non-zero.
- Teach X86 cmov optimization to eliminate the cmov from ctlz, cttz extension
when the source of X86ISD::BSR / X86ISD::BSF is proven to be non-zero.
rdar://9490949
llvm-svn: 131948
The following improvements are accomplished as a result of applying this patch:
- Fixed frame objects' offsets (relative to either the virtual frame pointer or
the stack pointer) are set before instruction selection is completed. There is
no need to wait until Prologue/Epilogue Insertion is run to set them.
- Calculation of final offsets of fixed frame objects is straightforward. It is
no longer necessary to assign negative offsets to fixed objects for incoming
arguments in order to distinguish them from the others.
- Since a fixed object has its relative offset set during instruction
selection, there is no need to conservatively set its alignment to 4.
- It is no longer necessary to reorder non-fixed frame objects in
MipsFrameLowering::adjustMipsStackFrame.
llvm-svn: 131915
x86_64 sibcall logic. I've filed PR9943 for the sibcall problem, and
this patch alters the testcase to work around the flaw. When PR9943
is fixed, this patch should be reverted.
llvm-svn: 131557
When instructions are deleted, they leave tombstone SlotIndex entries.
The isZeroLength method should ignore these null indexes.
This causes RABasic to sometimes spill a callee-saved register in the
abi-isel.ll test, so don't run that test with -regalloc=basic. Prioritizing
register allocation according to spill weight can cause more registers to be
used.
llvm-svn: 131436
by non-CMP expressions. The executable test case (129821) would test
this as well, if we had an "-O0 -disable-arm-fast-isel" LLVM-GCC
tester. Alas, the ARM assembly would be very difficult to check with
FileCheck.
The thumb2-cbnz.ll test is affected; it generates larger code (tst.w
vs. cmp #0), but I believe the new version is correct.
rdar://problem/9298790
llvm-svn: 131261
If there is a store after the load node, then there is a chain, which means
that there is another user. Thus, asking hasOneUser would fail. Instead we
ask hasNUsesOfValue on the 'data' value.
llvm-svn: 131183
at the start of basic blocks to their common predecessor. It's actually quite
common (e.g. about 50 times in JM/lencod) and has shown to be a nice code size
benefit. e.g.
pushq %rax
testl %edi, %edi
jne LBB0_2
## BB#1:
xorb %al, %al
popq %rdx
ret
LBB0_2:
xorb %al, %al
callq _foo
popq %rdx
ret
=>
pushq %rax
xorb %al, %al
testl %edi, %edi
je LBB0_2
## BB#1:
callq _foo
LBB0_2:
popq %rdx
ret
rdar://9145558
llvm-svn: 131172
This can't be just an assertion, users can always write impossible inline
assembly. Such an assembly statement should be included in the error message.
llvm-svn: 131024
Most of these tests require a single mov instruction that can come either before
or after a 2-addr instruction. -join-physregs changes the behavior, but the
results are equivalent.
llvm-svn: 130891
landing pad as its successor.
SjLj exception handling jumps to the correct landing pad via a switch statement
that's generated right before code-gen. Loosen the constraint in the machine
instruction verifier to allow for this. Note, this isn't the most rigorous check
since we cannot determine where that switch statement came from. But it's
marginally better than turning this check off when SjLj exceptions are used.
<rdar://problem/9187612>
llvm-svn: 130881
Original message:
Teach MachineCSE how to do simple cross-block CSE involving physregs. This allows, for example, eliminating duplicate cmpl's on x86. Part of rdar://problem/8259436 .
llvm-svn: 130877
These tests all follow the same pattern:
mov r2, r0
movs r0, #0
$CMP r2, r1
it eq
moveq r0, #1
bx lr
The first 'mov' can be eliminated by rematerializing 'movs r0, #0' below the
test instruction:
$CMP r0, r1
mov.w r0, #0
it eq
moveq r0, #1
bx lr
So far, only physreg coalescing can do that. The register allocators won't yet
split live ranges just to eliminate copies. They can learn, but this particular
problem is not likely to show up in real code. It only appears because r0 is
used for both the function argument and return value.
llvm-svn: 130858
model constants which can be added to base registers via add-immediate
instructions which don't require an additional register to materialize
the immediate.
llvm-svn: 130743
Currently the output should be almost identical to the one produced by CodeGen
to make the transition easier.
