18:49:43 +00:00
f28c810036
Oops. Forgot these.
...
llvm-svn: 45036
2007-12-14 18:25:34 +00:00
8a332b235d
Add explicit keywords, and fix a minor typo that they uncovered.
...
llvm-svn: 45034
2007-12-14 15:41:34 +00:00
11352df8e4
Don't redirect stderr when it isn't needed.
...
llvm-svn: 45033
2007-12-14 15:15:11 +00:00
6869bfc351
Fix a typo in a comment.
...
llvm-svn: 45032
2007-12-14 15:13:08 +00:00
4069a593cc
Fix a typo in a comment.
...
llvm-svn: 45031
2007-12-14 15:11:58 +00:00
9d2e9e376f
Fix Intel asm syntax for the bsr and bsf instructions.
...
llvm-svn: 45030
2007-12-14 15:10:00 +00:00
0e6408124e
Fix ctlz and cttz. llvm definition requires them to return number of bits in of the src type when value is zero.
...
llvm-svn: 45029
2007-12-14 08:30:15 +00:00
0fcf56f8f5
Bug fix. Must also match ResNo when matching an operand with a user.
...
llvm-svn: 45028
2007-12-14 08:25:15 +00:00
f924f34b6b
All MMX shift instructions took a <2 x i32> vector as the shift amount parameter. Change this to be <1 x i64> instead, which matches the assembler instruction.
...
llvm-svn: 45027
2007-12-14 06:38:54 +00:00
e9fbc3f014
Implement ctlz and cttz with bsr and bsf.
...
llvm-svn: 45024
2007-12-14 02:13:44 +00:00
cb77f04e1f
Add flags to indicate that there are "never" side effects or that there "may be"
...
side effects for machine instructions.
llvm-svn: 45022
2007-12-14 01:48:59 +00:00
d4551fd0a4
Make it more clear that some things that can't be done in .td files can
...
still be done in the LLVM code generator. And update the summary for the
X86 target.
llvm-svn: 45013
2007-12-13 20:43:47 +00:00
86d51b896e
Make these loops follow GetGEPOperands() behavior.
...
Let: %q = GEP %p, X, ...
If %p is a GEP, we can chase baseptr further, only if X==0.
llvm-svn: 44999
2007-12-13 16:22:58 +00:00
2011df4e39
Fix typo.
...
llvm-svn: 44997
2007-12-13 07:50:36 +00:00
53b677e4e8
Add register pairs to the list to check for local interferences.
...
llvm-svn: 44987
2007-12-13 05:53:03 +00:00
1f93edd08a
Remove ugly and horrible code. It's not necessary for correctness, and can be added back later if it causes code quality issues.
...
llvm-svn: 44986
2007-12-13 05:43:37 +00:00
37c36ed79a
Be extra careful with extension use optimation. Now turned on by default.
...
llvm-svn: 44981
2007-12-13 03:32:53 +00:00
23ffde507f
Apply CBE/MSIL patch to autoconf
...
llvm-svn: 44980
2007-12-13 02:24:45 +00:00
938d8cb8d9
Add install-libs target which only installs libraries, not tools
...
llvm-svn: 44979
2007-12-13 02:17:17 +00:00
461d8419a7
Remove this testcase as it will always fail on
...
platform like Darwin.
llvm-svn: 44977
2007-12-13 02:03:57 +00:00
272ee458cf
Do not build CBackend and MSIL regardless of configured targets
...
llvm-svn: 44973
2007-12-13 01:18:52 +00:00
827d30db19
Fold some and + shift in x86 addressing mode.
...
llvm-svn: 44970
2007-12-13 00:43:27 +00:00
687567bca5
Oops. Forgot these.
...
llvm-svn: 44969
2007-12-13 00:42:35 +00:00
f0fbf077d2
Fix for edge profiling, patch by 'Marc' for PR1857
...
llvm-svn: 44965
2007-12-13 00:04:46 +00:00
7df585d9c4
Cosmetic change.
...
llvm-svn: 44961
2007-12-12 23:15:59 +00:00
6e68381e02
Implicit def instructions, e.g. X86::IMPLICIT_DEF_GR32, are always re-materializable and they should not be spilled.
...
llvm-svn: 44960
2007-12-12 23:12:09 +00:00
fde556745b
Remove host endianness info from TargetData and
...
put it in a new header System/Host.h instead.
Instead of getting the endianness from configure,
calculate it directly.
llvm-svn: 44959
2007-12-12 23:03:45 +00:00
015982aa32
Remove a forward-declaration for a non-existant class.
...
llvm-svn: 44955
2007-12-12 22:25:09 +00:00
7a7742c2fe
Allow vector integer constants to be created with
...
