with SUBREG_TO_REG, teach SimpleRegisterCoalescing to coalesce
SUBREG_TO_REG instructions (which are similar to INSERT_SUBREG
instructions), and teach the DAGCombiner to take advantage of this on
targets which support it. This eliminates many redundant
zero-extension operations on x86-64.
This adds a new TargetLowering hook, isZExtFree. It's similar to
isTruncateFree, except it only applies to actual definitions, and not
no-op truncates which may not zero the high bits.
Also, this adds a new optimization to SimplifyDemandedBits: transform
operations like x+y into (zext (add (trunc x), (trunc y))) on targets
where all the casts are no-ops. In contexts where the high part of the
add is explicitly masked off, this allows the mask operation to be
eliminated. Fix the DAGCombiner to avoid undoing these transformations
to eliminate casts on targets where the casts are no-ops.
Also, this adds a new two-address lowering heuristic. Since
two-address lowering runs before coalescing, it helps to be able to
look through copies when deciding whether commuting and/or
three-address conversion are profitable.
Also, fix a bug in LiveInterval::MergeInClobberRanges. It didn't handle
the case that a clobber range extended both before and beyond an
existing live range. In that case, multiple live ranges need to be
added. This was exposed by the new subreg coalescing code.
Remove 2008-05-06-SpillerBug.ll. It was bugpoint-reduced, and the
spiller behavior it was looking for no longer occurrs with the new
instruction selection.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68576 91177308-0d34-0410-b5e6-96231b3b80d8
builds.
--- Reverse-merging (from foreign repository) r68552 into '.':
U test/CodeGen/X86/tls8.ll
U test/CodeGen/X86/tls10.ll
U test/CodeGen/X86/tls2.ll
U test/CodeGen/X86/tls6.ll
U lib/Target/X86/X86Instr64bit.td
U lib/Target/X86/X86InstrSSE.td
U lib/Target/X86/X86InstrInfo.td
U lib/Target/X86/X86RegisterInfo.cpp
U lib/Target/X86/X86ISelLowering.cpp
U lib/Target/X86/X86CodeEmitter.cpp
U lib/Target/X86/X86FastISel.cpp
U lib/Target/X86/X86InstrInfo.h
U lib/Target/X86/X86ISelDAGToDAG.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.h
U lib/Target/X86/AsmPrinter/X86IntelAsmPrinter.h
U lib/Target/X86/X86ISelLowering.h
U lib/Target/X86/X86InstrInfo.cpp
U lib/Target/X86/X86InstrBuilder.h
U lib/Target/X86/X86RegisterInfo.td
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68560 91177308-0d34-0410-b5e6-96231b3b80d8
This introduces a small regression on the generated code
quality in the case we are just computing addresses, not
loading values.
Will work on it and on X86-64 support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68552 91177308-0d34-0410-b5e6-96231b3b80d8
entered via fall-through. Don't miss fallthroughs from blocks
terminated by conditional branches. Also, move
isOnlyReachableByFallthrough out of line.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68129 91177308-0d34-0410-b5e6-96231b3b80d8
only reachable via fall-through edges. This dramatically reduces the
number of labels printed, and thus also the number of labels the
assembler must parse and remember.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@68073 91177308-0d34-0410-b5e6-96231b3b80d8
x * 40
=>
shlq $3, %rdi
leaq (%rdi,%rdi,4), %rax
This has the added benefit of allowing more multiply to be folded into addressing mode. e.g.
a * 24 + b
=>
leaq (%rdi,%rdi,2), %rax
leaq (%rsi,%rax,8), %rax
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@67917 91177308-0d34-0410-b5e6-96231b3b80d8
%a = ...
%b = and i32 %a, 2
%c = srl i32 %b, 1
%d = br i32 %c,
into
%a = ...
%b = and %a, 2
%c = X86ISD::CMP %b, 0
%d = X86ISD::BRCOND %c ...
This applies only when the AND constant value has one bit set and the SRL
constant is equal to the log2 of the AND constant. The back-end is smart enough
to convert the result into a TEST/JMP sequence.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@67728 91177308-0d34-0410-b5e6-96231b3b80d8
to be returned in DL. LLVM's multiple-return-value support is
not ABI-conforming; front-ends that wish to have code emitted
that conforms to an ABI are currently expected to make
arrangements for this on their own rather than assuming that
multiple-return-values will automatically do the right thing.
