usage of the shuffle bitmask. Both work in 128-bit lanes without
crossing, but in the former the mask of the high part is the same
used by the low part while in the later both lanes have independent
masks. Handle this properly and and add support for vpermilpd.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136200 91177308-0d34-0410-b5e6-96231b3b80d8
On x86 we can't encode an immediate LHS of a sub directly. If the RHS comes from a XOR with a constant we can
fold the negation into the xor and add one to the immediate of the sub. Then we can turn the sub into an add,
which can be commuted and encoded efficiently.
This code is generated for __builtin_clz and friends.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136167 91177308-0d34-0410-b5e6-96231b3b80d8
different from the previous 128-bit because they work in lanes.
Update a few comments and add testcases
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136157 91177308-0d34-0410-b5e6-96231b3b80d8
shuffle before inserting on a 256-bit vector.
- Add AVX versions of movd/movq instructions
- Introduce a few COPY patterns to match insert_subvector instructions.
This turns a trivial insert_subvector instruction into a register copy,
coalescing the xmm into a ymm and avoid emiting on more instruction.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@136002 91177308-0d34-0410-b5e6-96231b3b80d8
Addresses PR10466, although the crash from that PR only triggers in cases where DAGCombine misses optimizing a shuffle.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135980 91177308-0d34-0410-b5e6-96231b3b80d8
of doing the RAUW calls for the overflow value itself. This makes
it more consistent with how the rest of LegalizeDAG works.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135788 91177308-0d34-0410-b5e6-96231b3b80d8
the way to go. Doing this here will prevent several node matches later,
and would have to force looking all the way through several
VINSERTF128/VEXTRACTF128 chains to optimize simple things.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135730 91177308-0d34-0410-b5e6-96231b3b80d8
and was actually very wrong, fix it and make it simpler. Also remove the
ConcatVectors function, which is unused now.
- Fix a introduction of useless nodes in r126664 and r126264. The
VUNPCKL* should never be introduced cause we don't want duplicate
nodes for 128 AVX and non-AVX modes, the actual instruction
difference only exists during isel, but not for target specific DAG
nodes. We only introduce V* target nodes when there is no 128-bit
version already there.
- Fix a fragile test and make it more useful.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135729 91177308-0d34-0410-b5e6-96231b3b80d8
- Add more bitcasts for v16i16
- Since 135661 and 135662 already added the splat logic,
just add one more splat test for v16i16
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135663 91177308-0d34-0410-b5e6-96231b3b80d8
instruction introduced in AVX, which can operate on 128 and 256-bit vectors.
It considers a 256-bit vector as two independent 128-bit lanes. It can permute
any 32 or 64 elements inside a lane, and restricts the second lane to
have the same permutation of the first one. With the improved splat support
introduced early today, adding codegen for this instruction enable more
efficient 256-bit code:
Instead of:
vextractf128 $0, %ymm0, %xmm0
punpcklbw %xmm0, %xmm0
punpckhbw %xmm0, %xmm0
vinsertf128 $0, %xmm0, %ymm0, %ymm1
vinsertf128 $1, %xmm0, %ymm1, %ymm0
vextractf128 $1, %ymm0, %xmm1
shufps $1, %xmm1, %xmm1
movss %xmm1, 28(%rsp)
movss %xmm1, 24(%rsp)
movss %xmm1, 20(%rsp)
movss %xmm1, 16(%rsp)
vextractf128 $0, %ymm0, %xmm0
shufps $1, %xmm0, %xmm0
movss %xmm0, 12(%rsp)
movss %xmm0, 8(%rsp)
movss %xmm0, 4(%rsp)
movss %xmm0, (%rsp)
vmovaps (%rsp), %ymm0
We get:
vextractf128 $0, %ymm0, %xmm0
punpcklbw %xmm0, %xmm0
punpckhbw %xmm0, %xmm0
vinsertf128 $0, %xmm0, %ymm0, %ymm1
vinsertf128 $1, %xmm0, %ymm1, %ymm0
vpermilps $85, %ymm0, %ymm0
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135662 91177308-0d34-0410-b5e6-96231b3b80d8
refactor the code and add a bunch of comments. The final shuffle
emitted by handling 256-bit types is suitable for the VPERM shuffle
instruction which is going to be introduced in a next commit (with
a testcase which cover this commit)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135661 91177308-0d34-0410-b5e6-96231b3b80d8
There is still a bit more refactoring left to do in Targets. But we are now very
close to fixing all the layering issues in MC.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135611 91177308-0d34-0410-b5e6-96231b3b80d8
to MCRegisterInfo. Also initialize the mapping at construction time.
