It adds register mask operands to x86 call instructions. Once all the
backend passes support register mask operands, this will be permanently
enabled.
llvm-svn: 148438
When set, this bit indicates that a register is completely defined by
the value of its sub-registers.
Use the CoveredBySubRegs property to infer which super-registers are
call-preserved given a list of callee-saved registers. For example, the
ARM registers D8-D15 are callee-saved. This now automatically implies
that Q4-Q7 are call-preserved.
Conversely, Win64 callees save XMM6-XMM15, but the corresponding
YMM6-YMM15 registers are not call-preserved because they are not fully
defined by their sub-registers.
llvm-svn: 148363
In CanXFormVExtractWithShuffleIntoLoad we assumed that EXTRACT_VECTOR_ELT can be later handled by the DAGCombiner.
However, in some cases on AVX, the EXTRACT_VECTOR_ELT is legalized to EXTRACT_SUBVECTOR + EXTRACT_VECTOR_ELT, which
currently is not handled by the DAGCombiner. In this patch I added a check that we only extract from the XMM part.
llvm-svn: 148298
We know that the blend instructions only use the MSB, so if the mask is
sign-extended then we can convert it into a SHL instruction. This is a
common pattern because the type-legalizer sign-extends the i1 type which
is used by the LLVM-IR for the condition.
Added a new optimization in SimplifyDemandedBits for SIGN_EXTEND_INREG -> SHL.
llvm-svn: 148225
The registers are placed into the saved registers list in the reverse order,
which is why the original loop was written to loop backwards.
llvm-svn: 148064
lc: X86ISelLowering.cpp:6480: llvm::SDValue llvm::X86TargetLowering::LowerVECTOR_SHUFFLE(llvm::SDValue, llvm::SelectionDAG&) const: Assertion `V1.getOpcode() != ISD::UNDEF&& "Op 1 of shuffle should not be undef"' failed.
Added a test.
llvm-svn: 148044
Restore the (obviously wrong) behavior from before r147938 without relying on
undefined behavior. Add a fat FIXME note.
This should fix nightly tester failures.
llvm-svn: 148030
In att style asm syntax memory operand size is derived from suffix attached with mnemonic. In intel style asm syntax it is part of memory operand hence predicate method check is required to select appropriate instruction.
llvm-svn: 148006
same pattern. We already had this pattern is a few places, but others
tried to make a rough approximation of an actual DAG structure. As not
everywhere went to this trouble, nothing could rely on this being done.
In fact, I've checked all references to these node Ids, and the ones
that are using the topo-sort properties are actually satisfied with
a strict-weak-ordering. The requirement appears to be that Use >= Def.
I've added a big blurb of comments to this bit of the transform to
clarify why the order is so important for the next reader of the code.
I'm starting with this change as it is very small, and trivially
reverted if something breaks or the >= above really does need to be >.
If that proves the case, we can hide the problem by reverting this
patch, but the problem exists elsewhere as well, and so a more
comprehensive solution will be needed.
llvm-svn: 148001
hoped this would revive one of the llvm-gcc selfhost build bots, but it
didn't so it doesn't appear that my transform is the culprit.
If anyone else is seeing failures, please let me know!
llvm-svn: 147957
strange build bot failures that look like a miscompile into an infloop.
I'll investigate this tomorrow, but I'd both like to know whether my
patch is the culprit, and get the bots back to green.
llvm-svn: 147945
mask+shift pairs at the beginning of the ISD::AND case block, and then
hoist the final pattern into a helper function, simplifying and
reflowing it appropriately. This should have no observable behavior
change, but several simplifications fell out of this such as directly
computing the new mask constant, etc.
llvm-svn: 147939
extracts and scaled addressing modes into its own helper function. No
functionality changed here, just hoisting and layout fixes falling out
of that hoisting.
llvm-svn: 147937
detect a pattern which can be implemented with a small 'shl' embedded in
the addressing mode scale. This happens in real code as follows:
unsigned x = my_accelerator_table[input >> 11];
Here we have some lookup table that we look into using the high bits of
'input'. Each entity in the table is 4-bytes, which means this
implicitly gets turned into (once lowered out of a GEP):
*(unsigned*)((char*)my_accelerator_table + ((input >> 11) << 2));
The shift right followed by a shift left is canonicalized to a smaller
shift right and masking off the low bits. That hides the shift right
which x86 has an addressing mode designed to support. We now detect
masks of this form, and produce the longer shift right followed by the
proper addressing mode. In addition to saving a (rather large)
instruction, this also reduces stalls in Intel chips on benchmarks I've
measured.
