7414e21ec0
Update a comment to reflect the current code.
...
llvm-svn: 43463
2007-10-29 19:32:39 +00:00
f5feb01056
Remove an unused function argument.
...
llvm-svn: 43462
2007-10-29 19:31:25 +00:00
50d42224d0
Fix a typo in a comment.
...
llvm-svn: 43461
2007-10-29 19:26:14 +00:00
8e8adada83
Avoid calling ValidStride when not all uses are addresses.
...
llvm-svn: 43460
2007-10-29 19:23:53 +00:00
ef23e34bba
Fixed warning concerning implicit conversion from a NULL pointer
...
constant to an unsigned int. We now just directly assign the literal 0.
llvm-svn: 43459
2007-10-29 18:43:39 +00:00
7b3f7feaea
Avoid doing something dumb like rewriting using a 64-bit iv in 32-bit mode.
...
llvm-svn: 43446
2007-10-29 07:57:50 +00:00
909a54ccd4
add a note.
...
llvm-svn: 43444
2007-10-29 06:19:48 +00:00
b6fc40e60f
I am not sure this is a good idea to be an option, but rename the option for
...
now. It conflicts with clang's -pedantic flag.
llvm-svn: 43431
2007-10-29 03:14:55 +00:00
5e99fd8c0d
Add support for the x86-64 'q' regigster modifier, and add support for the
...
b/h/w/k/q inline asm memory modifiers, which are just ignored. This fixes
PR1748 and CodeGen/X86/2007-10-28-inlineasm-q-modifier.ll
llvm-svn: 43430
2007-10-29 03:09:07 +00:00
9a641510bd
Fix PR1749 and InstCombine/2007-10-28-EmptyField.ll by handling
...
zero-length fields better.
llvm-svn: 43427
2007-10-29 02:40:02 +00:00
4a15e04aee
Fix PR1752 and LoopSimplify/2007-10-28-InvokeCrash.ll: terminators
...
can have uses too. Wouldn't it be nice if invoke didn't exist? :)
llvm-svn: 43426
2007-10-29 02:30:37 +00:00
8be9bc64ce
Fixed assertion in Deserializer::~Deserializer that checks for
...
pointers that were not backpatched (previously checked the wrong invariant).
llvm-svn: 43425
2007-10-28 23:38:38 +00:00
9dced3f9ae
Add 'pedantic' mode to verifier rejecting syntactically valid, but 'bad' due to other reasons code
...
llvm-svn: 43424
2007-10-28 22:50:32 +00:00
387204749b
Updated backpatching logic during object deserialization to perform
...
eager backpatching instead of waithing until all objects have been
deserialized. This allows us to reduce the memory footprint needed
for backpatching.
llvm-svn: 43422
2007-10-28 21:17:59 +00:00
1826deda68
The guaranteed alignment of ptr+offset is only the minimum of
...
of offset and the alignment of ptr if these are both powers of
2. While the ptr alignment is guaranteed to be a power of 2,
there is no reason to think that offset is. For example, if
offset is 12 (the size of a long double on x86-32 linux) and
the alignment of ptr is 8, then the alignment of ptr+offset
will in general be 4, not 8. Introduce a function MinAlign,
lifted from gcc, for computing the minimum guaranteed alignment.
I've tried to fix up everywhere under lib/CodeGen/SelectionDAG/.
I also changed some places that weren't wrong (because both values
were a power of 2), as a defensive change against people copying
and pasting the code.
Hopefully someone who cares about alignment will review the rest
of LLVM and fix up the remaining places. Since I'm on x86 I'm
not very motivated to do this myself...
llvm-svn: 43421
2007-10-28 12:59:45 +00:00
c826ac533b
New entry.
...
llvm-svn: 43420
2007-10-28 04:01:09 +00:00
9dbe99dcd6
A number of LSR fixes:
...
- ChangeCompareStride only reuse stride that is larger than current stride. It
will let the general reuse mechanism to try to reuse a smaller stride.
- Watch out for multiplication overflow in ChangeCompareStride.
- Replace std::set with SmallPtrSet.
llvm-svn: 43408
2007-10-26 23:08:19 +00:00
2d95094bea
Added serialization support for APInt.
