- After moving logic recognizing vector shift with scalar amount from
DAG combining into DAG lowering, we declare to customize all vector
shifts even vector shift on AVX is legal. As a result, the cost model
needs special tuning to identify these legal cases.
llvm-svn: 177586
an X86Operand, but also performs a Sema lookup and adds the sizing directive
when appropriate. Use this when parsing a bracketed statement. This is
necessary to get the instruction matching correct as well. Test case coming
on clang side.
rdar://13455408
llvm-svn: 177439
def : Pat<(load (i64 (X86Wrapper tglobaltlsaddr :$dst))),
(MOV64rm tglobaltlsaddr :$dst)>;
This pattern is invalid because the MOV64rm instruction expects a
source operand of type "i64mem", which is a subclass of X86MemOperand
and thus actually consists of five MI operands, but the Pat provides
only a single MI operand ("tglobaltlsaddr" matches an SDnode of
type ISD::TargetGlobalTLSAddress and provides a single output).
Thus, if the pattern were ever matched, subsequent uses of the MOV64rm
instruction pattern would access uninitialized memory. In addition,
with the TableGen patch I'm about to check in, this would actually be
reported as a build-time error.
Fortunately, the pattern does in fact never match, for at least two
independent reasons.
First, the code generator actually never generates a pattern of the
form (load (X86Wrapper (tglobaltlsaddr))). For most combinations of
TLS and code models, (tglobaltlsaddr) represents just an offset that
needs to be added to some base register, so it is never directly
dereferenced. The only exception is the initial-exec model, where
(tglobaltlsaddr) refers to the (pc-relative) address of a GOT slot,
which *is* in fact directly dereferenced: but in that case, the
X86WrapperRIP node is used, not X86Wrapper, so the Pat doesn't match.
Second, even if some patterns along those lines *were* ever generated,
we should not need an extra Pat pattern to match it. Instead, the
original MOV64rm instruction pattern ought to match directly, since
it uses an "addr" operand, which is implemented via the SelectAddr
C++ routine; this routine is supposed to accept the full range of
input DAGs that may be implemented by a single mov instruction,
including those cases involving ISD::TargetGlobalTLSAddress (and
actually does so e.g. in the initial-exec case as above).
To avoid build breaks (due to the above-mentioned error) after the
TableGen patch is checked in, I'm removing this Pat here.
llvm-svn: 177426
We hitch a ride with the existing OpndItins class that was used to add
instruction itinerary classes in the many multiclasses in this file.
Use the link provided by the X86FoldableSchedWrite.Folded to find the
right SchedWrite for folded loads.
llvm-svn: 177326
This new-style scheduling information is going to replace the
instruction iteneraries.
This also serves as a test case for Andy's fix in r177317.
llvm-svn: 177323
MinGW is almost completely compatible to MSVC, with the exception of the _tls_array global not being available.
Patch by David Nadlinger!
llvm-svn: 177257
Since almost all X86 instructions can fold loads, use a multiclass to
define register/memory pairs of SchedWrites.
An X86FoldableSchedWrite represents the register version of an
instruction. It holds a reference to the SchedWrite to use when the
instruction folds a load.
This will be used inside multiclasses that define rr and rm instruction
versions together.
llvm-svn: 177210
The new InstrSchedModel is easier to use than the instruction
itineraries. It will be used to model instruction latency and throughput
in modern Intel microarchitectures like Sandy Bridge.
InstrSchedModel should be able to coexist with instruction itinerary
classes, but for cleanliness we should switch the Atom processor model
to the new InstrSchedModel as well.
llvm-svn: 177122
LegalizeDAG.cpp uses the value of the comparison operands when checking
the legality of BR_CC, so DAGCombiner should do the same.
v2:
- Expand more BR_CC value types for NVPTX
v3:
- Expand correct BR_CC value types for Hexagon, Mips, and XCore.
llvm-svn: 176694
That can usually be lowered efficiently and is common in sandybridge code.
It would be nice to do this in DAGCombiner but we can't insert arbitrary
BUILD_VECTORs this late.
Fixes PR15462.
llvm-svn: 176634
- Phi nodes should be replaced/updated after lowering CMOV into branch
because 'mainMBB' updating operand in Phi node is changed.
