Gordon Henriksen [Wed, 22 Oct 2008 12:40:55 +0000 (12:40 +0000)]
Regenerate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57975
91177308-0d34-0410-b5e6-
96231b3b80d8
Gordon Henriksen [Wed, 22 Oct 2008 12:40:40 +0000 (12:40 +0000)]
[PR2886] Don't look for ocaml's .opt executables; something in the autoconf/test machinery doesn't handle the period properly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57974
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 11:49:09 +0000 (11:49 +0000)]
LegalizeTypes soft-float support for fpow.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57973
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 11:24:12 +0000 (11:24 +0000)]
Get this working with LegalizeTypes: (1) don't
assume that i64 has been turned into a BUILD_PAIR
node (when called from LegalizeTypes this hasn't
happened yet) and don't use a vector shuffle mask
with an illegal element type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57972
91177308-0d34-0410-b5e6-
96231b3b80d8
Torok Edwin [Wed, 22 Oct 2008 09:56:27 +0000 (09:56 +0000)]
Regenerate configure
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57971
91177308-0d34-0410-b5e6-
96231b3b80d8
Argyrios Kyrtzidis [Wed, 22 Oct 2008 09:54:13 +0000 (09:54 +0000)]
Unbreak LLVM on the MSVC compiler:
-Bring in int64_t for TableGen/Record.h and TableGen/TGLexer.h
-Define strtoull
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57970
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 09:42:14 +0000 (09:42 +0000)]
Use spaces rather than tab.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57969
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 09:23:20 +0000 (09:23 +0000)]
Be nice to CellSPU: for this target getSetCCResultType
may return i8, which can result in SELECT nodes for
which the type of the condition is i8, but there are
no patterns for select with i8 condition. Tweak the
LegalizeTypes logic to avoid this as much as possible.
This isn't a real fix because it is still perfectly
possible to end up with such select nodes - CellSPU
needs to be fixed IMHO.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57968
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 09:07:29 +0000 (09:07 +0000)]
Port from LegalizeDAG the logic to only generate
ADDC/ADDE/SUBC/SUBE if the target supports it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57967
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 09:06:24 +0000 (09:06 +0000)]
Add some comments explaining the meaning of a boolean
that is not of type MVT::i1 in SELECT and SETCC nodes.
Relax the LegalizeTypes SELECT condition promotion
sanity checks to allow other condition types than i1.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57966
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Wed, 22 Oct 2008 09:00:33 +0000 (09:00 +0000)]
Temporarily allow the operands of a BUILD_VECTOR
to have a different type to the vector element
type. This should be fairly harmless because in
the past guys like this were being built all over
the place (and were cleaned up when I added this
check). The reason for relaxing this check is
that it helps LegalizeTypes legalize vector
shuffles: the mask is a BUILD_VECTOR that it is
*not always possible* to legalize while keeping it
a BUILD_VECTOR (vector_shuffle requires the mask
to be a BUILD_VECTOR, as opposed to a vector with
the right vector type). With this check it is even
harder to legalize the mask - turning the check off
means that LegalizeTypes manages to legalize almost
all vector shuffles encountered in practice. The
correct solution is to change vector_shuffle to be a
variadic node with the mask built into it as operands.
While waiting for that change, this hack stops the
problem with vector_shuffle from blocking the turning
on of LegalizeTypes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57965
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Wed, 22 Oct 2008 04:53:16 +0000 (04:53 +0000)]
Fix PR2907 by digging through constant expressions to find FP constants that
are their operands.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57956
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Wed, 22 Oct 2008 03:25:22 +0000 (03:25 +0000)]
Move Print*Pass to use raw_ostream.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57946
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Wed, 22 Oct 2008 02:56:07 +0000 (02:56 +0000)]
CMake: `make install' for libraries, executables and header files.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57945
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Wed, 22 Oct 2008 02:52:59 +0000 (02:52 +0000)]
CMake: use add_llvm_example for HowToUseJIT.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57944
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Wed, 22 Oct 2008 02:51:53 +0000 (02:51 +0000)]
CMake: Turned some libraries into partially linked objects. Corrected
names of LLVMCore and ARMCodeGen.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57943
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Wed, 22 Oct 2008 00:02:32 +0000 (00:02 +0000)]
Adjust comments for pedantic satisfaction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57940
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Tue, 21 Oct 2008 23:57:52 +0000 (23:57 +0000)]
Fix comment to name "TokenFactor" instead of "Token factor".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57939
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Tue, 21 Oct 2008 23:52:03 +0000 (23:52 +0000)]
CMake: updated lib/VMCore/CMakeLists.txt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57937
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Tue, 21 Oct 2008 23:33:38 +0000 (23:33 +0000)]
Privatize PrintModulePass and PrintFunctionPass and add
createPrintModulePass and createPrintFunctionPass.
