when X has multiple uses. This is useful for exposing secondary optimizations,
but the X86 backend isn't ready for this when X has a single use. For example,
this can disable load folding.
This is inching towards resolving PR6627.
llvm-svn: 130238
The hook will be used by the register allocator when recomputing register
classes after removing constraints.
Thumb1 code doesn't allow anything larger than tGPR, and x86 needs to ensure
that the spill size doesn't change.
llvm-svn: 130228
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
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
On the x86-64 and thumb2 targets, some registers are more expensive to encode
than others in the same register class.
Add a CostPerUse field to the TableGen register description, and make it
available from TRI->getCostPerUse. This represents the cost of a REX prefix or a
32-bit instruction encoding required by choosing a high register.
Teach the greedy register allocator to prefer cheap registers for busy live
ranges (as indicated by spill weight).
llvm-svn: 129864
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
llvm is built with unsigned chars where an immediate such as 0xff would be zero
extended to 64-bits, turning "cmp $0xff,%eax" into
"cmp $0xffffffffffffffff,%eax".
llvm-svn: 129845
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
is, it assumes addresses are 64-bit aligned (which should be the more common
case). If the alignment is found not to be aligned, then getOperandLatency()
would adjust the operand latency computation by one to compensate for it.
rdar://9294833
llvm-svn: 129742
the generated FastISel. X86 doesn't need to generate code to match ADD16ri8
since ADD16ri will do just fine. This is a small codesize win in the generated
instruction selector.
llvm-svn: 129692
simplifying them and exposing more information to tblgen. It would be nice
if other target authors adopted this as well, particularly arm since it has fastisel.
llvm-svn: 129676
kind of predicate: one that is specific to imm nodes. The predicate function
specified here just checks an int64_t directly instead of messing around with
SDNode's. The virtue of this is that it means that fastisel and other things
can reason about these predicates.
llvm-svn: 129675
structure and fix some fixmes. We now have a TreePredicateFn class
that handles all of the decoding of these things. This is an internal
cleanup that has no impact on the code generated by tblgen.
llvm-svn: 129670
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
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
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
The ARMARM specifies these instructions as unpredictable when storing the
writeback register. This shouldn't affect code generation much since storing a
pointer to itself is quite rare.
llvm-svn: 129409
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
disassembler API. Hooked this up to the ARM target so such tools as Darwin's
otool(1) can now print things like branch targets for example this:
blx _puts
instead of this:
blx #-36
And even print the expression encoded in the Mach-O relocation entried for
things like this:
movt r0, :upper16:((_foo-_bar)+1234)
llvm-svn: 129284
The previous cleanup of LDRD got overzealous and removed it, causing post-RA
scheduling to get overzealous in breaking antidependencies and invalidate these instructions. Hilarity and invalid assembly ensued.
rdar://9244161
llvm-svn: 129144
with the newer, cleaner model. It uses the IAPrinter class to hold the
information that is needed to match an instruction with its alias. This also
takes into account the available features of the platform.
There is one bit of ugliness. The way the logic determines if a pattern is
unique is O(N**2), which is gross. But in reality, the number of items it's
checking against isn't large. So while it's N**2, it shouldn't be a massive time
sink.
llvm-svn: 129110
Since these "Advanced SIMD and VFP" instructions have more specfic encoding bits
specified, if coproc == 10 or 11, we should reject the insn as invalid.
rdar://problem/9239922
rdar://problem/9239596
llvm-svn: 129027
Add more complete sanity check for LdStFrm instructions where if IBit (Inst{25})
is 1, Inst{4} should be 0. Otherwise, we should reject the insn as invalid.
rdar://problem/9239347
rdar://problem/9239467
llvm-svn: 128977
Qd -> bit[12] == 0
Qn -> bit[16] == 0
Qm -> bit[0] == 0
If one of these bits is 1, the instruction is UNDEFINED.
rdar://problem/9238399
rdar://problem/9238445
llvm-svn: 128949
For register-controlled shifts, we should check that the encoding constraint
Inst{7} = 0 and Inst{4} = 1 is satisfied.
rdar://problem/9237693
llvm-svn: 128941
An alternative syntax is available for a modified immediate constant that permits the programmer to specify
the encoding directly. In this syntax, #<const> is instead written as #<byte>,#<rot>, where:
<byte> is the numeric value of abcdefgh, in the range 0-255
<rot> is twice the numeric value of rotation, an even number in the range 0-30.
llvm-svn: 128897
It needed to be moved closer to the setjmp statement, because the code directly
after the setjmp needs to know about values that are on the stack. Also, the
'bitcast' of the function context was causing a dead load. This wouldn't be too
horrible, except that at -O0 it wasn't optimized out, and because it wasn't
using the correct base pointer (if there is a VLA), it would try to access a
value from a garbage address.