- Add EFLAGS in livein before lowering the 2nd CMOV. It's necessary as
we will reuse the EFLAGS generated before the 1st lowered CMOV, which
won't clobber EFLAGS. However, we need explicitly specify that.
- '-attr=-cmov' test case are added.
llvm-svn: 176598
- Clear 'mayStore' flag when loading from the atomic variable before the
spin loop
- Clear kill flag from one use to multiple use in registers forming the
address to that atomic variable
- don't use a physical register as live-in register in BB (neither entry
nor landing pad.) by copying it into virtual register
(patch by Cameron Zwarich)
llvm-svn: 176538
one-byte NOPs. If the processor actually executes those NOPs, as it sometimes
does with aligned bundling, this can have a performance impact. From my
micro-benchmarks run on my one machine, a 15-byte NOP followed by twelve
one-byte NOPs is about 20% worse than a 15 followed by a 12. This patch
changes NOP emission to emit as many 15-byte (the maximum) as possible followed
by at most one shorter NOP.
llvm-svn: 176464
* Only apply divide bypass optimization when not optimizing for size.
* Fixed bug caused by constant for 0 value of type Int32,
used dividend type to generate the constant instead.
* For atom x86-64 apply the divide bypass to use 16-bit divides instead of
64-bit divides when operand values are small enough.
* Added lit tests for 64-bit divide bypass.
Patch by Tyler Nowicki!
llvm-svn: 176442
This matters for example in following matrix multiply:
int **mmult(int rows, int cols, int **m1, int **m2, int **m3) {
int i, j, k, val;
for (i=0; i<rows; i++) {
for (j=0; j<cols; j++) {
val = 0;
for (k=0; k<cols; k++) {
val += m1[i][k] * m2[k][j];
}
m3[i][j] = val;
}
}
return(m3);
}
Taken from the test-suite benchmark Shootout.
We estimate the cost of the multiply to be 2 while we generate 9 instructions
for it and end up being quite a bit slower than the scalar version (48% on my
machine).
Also, properly differentiate between avx1 and avx2. On avx-1 we still split the
vector into 2 128bits and handle the subvector muls like above with 9
instructions.
Only on avx-2 will we have a cost of 9 for v4i64.
I changed the test case in test/Transforms/LoopVectorize/X86/avx1.ll to use an
add instead of a mul because with a mul we now no longer vectorize. I did
verify that the mul would be indeed more expensive when vectorized with 3
kernels:
for (i ...)
r += a[i] * 3;
for (i ...)
m1[i] = m1[i] * 3; // This matches the test case in avx1.ll
and a matrix multiply.
In each case the vectorized version was considerably slower.
radar://13304919
llvm-svn: 176403
- ISD::SHL/SRL/SRA must have either both scalar or both vector operands
but TLI.getShiftAmountTy() so far only return scalar type. As a
result, backend logic assuming that breaks.
- Rename the original TLI.getShiftAmountTy() to
TLI.getScalarShiftAmountTy() and re-define TLI.getShiftAmountTy() to
return target-specificed scalar type or the same vector type as the
1st operand.
- Fix most TICG logic assuming TLI.getShiftAmountTy() a simple scalar
type.
llvm-svn: 176364
fewer scalar integer (i32 or i64) arguments. It completely eliminates the need
for SDISel for trivial functions.
Also, add the new llc -fast-isel-abort-args option, which is similar to
-fast-isel-abort option, but for formal argument lowering.
llvm-svn: 176052
to TargetFrameLowering, where it belongs. Incidentally, this allows us
to delete some duplicated (and slightly different!) code in TRI.
There are potentially other layering problems that can be cleaned up
as a result, or in a similar manner.
The refactoring was OK'd by Anton Korobeynikov on llvmdev.
Note: this touches the target interfaces, so out-of-tree targets may
be affected.
llvm-svn: 175788
exists solely to enable it to call itself for i8 with some registers.