...
llvm-svn: 43405
2007-10-26 21:50:10 +00:00
6d15b32c15
- Remove the hacky code that forces a memcpy. Alignment is taken care of in the
...
FE.
- Explicitly pass in the alignment of the load & store.
- XFAIL 2007-10-23-UnalignedMemcpy.ll because llc has a bug that crashes on
unaligned pointers.
llvm-svn: 43398
2007-10-26 20:24:42 +00:00
d78a3e5555
Fix a crash. Make sure TLI is not null.
...
llvm-svn: 43384
2007-10-26 17:24:46 +00:00
d07d6a411c
Fix off-by-one stack offset computations (dwarf information) for callee-saved
...
registers in case, when FP pointer was eliminated. This should fixes misc. random
EH-related crahses, when stuff is compiled with -fomit-frame-pointer.
Thanks Duncan for nailing this bug!
llvm-svn: 43381
2007-10-26 09:13:24 +00:00
18063916b5
clo/clz aren't supported on mips I. Keep them around for when we'll
...
want them later (mips32/64).
llvm-svn: 43380
2007-10-26 04:00:13 +00:00
b7971bbdd7
Make a comment better.
...
llvm-svn: 43379
2007-10-26 03:47:14 +00:00
78c63ac41e
More fleshing out of docs/Passes.html, plus some typo fixes and
...
improved wording in source files.
llvm-svn: 43377
2007-10-26 03:03:51 +00:00
7f3d02471d
Loosen up iv reuse to allow reuse of the same stride but a larger type when truncating from the larger type to smaller type is free.
...
e.g.
Turns this loop:
LBB1_1: # entry.bb_crit_edge
xorl %ecx, %ecx
xorw %dx, %dx
movw %dx, %si
LBB1_2: # bb
movl L_X$non_lazy_ptr, %edi
movw %si, (%edi)
movl L_Y$non_lazy_ptr, %edi
movw %dx, (%edi)
addw $4, %dx
incw %si
incl %ecx
cmpl %eax, %ecx
jne LBB1_2 # bb
into
LBB1_1: # entry.bb_crit_edge
xorl %ecx, %ecx
xorw %dx, %dx
LBB1_2: # bb
movl L_X$non_lazy_ptr, %esi
movw %cx, (%esi)
movl L_Y$non_lazy_ptr, %esi
movw %dx, (%esi)
addw $4, %dx
incl %ecx
cmpl %eax, %ecx
jne LBB1_2 # bb
llvm-svn: 43375
2007-10-26 01:56:11 +00:00
1207cd6b1b
Silenced a VC++ warning.
...
llvm-svn: 43374
2007-10-25 23:49:14 +00:00
542e5f29b7
Updated backpatching during object deserialization to support "smart"
...
pointers that employ unused bits in a pointer to store extra data.
llvm-svn: 43373
2007-10-25 23:40:35 +00:00
fc69d322f2
Clarified operator precedence.
...
Silenced VC++ warning.
llvm-svn: 43372
2007-10-25 23:15:31 +00:00
7e16ce5270
Disambiguated variable name to comply with VC++'s archaic variable scoping rules.
...
llvm-svn: 43369
2007-10-25 22:57:48 +00:00
29e29e63bd
Do not rewrite compare instruction using iv of a different stride if the new
...
stride may be rewritten using the stride of the compare instruction.
llvm-svn: 43367
2007-10-25 22:45:20 +00:00
65e804a9c3
Support non-POSIX hosts by removing use of strncasecmp.
...
llvm-svn: 43364
2007-10-25 21:54:43 +00:00
f73340efb9
Changed XXX to FIXME, and added comment to the README file
...
llvm-svn: 43359
2007-10-25 19:49:32 +00:00
a83e6bc246
Added special treatment of serializing NULL pointers.
...
llvm-svn: 43357
2007-10-25 18:42:52 +00:00
5a38108374
Remove code that's commented out.
...
llvm-svn: 43356
2007-10-25 18:38:24 +00:00
5f7ed00d44
Added comment explaining why we are doing this check.
...
llvm-svn: 43353
2007-10-25 18:23:45 +00:00
d385f0759c
Small formatting changes. Add a sanity check.
...
Use NVT rather than looking it up, since we have
it to hand.
llvm-svn: 43341
2007-10-25 12:35:51 +00:00
a8f4ba6eb9
Promote SETCC operands.