In order for all of this to work, one part of the DAG needs to be
canonicalized *still further* than it currently is. This involves
removing pointless 'trunc' nodes between a zextload and a zext. Without
that, we end up generating spurious masks and hiding the pattern.
llvm-svn: 147936
As the comment around 7746 says, it's better to use the x87 extended precision
here than SSE. And the generic code doesn't know how to do that. It also regains
the speed lost for the uint64_to_float.c testcase.
<rdar://problem/10669858>
llvm-svn: 147869
this substraction will result in small negative numbers at worst which
become very large positive numbers on assignment and are thus caught by
the <=4 check on the next line. The >0 check clearly intended to catch
these as negative numbers.
Spotted by inspection, and impossible to trigger given the shift widths
that can be used.
llvm-svn: 147773
Testing: passed 'make check' including LIT tests for all sequences being handled (both SSE and AVX)
Reviewers: Evan Cheng, David Blaikie, Bruno Lopes, Elena Demikhovsky, Chad Rosier, Anton Korobeynikov
llvm-svn: 147601
This small bit of ASM code is sufficient to do what the old algorithm did:
movq %rax, %xmm0
punpckldq (c0), %xmm0 // c0: (uint4){ 0x43300000U, 0x45300000U, 0U, 0U }
subpd (c1), %xmm0 // c1: (double2){ 0x1.0p52, 0x1.0p52 * 0x1.0p32 }
#ifdef __SSE3__
haddpd %xmm0, %xmm0
#else
pshufd $0x4e, %xmm0, %xmm1
addpd %xmm1, %xmm0
#endif
It's arguably faster. One caveat, the 'haddpd' instruction isn't very fast on
all processors.
<rdar://problem/7719814>
llvm-svn: 147593
(x > y) ? x : y
=>
(x >= y) ? x : y
So for something like
(x - y) > 0 : (x - y) ? 0
It will be
(x - y) >= 0 : (x - y) ? 0
This makes is possible to test sign-bit and eliminate a comparison against
zero. e.g.
subl %esi, %edi
testl %edi, %edi
movl $0, %eax
cmovgl %edi, %eax
=>
xorl %eax, %eax
subl %esi, $edi
cmovsl %eax, %edi
rdar://10633221
llvm-svn: 147512
LZCNT instructions are available. Force promotion to i32 to get
a smaller encoding since the fix-ups necessary are just as complex for
either promoted type
We can't do standard promotion for CTLZ when lowering through BSR
because it results in poor code surrounding the 'xor' at the end of this
instruction. Essentially, if we promote the entire CTLZ node to i32, we
end up doing the xor on a 32-bit CTLZ implementation, and then
subtracting appropriately to get back to an i8 value. Instead, our
custom logic just uses the knowledge of the incoming size to compute
a perfect xor. I'd love to know of a way to fix this, but so far I'm
drawing a blank. I suspect the legalizer could be more clever and/or it
could collude with the DAG combiner, but how... ;]
llvm-svn: 147251
'bsf' instructions here.
This one is actually debatable to my eyes. It's not clear that any chip
implementing 'tzcnt' would have a slow 'bsf' for any reason, and unless
EFLAGS or a zero input matters, 'tzcnt' is just a longer encoding.
Still, this restores the old behavior with 'tzcnt' enabled for now.
llvm-svn: 147246
X86ISelLowering C++ code. Because this is lowered via an xor wrapped
around a bsr, we want the dagcombine which runs after isel lowering to
have a chance to clean things up. In particular, it is very common to
see code which looks like:
(sizeof(x)*8 - 1) ^ __builtin_clz(x)
Which is trying to compute the most significant bit of 'x'. That's
actually the value computed directly by the 'bsr' instruction, but if we
match it too late, we'll get completely redundant xor instructions.
The more naive code for the above (subtracting rather than using an xor)
still isn't handled correctly due to the dagcombine getting confused.
Also, while here fix an issue spotted by inspection: we should have been
expanding the zero-undef variants to the normal variants when there is
an 'lzcnt' instruction. Do so, and test for this. We don't want to
generate unnecessary 'bsr' instructions.
These two changes fix some regressions in encoding and decoding
benchmarks. However, there is still a *lot* to be improve on in this
type of code.
llvm-svn: 147244
use the zero-undefined variants of CTTZ and CTLZ. These are just simple
patterns for now, there is more to be done to make real world code using
these constructs be optimized and codegen'ed properly on X86.
