expansion is the same as that used by LegalizeDAG.
The resulting code sucks in terms of performance/codesize on x86-32 for a
64-bit operation; I haven't looked into whether different expansions might be
better in general.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@105378 91177308-0d34-0410-b5e6-96231b3b80d8
that are too large. This causes the freebsd bootloader to be too
large apparently.
It's unclear if this should be an -Os or -Oz thing. Thoughts welcome.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@105228 91177308-0d34-0410-b5e6-96231b3b80d8
shouldn't have a TargetLoweringInfo member. And FunctionLoweringInfo::set
doesn't needs its EnableFastISel argument.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@105101 91177308-0d34-0410-b5e6-96231b3b80d8
implementing pop with a linear search for a "best" element. The priority
queue was a neat idea, but in practice the comparison functions depend
on dynamic information.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104718 91177308-0d34-0410-b5e6-96231b3b80d8
Mon Ping provided; unfortunately bugpoint failed to
reduce it, but I think it's important to have a test for
this in the suite. 8023512.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104624 91177308-0d34-0410-b5e6-96231b3b80d8
so that it will continue to test what it was meant to test when I commit a
separate change for better support of BUILD_VECTOR and VECTOR_SHUFFLE for Neon.
Fix a DAG combiner crash exposed by this test change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104380 91177308-0d34-0410-b5e6-96231b3b80d8
pipeline stall. It's useful for targets like ARM cortex-a8. NEON has a lot
of long latency instructions so a strict register pressure reduction
scheduler does not work well.
Early experiments show this speeds up some NEON loops by over 30%.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104216 91177308-0d34-0410-b5e6-96231b3b80d8
test/Codegen/ARM/reg_sequence.ll but it doesn't affect the generated code
because the coalescer cleans it up. Radar 7998853.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104185 91177308-0d34-0410-b5e6-96231b3b80d8
need to be promoted. The BUILD_VECTOR and EXTRACT_VECTOR_ELT nodes generated
here already allow the promoted type to be used without further changes, so
just do the promotion. This fixes part of pr7167.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104141 91177308-0d34-0410-b5e6-96231b3b80d8
The trouble arises when the result of a vector cmp + sext is then and'ed with all ones. Instcombine will turn it into a vector cmp + zext, dag combiner will miss turning it into a vsetcc and hell breaks loose after that.
Teach dag combine to turn a vector cpm + zest into a vsetcc + and 1. This fixes rdar://7923010.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@104094 91177308-0d34-0410-b5e6-96231b3b80d8
- Change the logic DisableFramePointerElim() to check for the
-disable-non-leaf-fp-elim before -disable-fp-elim.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103990 91177308-0d34-0410-b5e6-96231b3b80d8
The implementation in LegalizeIntegerTypes to handle this as
sint64->float + appropriate power of 2 is subject to double rounding,
considered incorrect by numerics people. Use this implementation only
when it is safe. This leads to using library calls in some cases
that produced inline code before, but it's correct now.
(EVTToAPFloatSemantics belongs somewhere else, any suggestions?)
Add a correctly rounding (though not particularly fast) conversion
that uses X87 80-bit computations for x86-32.
7885399, 5901940. This shows up in gcc.c-torture/execute/ieee/rbug.c
in the gcc testsuite on some platforms.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103883 91177308-0d34-0410-b5e6-96231b3b80d8
the variable actually tracks.
N.B., several back-ends are using "HasCalls" as being synonymous for something
that adjusts the stack. This isn't 100% correct and should be looked into.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103802 91177308-0d34-0410-b5e6-96231b3b80d8
create separate virtual registers for CopyFromReg values, so uses of
them don't necessarily kill the value.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103519 91177308-0d34-0410-b5e6-96231b3b80d8
to LLVM_LIBRARY_VISIBILITY and introduce LLVM_GLOBAL_VISIBILITY, which is
the opposite, for future use by dragonegg.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103495 91177308-0d34-0410-b5e6-96231b3b80d8
Move EmitTargetCodeForMemcpy, EmitTargetCodeForMemset, and
EmitTargetCodeForMemmove out of TargetLowering and into
SelectionDAGInfo to exercise this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103481 91177308-0d34-0410-b5e6-96231b3b80d8
getConstantFP to accept the two supported long double
target types. This was not the original intent, but
there are other places that assume this works and it's
easy enough to do.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@103299 91177308-0d34-0410-b5e6-96231b3b80d8
in registers into a separate function to de-couple it from the
top-down-specific logic in getRegForValue.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102975 91177308-0d34-0410-b5e6-96231b3b80d8
code, and to eliminate the need for the SelectionDAGBuilder
state to be live during CodeGenAndEmitDAG calls.