...
llvm-svn: 43340
2007-10-25 12:32:31 +00:00
cf0da03312
Correctly extract the ValueType from a VTSDNode.
...
llvm-svn: 43339
2007-10-25 12:30:51 +00:00
133694db06
If a loop termination compare instruction is the only use of its stride,
...
and the compaison is against a constant value, try eliminate the stride
by moving the compare instruction to another stride and change its
constant operand accordingly. e.g.
loop:
...
v1 = v1 + 3
v2 = v2 + 1
if (v2 < 10) goto loop
=>
loop:
...
v1 = v1 + 3
if (v1 < 30) goto loop
llvm-svn: 43336
2007-10-25 09:11:16 +00:00
7827a3f366
Fix for PR1741.
...
llvm-svn: 43326
2007-10-25 02:36:18 +00:00
d5cb7b055d
Implemented prototype serialization of pointers, including support
...
for backpatching.
Added Deserialize::ReadVal.
llvm-svn: 43319
2007-10-25 00:10:21 +00:00
a4a972e32d
Another expansion for i64 multiply, suitable for PPC.
...
llvm-svn: 43314
2007-10-24 22:26:08 +00:00
4d06391c47
Fix off by 1 bug in printf->puts lowering.
...
llvm-svn: 43309
2007-10-24 20:14:50 +00:00
83610ae6f4
Split Serialization.h into separate headers: Serialize.h and
...
Deserialize.h Serialization.h now includes trait speciailizations for
unsigned long, etc.
llvm-svn: 43307
2007-10-24 19:06:40 +00:00
55b8302dfe
simplify some code by using the new isNaN predicate
...
llvm-svn: 43305
2007-10-24 18:54:45 +00:00
c62877e9da
Implement a couple of foldings for ordered and unordered comparisons,
...
implementing cases related to PR1738.
llvm-svn: 43289
2007-10-24 05:38:08 +00:00
38ccabcae9
Fix comment and use the "Size" variable that's already provided.
...
llvm-svn: 43271
2007-10-23 23:36:57 +00:00
e3b859298a
If there's an unaligned memcpy to/from the stack, don't lower it. Just call the
...
memcpy library function instead.
llvm-svn: 43270
2007-10-23 23:32:40 +00:00
10f4152471
Disable a couple more things for ppcf128.
...
llvm-svn: 43267
2007-10-23 23:20:14 +00:00
6f149c0571
This broke lots. Reverting.
...
llvm-svn: 43264
2007-10-23 22:04:26 +00:00
8971440e56
Lowering a memcpy to the stack is killing PPC. The ARM and X86 backends already
...
have their own custom memcpy lowering code. This code needs to be factored out
into a target-independent lowering method with hooks to the backend. In the
meantime, just call memcpy if we're trying to copy onto a stack.
llvm-svn: 43262
2007-10-23 21:30:25 +00:00
bd3501887f
Added preliminary implementation of generic object serialization to bitcode.
...
llvm-svn: 43261
2007-10-23 21:29:33 +00:00
9c614117da
Make DomTree and PostDomTree thin wrappers around DomTreeBase, rather than inheriting from it.
...
llvm-svn: 43259
2007-10-23 20:58:37 +00:00
5d7032bb08
It's possible to commute instrctions with more than 3 operands.
...
llvm-svn: 43256
2007-10-23 20:14:40 +00:00
847d42a85c
isSubRegOf() is a dup of isSubRegister.
...
llvm-svn: 43249
2007-10-23 06:51:50 +00:00
ec271b104c
Temporary solution: added a different set of BCTRL_Macho / BCTRL_ELF with right callee-saved defs set for ppc64.
...
llvm-svn: 43248
2007-10-23 06:42:42 +00:00
1f2dd35898
Fix memcpy lowering when addresses are 4-byte aligned but size is not multiple of 4.
...
llvm-svn: 43234
2007-10-22 22:11:27 +00:00
d09f1c40a2
The #include <iterator> isn't needed in this header.
...
llvm-svn: 43232
2007-10-22 20:44:10 +00:00
e0c3d9f338
Strength reduction improvements.
...
- Avoid attempting stride-reuse in the case that there are users that
aren't addresses. In that case, there will be places where the
multiplications won't be folded away, so it's better to try to
strength-reduce them.