The existing tests are spiffed up to check that we no longer generate
unnecessary cmov instructions, and that we generate the very important
'xor' to transform bsr which counts the index of the most significant
one bit to the number of leading (most significant) zero bits. Also they
now check that when the variant with defined zero result is used, the
cmov is still produced.
llvm-svn: 146974
Use information computed while inferring new register classes to emit
accurate, table-driven implementations of getMatchingSuperRegClass().
Delete the old manual, error-prone implementations in the targets.
llvm-svn: 146873
the compact unwind claiming that one register was saved before another, which
isn't all that great in general. Process them in the natural order. Reverse the
list only when necessary for the algorithm.
llvm-svn: 146612
to finalize MI bundles (i.e. add BUNDLE instruction and computing register def
and use lists of the BUNDLE instruction) and a pass to unpack bundles.
- Teach more of MachineBasic and MachineInstr methods to be bundle aware.
- Switch Thumb2 IT block to MI bundles and delete the hazard recognizer hack to
prevent IT blocks from being broken apart.
llvm-svn: 146542
undefined result. This adds new ISD nodes for the new semantics,
selecting them when the LLVM intrinsic indicates that the undef behavior
is desired. The new nodes expand trivially to the old nodes, so targets
don't actually need to do anything to support these new nodes besides
indicating that they should be expanded. I've done this for all the
operand types that I could figure out for all the targets. Owners of
various targets, please review and let me know if any of these are
incorrect.
Note that the expand behavior is *conservatively correct*, and exactly
matches LLVM's current behavior with these operations. Ideally this
patch will not change behavior in any way. For example the regtest suite
finds the exact same instruction sequences coming out of the code
generator. That's why there are no new tests here -- all of this is
being exercised by the existing test suite.
Thanks to Duncan Sands for reviewing the various bits of this patch and
helping me get the wrinkles ironed out with expanding for each target.
Also thanks to Chris for clarifying through all the discussions that
this is indeed the approach he was looking for. That said, there are
likely still rough spots. Further review much appreciated.
llvm-svn: 146466
subdirectories to traverse into.
- Originally I wanted to avoid this and just autoscan, but this has one key
flaw in that new subdirectories can not automatically trigger a rerun of the
llvm-build tool. This is particularly a pain when switching back and forth
between trees where one has added a subdirectory, as the dependencies will
tend to be wrong. This will also eliminates FIXME implicitly.
llvm-svn: 146436
does. The _GLOBAL_OFFSET_TABLE_ is still magical in that we get a R_386_GOTPC,
but it doesn't change the immediate in the same way as when the expression
has no right hand side symbol.
llvm-svn: 146311
if (HasAVX)
X86SSELevel = NoMMXSSE;
This is so patterns that are predicated on hasSSE3, etc. would not be selected when avx is available. Instead, the AVX variant is selected.
However, this breaks instructions which do not have AVX variants.
The right way to fix this is for the SSE but not-AVX patterns to predicate on something like hasSSE3() && !hasAVX().
Then we can take out the hack in X86Subtarget.cpp. Patterns which do not have AVX variants do not need to change.
However, we need to audit all the patterns before we make the change. This patch is workaround that fixes one specific case,
the prefetch instructions. rdar://10538297
llvm-svn: 146163
generator to it. For non-bundle instructions, these behave exactly the same
as the MC layer API.
For properties like mayLoad / mayStore, look into the bundle and if any of the
bundled instructions has the property it would return true.
For properties like isPredicable, only return true if *all* of the bundled
instructions have the property.
For properties like canFoldAsLoad, isCompare, conservatively return false for
bundles.
llvm-svn: 146026
This was actually a bit of a mess. TLI.setPrefLoopAlignment was clearly
documented as taking log2(bytes) units, but the x86 target would still
set a preferred loop alignment of '16'.
CodePlacementOpt passed this number on to the basic block, and
AsmPrinter interpreted it as bytes.
Now both MachineFunction and MachineBasicBlock use logarithmic
alignments.
Obviously, MachineConstantPool still measures alignments in bytes, so we
can emulate the thrill of using as.
llvm-svn: 145889
Whether a fixup needs relaxation for the associated instruction is a
target-specific function, as the FIXME indicated. Create a hook for that
and use it.
llvm-svn: 145881
libgcc sets the stack limit field in TCB to 256 bytes above the actual
allocated stack limit. This means if the function's stack frame needs
less than 256 bytes, we can just compare the stack pointer with the
stack limit. This should result in lesser calls to __morestack.
llvm-svn: 145766
Currently LLVM pads the call to __morestack with a add and sub of 8
bytes to esp. This isn't correct since __morestack expects the call
to be followed directly by a ret.