<rdar://problem/9130540>
llvm-svn: 128873
rdar://problem/9229922 ARM disassembler discrepancy: erroneously accepting RFE
Also LDC/STC instructions are predicated while LDC2/STC2 instructions are not, fixed while
doing regression testings.
llvm-svn: 128859
also fix the encoding of the later.
- Add a new encoding bit to describe the index mode used in AM3.
- Teach printAddrMode3Operand to check by the addressing mode which
index mode to print.
- Testcases.
llvm-svn: 128832
Define most shift masks incrementally to reduce the redundant
hard-coding. Introduce new shift for the VEX flags to replace the
magic constant 32 in various places.
llvm-svn: 128822
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
all LDR/STR changes and left them to a future patch. Passing all
checks now.
- Implement asm parsing support for LDRT, LDRBT, STRT, STRBT and
fix the encoding wherever is possible.
- Add a new encoding bit to describe the index mode used and teach
printAddrMode2Operand to check by the addressing mode which index
mode to print.
- Testcases
llvm-svn: 128689
- Implement asm parsing support for LDRT, LDRBT, STRT, STRBT and
{STR,LDC}{2}_{PRE,POST} fixing the encoding wherever is possible.
- Move all instructions which use am2offset without a pattern to use
addrmode2.
- Add a new encoding bit to describe the index mode used and teach
printAddrMode2Operand to check by the addressing mode which index
mode to print.
- Testcases
llvm-svn: 128632
{STR,LDC}{2}_PRE.
- Fixed the encoding in some places.
- Some of those instructions were using am2offset and now use addrmode2.
Codegen isn't affected, instructions which use SelectAddrMode2Offset were not
touched.
- Teach printAddrMode2Operand to check by the addressing mode which index
mode to print.
- This is a work in progress, more work to come. The idea is to change places
which use am2offset to use addrmode2 instead, as to unify assembly parser.
- Add testcases for assembly parser
llvm-svn: 128585
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
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
These instructions were changed to not embed the addressing mode within the MC instructions
We also need to update the corresponding assert stmt. Also add a test case.
llvm-svn: 128240
Set the encoding bits to {0,?,?,0}, not 0. Plus delegate the disassembly of ADR to
the more generic ADDri/SUBri instructions, and add a test case for that.
llvm-svn: 128234
The MC asm lexer wasn't honoring a non-default (anything but ';') statement
separator. Fix that, and generalize a bit to support multi-character
statement separators.
llvm-svn: 128227
entries being compared may not be ARMConstantPoolValue. Without checking
whether they are ARMConstantPoolValue first, and if the stars and moons
are aligned properly, the equality test may return true (when the first few
words of two Constants' values happen to be identical) and very bad things can
happen.
rdar://9125354
llvm-svn: 128203
These instructions were changed to not embed the addressing mode within the MC instructions
We also need to update the corresponding assert stmt. Also add two test cases.
llvm-svn: 128191
were incomplete. The assert stmt needs to be updated and the operand index incrment is wrong.
Fix the bad logic and add some sanity checking to detect bad instruction encoding;
and add a test case.
llvm-svn: 128186
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
VFP Load/Store Multiple Instructions used to embed the IA/DB addressing mode within the
MC instruction; that has been changed so that now, for example, VSTMDDB_UPD and VSTMDIA_UPD
are two instructions. Update the ARMDisassemblerCore.cpp's DisassembleVFPLdStMulFrm()
to reflect the change.
Also add a test case.
llvm-svn: 128103
the alias of an InstAlias instead of the thing being aliased. Because we need to
know the features that are valid for an InstAlias.
This is part of a work-in-progress.
llvm-svn: 127986
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
The relevant instruction table entries were changed sometime ago to no longer take
<Rt2> as an operand. Modify ARMDisassemblerCore.cpp to accomodate the change and
add a test case.
llvm-svn: 127935
Eli Friedman