This commit doesn't fundamentally change this situation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@67588 91177308-0d34-0410-b5e6-96231b3b80d8
not safe in general because the immediate could be an arbitrary
value that does not fit in a 32-bit pcrel displacement.
Conservatively fall back to loading the value into a register
and calling through it.
We still do the optzn on X86-32.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@67142 91177308-0d34-0410-b5e6-96231b3b80d8
operand is a signed 32-bit immediate. Unlike with the 8-bit
signed immediate case, it isn't actually smaller to fold a
32-bit signed immediate instead of a load. In fact, it's
larger in the case of 32-bit unsigned immediates, because
they can be materialized with movl instead of movq.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@67001 91177308-0d34-0410-b5e6-96231b3b80d8
ptrtoint and inttoptr in X86FastISel. These casts aren't always
handled in the generic FastISel code because X86 sometimes needs
custom code to do truncation and zero-extension.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66988 91177308-0d34-0410-b5e6-96231b3b80d8
by inserting explicit zero extensions where necessary. Included
is a testcase where SelectionDAG produces a virtual register
holding an i1 value which FastISel previously mistakenly assumed
to be zero-extended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66941 91177308-0d34-0410-b5e6-96231b3b80d8
codegen to the same thing as integer truncates to i8 (the top bits are
just undefined). This implements rdar://6667338
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66902 91177308-0d34-0410-b5e6-96231b3b80d8
1. ConstantPoolSDNode alignment field is log2 value of the alignment requirement. This is not consistent with other SDNode variants.
2. MachineConstantPool alignment field is also a log2 value.
3. However, some places are creating ConstantPoolSDNode with alignment value rather than log2 values. This creates entries with artificially large alignments, e.g. 256 for SSE vector values.
4. Constant pool entry offsets are computed when they are created. However, asm printer group them by sections. That means the offsets are no longer valid. However, asm printer uses them to determine size of padding between entries.
5. Asm printer uses expensive data structure multimap to track constant pool entries by sections.
6. Asm printer iterate over SmallPtrSet when it's emitting constant pool entries. This is non-deterministic.
Solutions:
1. ConstantPoolSDNode alignment field is changed to keep non-log2 value.
2. MachineConstantPool alignment field is also changed to keep non-log2 value.
3. Functions that create ConstantPool nodes are passing in non-log2 alignments.
4. MachineConstantPoolEntry no longer keeps an offset field. It's replaced with an alignment field. Offsets are not computed when constant pool entries are created. They are computed on the fly in asm printer and JIT.
5. Asm printer uses cheaper data structure to group constant pool entries.
6. Asm printer compute entry offsets after grouping is done.
7. Change JIT code to compute entry offsets on the fly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66875 91177308-0d34-0410-b5e6-96231b3b80d8
for i32/i64 expressions (we could also do i16 on cpus where
i16 lea is fast, but I didn't add this). On the example, we now
generate:
_test:
movl 4(%esp), %eax
cmpl $42, (%eax)
setl %al
movzbl %al, %eax
leal 4(%eax,%eax,8), %eax
ret
instead of:
_test:
movl 4(%esp), %eax
cmpl $41, (%eax)
movl $4, %ecx
movl $13, %eax
cmovg %ecx, %eax
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66869 91177308-0d34-0410-b5e6-96231b3b80d8
operands can't both be fully folded at the same time. For example,
in the included testcase, a global variable is being added with
an add of two values. The global variable wants RIP-relative
addressing, so it can't share the address with another base
register, but it's still possible to fold the initial add.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66865 91177308-0d34-0410-b5e6-96231b3b80d8
related transformations out of target-specific dag combine into the
ARM backend. These were added by Evan in r37685 with no testcases
and only seems to help ARM (e.g. test/CodeGen/ARM/select_xform.ll).
Add some simple X86-specific (for now) DAG combines that turn things
like cond ? 8 : 0 -> (zext(cond) << 3). This happens frequently
with the recently added cp constant select optimization, but is a
very general xform. For example, we now compile the second example
in const-select.ll to:
_test:
movsd LCPI2_0, %xmm0
ucomisd 8(%esp), %xmm0
seta %al
movzbl %al, %eax
movl 4(%esp), %ecx
movsbl (%ecx,%eax,4), %eax
ret
instead of:
_test:
movl 4(%esp), %eax
leal 4(%eax), %ecx
movsd LCPI2_0, %xmm0
ucomisd 8(%esp), %xmm0
cmovbe %eax, %ecx
movsbl (%ecx), %eax
ret
This passes multisource and dejagnu.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66779 91177308-0d34-0410-b5e6-96231b3b80d8
and extern_weak_odr. These are the same as the non-odr versions,
except that they indicate that the global will only be overridden
by an *equivalent* global. In C, a function with weak linkage can
be overridden by a function which behaves completely differently.