This commit also adjusts the relevant test-case.
llvm-svn: 145765
the same value) to this variable. This code could be refactored, but it doesn't
matter since the old JIT is going away. Add tsan annotations to ignore the
race.
llvm-svn: 145745
change, now you need a TargetOptions object to create a TargetMachine. Clang
patch to follow.
One small functionality change in PTX. PTX had commented out the machine
verifier parts in their copy of printAndVerify. That now calls the version in
LLVMTargetMachine. Users of PTX who need verification disabled should rely on
not passing the command-line flag to enable it.
llvm-svn: 145714
Like V_SET0, these instructions are expanded by ExpandPostRA to xorps /
vxorps so they can participate in execution domain swizzling.
This also makes the AVX variants redundant.
llvm-svn: 145440
as MC is the only assembler we support.
This splits MS/Windows and GNU/Windows ASM infos into two seperate classes.
While there is currently only one difference, full MS C++ ABI support will
require many more.
llvm-svn: 145409
Before:
movabsq $4294967296, %rax ## encoding: [0x48,0xb8,0x00,0x00,0x00,0x00,0x01,0x00,0x00,0x00]
testq %rax, %rdi ## encoding: [0x48,0x85,0xf8]
jne LBB0_2 ## encoding: [0x75,A]
After:
btq $32, %rdi ## encoding: [0x48,0x0f,0xba,0xe7,0x20]
jb LBB0_2 ## encoding: [0x72,A]
btq is usually slower than testq because it doesn't fuse with the jump, but here we're better off
saving one register and a giant movabsq.
llvm-svn: 145103
This was a bug in keeping track of the available domains when merging
domain values.
The wrong domain mask caused ExecutionDepsFix to try to move VANDPSYrr
to the integer domain which is only available in AVX2.
Also add an assertion to catch future attempts at emitting AVX2
instructions.
llvm-svn: 145096
and code model. This eliminates the need to pass OptLevel flag all over the
place and makes it possible for any codegen pass to use this information.
llvm-svn: 144788
Two new TargetInstrInfo hooks lets the target tell ExecutionDepsFix
about instructions with partial register updates causing false unwanted
dependencies.
The ExecutionDepsFix pass will break the false dependencies if the
updated register was written in the previoius N instructions.
The small loop added to sse-domains.ll runs twice as fast with
dependency-breaking instructions inserted.
llvm-svn: 144602
handle defining the "magic" target related components (like native,
nativecodegen, and engine).
- We still require these components to be in the project (currently in
lib/Target) so that we have a place to document them and hopefully make it
more obvious that they are "magic".
llvm-svn: 144253
When this field is true it means that the load is from constant (runt-time or compile-time) and so can be hoisted from loops or moved around other memory accesses
llvm-svn: 144100
The xorps instruction is smaller than pxor, so prefer that encoding.
The ExecutionDepsFix pass will switch the encoding to pxor and xorpd
when appropriate.
llvm-svn: 143996
On x86: (shl V, 1) -> add V,V
Hardware support for vector-shift is sparse and in many cases we scalarize the
result. Additionally, on sandybridge padd is faster than shl.
llvm-svn: 143311
fixes: Use a separate register, instead of SP, as the
calling-convention resource, to avoid spurious conflicts with
actual uses of SP. Also, fix unscheduling of calling sequences,
which can be triggered by pseudo-two-address dependencies.
llvm-svn: 143206
it fixes the dragonegg self-host (it looks like gcc is miscompiled).
Original commit messages:
Eliminate LegalizeOps' LegalizedNodes map and have it just call RAUW
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.
Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.
Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.
Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.
This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.
Delete #if 0 code accidentally left in.
llvm-svn: 143188
on every node as it legalizes them. This makes it easier to use
hasOneUse() heuristics, since unneeded nodes can be removed from the
DAG earlier.
Make LegalizeOps visit the DAG in an operands-last order. It previously
used operands-first, because LegalizeTypes has to go operands-first, and
LegalizeTypes used to be part of LegalizeOps, but they're now split.
The operands-last order is more natural for several legalization tasks.
For example, it allows lowering code for nodes with floating-point or
vector constants to see those constants directly instead of seeing the
lowered form (often constant-pool loads). This makes some things
somewhat more complicated today, though it ought to allow things to be
simpler in the future. It also fixes some bugs exposed by Legalizing
using RAUW aggressively.