Call SDB->clear() before CodeGenAndEmitDAG calls instead of
before it, and move the CurDAG->clear() out of SelectionDAGBuilder,
which doesn't own the DAG, and into CodeGenAndEmitDAG.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102814 91177308-0d34-0410-b5e6-96231b3b80d8
indexes could be of a different value type. Or not even using the same SDNode
for the constant (weird, I know). Compare the actual values instead of the
pointers.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102791 91177308-0d34-0410-b5e6-96231b3b80d8
instruction selection is done; it's confusing to see parts of it printed,
while other parts are omitted, along the way.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102771 91177308-0d34-0410-b5e6-96231b3b80d8
- Catch more further dag combine opportunities as result of operand promotion, e.g. (i32 anyext (i16 trunc (i32 x))) -> (i32 x)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102455 91177308-0d34-0410-b5e6-96231b3b80d8
of the dbg testsuite regressions. I don't think this is
really the right fix; this change exposed an existing problem
upstream somewhere.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102410 91177308-0d34-0410-b5e6-96231b3b80d8
form of DEBUG_VALUE, as it doesn't have reasonable default
behavior for unsupported targets. Add a new hook instead.
No functional change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102320 91177308-0d34-0410-b5e6-96231b3b80d8
FunctionLoweringInfo, as it isn't SelectionDAG-specific. This isn't
completely natural, as PHI node state is not per-function but rather
per-basic-block, however there's currently no other convenient
per-basic-block state to group it with.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@102109 91177308-0d34-0410-b5e6-96231b3b80d8
user-defined operations that use MMX register types, but
the compiler shouldn't generate them on its own. This adds
a Synthesizable abstraction to represent this, and changes
the vector widening computation so it won't produce MMX types.
(The motivation is to remove noise from the ABI compatibility
part of the gcc test suite, which has some breakage right now.)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101951 91177308-0d34-0410-b5e6-96231b3b80d8
into SelectionDAGBuilder. This avoids a separate pass over the
instructions, and has the side effect of providing debug location
information to the copy.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101906 91177308-0d34-0410-b5e6-96231b3b80d8
where multiple blocks are emitted; functions which do this need to return
the new BB so that their callers can stay current.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101843 91177308-0d34-0410-b5e6-96231b3b80d8
const_casts, and it reinforces the design of the Target classes being
immutable.
SelectionDAGISel::IsLegalToFold is now a static member function, because
PIC16 uses it in an unconventional way. There is more room for API
cleanup here.
And PIC16's AsmPrinter no longer uses TargetLowering.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101635 91177308-0d34-0410-b5e6-96231b3b80d8
Probably the best way to know that all getOperand() calls have been handled
is to replace that API instead of updating.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101579 91177308-0d34-0410-b5e6-96231b3b80d8
with a fix for self-hosting
rotate CallInst operands, i.e. move callee to the back
of the operand array
the motivation for this patch are laid out in my mail to llvm-commits:
more efficient access to operands and callee, faster callgraph-construction,
smaller compiler binary
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101465 91177308-0d34-0410-b5e6-96231b3b80d8
with a fix
rotate CallInst operands, i.e. move callee to the back
of the operand array
the motivation for this patch are laid out in my mail to llvm-commits:
more efficient access to operands and callee, faster callgraph-construction,
smaller compiler binary
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101397 91177308-0d34-0410-b5e6-96231b3b80d8
of the operand array
the motivation for this patch are laid out in my mail to llvm-commits:
more efficient access to operands and callee, faster callgraph-construction,
smaller compiler binary
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101364 91177308-0d34-0410-b5e6-96231b3b80d8
This doesn't occur much at all, it only seems to formed in the case
when the trunc optimization kicks in due to phase ordering. In that
case it is saves a few bytes on x86-32.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101350 91177308-0d34-0410-b5e6-96231b3b80d8
a load/or/and/store sequence into a narrower store when it is
safe. Daniel tells me that clang will start producing this sort
of thing with bitfields, and this does trigger a few dozen times
on 176.gcc produced by llvm-gcc even now.