SelectionDAG::getConstant, in the same way as vector floating-point
constants. This allows the legalize expansion code for @llvm.ctpop and
friends to be usable with vector types.
llvm-svn: 44954
2007-12-12 22:21:26 +00:00
7ab71379b6
Use vector for child storage instead of map. This will also make
...
our life during future GraphTraits'ing slightly easier.
llvm-svn: 44952
2007-12-12 19:08:44 +00:00
9ab0943b86
Revert r44626, which turned off the use of readonly
...
and readnone for functions with bodies because it
broke llvm-gcc-4.2 bootstrap. It turns out that,
because of LLVM's array_ref hack, gcc was computing
pure/const attributes wrong (now fixed by turning
off the gcc ipa-pure-const pass).
llvm-svn: 44937
2007-12-12 16:01:40 +00:00
309e5a723b
1. "Upgrage" comments.
...
2. Using zero-extended value of Scale and unsigned division is safe provided
that Scale doesn't have the sign bit set.
Previously these 2 instructions:
%p = bitcast [100 x {i8,i8,i8}]* %x to i8*
%q = getelementptr i8* %p, i32 -4
were combined into:
%q = getelementptr [100 x { i8, i8, i8 }]* %x, i32 0,
i32 1431655764, i32 0
what was incorrect.
llvm-svn: 44936
2007-12-12 15:21:32 +00:00
17c405d29e
Regenerate.
...
llvm-svn: 44934
2007-12-12 08:45:45 +00:00
25f5076612
Implement part of review feedback for address spaces.
...
llvm-svn: 44933
2007-12-12 08:44:39 +00:00
0f42730722
Use shuffles to implement insert_vector_elt for i32, i64, f32, and f64.
...
llvm-svn: 44929
2007-12-12 07:55:34 +00:00
0a1254f634
Add a test case for -optimize-ext-uses.
...
llvm-svn: 44928
2007-12-12 07:54:08 +00:00
2a98956796
Lower a build_vector with all constants into a constpool load unless it can be done with a move to low part.
...
llvm-svn: 44921
2007-12-12 06:45:40 +00:00
3115ba39e8
Add a guard to cxxabi header as other platform may
...
not support it.
llvm-svn: 44914
2007-12-12 06:16:47 +00:00
ed5d693d67
Fixed PR1629.
...
Make lli interpreter correctly call external functions sin()/cos(),
__cxa_guard_acquire() and __cxa_guard_release().
llvm-svn: 44910
2007-12-12 04:55:43 +00:00
b3ba3d9827
Fix a bug in my previous patch, thanks to Jay Foad for
...
pointing this out and correcting the patch!
llvm-svn: 44907
2007-12-12 03:56:54 +00:00
63d33cfd2b
Don't muck with phi nodes; bug fixes.
...
llvm-svn: 44905
2007-12-12 02:53:41 +00:00
4a8bc7e105
Correct typo for Linux: s/esp/%rsp/
...
llvm-svn: 44904
2007-12-12 02:38:28 +00:00
bf85dd4f61
Bit masks conflicted. Needed to bump them by one.
...
llvm-svn: 44903
2007-12-12 01:51:58 +00:00
499e5bffcf
Forgot to remove a register from the PHI-union after I'd determined that it
...
interfered with other registers. Seems like that might be a good thing to do. :-)
llvm-svn: 44902
2007-12-12 01:25:08 +00:00
0a68fe2ffc
Add (very basic) bindings for ModuleProvider.
...
llvm-svn: 44899
2007-12-12 01:04:30 +00:00
7bc8942532
Bug fix. Only safe to perform extension uses optimization if the source of extension is also defined in the same BB as the extension.
...
llvm-svn: 44896
2007-12-12 00:51:06 +00:00
0af43a1895
Changes from Curtis Dunham implementing lazy cycle detection algorithm.
...
Changes from me implementing different way of representing points-to anything.
Changes from me that improve slightly on LCD.
llvm-svn: 44895
2007-12-12 00:37:04 +00:00
6766d2fa4f
If deleting a reload instruction due to reuse (value is available in register R and reload is targeting R), make sure to invalidate the kill information of the last kill.
...
llvm-svn: 44894
2007-12-11 23:36:57 +00:00
38236ef6cb
Need to grow the indexed map. Added debug statements.
...
llvm-svn: 44892
2007-12-11 23:27:51 +00:00
642e15a7cb
Simplify slightly.
...
llvm-svn: 44881
2007-12-11 22:22:22 +00:00
7f01ebefc6
Remove Trie::Edge class. Now edge labels are stored into nodes itself.
...
llvm-svn: 44880
2007-12-11 21:55:38 +00:00
f24dd1c1eb
More progress on StrongPHIElimination. Now we actually USE the DomForest!
...
llvm-svn: 44877
2007-12-11 20:12:11 +00:00
b678ae7c38
Blark! How in the world did this work without this?!
...
llvm-svn: 44874
2007-12-11 19:40:06 +00:00
7717a8a37d
- Update the virtual reg to machine instruction map when hoisting.
...
- Fix subtle bug when creating initially creating this map.
llvm-svn: 44873
2007-12-11 19:17:04 +00:00
5143d898c8
Checking for "zero operands" during the "CanHoistInst()" method isn't necessary
...
because those with side effects will be caught by other checks in here.