Remove the part of LegalizeOps that attempted to patch up invalid chain
operands on libcalls generated by LegalizeTypes, since it doesn't work
with the new LegalizeOps traversal order. Instead, define what
LegalizeTypes is doing to be correct, and transfer the responsibility
of keeping calls from having overlapping calling sequences into the
scheduler.
Teach the scheduler to model callseq_begin/end pairs as having a
physical register definition/use to prevent calls from having
overlapping calling sequences. This is also somewhat complicated, though
there are ways it might be simplified in the future.
This addresses rdar://9816668, rdar://10043614, rdar://8434668, and others.
Please direct high-level questions about this patch to management.
llvm-svn: 143177
not depend on In32BitMode. Use the sysexitq mnemonic for the version with the
REX.W prefix and only allow it only In64BitMode. rdar://9738584
llvm-svn: 143112
MORESTACK_RET_RESTORE_R10; which are lowered to a RET and a RET
followed by a MOV respectively. Having a fake instruction prevents
the verifier from seeing a MachineBasicBlock end with a
non-terminator (MOV). It also prevents the rather eccentric case of a
MachineBasicBlock ending with RET but having successors nevertheless.
Patch by Sanjoy Das.
llvm-svn: 143062
the X86 asmparser to produce ranges in the one case that was annoying me, for example:
test.s:10:15: error: invalid operand for instruction
movl 0(%rax), 0(%edx)
^~~~~~~
It should be straight-forward to enhance filecheck, tblgen, and/or the .ll parser to use
ranges where appropriate if someone is interested.
llvm-svn: 142106
TableGen infers unmodeled side effects on instructions without a
pattern. Fix some instruction definitions where that was overlooked.
Also raise an error if a rematerializable instruction has unmodeled side
effects. That doen't make any sense.
llvm-svn: 141929
TableGen will mark any pattern-less instruction as having unmodeled side
effects. This is extra bad for V_SET0 which gets rematerialized a lot.
This was part of the cause for PR11125, but the real bug was fixed
in r141923.
llvm-svn: 141924
http://lab.llvm.org:8011/builders/llvm-x86_64-linux/builds/101
--- Reverse-merging r141854 into '.':
U test/MC/Disassembler/X86/x86-32.txt
U test/MC/Disassembler/X86/simple-tests.txt
D test/CodeGen/X86/bmi.ll
U lib/Target/X86/X86InstrInfo.td
U lib/Target/X86/X86ISelLowering.cpp
U lib/Target/X86/X86.td
U lib/Target/X86/X86Subtarget.h
llvm-svn: 141857
promoting allocas to preferred alignments that exceed the natural
alignment. This avoids some potentially expensive dynamic stack realignments.
The natural stack alignment is set in target data strings via the "S<size>"
option. Size is in bits and must be a multiple of 8. The natural stack alignment
defaults to "unspecified" (represented by a zero value), and the "unspecified"
value does not prevent any alignment promotions. Target maintainers that care
about avoiding promotions should explicitly add the "S<size>" option to their
target data strings.
llvm-svn: 141599
A GR8_NOREX virtual register is created when extrating a sub_8bit_hi
sub-register:
%vreg2<def> = COPY %vreg1:sub_8bit_hi; GR8_NOREX:%vreg2 %GR64_ABCD:%vreg1
TEST8ri_NOREX %vreg2, 1, %EFLAGS<imp-def>; GR8_NOREX:%vreg2
If such a live range is ever split, its register class must not be
inflated to GR8. The sub-register copy can only target GR8_NOREX.
I dont have a test case for this theoretical bug.
llvm-svn: 141500
In 64-bit mode, sub_8bit_hi sub-registers can only be used by NOREX
instructions. The COPY created from the EXTRACT_SUBREG DAG node cannot
target all GR8 registers, only those in GR8_NOREX.
TO enforce this, we ensure that all instructions using the
EXTRACT_SUBREG are GR8_NOREX constrained.
This fixes PR11088.
llvm-svn: 141499
This instruction is explicitly encoded without an REX prefix, so both
operands but be *_NOREX.
Also add an assertion to copyPhysReg() that fires when the MOV8rr_NOREX
constraints are not satisfied.
This fixes a miscompilation in 20040709-2 in the gcc test suite.
llvm-svn: 141410
There are fewer registers with sub_8bit sub-registers in 32-bit mode
than in 64-bit mode. In 32-bit mode, sub_8bit behaves the same as
sub_8bit_hi.
llvm-svn: 141206
This uses less memory and it reduces the complexity of sub-class
operations:
- hasSubClassEq() and friends become O(1) instead of O(N).