This compiles code like CodeGen/X86/2009-05-28-DAGCombineCrash.ll
into:
movl %eax, 36(%rdi)
instead of:
movl $4294967295, %eax ## imm = 0xFFFFFFFF
andq 32(%rdi), %rax
shlq $32, %rcx
addq %rax, %rcx
movq %rcx, 32(%rdi)
and each of the testcases into a single store. Each of them used
to compile into craziness like this:
_test4:
movl $65535, %eax ## imm = 0xFFFF
andl (%rdi), %eax
shll $16, %esi
addl %eax, %esi
movl %esi, (%rdi)
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101343 91177308-0d34-0410-b5e6-96231b3b80d8
SelectionDAGBuilder. FunctionLoweringInfo isn't an ideal place for
them to live, but it's better than SelectionDAGBuilder for now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@101267 91177308-0d34-0410-b5e6-96231b3b80d8
so the user at least knows what inline asm is a problem. For example:
error: inline asm not supported yet: don't know how to handle tied indirect register inputs
pr8788-1.c:14:10: note: generated from here
asm ("\n" : "+r" (stack->regs)
^
Instead of:
fatal error: error in backend: Don't know how to handle tied indirect register inputs yet!
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100731 91177308-0d34-0410-b5e6-96231b3b80d8
Have the asmprinter use the mdnode to scavenge a source location if
present. Document this nonsense in langref.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100607 91177308-0d34-0410-b5e6-96231b3b80d8
1. Introduce some enums and accessors in the InlineAsm class
that eliminate a ton of magic numbers when handling inline
asm SDNode.
2. Add a new MDNodeSDNode selection dag node type that holds
a MDNode (shocking!)
3. Add a new argument to ISD::INLINEASM nodes that hold !srcloc
metadata, propagating it to the instruction emitter, which
drops it.
No functionality change.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100605 91177308-0d34-0410-b5e6-96231b3b80d8
Added support for address spaces and added a isVolatile field to memcpy, memmove, and memset,
e.g., llvm.memcpy.i32(i8*, i8*, i32, i32) -> llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100304 91177308-0d34-0410-b5e6-96231b3b80d8
representation. This eliminates the 'DILocation' MDNodes for
file/line/col tuples from -O0 -g codegen.
This remove the old DebugLoc class, making it a typedef for DebugLoc,
I'll rename NewDebugLoc next.
I didn't update the JIT to use the new apis, so it will continue to
work, but be as slow as before. Someone should eventually do this
or, better yet, rip out the JIT debug info stuff and build the JIT
on top of MC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100209 91177308-0d34-0410-b5e6-96231b3b80d8
Added support for address spaces and added a isVolatile field to memcpy, memmove, and memset,
e.g., llvm.memcpy.i32(i8*, i8*, i32, i32) -> llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100191 91177308-0d34-0410-b5e6-96231b3b80d8
- Do not try to infer GV alignment unless its type is sized. It's not possible to infer alignment if it has opaque type.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100118 91177308-0d34-0410-b5e6-96231b3b80d8
1. Makes it possible to lower with floating point loads and stores.
2. Avoid unaligned loads / stores unless it's fast.
3. Fix some memcpy lowering logic bug related to when to optimize a
load from constant string into a constant.