The only two differences I know of are:
* Some files get an extra advance loc of size 0. This will be fixed when
relaxations are enabled.
* The optimization of declaring an EH symbol as an external variable is not
implemented. This is a subset of adding the nounwind attribute, so we if really
this at -O0 we should probably do it at the IL level.
llvm-svn: 130623
successors) and use inverse depth first search to traverse the BBs. However
that doesn't work when the CFG has infinite loops. Simply do a linear
traversal of all BBs work just fine.
rdar://9344645
llvm-svn: 130324
more callee-saved registers and introduce copies. Only allows it if scheduling
a node above calls would end up lessen register pressure.
Call operands also has added ABI restrictions for register allocation, so be
extra careful with hoisting them above calls.
rdar://9329627
llvm-svn: 130245
Fixes Thumb2 ADCS and SBCS lowering: <rdar://problem/9275821>.
t2ADCS/t2SBCS are now pseudo instructions, consistent with ARM, so the
assembly printer correctly prints the 's' suffix.
Fixes Thumb2 adde -> SBC matching to check for live/dead carry flags.
Fixes the internal ARM machine opcode mnemonic for ADCS/SBCS.
Fixes ARM SBC lowering to check for live carry (potential bug).
llvm-svn: 130048
fix bugs exposed by the gcc dejagnu testsuite:
1. The load may actually be used by a dead instruction, which
would cause an assert.
2. The load may not be used by the current chain of instructions,
and we could move it past a side-effecting instruction. Change
how we process uses to define the problem away.
llvm-svn: 130018
On x86 this allows to fold a load into the cmp, greatly reducing register pressure.
movzbl (%rdi), %eax
cmpl $47, %eax
->
cmpb $47, (%rdi)
This shaves 8k off gcc.o on i386. I'll leave applying the patch in README.txt to Chris :)
llvm-svn: 130005
This tends to happen a lot with bitfield code generated by clang. A simple example for x86_64 is
uint64_t foo(uint64_t x) { return (x&1) << 42; }
which used to compile into bloated code:
shlq $42, %rdi ## encoding: [0x48,0xc1,0xe7,0x2a]
movabsq $4398046511104, %rax ## encoding: [0x48,0xb8,0x00,0x00,0x00,0x00,0x00,0x04,0x00,0x00]
andq %rdi, %rax ## encoding: [0x48,0x21,0xf8]
ret ## encoding: [0xc3]
with this patch we can fold the immediate into the and:
andq $1, %rdi ## encoding: [0x48,0x83,0xe7,0x01]
movq %rdi, %rax ## encoding: [0x48,0x89,0xf8]
shlq $42, %rax ## encoding: [0x48,0xc1,0xe0,0x2a]
ret ## encoding: [0xc3]
It's possible to save another byte by using 'andl' instead of 'andq' but I currently see no way of doing
that without making this code even more complicated. See the TODOs in the code.
llvm-svn: 129990
add <rd>, sp, #<imm8>
ldr <rd>, [sp, #<imm8>]
When the offset from sp is multiple of 4 and in range of 0-1020.
This saves code size by utilizing 16-bit instructions.
rdar://9321541
llvm-svn: 129971
This patch depends on the prior fix r129908 that changes to use std::find,
rather than std::binary_search, on unordered array.
Patch by Dan Bailey
llvm-svn: 129909
used by Clang. To help Clang integration, the PTX target has been split
into two targets: ptx32 and ptx64, depending on the desired pointer size.