- Several SSE intrinsics have operands that strength-reduction can
treat as addresses. The previous item makes this more visible, as
any non-address use of an IV can inhibit stride-reuse.
- Make ValidStride aware of whether there's likely to be a base
register in the address computation. This prevents it from thinking
that things like stride 9 are valid on x86 when the base register is
already occupied.
Also, XFAIL the 2007-08-10-LEA16Use32.ll test; the new logic to avoid
stride-reuse elimintes the LEA in the loop, so the test is no longer
testing what it was intended to test.
llvm-svn: 43231
2007-10-22 20:40:42 +00:00
bf474959a3
Fix the folding of multiplication into addresses on x86, which was broken
...
by the recent {U,S}MUL_LOHI changes.
llvm-svn: 43230
2007-10-22 20:22:24 +00:00
bdbed66333
Use ptr type in the immediate field of a BxA instruction so we don't end up selecting 32-bit call instruction for ppc64.
...
llvm-svn: 43228
2007-10-22 19:46:19 +00:00
5163a8f53e
Add missing paratheses.
...
llvm-svn: 43227
2007-10-22 19:42:28 +00:00
941db4da0a
Support for expanding extending loads of integers with
...
funky bit-widths.
llvm-svn: 43225
2007-10-22 19:00:05 +00:00
a37eaf2bf9
Move the SCEV object factors from being static members of the individual
...
SCEV subclasses to being non-static member functions of the ScalarEvolution
class.
llvm-svn: 43224
2007-10-22 18:31:58 +00:00
8fc995069b
Fix up the logic for result expanding the various extension
...
operations so they work right for integers with funky
bit-widths. For example, consider extending i48 to i64
on a 32 bit machine. The i64 result is expanded to 2 x i32.
We know that the i48 operand will be promoted to i64, then
also expanded to 2 x i32. If we had the expanded promoted
operand to hand, then expanding the result would be trivial.
Unfortunately at this stage we can only get hold of the
promoted operand. So instead we kind of hand-expand, doing
explicit shifting and truncating to get the top and bottom
halves of the i64 operand into 2 x i32, which are then used
to expand the result. This is harmless, because when the
promoted operand is finally expanded all this bit fiddling
turns into trivial operations which are eliminated either
by the expansion code itself or the DAG combiner.
llvm-svn: 43223
2007-10-22 18:26:21 +00:00
c92446af1f
Fix an unfolding bug.
...
llvm-svn: 43212
2007-10-22 03:03:20 +00:00
8557603781
- Only perform the unfolding optimization when the folding in question is modref.
...
- Remove a bogus assertion.
llvm-svn: 43211
2007-10-22 03:01:44 +00:00
fd6f3257b8
add a mechanism for the JIT to invoke a function to lazily create functions as they are referenced.
...
llvm-svn: 43210
2007-10-22 02:50:12 +00:00
bf5e958ba0
llvm-gcc3 is dead, along with it __main.
...
llvm-svn: 43209
2007-10-22 02:39:47 +00:00
7499a3b092
Reg2Mem cleanup and optimizations:
...
- enable phi instructions demotion to stack
- create alloca instructions in the entry block
llvm-svn: 43208
2007-10-21 23:05:16 +00:00
edaf0b4651
LoadLibraryPermanently doesn't throw.
...
llvm-svn: 43207
2007-10-21 22:58:11 +00:00
b5163bb9f0
Add a convenience method for creating EE's.
...
llvm-svn: 43206
2007-10-21 22:57:11 +00:00
8ee70112ea
Allow for copysign having f80 second argument.
...
Fixes 5550319.
llvm-svn: 43205
2007-10-21 01:07:44 +00:00
36f06c80e6
Add promote operand support for [su]int_to_fp.
...
llvm-svn: 43204
2007-10-20 22:57:56 +00:00
2ba4b148f3
Add result promotion of FP_TO_*INT, fixing CodeGen/X86/trunc-to-bool.ll
...
with the new legalizer.
llvm-svn: 43199
2007-10-20 04:32:38 +00:00
1c87f0c620
simplify some code.
...
llvm-svn: 43198
2007-10-20 04:09:48 +00:00
2bcac640b7
Implement promote and expand for operands of memcpy and friends.