4. Adjust x86 memcpy lowering threshold to make it more sane.
5. Fix x86 target hook so it uses vector and floating point memory
ops more effectively.
rdar://7774704
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100090 91177308-0d34-0410-b5e6-96231b3b80d8
instructions. In addition to being a convenience,
they are faster than the old apis, particularly when
not going from an MDKindID like people should be
doing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99982 91177308-0d34-0410-b5e6-96231b3b80d8
e.g., llvm.memcpy.i32(i8*, i8*, i32, i32) -> llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
A update of langref will occur in a subsequent checkin.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99928 91177308-0d34-0410-b5e6-96231b3b80d8
and those derived from them. These are obnoxious because
they were written as: PatLeaf<(bitconvert). Not having an
argument was foiling adding better type checking for operand
count matching up with what was required (in this case,
bitconvert always requires an operand!)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99759 91177308-0d34-0410-b5e6-96231b3b80d8
1, 1 cases which are by-far the most frequent. This shrinks the X86
isel table from 77014 -> 74657 bytes.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99740 91177308-0d34-0410-b5e6-96231b3b80d8
the custom insertion hook deletes the instruction, then we try to set dead
flags on it. Neither the code that I added nor the code that was there
before was safe.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99538 91177308-0d34-0410-b5e6-96231b3b80d8
bytes instead of one byte. This is important because
we're running up to too many opcodes to fit in a byte
and it is aggrevated by FIRST_TARGET_MEMORY_OPCODE
making the numbering sparse. This just bites the
bullet and bloats out the table. In practice, this
increases the size of the x86 isel table from 74.5K
to 76K. I think we'll cope :)
This fixes rdar://7791648
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99494 91177308-0d34-0410-b5e6-96231b3b80d8
happening.
Enhance scheduling to set the DEAD flag on implicit defs
more aggressively. Before, we'd set an implicit def operand
to dead if it were present in the SDNode corresponding to
the machineinstr but had no use. Now we do it in this case
AND if the implicit def does not exist in the SDNode at all.
This exposes a couple of problems: one is the FIXME, which
causes a live intervals crash on CodeGen/X86/sibcall.ll.
The second is that it makes machinecse and licm more
aggressive (which is a good thing) but also exposes a case
where licm hoists a set0 and then it doesn't get resunk.
Talking to codegen folks about both these issues, but I need
this patch in in the meantime.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99485 91177308-0d34-0410-b5e6-96231b3b80d8
Here is a theoretical example that illustrates why the placement is important.
tmp1 =
store tmp1 -> x
...
tmp2 = add ...
...
call
...
store tmp2 -> x
Now mem2reg comes along:
tmp1 =
dbg_value (tmp1 -> x)
...
tmp2 = add ...
...
call
...
dbg_value (tmp2 -> x)
When the debugger examine the value of x after the add instruction but before the call, it should have the value of tmp1.
Furthermore, for dbg_value's that reference constants, they should not be emitted at the beginning of the block (since they do not have "producers").
This patch also cleans up how SDISel manages DbgValue nodes. It allow a SDNode to be referenced by multiple SDDbgValue nodes. When a SDNode is deleted, it uses the information to find the SDDbgValues and invalidate them. They are not deleted until the corresponding SelectionDAG is destroyed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99469 91177308-0d34-0410-b5e6-96231b3b80d8
2006-07-19-stwbrx-crash.ll for me, but it's the only likely
patch in the blame list of several bots. Lets see if this
fixes it.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@99453 91177308-0d34-0410-b5e6-96231b3b80d8
always create a new jump table. The intention was to avoid merging jump
tables in SelectionDAGBuilder, and to wait for the branch folding pass to
merge tables. Unfortunately, the same getJumpTableIndex() method is also
used to merge tables in branch folding, so as a result of this change
branch tables are never merged. Worse, the branch folding code is expecting
getJumpTableIndex to always return the index of an existing table, but with
this change, it never does so. In at least some cases, e.g., pr6543, this
creates references to non-existent tables.