- Add GCCBuiltin class to all intrinsics
- Split PTX target into ptx32 and ptx64
llvm-svn: 129851
manually and pass all (now) 4 arguments to the mul libcall. Add a new
ExpandLibCall for just this (copied gratuitously from type legalization).
Fixes rdar://9292577
llvm-svn: 129842
- As before, there is a minor semantic change here (evidenced by the test
change) for Darwin triples that have no version component. I debated changing
the default behavior of isOSVersionLT, but decided it made more sense for
triples to be explicit.
llvm-svn: 129805
- There is a minor semantic change here (evidenced by the test change) for
Darwin triples that have no version component. I debated changing the default
behavior of isOSVersionLT, but decided it made more sense for triples to be
explicit.
llvm-svn: 129802
Making use of VFP / NEON floating point multiply-accumulate / subtraction is
difficult on current ARM implementations for a few reasons.
1. Even though a single vmla has latency that is one cycle shorter than a pair
of vmul + vadd, a RAW hazard during the first (4? on Cortex-a8) can cause
additional pipeline stall. So it's frequently better to single codegen
vmul + vadd.
2. A vmla folowed by a vmul, vmadd, or vsub causes the second fp instruction to
stall for 4 cycles. We need to schedule them apart.
3. A vmla followed vmla is a special case. Obvious issuing back to back RAW
vmla + vmla is very bad. But this isn't ideal either:
vmul
vadd
vmla
Instead, we want to expand the second vmla:
vmla
vmul
vadd
Even with the 4 cycle vmul stall, the second sequence is still 2 cycles
faster.
Up to now, isel simply avoid codegen'ing fp vmla / vmls. This works well enough
but it isn't the optimial solution. This patch attempts to make it possible to
use vmla / vmls in cases where it is profitable.
A. Add missing isel predicates which cause vmla to be codegen'ed.
B. Make sure the fmul in (fadd (fmul)) has a single use. We don't want to
compute a fmul and a fmla.
C. Add additional isel checks for vmla, avoid cases where vmla is feeding into
fp instructions (except for the #3 exceptional case).
D. Add ARM hazard recognizer to model the vmla / vmls hazards.
E. Add a special pre-regalloc case to expand vmla / vmls when it's likely the
vmla / vmls will trigger one of the special hazards.
Enable these fp vmlx codegen changes for Cortex-A9.
llvm-svn: 129775
Add a avoidWriteAfterWrite() target hook to identify register classes that
suffer from write-after-write hazards. For those register classes, try to avoid
writing the same register in two consecutive instructions.
This is currently disabled by default. We should not spill to avoid hazards!
The command line flag -avoid-waw-hazard can be used to enable waw avoidance.
llvm-svn: 129772
when they are a truncate from something else. This eliminates fully half of all the
fastisel rejections on a test c++ file I'm working with, which should make a substantial
improvement for -O0 compile of c++ code.
This fixed rdar://9297003 - fast isel bails out on all functions taking bools
llvm-svn: 129752
Before we would bail out on i1 arguments all together, now we just bail on
non-constant ones. Also, we used to emit extraneous code. e.g. test12 was:
movb $0, %al
movzbl %al, %edi
callq _test12
and test13 was:
movb $0, %al
xorl %edi, %edi
movb %al, 7(%rsp)
callq _test13f
Now we get:
movl $0, %edi
callq _test12
and:
movl $0, %edi
callq _test13f
llvm-svn: 129751
registers for fast allocation a different way. This has us updating
used registers only when we're using that exact register.
Fixes rdar://9207598
llvm-svn: 129711
value constraints on them (when defined as ImmLeaf's). This is particularly important
for X86-64, where almost all reg/imm instructions take a i64immSExt32 immediate operand,
which has a value constraint. Before this patch we ended up iseling the examples into
such amazing code as:
movabsq $7, %rax
imulq %rax, %rdi
movq %rdi, %rax
ret
now we produce:
imulq $7, %rdi, %rax
ret
This dramatically shrinks the generated code at -O0 on x86-64.
llvm-svn: 129691
2. implement rdar://9289501 - fast isel should fold trivial multiplies to shifts
3. teach tblgen to handle shift immediates that are different sizes than the
shifted operands, eliminating some code from the X86 fast isel backend.