- So clients who compile w/o RTTI can use them.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57933
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Tue, 21 Oct 2008 23:07:49 +0000 (23:07 +0000)]
Add comments to explain uint64->f64 algorithm,
well, sort of. (Algorithm by Ian Ollmann.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57932
91177308-0d34-0410-b5e6-
96231b3b80d8
Dale Johannesen [Tue, 21 Oct 2008 20:50:01 +0000 (20:50 +0000)]
Add an SSE2 algorithm for uint64->f64 conversion.
The same one Apple gcc uses, faster. Also gets the
extreme case in gcc.c-torture/execute/ieee/rbug.c
correct which we weren't before; this is not
sufficient to get the test to pass though, there
is another bug.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57926
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 20:00:42 +0000 (20:00 +0000)]
Fix SelectionDAGBuild lowering of Select instructions to
handle first-class aggregate values. Also, fix a bug in
the Ret handling for empty aggregates.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57925
91177308-0d34-0410-b5e6-
96231b3b80d8
Daniel Dunbar [Tue, 21 Oct 2008 19:53:10 +0000 (19:53 +0000)]
Clear raw_fd_ostream error string on success and explain behavior in
documentation.
Add C++ header marker.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57923
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 19:50:00 +0000 (19:50 +0000)]
Updated email address.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57920
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 18:24:51 +0000 (18:24 +0000)]
Implement the optimized FCMP_OEQ/FCMP_UNE code for x86 fast-isel.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57915
91177308-0d34-0410-b5e6-
96231b3b80d8
Torok Edwin [Tue, 21 Oct 2008 17:21:32 +0000 (17:21 +0000)]
Fix make check on Solaris 10/x86: the default grep is not GNU grep, same for as.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57912
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Tue, 21 Oct 2008 16:54:12 +0000 (16:54 +0000)]
use pre-UAL mnemonics for push/pop for compilaton callback function
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57911
91177308-0d34-0410-b5e6-
96231b3b80d8
Nuno Lopes [Tue, 21 Oct 2008 11:42:16 +0000 (11:42 +0000)]
fix a tricky bug in the JIT global variable emitter, that was triggered when JITing a variable independently of a function. This lead to sharing memory memory between functions and GVs thus changing the value of a GV could change the code in execution. more details on the ML.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57900
91177308-0d34-0410-b5e6-
96231b3b80d8
Ted Kremenek [Tue, 21 Oct 2008 05:59:33 +0000 (05:59 +0000)]
constify some methods and variables in ImmutableList.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57894
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 21 Oct 2008 03:55:19 +0000 (03:55 +0000)]
really fix run line
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57889
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 21 Oct 2008 03:54:49 +0000 (03:54 +0000)]
fix run line
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57888
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 21 Oct 2008 03:49:19 +0000 (03:49 +0000)]
remove some unneeded eh generation
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57887
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:41:46 +0000 (03:41 +0000)]
Disable constant-offset folding for PowerPC, as the PowerPC target
isn't yet prepared for it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57886
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:38:42 +0000 (03:38 +0000)]
Don't create TargetGlobalAddress nodes with offsets that don't fit
in the 32-bit signed offset field of addresses. Even though this
may be intended, some linkers refuse to relocate code where the
relocated address computation overflows.