The proposed patch simplifies the function somewhat to make the High
bit only meaningful for the i8 mode, which makes sense. No functional
difference (getX86SubSuperRegister is not getting called from anywhere
outside with i64 and High=true).
llvm-svn: 175762
sext <4 x i1> to <4 x i64>
sext <4 x i8> to <4 x i64>
sext <4 x i16> to <4 x i64>
I'm running Combine on SIGN_EXTEND_IN_REG and revert SEXT patterns:
(sext_in_reg (v4i64 anyext (v4i32 x )), ExtraVT) -> (v4i64 sext (v4i32 sext_in_reg (v4i32 x , ExtraVT)))
The sext_in_reg (v4i32 x) may be lowered to shl+sar operations.
The "sar" does not exist on 64-bit operation, so lowering sext_in_reg (v4i64 x) has no vector solution.
I also added a cost of this operations to the AVX costs table.
llvm-svn: 175619
MS-style inline assembly.
This is a follow-on to r175334. Forcing a FP to be emitted doesn't ensure it
will be used. Therefore, force the base pointer as well. We now treat MS
inline assembly in the same way we treat functions with dynamic stack
realignment and VLAs. This guarantees the BP will be used to reference
parameters and locals.
rdar://13218191
llvm-svn: 175576
If the frame pointer is omitted, and any stack changes occur in the inline
assembly, e.g.: "pusha", then any C local variable or C argument references
will be incorrect.
I pass no judgement on anyone who would do such a thing. ;)
rdar://13218191
llvm-svn: 175334
If two functions require different features (e.g., `-mno-sse' vs. `-msse') then
we want to honor that, especially during LTO. We can do that by resetting the
subtarget's features depending upon the 'target-feature' attribute.
llvm-svn: 175314
blocks. We still don't have consensus if we should try to change clang or
the standard, but llvm should work with compilers that implement the current
standard and mangle those functions.
llvm-svn: 175267
This happens when there is both stack realignment and a dynamic alloca in the
function. If we overwrite %esi (rep;movsl uses fixed registers) we'll lose the
base pointer and the next register spill will write into oblivion.
Fixes PR15249 and unbreaks firefox on i386/freebsd. Mozilla uses dynamic allocas
and freebsd a 4 byte stack alignment.
llvm-svn: 175057
account. Atoms use LEA for updating SP in prologs/epilogs, and the
exact LEA opcode depends on the data model.
Also reapplying the test case which was added and then reverted
(because of Atom failures), this time specifying explicitly the CPU in
addition to the triple. The test case now checks all variations (data
mode, cpu Atom vs. Core).
llvm-svn: 174542
pointer in function prologs/epilogs. The opcodes should depend on the
data model (LP64 vs. ILP32) rather than the architecture bit-ness.
llvm-svn: 174446
This change lets us bootstrap LLVM/Clang under ASan and MSan. It contains
fixes for 2 issues:
- X86JIT reads return address from stack, which MSan does not know is
initialized.
- bugpoint tests run binaries with RLIMIT_AS. This does not work with certain
Sanitizers.
We are no longer including config.h in Compiler.h with this change.
llvm-svn: 174306
1) allows the use of RIP-relative addressing in 32-bit LEA instructions under
x86-64 (ILP32 and LP64)
2) separates the size of address registers in 64-bit LEA instructions from
control by ILP32/LP64.
llvm-svn: 174208
conditions are met:
1. They share the same operand and are in the same BB.
2. Both outputs are used.
3. The target has a native instruction that maps to ISD::FSINCOS node or
the target provides a sincos library call.
Implemented the generic optimization in sdisel and enabled it for
Mac OSX. Also added an additional optimization for x86_64 Mac OSX by
using an alternative entry point __sincos_stret which returns the two
results in xmm0 / xmm1.
rdar://13087969
PR13204
llvm-svn: 173755
This catches many cases where we can emit a more efficient shuffle for a
specific mask or when the mask contains undefs. Once the splat is lowered to
unpacks we can't do that anymore.
There is a possibility of moving the promotion after pshufb matching, but I'm
not sure if pshufb with a mask loaded from memory is faster than 3 shuffles, so
I avoided that for now.
llvm-svn: 173569
(defined by the x32 ABI) mode, in which case its pointers are 32-bits
in size. This knowledge is also added to X86RegisterInfo that now
returns the appropriate registers in getPointerRegClass.
There are many outcomes to this change. In order to keep the patches
separate and manageable, we start by focusing on some simple testable
cases. The patch adds a test with passing a pointer to a function -
focusing on the difference between the two data models for x86-64.