4. Have FastISel::SelectBinaryOp use (the poorly named) FastEmit_ri_ function
instead of FastEmit_ri to simplify code.
llvm-svn: 129666
when we have a global variable base an an index. Instead, just give up on
folding the global variable.
Before we'd geenrate:
_test: ## @test
## BB#0:
movq _rtx_length@GOTPCREL(%rip), %rax
leaq (%rax), %rax
addq %rdi, %rax
movzbl (%rax), %eax
ret
now we generate:
_test: ## @test
## BB#0:
movq _rtx_length@GOTPCREL(%rip), %rax
movzbl (%rax,%rdi), %eax
ret
The difference is even more significant when there is a scale
involved.
This fixes rdar://9289558 - total fail with addr mode formation at -O0/x86-64
llvm-svn: 129664
less trivial things) into a dummy lea. Before we generated:
_test: ## @test
movq _G@GOTPCREL(%rip), %rax
leaq (%rax), %rax
ret
now we produce:
_test: ## @test
movq _G@GOTPCREL(%rip), %rax
ret
This is part of rdar://9289558
llvm-svn: 129662
The basic issue here is that bottom-up isel is matching the branch
and compare, and was failing to fold the load into the branch/compare
combo. Fixing this (by allowing folding into any instruction of a
sequence that is selected) allows us to produce things like:
cmpb $0, 52(%rax)
je LBB4_2
instead of:
movb 52(%rax), %cl
cmpb $0, %cl
je LBB4_2
This makes the generated -O0 code run a bit faster, but also speeds up
compile time by putting less pressure on the register allocator and
generating less code.
This was one of the biggest classes of missing load folding. Implementing
this shrinks 176.gcc's c-decl.s (as a random example) by about 4% in (verbose-asm)
line count.
llvm-svn: 129656
Change ELF systems to use CFI for producing the EH tables. This reduces the
size of the clang binary in Debug builds from 690MB to 679MB.
llvm-svn: 129571
This is done by pushing physical register definitions close to their
use, which happens to handle flag definitions if they're not glued to
the branch. This seems to be generally a good thing though, so I
didn't need to add a target hook yet.
The primary motivation is to generate code closer to what people
expect and rule out missed opportunity from enabling macro-op
fusion. As a side benefit, we get several 2-5% gains on x86
benchmarks. There is one regression:
SingleSource/Benchmarks/Shootout/lists slows down be -10%. But this is
an independent scheduler bug that will be tracked separately.
See rdar://problem/9283108.
Incidentally, pre-RA scheduling is only half the solution. Fixing the
later passes is tracked by:
<rdar://problem/8932804> [pre-RA-sched] on x86, attempt to schedule CMP/TEST adjacent with condition jump
Fixes:
<rdar://problem/9262453> Scheduler unnecessary break of cmp/jump fusion
llvm-svn: 129508
ignored. There was a test to catch this, but it was just blindly updated in
a large change. This fixes another part of <rdar://problem/9275290>.
llvm-svn: 129466
the max itself, so it is not easy to write a test case for this, but I added a
test case that would fail if the code in AsmPrinter were removed.
llvm-svn: 129432
Additional fixes:
Do something reasonable for subtargets with generic
itineraries by handle node latency the same as for an empty
itinerary. Now nodes default to unit latency unless an itinerary
explicitly specifies a zero cycle stage or it is a TokenFactor chain.
Original fixes:
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make the ndoe latency adjustments work, I also
needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.
llvm-svn: 129421
Now that we have a first-class way to represent unaligned loads, the unaligned
load intrinsics are superfluous.