Also, fix the sign-extension of constant offsets to use the
actual pointer size, rather than the size of the GlobalAddress
node, which may be different, for example on x86-64 where MVT::i32
is used when the address is being fit into the 32-bit displacement
field.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57885
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:29:32 +0000 (03:29 +0000)]
Optimized FCMP_OEQ and FCMP_UNE for x86.
Where previously LLVM might emit code like this:
ucomisd %xmm1, %xmm0
setne %al
setp %cl
orb %al, %cl
jne .LBB4_2
it now emits this:
ucomisd %xmm1, %xmm0
jne .LBB4_2
jp .LBB4_2
It has fewer instructions and uses fewer registers, but it does
have more branches. And in the case that this code is followed by
a non-fallthrough edge, it may be followed by a jmp instruction,
resulting in three branch instructions in sequence. Some effort
is made to avoid this situation.
To achieve this, X86ISelLowering.cpp now recognizes FCMP_OEQ and
FCMP_UNE in lowered form, and replace them with code that emits
two branches, except in the case where it would require converting
a fall-through edge to an explicit branch.
Also, X86InstrInfo.cpp's branch analysis and transform code now
knows now to handle blocks with multiple conditional branches. It
uses loops instead of having fixed checks for up to two
instructions. It can now analyze and transform code generated
from FCMP_OEQ and FCMP_UNE.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57873
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:24:31 +0000 (03:24 +0000)]
When the coalescer is doing rematerializing, have it remove
the copy instruction from the instruction list before asking the
target to create the new instruction. This gets the old instruction
out of the way so that it doesn't interfere with the target's
rematerialization code. In the case of x86, this helps it find
more cases where EFLAGS is not live.
Also, in the X86InstrInfo.cpp, teach isSafeToClobberEFLAGS to check
to see if it reached the end of the block after scanning each
instruction, instead of just before. This lets it notice when the
end of the block is only two instructions away, without doing any
additional scanning.
These changes allow rematerialization to clobber EFLAGS in more
cases, for example using xor instead of mov to set the return value
to zero in the included testcase.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57872
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:12:54 +0000 (03:12 +0000)]
Make the NaN test come second, heuristically assuming
that NaNs are less common.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57871
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 03:10:28 +0000 (03:10 +0000)]
Use Function::getEntryBlock() instead of Function::front(), for clarity.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57870
91177308-0d34-0410-b5e6-
96231b3b80d8
Oscar Fuentes [Tue, 21 Oct 2008 02:37:50 +0000 (02:37 +0000)]
CMake: updated lib/CodeGen/CMakeLists.txt
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57869
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Tue, 21 Oct 2008 01:08:07 +0000 (01:08 +0000)]
Fix a bug that prevented llvm-extract -delete from working.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57864
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Tue, 21 Oct 2008 00:45:36 +0000 (00:45 +0000)]
Fix gcc.c-torture/compile/920520-1.c by inserting bitconverts
for strange asm conditions earlier. In this case, we have a
double being passed in an integer reg class. Convert to like
sized integer register so that we allocate the right number
for the class (two i32's for the f64 in this case).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57862
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 20 Oct 2008 21:44:59 +0000 (21:44 +0000)]
Add skeleton for the pre-register allocation live interval splitting pass.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57847
91177308-0d34-0410-b5e6-
96231b3b80d8
Jim Grosbach [Mon, 20 Oct 2008 21:39:23 +0000 (21:39 +0000)]
Update the stub and callback code to handle lazy compilation. The stub
is re-written by the callback to branch directly to the compiled code
in future invocations.