...
This fixes CodeGen/X86/mem*.ll.
llvm-svn: 43197
2007-10-20 04:07:07 +00:00
f12967124c
Added missing curly braces which renders the if clause useless in debug build.
...
llvm-svn: 43196
2007-10-20 04:01:47 +00:00
771188cf60
Fix a few places vector operations were not getting
...
the operand's type from the right place.
llvm-svn: 43195
2007-10-20 00:07:52 +00:00
45e096c77e
Resolve unfold tables ambiguity.
...
llvm-svn: 43194
2007-10-19 23:50:58 +00:00
35ff79370b
Local spiller optimization:
...
Turn a store folding instruction into a load folding instruction. e.g.
xorl %edi, %eax
movl %eax, -32(%ebp)
movl -36(%ebp), %eax
orl %eax, -32(%ebp)
=>
xorl %edi, %eax
orl -36(%ebp), %eax
mov %eax, -32(%ebp)
This enables the unfolding optimization for a subsequent instruction which will
also eliminate the newly introduced store instruction.
llvm-svn: 43192
2007-10-19 21:23:22 +00:00
ac5c93040f
Don't branch fold inline asm statements.
...
llvm-svn: 43191
2007-10-19 21:09:55 +00:00
a87c9e4b75
Add support for a few more nodes.
...
llvm-svn: 43190
2007-10-19 20:29:48 +00:00
6802d0c96f
Redo "last ppc long double fix" as Chris wants.
...
llvm-svn: 43189
2007-10-19 20:29:00 +00:00
064c31ebac
Fix a really nasty vector miscompilation bill recently introduced.
...
llvm-svn: 43181
2007-10-19 16:47:35 +00:00
3ea519e56d
rename ExpandOperation to ExpandOperationResult, as suggested
...
by Duncan
llvm-svn: 43177
2007-10-19 15:28:47 +00:00
18a831d783
split LowerMEMCPY into LowerMEMCPYCall and LowerMEMCPYInline in the ARM backend.
...
llvm-svn: 43176
2007-10-19 14:35:17 +00:00
a9953e4d0a
Support for expanding ADDE and SUBE.
...
llvm-svn: 43175
2007-10-19 13:06:17 +00:00
d9834b29dd
If the value types are equal then this routine
...
asserts in later checks rather than producing
the ordinary load it is supposed to. Avoid all
such hassles by directly returning an ordinary
load in this case.
llvm-svn: 43174
2007-10-19 13:05:40 +00:00
846c19dd70
Add support for byval function whose argument is not 32 bit aligned.
...
To do this it is necessary to add a "always inline" argument to the
memcpy node. For completeness I have also added this node to memmove
and memset. I have also added getMem* functions, because the extra
argument makes it cumbersome to use getNode and because I get confused
by it :-)
llvm-svn: 43172
2007-10-19 10:41:11 +00:00
e5a6448533
Implement a few new operations.
...
llvm-svn: 43171
2007-10-19 04:46:45 +00:00
e31365eecc
Implement expansion of SINT_TO_FP and UINT_TO_FP operands.
...
llvm-svn: 43170
2007-10-19 04:32:47 +00:00
9081d08083
implement support for custom expansion of any node type, in one place.
...
llvm-svn: 43169
2007-10-19 04:14:36 +00:00
b193576bc6
comment fixes
...
llvm-svn: 43168
2007-10-19 04:08:28 +00:00
d01b8ea4a5
Make use of TLI.ExpandOperation, remove softfloat stuff.
...
llvm-svn: 43167
2007-10-19 03:58:25 +00:00
3c7ee41c78
add expand support for bit_convert result, even allowing custom expansion.
...
llvm-svn: 43166
2007-10-19 03:33:14 +00:00
579db81f1c
add a new target hook.
...
llvm-svn: 43165
2007-10-19 03:31:45 +00:00
5d979d57ae
Add an easy microoptimization I noticed.
...
llvm-svn: 43164
2007-10-19 03:29:26 +00:00
de16ad1446
Negative indices aren't allowed here.
...
llvm-svn: 43161
2007-10-19 01:10:49 +00:00
10432e5a67
More ppcf128 issues (maybe the last)?
...
llvm-svn: 43160
2007-10-19 00:59:18 +00:00
Dan Gohman