First part of <rdar://problem/8460511>.
llvm-svn: 129401
UnitsSharePred was a source of randomness in the scheduler: node
priority depended on the queue data structure. I rewrote the recent
VRegCycle heuristics to completely replace the old heuristic without
any randomness. To make these heuristic adjustments to node latency work,
I also needed to do something a little more reasonable with TokenFactor. I
gave it zero latency to its consumers and always schedule it as low as
possible.
llvm-svn: 129383
induction variable. The preRA scheduler is unaware of induction vars,
so we look for potential "virtual register cycles" instead.
Fixes <rdar://problem/8946719> Bad scheduling prevents coalescing
llvm-svn: 129100
There can be multiple defs for a single virtual register when they are defining
sub-registers.
The missing <dead> flag was stopping the inline spiller from eliminating dead
code after rematerialization.
llvm-svn: 128888
When a virtual register has a single value that is defined as a copy of a
reserved register, permit that copy to be joined. These virtual register are
usually copies of the stack pointer:
%vreg75<def> = COPY %ESP; GR32:%vreg75
MOV32mr %vreg75, 1, %noreg, 0, %noreg, %vreg74<kill>
MOV32mi %vreg75, 1, %noreg, 8, %noreg, 0
MOV32mi %vreg75<kill>, 1, %noreg, 4, %noreg, 0
CALLpcrel32 ...
Coalescing these virtual registers early decreases register pressure.
Previously, they were coalesced by RALinScan::attemptTrivialCoalescing after
register allocation was completed.
The lower register pressure causes the mcinst-lowering-cmp0.ll test case to fail
because it depends on linear scan spilling a particular register.
I am deleting 2008-08-05-SpillerBug.ll because it is counting the number of
instructions emitted, and its revision history shows the 'correct' count being
edited many times.
llvm-svn: 128845
The code inserted by PPCTargetLowering::EmitInstrWithCustomInserter for ppc64 is
wrong, and I don't know how to fix it. It seems to be using the correct register
classes for pointers, but it inserts all 32-bit instructions.
llvm-svn: 128835
registers that arise from argument shuffling with the soft float ABI. These
instructions are particularly slow on Cortex A8. This fixes one half of
<rdar://problem/8674845>.
llvm-svn: 128759
This way, shrinkToUses() will ignore the instruction that is about to be
deleted, and we avoid leaving invalid live ranges that SplitKit doesn't like.
Fix a misunderstanding in MachineVerifier about <def,undef> operands. The
<undef> flag is valid on def operands where it has the same meaning as <undef>
on a use operand. It only applies to sub-register defines which also read the
full register.
llvm-svn: 128642
The rematerialized instruction may require a more constrained register class
than the register being spilled. In the test case, the spilled register has been
inflated to the DPR register class, but we are rematerializing a load of the
ssub_0 sub-register which only exists for DPR_VFP2 registers.
The register class is reinflated after spilling, so the conservative choice is
only temporary.
llvm-svn: 128610
was lowering them to sext / uxt + mul instructions. Unfortunately the
optimization passes may hoist the extensions out of the loop and separate them.
When that happens, the long multiplication instructions can be broken into
several scalar instructions, causing significant performance issue.
Note the vmla and vmls intrinsics are not added back. Frontend will codegen them
as intrinsics vmull* + add / sub. Also note the isel optimizations for catching
mul + sext / zext are not changed either.
First part of rdar://8832507, rdar://9203134
llvm-svn: 128502
isel lowering to fold the zero-extend's and take advantage of no-stall
back to back vmul + vmla:
vmull q0, d4, d6
vmlal q0, d5, d6
is faster than
vaddl q0, d4, d5
vmovl q1, d6
vmul q0, q0, q1
This allows us to vmull + vmlal for:
f = vmull_u8( vget_high_u8(s), c);
f = vmlal_u8(f, vget_low_u8(s), c);
rdar://9197392
llvm-svn: 128444
Correctly terminate the range of register DBG_VALUEs when the register is
clobbered or when the basic block ends.