Added back in range-based memory permission functions for the updating of
the stub on Darwin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57846
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Mon, 20 Oct 2008 21:30:12 +0000 (21:30 +0000)]
Fast-isel no longer an experiment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57845
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 20 Oct 2008 20:03:28 +0000 (20:03 +0000)]
Add a register class -> virtual registers map.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57844
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Mon, 20 Oct 2008 20:02:17 +0000 (20:02 +0000)]
This forward declaration is unnecessary.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57843
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:31:21 +0000 (16:31 +0000)]
Support operations like fp_to_uint with a vector
result type when the result type is legal but
not the operand type. Add additional support
for EXTRACT_SUBVECTOR and CONCAT_VECTORS,
needed to handle such cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57840
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:24:25 +0000 (16:24 +0000)]
Teach getTypeToTransformTo to return something
sensible for vectors being scalarized. Note
that this method can't return anything very
sensible when splitting non-power-of-two vectors.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57839
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:17:42 +0000 (16:17 +0000)]
LegalizeTypes support for atomic operation promotion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57838
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:14:43 +0000 (16:14 +0000)]
Use DAG.getIntPtrConstant rather than DAG.getConstant
with TLI.getPointerTy for a small simplification.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57837
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:13:04 +0000 (16:13 +0000)]
Always use either MVT::i1 or getSetCCResultType for
the condition of a SELECT node. Make sure that the
correct extension type (any-, sign- or zero-extend)
is used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57836
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:06:47 +0000 (16:06 +0000)]
Formatting - no functional change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57834
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 16:04:57 +0000 (16:04 +0000)]
Don't use a random type for the select condition,
use an MVT::i1 and simplify the code while there.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57833
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Mon, 20 Oct 2008 15:58:02 +0000 (15:58 +0000)]
Fix a typo in a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57832
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Mon, 20 Oct 2008 15:56:33 +0000 (15:56 +0000)]
Have X86 custom lowering for LegalizeTypes use
LowerOperation if it doesn't know what else to do.
This methods should probably be factorized some,
but this is good enough for the moment. Have
LowerATOMIC_BINARY_64 use EXTRACT_ELEMENT rather
than assuming the operand is a BUILD_PAIR (if it
is then getNode will automagically simplify the
EXTRACT_ELEMENT). This way LowerATOMIC_BINARY_64
usable from LegalizeTypes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57831
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthijs Kooijman [Mon, 20 Oct 2008 11:24:57 +0000 (11:24 +0000)]
Fix typo in a comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57829
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthijs Kooijman [Mon, 20 Oct 2008 11:23:18 +0000 (11:23 +0000)]
Remove another stale comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57828
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthijs Kooijman [Mon, 20 Oct 2008 11:21:12 +0000 (11:21 +0000)]
Remove an inappropriate (probably outdated) comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57827
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthijs Kooijman [Mon, 20 Oct 2008 08:45:34 +0000 (08:45 +0000)]
Fix spelling error.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57820
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Sun, 19 Oct 2008 20:51:12 +0000 (20:51 +0000)]
Set N->OperandList to 0 after deletion. Otherwise, it's possible that it will
be either deleted or referenced afterwards.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57786
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Sun, 19 Oct 2008 20:34:04 +0000 (20:34 +0000)]
Fix comment. Other formatting changes. No functionality changes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57785
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sun, 19 Oct 2008 15:00:25 +0000 (15:00 +0000)]
Vector shuffle mask elements may be "undef". Handle
this everywhere in LegalizeTypes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57783
91177308-0d34-0410-b5e6-
96231b3b80d8
Duncan Sands [Sun, 19 Oct 2008 14:58:05 +0000 (14:58 +0000)]
Use a legal integer type for vector shuffle mask
elements. Otherwise LegalizeTypes will, reasonably
enough, legalize the mask, which may result in it
no longer being a BUILD_VECTOR node (LegalizeDAG
simply ignores the legality or not of vector masks).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57782
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Sat, 18 Oct 2008 18:49:30 +0000 (18:49 +0000)]
Reapply r57699 with a fix to not crash on asms with multiple results. Unlike
the previous patch this one actually passes make check.
"Fix PR2356 on PowerPC: if we have an input and output that are tied together
that have different sizes (e.g. i32 and i64) make sure to reserve registers for
the bigger operand."