This means that IP passes have to skip weak functions, since any
deductions made from the function definition might be wrong, since
the definition could be replaced by something completely different
at link time. This is not allowed in C++, thanks to the ODR
(One-Definition-Rule): if a function is replaced by another at
link-time, then the new function must be the same as the original
function. If a language knows that a function or other global can
only be overridden by an equivalent global, it can give it the
weak_odr linkage type, and the optimizers will understand that it
is alright to make deductions based on the function body. The
code generators on the other hand map weak and weak_odr linkage
to the same thing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66339 91177308-0d34-0410-b5e6-96231b3b80d8
the same say the "test" instruction does in overflow cases,
so eliminating the test is only safe when those bits aren't
needed, as is the case for COND_E and COND_NE, or if it
can be proven that no overflow will occur. For now, just
restrict the optimization to COND_E and COND_NE and don't
do any overflow analysis.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66318 91177308-0d34-0410-b5e6-96231b3b80d8
INC64_32r and INC64_16r, because these instructions are encoded
differently on x86-64. This fixes JIT regressions on x86-64 in
kimwitu++ and others.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66207 91177308-0d34-0410-b5e6-96231b3b80d8
of MachineInstr def operands must be subtracted out. This bug
was uncovered by the recent x86 EFLAGS optimization. Before
that, the only instructions that ever needed unfolding were
things like CMP32rm, where NumDefs is zero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@66056 91177308-0d34-0410-b5e6-96231b3b80d8
them are generic changes.
- Use the "fast" flag that's already being passed into the asm printers instead
of shoving it into the DwarfWriter.
- Instead of calling "MI->getParent()->getParent()" for every MI, set the
machine function when calling "runOnMachineFunction" in the asm printers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65379 91177308-0d34-0410-b5e6-96231b3b80d8
instruction. The class also consolidates the code for detecting constant
splats that's shared across PowerPC and the CellSPU backends (and might be
useful for other backends.) Also introduces SelectionDAG::getBUID_VECTOR() for
generating new BUILD_VECTOR nodes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65296 91177308-0d34-0410-b5e6-96231b3b80d8
Now we're using one gross, but quite robust hack :) (previous ones
did not work, for example, when ext_weak symbol was used deep inside
constant expression in the initializer).
The proper fix of this problem will require some quite huge asmprinter
changes and that's why was postponed. This fixes PR3629 by the way :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@65230 91177308-0d34-0410-b5e6-96231b3b80d8
that has not been JIT'd yet, the callee is put on a list of pending functions
to JIT. The call is directed through a stub, which is updated with the address
of the function after it has been JIT'd. A new interface for allocating and
updating empty stubs is provided.
Add support for removing the ModuleProvider the JIT was created with, which
would otherwise invalidate the JIT's PassManager, which is initialized with the
ModuleProvider's Module.
Add support under a new ExecutionEngine flag for emitting the infomration
necessary to update Function and GlobalVariable stubs after JITing them, by
recording the address of the stub and the name of the GlobalValue. This allows
code to be copied from one address space to another, where libraries may live
at different virtual addresses, and have the stubs updated with their new
correct target addresses.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64906 91177308-0d34-0410-b5e6-96231b3b80d8
(Note: Eventually, commits like this will be handled via a pre-commit hook that
does this automagically, as well as expand tabs to spaces and look for 80-col
violations.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64827 91177308-0d34-0410-b5e6-96231b3b80d8
U include/llvm/CodeGen/DebugLoc.h
U lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
U lib/CodeGen/SelectionDAG/SelectionDAGBuild.cpp
U lib/Target/X86/AsmPrinter/X86ATTAsmPrinter.cpp
Enable debug location generation at -Os. This goes with the reapplication of the
r63639 patch.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64715 91177308-0d34-0410-b5e6-96231b3b80d8
in inline asm as signed (what gcc does). Add partial support
for x86-specific "e" and "Z" constraints, with appropriate
signedness for printing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64400 91177308-0d34-0410-b5e6-96231b3b80d8
suprise to some callers, e.g. register coalescer. For now, add an parameter
that tells AnalyzeBranch whether it's safe to modify the mbb. A better
solution is out there, but I don't have time to deal with it right now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@64124 91177308-0d34-0410-b5e6-96231b3b80d8
Many targets build placeholder nodes for special operands, e.g.