I've fixed the problem by adding a new createJumpTableIndex function, which
will always create a new table, and I've changed getJumpTableIndex to only
look at existing tables.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98845 91177308-0d34-0410-b5e6-96231b3b80d8
label instead of trying to form one based on the BB name (which
causes collisions if the name is empty). This fixes PR6608
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98495 91177308-0d34-0410-b5e6-96231b3b80d8
an MCSymbol. Make the EH_LABEL MachineInstr hold its label
with an MCSymbol instead of ID. Fix a bug in MMI.cpp which
would return labels named "Label4" instead of "label4".
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98463 91177308-0d34-0410-b5e6-96231b3b80d8
instead of label ID's. This cleans up and regularizes a bunch
of code and makes way for future progress.
Unfortunately, this pointed out to me that JITDwarfEmitter.cpp
is largely copy and paste from DwarfException/MachineModuleInfo
and other places. This is very sad and disturbing. :(
One major change here is that TidyLandingPads moved from being
called in DwarfException::BeginFunction to being called in
DwarfException::EndFunction. There should not be any
functionality change from doing this, but I'm not an EH expert.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98459 91177308-0d34-0410-b5e6-96231b3b80d8
ppc calls copysignl with a 128 bit ppc long double, resulting
in a node that the type legalizer doesn't know how to expand.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98357 91177308-0d34-0410-b5e6-96231b3b80d8
(it seems that FreeBSD doesn't have copysignl). Done by
removing a bunch of assumptions from the code. This may also
help with sparc 128 bit floats.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@98346 91177308-0d34-0410-b5e6-96231b3b80d8
node which has a flag. That flag in turn was used by an
already-selected adde which turned into an ADC32ri8 which
used a selected load which was chained to the load we
folded. This flag use caused us to form a cycle. Fix
this by not ignoring chains in IsLegalToFold even in
cases where the isel thinks it can.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97791 91177308-0d34-0410-b5e6-96231b3b80d8
rounding correctly. This implementation is a generalization of
the x86_64 code in compiler-rt.
This fixes rdar://7683708.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97778 91177308-0d34-0410-b5e6-96231b3b80d8
as the very last thing before node emission. This should
dramatically reduce the number of times we do 'MatchAddress'
on X86, speeding up compile time. This also improves comments
in the tables and shrinks the table a bit, now down to
80506 bytes for x86.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97703 91177308-0d34-0410-b5e6-96231b3b80d8
CSE and recursive RAUW calls delete a node from the use list,
invalidating the use list iterator. There's currently no known
way to reproduce this in an unmodified LLVM, however there's no
fundamental reason why a SelectionDAG couldn't be formed which
would trigger this case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97665 91177308-0d34-0410-b5e6-96231b3b80d8
entry we're about to process is obviously going to fail, don't
bother pushing a scope only to have it immediately be popped.
This avoids a lot of scope stack traffic in common cases.
Unfortunately, this requires duplicating some of the predicate
dispatch. To avoid duplicating the actual logic I pulled each
predicate out to its own static function which gets used in
both places.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97651 91177308-0d34-0410-b5e6-96231b3b80d8
SwitchOpcodeMatcher) and have DAGISelMatcherOpt form it. This
speeds up selection, particularly for X86 which has lots of
variants of instructions with only type differences.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97645 91177308-0d34-0410-b5e6-96231b3b80d8
long test(long x) { return (x & 123124) | 3; }
Currently compiles to:
_test:
orl $3, %edi
movq %rdi, %rax
andq $123127, %rax
ret
This is because instruction and DAG combiners canonicalize
(or (and x, C), D) -> (and (or, D), (C | D))
However, this is only profitable if (C & D) != 0. It gets in the way of the
3-addressification because the input bits are known to be zero.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97616 91177308-0d34-0410-b5e6-96231b3b80d8
CopyToReg/CopyFromReg/INLINEASM. These are annoying because
they have the same opcode before an after isel. Fix this by
setting their NodeID to -1 to indicate that they are selected,
just like what automatically happens when selecting things that
end up being machine nodes.