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57771
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Sat, 18 Oct 2008 18:22:42 +0000 (18:22 +0000)]
Don't truncate GlobalAddress offsets to int in debug output.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57770
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 18 Oct 2008 05:21:37 +0000 (05:21 +0000)]
By min, I mean max.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57766
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Sat, 18 Oct 2008 05:18:55 +0000 (05:18 +0000)]
When creating intervals, leave min(1, numdefs) holes after each instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57765
91177308-0d34-0410-b5e6-
96231b3b80d8
Mon P Wang [Sat, 18 Oct 2008 02:48:13 +0000 (02:48 +0000)]
Make llvm memory barrier available as an intrinsic
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57750
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Sat, 18 Oct 2008 02:06:02 +0000 (02:06 +0000)]
Teach DAGCombine to fold constant offsets into GlobalAddress nodes,
and add a TargetLowering hook for it to use to determine when this
is legal (i.e. not in PIC mode, etc.)
This allows instruction selection to emit folded constant offsets
in more cases, such as the included testcase, eliminating the need
for explicit arithmetic instructions.
This eliminates the need for the C++ code in X86ISelDAGToDAG.cpp
that attempted to achieve the same effect, but wasn't as effective.
Also, fix handling of offsets in GlobalAddressSDNodes in several
places, including changing GlobalAddressSDNode's offset from
int to int64_t.
The Mips, Alpha, Sparc, and CellSPU targets appear to be
unaware of GlobalAddress offsets currently, so set the hook to
false on those targets.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57748
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Sat, 18 Oct 2008 01:03:45 +0000 (01:03 +0000)]
Revert r57699. It's causing regressions in
test/CodeGen/X86/2008-09-17-inline-asm-1.ll
and a few others, and it breaks the llvm-gcc build.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57747
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 21:42:45 +0000 (21:42 +0000)]
Use the opcode predicates, instead of duplicating the code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57735
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 21:39:27 +0000 (21:39 +0000)]
This is now partly done.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57734
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 21:38:40 +0000 (21:38 +0000)]
This is done.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57733
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 21:16:08 +0000 (21:16 +0000)]
Factor out the code for mapping LLVM IR condition opcodes to
ISD condition opcodes into helper functions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57726
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 21:02:22 +0000 (21:02 +0000)]
Add implicit defs of XMM8 to XMM15 on 32-bit call instructions. While this is not technically true, it tells tblgen that these instructions "clobber" the entire XMM register file.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57723
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 21:00:09 +0000 (21:00 +0000)]
Add RCBarriers to TargetInstrDesc. It's a list of register classes the given instruction can "clobber". For example, on x86 the call instruction can modify all of the XMM and fp stack registers.
TableGen has been taught to generate the lists from instruction definitions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57722
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 20:56:41 +0000 (20:56 +0000)]
Fix PR2898. Spiller delete a store for reuse before it knows for sure the reuse happened.
Patch by Lang Hames!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57720
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 19:59:51 +0000 (19:59 +0000)]
add support for 128 bit aggregates.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57715
91177308-0d34-0410-b5e6-
96231b3b80d8
Bill Wendling [Fri, 17 Oct 2008 18:48:57 +0000 (18:48 +0000)]
The Dwarf writer was comparing mangled and unmangled names for C++ code when we
have an unreachable block in a function. This was triggering the assert. This is
a horrid hack to cover this up.