Also, simplify the check for a BB in a sub loop.
llvm-svn: 44871
2007-12-11 18:45:11 +00:00
6dc8b4ed13
Allow the JIT to encode MMX instructions
...
llvm-svn: 44869
2007-12-11 18:06:14 +00:00
5bf9f01349
Use correct member access operator.
...
(my test commit as well)
llvm-svn: 44868
2007-12-11 17:46:25 +00:00
5e1f8472d2
Rename these tests to use the appropriate suffixes.
...
llvm-svn: 44867
2007-12-11 15:55:52 +00:00
b582867567
Use not instead of ignore when an exit status is expected to always
...
be non-zero.
llvm-svn: 44866
2007-12-11 15:50:23 +00:00
cfed707c12
Don't redirect stderr when it isn't needed.
...
llvm-svn: 44865
2007-12-11 15:41:11 +00:00
f41217d4fa
Fix compilation.
...
llvm-svn: 44864
2007-12-11 12:20:47 +00:00
595798f041
Update credits.
...
llvm-svn: 44861
2007-12-11 09:32:07 +00:00
308121c757
Add information on address space qualifiers for pointer types and global
...
declarations to the LangRef.
llvm-svn: 44860
2007-12-11 09:31:00 +00:00
e92cba4535
Regenerate.
...
llvm-svn: 44859
2007-12-11 09:02:08 +00:00
54dd24c2a7
Implement address space attribute for LLVM pointer types. Address spaces are
...
regions of memory that have a target specific relationship, as described in the
Embedded C Technical Report.
This also implements the 2007-12-11-AddressSpaces test,
which demonstrates how address space attributes can be used in LLVM IR.
In addition, this patch changes the bitcode signature for stores (in a backwards
compatible manner), such that the pointer type, rather than the pointee type, is
encoded. This permits type information in the pointer (e.g. address space) to be
preserved for stores.
LangRef updates are forthcoming.
llvm-svn: 44858
2007-12-11 08:59:05 +00:00
8e6a8f984c
Teach VMCore to constant fold shufflevectors with constant operands.
...
This allows us to compile:
#include <emmintrin.h>
typedef __m128i VSInt16;
typedef short vSInt16 __attribute__ ((__vector_size__ (16)));
VSInt16 t3() {
return (VSInt16)((vSInt16)_mm_set1_epi16(6518));
}
into:
_t3:
movaps LCPI1_0, %xmm0
ret
instead of:
_t3:
movl $6518, %eax
movd %eax, %xmm0
pextrw $0, %xmm0, %eax
xorps %xmm0, %xmm0
pinsrw $0, %eax, %xmm0
punpcklwd %xmm0, %xmm0
pshufd $0, %xmm0, %xmm0
ret
llvm-svn: 44856
2007-12-11 07:49:37 +00:00
6a6b3fb62b
Implement constant folding if vector<->vector bitcasts where the number
...
of source/dest elements changes. This implements
test/Transforms/InstCombine/bitcast-vector-fold.ll
llvm-svn: 44855
2007-12-11 07:29:44 +00:00
0b1df1d3da
Don't have APInt.cpp depend upon DerivedTypes.h. This helps with splitting the
...
Support libraries separately into their own module.
llvm-svn: 44852
2007-12-11 06:53:58 +00:00
3d23737c5f
Add first and really dirty version of generic Trie structure
...
llvm-svn: 44851
2007-12-11 06:53:44 +00:00
5c6399eee6
significantly simplify some code, no functionality change.
...
llvm-svn: 44850
2007-12-11 06:07:39 +00:00
e8ea0374ac
refactor some code, no functionality change.
...
llvm-svn: 44849
2007-12-11 05:55:02 +00:00
303417d242
Switch over to MachineLoopInfo.
...
llvm-svn: 44838
2007-12-11 02:09:15 +00:00
f54030231e
Pretty print shuffle mask operand.
...
llvm-svn: 44837
2007-12-11 02:08:35 +00:00
4fbf459549
- Improved v8i16 shuffle lowering. It now uses pshuflw and pshufhw as much as
...
possible before resorting to pextrw and pinsrw.
- Better codegen for v4i32 shuffles masquerading as v8i16 or v16i8 shuffles.
- Improves (i16 extract_vector_element 0) codegen by recognizing
(i32 extract_vector_element 0) does not require a pextrw.
llvm-svn: 44836
2007-12-11 01:46:18 +00:00
a55a67ae91
x86 doesn't actually want to custom lower v3i32
...
llvm-svn: 44835
2007-12-11 01:41:33 +00:00
7843c16f31
CollectorMetadata and Collector are rejiggered to get along with
...
per-function collector model. Collector is now the factory for
CollectorMetadata, so the latter may be subclassed.
llvm-svn: 44827
2007-12-11 00:30:17 +00:00
Christopher Lamb