- getCommonSubClass() becomes O(N) instead of O(N^2).
In the future, TableGen will infer register classes. This makes it
cheap to add them.
llvm-svn: 140898
This also enables domain swizzling for AVX code which required a few
trivial test changes.
The pass will be moved to lib/CodeGen shortly.
llvm-svn: 140659
I am going to unify the SSEDomainFix and NEONMoveFix passes into a
single target independent pass. They are essentially doing the same
thing.
llvm-svn: 140652
We already support GR64 <-> VR128 copies. All of these copies break
partial register dependencies by zeroing the high part of the target
register.
llvm-svn: 140348
- x87: no min or max.
- SSE1: min/max for single precision scalars and vectors.
- SSE2: min/max for single and double precision scalars and vectors.
- AVX: as SSE2, but also supports the wider ymm vectors. (this is covered by the isTypeLegal check)
llvm-svn: 140296
dag-combine optimization to implement the ext-load efficiently (using shuffles).
For example the type <4 x i8> is stored in memory as i32, but it needs to
find its way into a <4 x i32> register. Previously we scalarized the memory
access, now we use shuffles.
llvm-svn: 139995
maxps and maxpd). This broke the sse41-blend.ll testcase by causing
maxpd to be produced rather than a cmp+blend pair, which is the reason
I tweaked it. Gives a small speedup on doduc with dragonegg when the
GCC vectorizer is used.
llvm-svn: 139986
are declared with load patterns. This fix the crash in PR10941. No testcases,
since a fold is triggered and then converted back to the register form
afterwards.
llvm-svn: 139953
This PR basically reports a problem where a crash in generated code
happened due to %rbp being clobbered:
pushq %rbp
movq %rsp, %rbp
....
vmovmskps %ymm12, %ebp
....
movq %rbp, %rsp
popq %rbp
ret
Since Eric's r123367 commit, the default stack alignment for x86 32-bit
has changed to be 16-bytes. Since then, the MaxStackAlignmentHeuristicPass
hasn't been really used, but with AVX it becomes useful again, since per
ABI compliance we don't always align the stack to 256-bit, but only when
there are 256-bit incoming arguments.
ReserveFP was only used by this pass, but there's no RA target hook that
uses getReserveFP() to check for the presence of FP (since nothing was
triggering the pass to run, the uses of getReserveFP() were removed
through time without being noticed). Change this pass to use
setForceFramePointer, which is properly called by MachineFunction
hasFP method.
The testcase is very big and dependent on RA, not sure if it's worth
adding to test/CodeGen/X86.
llvm-svn: 139939
take into consideration the presence of AVX. This change, together with
the SSEDomainFix enabled for AVX, makes AVX codegen to always (hopefully)
emit the same code as SSE for 128-bit vector ops. I don't
have a testcase for this, but AVX now beats SSE in performance for
128-bit ops in the majority of programas in the llvm testsuite
llvm-svn: 139817
alignment check for 256-bit classes more strict. There're no testcases
but we catch more folding cases for AVX while running single and multi
sources in the llvm testsuite.
Since some 128-bit AVX instructions have different number of operands
than their SSE counterparts, they are placed in different tables.
256-bit AVX instructions should also be added in the table soon. And
there a few more 128-bit versions to handled, which should come in
the following commits.
llvm-svn: 139687
more strict about the alignment checking. This was found by inspection
and I don't have any testcases so far, although the llvm testsuite runs
without any problem.
llvm-svn: 139625
However with this fix it does now.
Basically the operand order for the x86 target specific node
is not the same as the instruction, but since the intrinsic need that
specific order at the instruction definition, just change the order
during legalization. Also, there were some wrong invertions of condition
codes, such as GE => LE, GT => LT, fix that too. Fix PR10907.
llvm-svn: 139528
assert("not implemented for target shuffle node");
to:
assert(0 && "not implemented for target shuffle node");
This causes a test failure in CodeGen/X86/palignr.ll which has
been marked as XFAIL for the time being.
Test failure filed at PR10901.
llvm-svn: 139454
single field (Flags), which is a bitwise OR of items from the TB_*
enum. This makes it easier to add new information in the future.
* Gives every static array an equivalent layout: { RegOp, MemOp, Flags }
* Adds a helper function, AddTableEntry, to avoid duplication of the
insertion code.