This patch eliminate TargetRegisterInfo from TargetAsmInfo. It's another step
towards fixing the layering violation.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135424 91177308-0d34-0410-b5e6-96231b3b80d8
1) Make non-legal 256-bit loads to be promoted to v4i64. This lets us
canonize the loads and handle things the same way we use to handle
for 128-bit registers. Despite of what one of the removed comments
explained, the load promotion would not mess with VPERM, it's only a
matter of doing the appropriate bitcasts when this instructions comes
to be introduced. Also make LOAD v8i32 legal.
2) Doing 1) exposed two bugs:
- v4i64 was being promoted to itself for several opcodes (introduced
in r124447 by David Greene) causing endless recursion and the stack to
explode.
- there was no support for allOnes BUILD_VECTORs and ANDNP would fail to
match because it was generating early target constant pools during
lowering.
3) The testcases are already checked-in, doing 1) exposed the
bugs in the current testcases.
4) Tidy up code to be more clear and explicit about AVX.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135313 91177308-0d34-0410-b5e6-96231b3b80d8
when determining validity of matching constraint. Allow i1
types access to the GR8 reg class for x86.
Fixes PR10352 and rdar://9777108
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135180 91177308-0d34-0410-b5e6-96231b3b80d8
During type legalization we often use the SIGN_EXTEND_INREG SDNode.
When this SDNode is legalized during the LegalizeVector phase, it is
scalarized because non-simple types are automatically marked to be expanded.
In this patch we add support for lowering SIGN_EXTEND_INREG manually.
This fixes CodeGen/X86/vec_sext.ll when running with the '-promote-elements'
flag.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@135144 91177308-0d34-0410-b5e6-96231b3b80d8
Drop the FpMov instructions, use plain COPY instead.
Drop the FpSET/GET instruction for accessing fixed stack positions.
Instead use normal COPY to/from ST registers around inline assembly, and
provide a single new FpPOP_RETVAL instruction that can access the return
value(s) from a call. This is still necessary since you cannot tell from
the CALL instruction alone if it returns anything on the FP stack. Teach
fast isel to use this.
This provides a much more robust way of handling fixed stack registers -
we can tolerate arbitrary FP stack instructions inserted around calls
and inline assembly. Live range splitting could sometimes break x87 code
by inserting spill code in unfortunate places.
As a bonus we handle floating point inline assembly correctly now.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@134018 91177308-0d34-0410-b5e6-96231b3b80d8
optimizations when emitting calls to the function; instead those calls may
use faster relocations which require the function to be immediately resolved
upon loading the dynamic object featuring the call. This is useful when it
is known that the function will be called frequently and pervasively and
therefore there is no merit in delaying binding of the function.
Currently only implemented for x86-64, where it turns into a call through
the global offset table.
Patch by Dan Gohman, who assures me that he's going to add LangRef documentation
for this once it's committed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133080 91177308-0d34-0410-b5e6-96231b3b80d8
floating-point comparison, generate a mask of 0s or 1s, and generally
DTRT with NaNs. Only profitable when the user wants a materialized 0
or 1 at runtime. rdar://problem/5993888
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@132404 91177308-0d34-0410-b5e6-96231b3b80d8
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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@131948 91177308-0d34-0410-b5e6-96231b3b80d8
I'm backing this out for the second time. It was supposed to be fixed by r128164, but the mingw self-host must be defeating the fix.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@128181 91177308-0d34-0410-b5e6-96231b3b80d8
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
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127953 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127951 91177308-0d34-0410-b5e6-96231b3b80d8
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.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127852 91177308-0d34-0410-b5e6-96231b3b80d8
rather than an int. Thankfully, this only causes LLVM to miss optimizations, not
generate incorrect code.