Another test is added for handling of 'sret' arguments (and
functionality is added in X86ISelLowering to make it work).
A note on naming: the "x32 ABI" document refers to the AMD64
architecture (in LLVM it's distinguished by being is64Bits() in the
x86 subtarget) with two variations: the LP64 (default) data model, and
the ILP32 data model. This patch adds predicates to the subtarget
which are consistent with this naming scheme.
llvm-svn: 173503
- Add list of physical registers clobbered in pseudo atomic insts
Physical registers are clobbered when pseudo atomic instructions are
expanded. Add them in clobber list to prevent DAG scheduler to
mis-schedule them after these insns are declared side-effect free.
- Add test case from Michael Kuperstein <michael.m.kuperstein@intel.com>
llvm-svn: 173200
Add the x32 environment kind to the triple, and separate the concept of
pointer size and callee save stack slot size, since they're not equal
on x32.
llvm-svn: 173175
Previously we tried to infer it from the bit width size, with an added
IsIEEE argument for the PPC/IEEE 128-bit case, which had a default
value. This default value allowed bugs to creep in, where it was
inappropriate.
llvm-svn: 173138
The optimization handles esoteric cases but adds a lot of complexity both to the X86 backend and to other backends.
This optimization disables an important canonicalization of chains of SEXT nodes and makes SEXT and ZEXT asymmetrical.
Disabling the canonicalization of consecutive SEXT nodes into a single node disables other DAG optimizations that assume
that there is only one SEXT node. The AVX mask optimizations is one example. Additionally this optimization does not update the cost model.
llvm-svn: 172968
Move the early if-conversion pass into this group.
ILP optimizations usually need to find the right balance between
register pressure and ILP using the MachineTraceMetrics analysis to
identify critical paths and estimate other costs. Such passes should run
together so they can share dominator tree and loop info analyses.
Besides if-conversion, future passes to run here here could include
expression height reduction and ARM's MLxExpansion pass.
llvm-svn: 172687
Moving the X86CostTable to a common place, so that other back-ends
can share the code. Also simplifying it a bit and commoning up
tables with one and two types on operations.
llvm-svn: 172658
PR 14848. The lowered sequence is based on the existing sequence the target-independent
DAG Combiner creates for the scalar case.
Patch by Zvi Rackover.
llvm-svn: 171953
This was an experimental option, but needs to be defined
per-target. e.g. PPC A2 needs to aggressively hide latency.
I converted some in-order scheduling tests to A2. Hal is working on
more test cases.
llvm-svn: 171946
The current Intel Atom microarchitecture has a feature whereby
when a function returns early then it is slightly faster to execute
a sequence of NOP instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction until
the return address is ready.
When compiling for X86 Atom only, this patch will run a pass,
called "X86PadShortFunction" which will add NOP instructions where less
than four cycles elapse between function entry and return.
It includes tests.
This patch has been updated to address Nadav's review comments
- Optimize only at >= O1 and don't do optimization if -Os is set
- Stores MachineBasicBlock* instead of BBNum
- Uses DenseMap instead of std::map
- Fixes placement of braces
Patch by Andy Zhang.
llvm-svn: 171879
This is necessary not only for representing empty ranges, but for handling
multibyte characters in the input. (If the end pointer in a range refers to
a multibyte character, should it point to the beginning or the end of the
character in a char array?) Some of the code in the asm parsers was already
assuming this anyway.
llvm-svn: 171765
a TargetMachine to construct (and thus isn't always available), to an
analysis group that supports layered implementations much like
AliasAnalysis does. This is a pretty massive change, with a few parts
that I was unable to easily separate (sorry), so I'll walk through it.
The first step of this conversion was to make TargetTransformInfo an
analysis group, and to sink the nonce implementations in
ScalarTargetTransformInfo and VectorTargetTranformInfo into
a NoTargetTransformInfo pass. This allows other passes to add a hard
requirement on TTI, and assume they will always get at least on
implementation.
The TargetTransformInfo analysis group leverages the delegation chaining
trick that AliasAnalysis uses, where the base class for the analysis
group delegates to the previous analysis *pass*, allowing all but tho
NoFoo analysis passes to only implement the parts of the interfaces they
support. It also introduces a new trick where each pass in the group
retains a pointer to the top-most pass that has been initialized. This
allows passes to implement one API in terms of another API and benefit
when some other pass above them in the stack has more precise results
for the second API.