* Renames TB_NOT_REVERSABLE to TB_NO_REVERSE.
* Adds TB_NO_FORWARD, which is analogous to TB_NO_REVERSE, except that
it prevents addition of the Reg->Mem entry. (This is going to be used
by Native Client, in the next CL).
Patch by David Meyer
llvm-svn: 139311
in Nadav's r139285 and r139287 commits.
1) Rename vsel.ll to a more descriptive name
2) Change the order of BLEND operands to "Op1, Op2, Cond", this is
necessary because PBLENDVB is already used in different places with
this order, and it was being emitted in the wrong way for vselect
3) Add AVX patterns and tests for the same SSE41 instructions
llvm-svn: 139305
with a vector condition); such selects become VSELECT codegen nodes.
This patch also removes VSETCC codegen nodes, unifying them with SETCC
nodes (codegen was actually often using SETCC for vector SETCC already).
This ensures that various DAG combiner optimizations kick in for vector
comparisons. Passes dragonegg bootstrap with no testsuite regressions
(nightly testsuite as well as "make check-all"). Patch mostly by
Nadav Rotem.
llvm-svn: 139159
init.trampoline and adjust.trampoline intrinsics, into two intrinsics
like in GCC. While having one combined intrinsic is tempting, it is
not natural because typically the trampoline initialization needs to
be done in one function, and the result of adjust trampoline is needed
in a different (nested) function. To get around this llvm-gcc hacks the
nested function lowering code to insert an additional parent variable
holding the adjust.trampoline result that can be accessed from the child
function. Dragonegg doesn't have the luxury of tweaking GCC code, so it
stored the result of adjust.trampoline in the memory GCC set aside for
the trampoline itself (this is always available in the child function),
and set up some new memory (using an alloca) to hold the trampoline.
Unfortunately this breaks Go which allocates trampoline memory on the
heap and wants to use it even after the parent has exited (!). Rather
than doing even more hacks to get Go working, it seemed best to just use
two intrinsics like in GCC. Patch mostly by Sanjoy Das.
llvm-svn: 139140
instructions are more aligned than the CPU requires, and adds some additional
directives, to follow in future patches. Patch by David Meyer!
llvm-svn: 139125
The explanation about a 0 argument being materialized as xor is no
longer valid. Rematerialization will check if EFLAGS is live before
clobbering it.
The code produced by X86TargetLowering::EmitLoweredSelect does not
clobber EFLAGS.
This causes one less testb instruction to be generated in the cmov.ll
test case.
llvm-svn: 139057
- Duplicate some store patterns to their AVX forms!
- Catched a bug while restricting the patterns subtarget, fix it
and update a testcase to check it properly
llvm-svn: 138851
code is inserted to first check if the current stacklet has enough
space. If so, space is allocated by simply decrementing the stack
pointer. Otherwise a runtime routine (__morestack_allocate_stack_space
in libgcc) is called which allocates the required memory from the
heap.
Patch by Sanjoy Das.
llvm-svn: 138818
from DYNAMIC_STACKALLOC.
Two new pseudo instructions (SEG_ALLOCA_32 and SEG_ALLOCA_64) which
will match X86SegAlloca (based on word size) are also added. They
will be custom emitted to inject the actual stack handling code.
Patch by Sanjoy Das.
llvm-svn: 138814
X86. Modify the pass added in the previous patch to call this new
code.
This new prologues generated will call a libgcc routine (__morestack)
to allocate more stack space from the heap when required
Patch by Sanjoy Das.
llvm-svn: 138812
explicit about which subtarget they refer to, and add AVX versions of
the ones we currently don't. Make the mask check more strict, to be
clear it won't be used to match to 256-bit versions!
llvm-svn: 138514
SSE transition penalty. The pass is enabled through the "x86-use-vzeroupper"
llc command line option. This is only the first step (very naive and
conservative one) to sketch out the idea, but proper DFA is coming next
to allow smarter decisions. Comments and ideas now and in further commits
will be very appreciated.
llvm-svn: 138317
instead of 2. They were already defined this way in their regular
version, but not for the intrinsics versions (*_Int), and that would work
for assembly emission but not for object code, since a MachineOperand
would be missing. This commit fix PR10697.
Also removed the {VSQRT,VRSQRT,VRCP}r_Int forms and match the intrinsic
via INSERT_SUBREG+EXTRACT_SUBREG patterns. The same couldn't be done for
memory versions because sse_load_f32/sse_load_f64 operand need special
handling and don't work like regular "addr" operands.