This just fixes the zext at the return. We still insert an i32 ZextAssert when
reading a function's arguments, but it is followed by a truncate and another i8
ZextAssert so it is not optimized.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127766 91177308-0d34-0410-b5e6-96231b3b80d8
corresponding testcases back to the previous versions.
Fixes some performance regressions only seen on 32-bit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127441 91177308-0d34-0410-b5e6-96231b3b80d8
testcases accordingly. Some are currently xfailed and will be filed
as bugs to be fixed or understood.
Performance results:
roughly neutral on SPEC
some micro benchmarks in the llvm suite are up between 100 and 150%, only
a pair of regressions that are due to be investigated
john-the-ripper saw:
10% improvement in traditional DES
8% improvement in BSDI DES
59% improvement in FreeBSD MD5
67% improvement in OpenBSD Blowfish
14% improvement in LM DES
Small compile time impact.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127208 91177308-0d34-0410-b5e6-96231b3b80d8
regs. This is the only change in this checkin that may affects the
default scheduler. With better register tracking and heuristics, it
doesn't make sense to artificially lower the register limit so much.
Added -sched-high-latency-cycles and X86InstrInfo::isHighLatencyDef to
give the scheduler a way to account for div and sqrt on targets that
don't have an itinerary. It is currently defaults to 10 (the actual
number doesn't matter much), but only takes effect on non-default
schedulers: list-hybrid and list-ilp.
Added several heuristics that can be individually disabled for the
non-default sched=list-ilp mode. This helps us determine how much
better we can do on a given benchmark than the default
scheduler. Certain compute intensive loops run much faster in this
mode with the right set of heuristics, and it doesn't seem to have
much negative impact elsewhere. Not all of the heuristics are needed,
but we still need to experiment to decide which should be disabled by
default for sched=list-ilp.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@127067 91177308-0d34-0410-b5e6-96231b3b80d8
missing patterns for them.
Add a SIMD test subdirectory to hold tests for SIMD instruction
selection correctness and quality.
'
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126845 91177308-0d34-0410-b5e6-96231b3b80d8
and 256-bit forms. Because the number of elements in a vector
does not determine the vector type (4 elements could be v4f32 or
v4f64), pass the full type of the vector to decode routines.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126664 91177308-0d34-0410-b5e6-96231b3b80d8
In other words, do not keep track of argument's location. The debugger (gdb) is not prepared to see line table entries for arguments. For the debugger, "second" line table entry marks beginning of function body.
This requires some coordination with debugger to get this working.
- The debugger needs to be aware of prolog_end attribute attached with line table entries.
- The compiler needs to accurately mark prolog_end in line table entries (at -O0 and at -O1+)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126155 91177308-0d34-0410-b5e6-96231b3b80d8
since one needs to be a register operand. Just use movss instead of forcing
an operand into a register.
Fixes PR9239
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@126072 91177308-0d34-0410-b5e6-96231b3b80d8
(LLVMX86Utils.a) to break cyclic library dependencies between
LLVMX86CodeGen.a and LLVMX86AsmParser.a. Previously this code was in
a header file and marked static but AVX requires some additional
functionality here that won't be used by all clients. Since including
unused static functions causes a gcc compiler warning, keeping it as a
header would break builds that use -Werror. Putting this in its own
library solves both problems at once.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125765 91177308-0d34-0410-b5e6-96231b3b80d8
have their low bits set to zero. This allows us to optimize
out explicit stack alignment code like in stack-align.ll:test4 when
it is redundant.