GlobalBaseReg on X86 and PPC for the PIC base. There's no
sensible way to associate debug info with these. I've left
them built with getNode calls with explicit DebugLoc::getUnknownLoc operands.
I'm not too happy about this but don't see a good improvement;
I considered adding a getPseudoOperand or something, but it
seems to me that'll just make it harder to read.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63992 91177308-0d34-0410-b5e6-96231b3b80d8
getCALLSEQ_{END,START} to permit passing no DebugLoc
there. UNDEF doesn't logically have DebugLoc; add
getUNDEF to encapsulate this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63978 91177308-0d34-0410-b5e6-96231b3b80d8
is given, override the subtarget settings and enable 64-bit support.
This restores the earlier behavior, and fixes regressions on
Non-64-bit-capable x86-32 hosts.
This isn't necessarily the best approach, but the most obvious
alternative is to require -mcpu=x86-64 or -mattr=+64bit to be used
with -march=x86-64 when the host doesn't have 64-bit support. This
makes things little more consistent, but it's less convenient, and
it has the practical drawback of requiring lots of test changes, so
I opted for the above approach for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63642 91177308-0d34-0410-b5e6-96231b3b80d8
SSE2, however it's possible to disable SSE2, and the subtarget support
code thinks that if 64-bit implies SSE2 and SSE2 is disabled then
64-bit should also be disabled. Instead, just mark all the 64-bit
subtargets as explicitly supporting SSE2.
Also, move the code that makes -march=x86-64 enable 64-bit support by
default to only apply when there is no explicit subtarget. If you
need to specify a subtarget and you want 64-bit code, you'll need to
select a subtarget that supports 64-bit code.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63575 91177308-0d34-0410-b5e6-96231b3b80d8
crashes or wrong code with codegen of large integers:
eliminate the legacy getIntegerVTBitMask and
getIntegerVTSignBit methods, which returned their
value as a uint64_t, so couldn't handle huge types.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63494 91177308-0d34-0410-b5e6-96231b3b80d8
dagcombines that help it match in several more cases. Add
several more cases to test/CodeGen/X86/bt.ll. This doesn't
yet include matching for BT with an immediate operand, it
just covers more register+register cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63266 91177308-0d34-0410-b5e6-96231b3b80d8
Don't use the Red Zone when dynamic stack realignment is needed.
This could be implemented, but most x86-64 ABIs don't require
dynamic stack realignment so it isn't urgent.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@63074 91177308-0d34-0410-b5e6-96231b3b80d8
tidy up SDUse and related code.
- Replace the operator= member functions with a set method, like
LLVM Use has, and variants setInitial and setNode, which take
care up updating use lists, like LLVM Use's does. This simplifies
code that calls these functions.
- getSDValue() is renamed to get(), as in LLVM Use, though most
places can either use the implicit conversion to SDValue or the
convenience functions instead.
- Fix some more node vs. value terminology issues.
Also, eliminate the one remaining use of SDOperandPtr, and
SDOperandPtr itself.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@62995 91177308-0d34-0410-b5e6-96231b3b80d8
corresponding to the "not" and "vnot" PatFrags. Use the new method
in some places where it seems appropriate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@62768 91177308-0d34-0410-b5e6-96231b3b80d8
we want to clear %ah to zero before a division, just use a
zero-extending mov to %al. This fixes PR3366.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@62691 91177308-0d34-0410-b5e6-96231b3b80d8
optimize it to a SINT_TO_FP when the sign bit is known zero. X86 isel should perform the optimization itself.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@62504 91177308-0d34-0410-b5e6-96231b3b80d8
X86. This code:
void f() {
uint32_t x;
float y = (float)x;
}
used to be:
movl %eax, -8(%ebp)
movl [2^52 double], -4(%ebp)
movsd -8(%ebp), %xmm0
subsd [2^52 double], %xmm0
cvtsd2ss %xmm0, %xmm0
Is now:
movsd [2^52 double], %xmm0
movsd %xmm0, %xmm1
movd %ecx, %xmm2
orps %xmm2, %xmm1
subsd %xmm0, %xmm1
cvtsd2ss %xmm1, %xmm0
This is faster on X86. Note that there's an extra load of %xmm0 into %xmm1. That
will be fixed in a later coalescer fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@62404 91177308-0d34-0410-b5e6-96231b3b80d8