Oh! for a good debug info architecture!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57714
91177308-0d34-0410-b5e6-
96231b3b80d8
Mon P Wang [Fri, 17 Oct 2008 18:22:58 +0000 (18:22 +0000)]
Added MemIntrinsicNode which is useful to represent target intrinsics that
touches memory and need an associated MemOperand
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57712
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 18:18:45 +0000 (18:18 +0000)]
Factor out the code for mapping LLVM IR condition opcodes to
ISD condition opcodes into helper functions.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57710
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 18:15:05 +0000 (18:15 +0000)]
add support for 128 bit inputs on both x86-64 and x86-32.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57709
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 17:59:52 +0000 (17:59 +0000)]
Fix a bug where the x86 backend would reject 64-bit r constraints when
in 32-bit mode instead of assigning a register pair. This has nothing to
do with PR2356, but I happened to notice it while working on it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57704
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 17:52:49 +0000 (17:52 +0000)]
Fix PR2356 on PowerPC: if we have an input and output that are tied together
that have different sizes (e.g. i32 and i64) make sure to reserve registers for
the bigger operand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57699
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 17:26:48 +0000 (17:26 +0000)]
remove an xfailed test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57695
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 17:26:19 +0000 (17:26 +0000)]
remove this test: it is xfailed anyway, and is failing for a reason
other than why it was xfailed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57694
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 17:14:20 +0000 (17:14 +0000)]
Fix lfence and mfence encoding. These look like MRM5r and MRM6r instructions except they do not have any operands. The RegModRM byte is encoded with register number 0.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57692
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 17:12:18 +0000 (17:12 +0000)]
getX86RegNum has long been moved to X86RegisterInfo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57691
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 17:05:25 +0000 (17:05 +0000)]
refactor some code into a helper method, no functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57690
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 16:47:46 +0000 (16:47 +0000)]
Keep track of *which* input constraint matches an output
constraint. Reject asms where an output has multiple
input constraints tied to it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57687
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 16:21:11 +0000 (16:21 +0000)]
add an assert so that PR2356 explodes instead of running off an
array. Improve some minor comments, refactor some helpers in
AsmOperandInfo. No functionality change for valid code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57686
91177308-0d34-0410-b5e6-
96231b3b80d8
Gabor Greif [Fri, 17 Oct 2008 14:43:58 +0000 (14:43 +0000)]
remove spurious space in link
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57677
91177308-0d34-0410-b5e6-
96231b3b80d8
Gabor Greif [Fri, 17 Oct 2008 08:31:36 +0000 (08:31 +0000)]
Add comment on how tagged pointers are
distinguished from normal (untagged) ones
as per review comment.
I am sufficiently unaquainted with doxygen to
defer the markup to someone with more experience.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57676
91177308-0d34-0410-b5e6-
96231b3b80d8
Evan Cheng [Fri, 17 Oct 2008 06:16:07 +0000 (06:16 +0000)]
Fix a very subtle spiller bug: UpdateKills should not forget to track defs of aliases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57673
91177308-0d34-0410-b5e6-
96231b3b80d8
Chris Lattner [Fri, 17 Oct 2008 06:11:48 +0000 (06:11 +0000)]
add some simple hacky long double support for the CBE. This
should work for intel long double, but ppc long double aborts
in convert.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57672
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 04:40:39 +0000 (04:40 +0000)]
Use INT64_C to emit constant values, to avoid problems with
constants that don't fit in an int. This fixes
"this decimal constant is unsigned only in ISO C90"
warnings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57668
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 01:33:43 +0000 (01:33 +0000)]
Fun x86 encoding tricks: when adding an immediate value of 128,
use a SUB instruction instead of an ADD, because -128 can be
encoded in an 8-bit signed immediate field, while +128 can't be.
This avoids the need for a 32-bit immediate field in this case.
A similar optimization applies to 64-bit adds with 0x80000000,
with the 32-bit signed immediate field.
To support this, teach tablegen how to handle 64-bit constants.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57663
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 01:23:35 +0000 (01:23 +0000)]
Define patterns for shld and shrd that match immediate
shift counts, and patterns that match dynamic shift counts
when the subtract is obscured by a truncate node.
Add DAGCombiner support for recognizing rotate patterns
when the shift counts are defined by truncate nodes.
Fix and simplify the code for commuting shld and shrd
instructions to work even when the given instruction doesn't
have a parent, and when the caller needs a new instruction.
These changes allow LLVM to use the shld, shrd, rol, and ror
instructions on x86 to replace equivalent code using two
shifts and an or in many more cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57662
91177308-0d34-0410-b5e6-
96231b3b80d8
Dan Gohman [Fri, 17 Oct 2008 00:56:52 +0000 (00:56 +0000)]
Use 0 instead of false to return a null pointer.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@57660
91177308-0d34-0410-b5e6-
96231b3b80d8