Doing this causes the code generator to start turning FI+cst into
FI|cst all over the place, which is general goodness (that is the
canonical form) except that various pieces of the code generator
don't handle OR aggressively. Fix this by introducing a new
SelectionDAG::isBaseWithConstantOffset predicate, and using it
in places that are looking for ADD(X,CST). The ARM backend in
particular was missing a lot of addressing mode folding opportunities
around OR.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125470 91177308-0d34-0410-b5e6-96231b3b80d8
anything but the simplest of cases, lower a 256-bit BUILD_VECTOR by
splitting it into 128-bit parts and recombining.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@125105 91177308-0d34-0410-b5e6-96231b3b80d8
This allows us to easily support 256-bit operations that don't have
native 256-bit support. This applies to integer operations, certain
types of shuffles and various othher things.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124910 91177308-0d34-0410-b5e6-96231b3b80d8
infrastructure. This makes lowering 256-bit vectors to 128-bit
vectors simple when 256-bit vector support is not available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124868 91177308-0d34-0410-b5e6-96231b3b80d8
matching EXTRACT_SUBVECTOR to VEXTRACTF128 along with support routines
to examine and translate index values. VINSERTF128 comes next. With
these two in place we can begin supporting more AVX operations as
INSERT/EXTRACT can be used as a fallback when 256-bit support is not
available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124797 91177308-0d34-0410-b5e6-96231b3b80d8
Reversing the operands allows us to fold, but doesn't force us to. Also, at
this point the DAG is still being optimized, so the check for hasOneUse is not
very precise.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124773 91177308-0d34-0410-b5e6-96231b3b80d8
default implementation for x86, going through the stack in a similr
fashion to how the codegen implements BUILD_VECTOR. Eventually this
will get matched to VINSERTF128 if AVX is available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124307 91177308-0d34-0410-b5e6-96231b3b80d8
implementation of EXTRACT_SUBVECTOR for x86, going through the stack
in a similr fashion to how the codegen implements BUILD_VECTOR.
Eventually this will get matched to VEXTRACTF128 if AVX is available.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@124292 91177308-0d34-0410-b5e6-96231b3b80d8
These functions not longer assert when passed 0, but simply return false instead.
No functional change intended.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@123155 91177308-0d34-0410-b5e6-96231b3b80d8
The theory is it's still faster than a pair of movq / a quad of movl. This
will probably hurt older chips like P4 but should run faster on current
and future Intel processors. rdar://8817010
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122955 91177308-0d34-0410-b5e6-96231b3b80d8
etc. takes an option OptSize. If OptSize is true, it would return
the inline limit for functions with attribute OptSize.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122952 91177308-0d34-0410-b5e6-96231b3b80d8
lowering to use it. Hopefully the pattern fragment is doing the right thing with XMM0, looks correct in testing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122277 91177308-0d34-0410-b5e6-96231b3b80d8
the same as setcc. Optimize ADDC(0,0,FLAGS) -> SET_CARRY(FLAGS). This is
a step towards finishing off PR5443. In the testcase in that bug we now get:
movq %rdi, %rax
addq %rsi, %rax
sbbq %rcx, %rcx
testb $1, %cl
setne %dl
ret
instead of:
movq %rdi, %rax
addq %rsi, %rax
movl $0, %ecx
adcq $0, %rcx
testq %rcx, %rcx
setne %dl
ret
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122219 91177308-0d34-0410-b5e6-96231b3b80d8
their carry depenedencies with MVT::Flag operands) and use clean and beautiful
EFLAGS dependences instead.
We do this by changing the modelling of SBB/ADC to have EFLAGS input and outputs
(which is what requires the previous scheduler change) and change X86 ISelLowering
to custom lower ADDC and friends down to X86ISD::ADD/ADC/SUB/SBB nodes.
With the previous series of changes, this causes no changes in the testsuite, woo.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122213 91177308-0d34-0410-b5e6-96231b3b80d8
consistently by moving it out of lowering into dag combine.
Add some missing patterns for matching away extended versions of setcc_c.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@122201 91177308-0d34-0410-b5e6-96231b3b80d8
the output to the correct register. Fixes a hidden problem uncovered
by the last patch where we'd try to DAG combine our MVT::Other node
oddly.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@121358 91177308-0d34-0410-b5e6-96231b3b80d8