With that done, give IsLegalToFold a new flag that causes it to
ignore chains. This lets the HandleMergeInputChains routine be
the one place that validates chains after a match is successful,
enabling the new hotness in chain processing. This smarter
chain processing eliminates the need for "PreprocessRMW" in the
X86 and MSP430 backends and enables MSP to start matching it's
multiple mem operand instructions more aggressively.
I currently #if out the dead code in the X86 backend and MSP
backend, I'll remove it for real in a follow-on patch.
The testcase changes are:
test/CodeGen/X86/sse3.ll: we generate better code
test/CodeGen/X86/store_op_load_fold2.ll: PreprocessRMW was
miscompiling this before, we now generate correct code
Convert it to filecheck while I'm at it.
test/CodeGen/MSP430/Inst16mm.ll: Add a testcase for mem/mem
folding to make anton happy. :)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97596 91177308-0d34-0410-b5e6-96231b3b80d8
was that we weren't properly handling the case when interior
nodes of a matched pattern become dead after updating chain
and flag uses. Now we handle this explicitly in
UpdateChainsAndFlags.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97561 91177308-0d34-0410-b5e6-96231b3b80d8
DoInstructionSelection. Inline "SelectRoot" into it from DAGISelHeader.
Sink some other stuff out of DAGISelHeader into SDISel.
Eliminate the various 'Indent' stuff from various targets, which dates
to when isel was recursive.
17 files changed, 114 insertions(+), 430 deletions(-)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97555 91177308-0d34-0410-b5e6-96231b3b80d8
stuff now that we don't care about emulating the old broken
behavior of the old isel. This eliminates the
'CheckChainCompatible' check (along with IsChainCompatible) which
did an incorrect and inefficient scan *up* the chain nodes which
happened as the pattern was being formed and does the validation
at the end in HandleMergeInputChains when it forms a structural
pattern. This scans "down" the graph, which means that it is
quickly bounded by nodes already selected. This also handles
token factors that get "trapped" in the dag.
Removing the CheckChainCompatible nodes also shrinks the
generated tables by about 6K for X86 (down to 83K).
There are two pieces remaining before I can nuke PreprocessRMW:
1. I xfailed a test because we're now producing worse code in a
case that has nothing to do with the change: it turns out that
our use of MorphNodeTo will leave dead nodes in the graph
which (depending on how the graph is walked) end up causing
bogus uses of chains and blocking matches. This is really
bad for other reasons, so I'll fix this in a follow-up patch.
2. CheckFoldableChainNode needs to be improved to handle the TF.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97539 91177308-0d34-0410-b5e6-96231b3b80d8
ComplexPattern at the root be generated multiple times, once
for each opcode they are part of. This encourages factoring
because the opcode checks get treated just like everything
else in the matcher.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97439 91177308-0d34-0410-b5e6-96231b3b80d8
to a scope where every child starts with a CheckOpcode, but
executes more efficiently. Enhance DAGISelMatcherOpt to
form it.
This also fixes a bug in CheckOpcode: apparently the SDNodeInfo
objects are not pointer comparable, we have to compare the
enum name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97438 91177308-0d34-0410-b5e6-96231b3b80d8
(which gets #included into the middle of each
target's DAGISel class) into a .cpp file where it is
only compiled once.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97425 91177308-0d34-0410-b5e6-96231b3b80d8
terms of store and load, which means bitcasting between scalar
integer and vector has endian-specific results, which undermines
this whole approach.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97137 91177308-0d34-0410-b5e6-96231b3b80d8
the number of value bits, not the number of bits of allocation for in-memory
storage.
Make getTypeStoreSize and getTypeAllocSize work consistently for arrays and
vectors.
Fix several places in CodeGen which compute offsets into in-memory vectors
to use TargetData information.
This fixes PR1784.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97064 91177308-0d34-0410-b5e6-96231b3b80d8
necessary to swap the operands to handle NaN and negative zero properly.
Also, reintroduce logic for checking for NaN conditions when forming
SSE min and max instructions, fixed to take into consideration NaNs and
negative zeros. This allows forming min and max instructions in more
cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@97025 91177308-0d34-0410-b5e6-96231b3b80d8