The second step of this conversion is to create a pass that implements
the TargetTransformInfo analysis using the target-independent
abstractions in the code generator. This replaces the
ScalarTargetTransformImpl and VectorTargetTransformImpl classes in
lib/Target with a single pass in lib/CodeGen called
BasicTargetTransformInfo. This class actually provides most of the TTI
functionality, basing it upon the TargetLowering abstraction and other
information in the target independent code generator.
The third step of the conversion adds support to all TargetMachines to
register custom analysis passes. This allows building those passes with
access to TargetLowering or other target-specific classes, and it also
allows each target to customize the set of analysis passes desired in
the pass manager. The baseline LLVMTargetMachine implements this
interface to add the BasicTTI pass to the pass manager, and all of the
tools that want to support target-aware TTI passes call this routine on
whatever target machine they end up with to add the appropriate passes.
The fourth step of the conversion created target-specific TTI analysis
passes for the X86 and ARM backends. These passes contain the custom
logic that was previously in their extensions of the
ScalarTargetTransformInfo and VectorTargetTransformInfo interfaces.
I separated them into their own file, as now all of the interface bits
are private and they just expose a function to create the pass itself.
Then I extended these target machines to set up a custom set of analysis
passes, first adding BasicTTI as a fallback, and then adding their
customized TTI implementations.
The fourth step required logic that was shared between the target
independent layer and the specific targets to move to a different
interface, as they no longer derive from each other. As a consequence,
a helper functions were added to TargetLowering representing the common
logic needed both in the target implementation and the codegen
implementation of the TTI pass. While technically this is the only
change that could have been committed separately, it would have been
a nightmare to extract.
The final step of the conversion was just to delete all the old
boilerplate. This got rid of the ScalarTargetTransformInfo and
VectorTargetTransformInfo classes, all of the support in all of the
targets for producing instances of them, and all of the support in the
tools for manually constructing a pass based around them.
Now that TTI is a relatively normal analysis group, two things become
straightforward. First, we can sink it into lib/Analysis which is a more
natural layer for it to live. Second, clients of this interface can
depend on it *always* being available which will simplify their code and
behavior. These (and other) simplifications will follow in subsequent
commits, this one is clearly big enough.
Finally, I'm very aware that much of the comments and documentation
needs to be updated. As soon as I had this working, and plausibly well
commented, I wanted to get it committed and in front of the build bots.
I'll be doing a few passes over documentation later if it sticks.
Commits to update DragonEgg and Clang will be made presently.
llvm-svn: 171681
cvtsi2* should parse with an 'l' or 'q' suffix or no suffix at all. No suffix should be treated the same as 'l' suffix. Printing should always print a suffix. Previously we didn't parse or print an 'l' suffix.
cvtt*2si/cvt*2si should parse with an 'l' or 'q' suffix or not suffix at all. No suffix should use the destination register size to choose encoding. Printing should not print a suffix.
Original 'l' suffix issue with cvtsi2* pointed out by Michael Kuperstein.
llvm-svn: 171668
URL: http://llvm.org/viewvc/llvm-project?rev=171524&view=rev
Log:
The current Intel Atom microarchitecture has a feature whereby when a function
returns early then it is slightly faster to execute a sequence of NOP
instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction
until the return address is ready.
When compiling for X86 Atom only, this patch will run a pass, called
"X86PadShortFunction" which will add NOP instructions where less than four
cycles elapse between function entry and return.
It includes tests.
Patch by Andy Zhang.
llvm-svn: 171603
returns early then it is slightly faster to execute a sequence of NOP
instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction
until the return address is ready.
When compiling for X86 Atom only, this patch will run a pass, called
"X86PadShortFunction" which will add NOP instructions where less than four
cycles elapse between function entry and return.
It includes tests.
Patch by Andy Zhang.
llvm-svn: 171524
1. Add code to estimate register pressure.
2. Add code to select the unroll factor based on register pressure.
3. Add bits to TargetTransformInfo to provide the number of registers.
llvm-svn: 171469
Jakob Stoklund Olesen