There are right now 114 "*_Int" and 98 "Int_*" forms! I'm slowly
removing them as I step through, but hope we can get rid of these
someday, they are really annoying :)
llvm-svn: 138012
match splats in the form (splat (scalar_to_vector (load ...))) whenever
the load can be folded. All the logic and instruction emission is
working but because of PR8156, there are no ways to match loads, cause
they can never be folded for splats. Thus, the tests are XFAILed, but
I've tested and exercised all the logic using a relaxed version for
checking the foldable loads, as if the bug was already fixed. This
should work out of the box once PR8156 gets fixed since MayFoldLoad will
work as expected.
llvm-svn: 137810
vinsertf128 $1 + vpermilps $0, remove the old code that used to first
do the splat in a 128-bit vector and then insert it into a larger one.
This is better because the handling code gets simpler and also makes a
better room for the upcoming vbroadcast!
llvm-svn: 137807
there is no support for native 256-bit shuffles, be more smart in some
cases, for example, when you can extract specific 128-bit parts and use
regular 128-bit shuffles for them. Example:
For this shuffle:
shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32>
<i32 1, i32 0, i32 7, i32 6>
This was expanded to:
vextractf128 $1, %ymm1, %xmm2
vpextrq $0, %xmm2, %rax
vmovd %rax, %xmm1
vpextrq $1, %xmm2, %rax
vmovd %rax, %xmm2
vpunpcklqdq %xmm1, %xmm2, %xmm1
vpextrq $0, %xmm0, %rax
vmovd %rax, %xmm2
vpextrq $1, %xmm0, %rax
vmovd %rax, %xmm0
vpunpcklqdq %xmm2, %xmm0, %xmm0
vinsertf128 $1, %xmm1, %ymm0, %ymm0
ret
Now we get:
vshufpd $1, %xmm0, %xmm0, %xmm0
vextractf128 $1, %ymm1, %xmm1
vshufpd $1, %xmm1, %xmm1, %xmm1
vinsertf128 $1, %xmm1, %ymm0, %ymm0
llvm-svn: 137733
Allow a target assembly parser to do context sensitive constraint checking
on a potential instruction match. This will be used, for example, to handle
Thumb2 IT block parsing.
llvm-svn: 137675
vectors. It operates on 128-bit elements instead of regular scalar
types. Recognize shuffles that are suitable for VPERM2F128 and teach
the x86 legalizer how to handle them.
llvm-svn: 137519
(for example, after integer operation), do not pack the registers into a YMM
before saving. Its better to save as two XMM registers.
Before:
vinsertf128 $1, %xmm3, %ymm0, %ymm3
vinsertf128 $0, %xmm1, %ymm3, %ymm1
vmovaps %ymm1, 416(%rsp)
After:
vmovaps %xmm3, 416+16(%rsp)
vmovaps %xmm1, 416(%rsp)
llvm-svn: 137308
data in-register prior to saving to memory. When we reorder the data in memory
we prevent the need to save multiple scalars to memory, making a single regular
store.
llvm-svn: 137238
def : Pat<(X86Movss VR128:$src1,
(bc_v4i32 (v2i64 (load addr:$src2)))),
(MOVLPSrm VR128:$src1, addr:$src2)>;
This matches a MOVSS dag with a MOVLPS instruction. However, MOVSS will replace only the low 32 bits of the register, while the MOVLPS instruction will replace the low 64 bits. A testcase is added and illustrates the bug and also modified the one that was already present. Patch by Tanya Lattner.
llvm-svn: 137227
X86FloatingPoint keeps track of pending ST registers for an upcoming
inline asm instruction with fixed stack register constraints. It does
this by remembering which FP register holds the value that should appear
at a fixed stack position for the inline asm.
When that FP register is killed before the inline asm, make sure to
duplicate it to a scratch register, so the ST register still has a live
FP reference.
This could happen when the same FP register was copied to two ST
registers, or when a spill instruction is inserted between the ST copy
and the inline asm.
This fixes PR10602.
llvm-svn: 137050
avoid returning early for v8i32 types, which would only be valid for
vector with all zeros. Also split the handling of zeros and ones into separate
checking logic since they are handled differently. This fixes PR10547
llvm-svn: 136642
working on x86 (at least for trivial testcases); other architectures will
need more work so that they actually emit the appropriate instructions for
orderings stricter than 'monotonic'. (As far as I can tell, the ARM, PPC,
Mips, and Alpha backends need such changes.)
llvm-svn: 136457
Craig Topper