The code is now ready to deal with variables that are sometimes in a register
and sometimes on the stack. We just need to teach emitDebugLoc to say 'stack
slot'.
llvm-svn: 128327
masks to match inversely for the code as is to work. For the example given
we actually want:
bfi r0, r2, #1, #1
not #0, however, given the way the pattern is written it's not possible
at the moment.
Fixes rdar://9177502
llvm-svn: 128320
The .dot directives don't need labels, that is a leftover from when we created
line number info manually.
Instructions following a DBG_VALUE can share its label since the DBG_VALUE
doesn't produce any code.
llvm-svn: 128284
int tries = INT_MAX;
while (tries > 0) {
tries--;
}
The check should be:
subs r4, #1
cmp r4, #0
bgt LBB0_1
The subs can set the overflow V bit when r4 is INT_MAX+1 (which loop
canonicalization apparently does in this case). cmp #0 would have cleared
it while not changing the N and Z bits. Since BGT is dependent on the V
bit, i.e. (N == V) && !Z, it is not safe to eliminate the cmp #0.
rdar://9172742
llvm-svn: 128179
This will extend the ranges of debug info variables in registers until they are
clobbered.
Fix 1: Don't mistake DBG_VALUE instructions referring to incoming arguments on
the stack with DBG_VALUE instructions referring to variables in the frame
pointer. This fixes the gdb test-suite failure.
Fix 2: Don't trace through copies to physical registers setting up call
arguments. These registers are call clobbered, and the source register is more
likely to be a callee-saved register that can be extended through the call
instruction.
llvm-svn: 128114
These ranges get completely jumbled by the post-ra scheduler, and it is not
really reasonable to expect it to make sense of them.
Instead, teach DwarfDebug to notice when user variables in registers are
clobbered, and terminate the ranges there.
llvm-svn: 128045
to have single return block (at least getting there) for optimizations. This
is general goodness but it would prevent some tailcall optimizations.
One specific case is code like this:
int f1(void);
int f2(void);
int f3(void);
int f4(void);
int f5(void);
int f6(void);
int foo(int x) {
switch(x) {
case 1: return f1();
case 2: return f2();
case 3: return f3();
case 4: return f4();
case 5: return f5();
case 6: return f6();
}
}
=>
LBB0_2: ## %sw.bb
callq _f1
popq %rbp
ret
LBB0_3: ## %sw.bb1
callq _f2
popq %rbp
ret
LBB0_4: ## %sw.bb3
callq _f3
popq %rbp
ret
This patch teaches codegenprep to duplicate returns when the return value
is a phi and where the phi operands are produced by tail calls followed by
an unconditional branch:
sw.bb7: ; preds = %entry
%call8 = tail call i32 @f5() nounwind
br label %return
sw.bb9: ; preds = %entry
%call10 = tail call i32 @f6() nounwind
br label %return
return:
%retval.0 = phi i32 [ %call10, %sw.bb9 ], [ %call8, %sw.bb7 ], ... [ 0, %entry ]
ret i32 %retval.0
This allows codegen to generate better code like this:
LBB0_2: ## %sw.bb
jmp _f1 ## TAILCALL
LBB0_3: ## %sw.bb1
jmp _f2 ## TAILCALL
LBB0_4: ## %sw.bb3
jmp _f3 ## TAILCALL
rdar://9147433
llvm-svn: 127953
not have native support for this operation (such as X86).
The legalized code uses two vector INT_TO_FP operations and is faster
than scalarizing.
llvm-svn: 127951
- Emit mad instead of mad.rn for shader model 1.0
- Emit explicit mov.u32 instructions for reading global variables
- (most PTX instructions cannot take global variable immediates)
llvm-svn: 127895
comparisons on x86. Essentially, the way this works is that SUB+SBB sets
the relevant flags the same way a double-width CMP would.
This is a substantial improvement over the generic lowering in LLVM. The output
is also shorter than the gcc-generated output; I haven't done any detailed
benchmarking, though.
llvm-svn: 127852
Cameron Zwarich