This reverts commit r209638 because it broke self-hosting on ppc64/Linux. (the
Clang-compiled TableGen would segfault because it jumped to an invalid address
from within _ZNK4llvm17ManagedStaticBase21RegisterManagedStaticEPFPvvEPFvS1_E
(which is within the command-line parameter registration process)).
llvm-svn: 209745
Add regression tests for the following transformation:
str X, [x20]
...
add x20, x20, #32
->
str X, [x20], #32
with X being either w0, x0, s0, d0 or q0.
llvm-svn: 209715
Add regression tests for the following transformation:
ldr X, [x20]
...
add x20, x20, #32
->
ldr X, [x20], #32
with X being either w0, x0, s0, d0 or q0.
llvm-svn: 209711
%higher and %highest can have non-zero values only for offsets greater
than 2GB, which is highly unlikely, if not impossible when compiling a
single function. This makes long branch for MIPS64 3 instructions smaller.
Differential Revision: http://llvm-reviews.chandlerc.com/D3281.diff
llvm-svn: 209678
In PPCISelLowering.cpp: PPCTargetLowering::LowerBUILD_VECTOR(), there
is an optimization for certain patterns to generate one or two vector
splats followed by a vector add or subtract. This operation is
represented by a VADD_SPLAT in the selection DAG. Prior to this
patch, it was possible for the VADD_SPLAT to be assigned the wrong
data type, causing incorrect code generation. This patch corrects the
problem.
Specifically, the code previously assigned the value type of the
BUILD_VECTOR node to the newly generated VADD_SPLAT node. This is
correct much of the time, but not always. The problem is that the
call to isConstantSplat() may return a SplatBitSize that is not the
same as the number of bits in the original element vector type. The
correct type to assign is a vector type with the same element bit size
as SplatBitSize.
The included test case shows an example of this, where the
BUILD_VECTOR node has a type of v16i8. The vector to be built is {0,
16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16, 0, 16}. isConstantSplat
detects that we can generate a splat of 16 for type v8i16, which is
the type we must assign to the VADD_SPLAT node. If we do not, we
generate a vspltisb of 8 and a vaddubm, which generates the incorrect
result {16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16, 16,
16}. The correct code generation is a vspltish of 8 and a vadduhm.
This patch also corrected code generation for
CodeGen/PowerPC/2008-07-10-SplatMiscompile.ll, which had been marked
as an XFAIL, so we can remove the XFAIL from the test case.
llvm-svn: 209662
Cortex-M4 only has single-precision floating point support, so any LLVM
"double" type will have been split into 2 i32s by now. Fortunately, the
consecutive-register framework turns out to be precisely what's needed to
reconstruct the double and follow AAPCS-VFP correctly!
rdar://problem/17012966
llvm-svn: 209650
Summary:
Implemented an InstCombine transformation that takes a blendv* intrinsic
call and translates it into an IR select, if the mask is constant.
This will eventually get lowered into blends with immediates if possible,
or pblendvb (with an option to further optimize if we can transform the
pblendvb into a blend+immediate instruction, depending on the selector).
It will also enable optimizations by the IR passes, which give up on
sight of the intrinsic.
Both the transformation and the lowering of its result to asm got shiny
new tests.
The transformation is a bit convoluted because of blendvp[sd]'s
definition:
Its mask is a floating point value! This forces us to convert it and get
the highest bit. I suppose this happened because the mask has type
__m128 in Intel's intrinsic and v4sf (for blendps) in gcc's builtin.
I will send an email to llvm-dev to discuss if we want to change this or
not.
Reviewers: grosbach, delena, nadav
Differential Revision: http://reviews.llvm.org/D3859
llvm-svn: 209643
This commit is debatable. There are two possible approaches, neither
of which is really satisfactory:
1. Use "@foo(i1 zeroext)" to mean an extension to 32-bits on Darwin,
and 8 bits otherwise.
2. Redefine "@foo(i1)" to mean that the i1 is extended by the caller
to 8 bits. This goes against the spirit of "zeroext" I think, but
it's a bit of a vague construct anyway (by definition you're going
to extend to the amount required by the ABI, that's why it's the
ABI!).
This implements option 2. The DAG machinery really isn't setup for the
first (there's a fairly strong assumption that "zeroext" goes to at
least the smallest register size), and even if it was the resulting
DAG looks like it would be inferior in many cases.
Theoretically we could add AssertZext nodes in the consumers of
ABI-passed values too now, but this actually seems to make the code
worse in practice by making truncation proceed in two steps. The code
produced is equally valid if we continue to assume only the low bit is
defined.
Should fix PR19850
llvm-svn: 209637
We can eliminate the custom C++ code in favour of some TableGen to
check the same things. Functionality should be identical, except for a
buffer overrun that was present in the C++ code and meant webkit
failed if any small argument needed to be passed on the stack.
llvm-svn: 209636
Add tests for the following transform:
str X, [x0, #32]
...
add x0, x0, #32
->
str X, [x0, #32]!
with X being either w1, x1, s0, d0 or q0.
llvm-svn: 209627
We have a couple of regression tests for load/store pairing, but (to my knowledge) there are no regression tests for the load/store + add/sub folding.
As a first step towards increased test coverage of this area, this commit adds a test for one instance of a load + add to pre-indexed load transformation.
llvm-svn: 209618
Currently we look at the Aliasee to decide what type of export
directive to use. It seems better to use the type of the alias
directly. This is similar to how we handle the alias having the
same address but other attributes (linkage, visibility) from the
aliasee.
With this patch it is now possible to do things like
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-pc-windows-msvc"
@foo = global [6 x i8] c"\B8*\00\00\00\C3", section ".text", align 16
@f = dllexport alias i32 (), [6 x i8]* @foo
!llvm.module.flags = !{!0}
!0 = metadata !{i32 6, metadata !"Linker Options", metadata !1}
!1 = metadata !{metadata !2, metadata !3}
!2 = metadata !{metadata !"/DEFAULTLIB:libcmt.lib"}
!3 = metadata !{metadata !"/DEFAULTLIB:oldnames.lib"}
llvm-svn: 209600
This commit starts with a "git mv ARM64 AArch64" and continues out
from there, renaming the C++ classes, intrinsics, and other
target-local objects for consistency.
"ARM64" test directories are also moved, and tests that began their
life in ARM64 use an arm64 triple, those from AArch64 use an aarch64
triple. Both should be equivalent though.
This finishes the AArch64 merge, and everyone should feel free to
continue committing as normal now.
llvm-svn: 209577
I'm doing this in two phases for a better "git blame" record. This
commit removes the previous AArch64 backend and redirects all
functionality to ARM64. It also deduplicates test-lines and removes
orphaned AArch64 tests.
The next step will be "git mv ARM64 AArch64" and rewire most of the
tests.
Hopefully LLVM is still functional, though it would be even better if
no-one ever had to care because the rename happens straight
afterwards.
llvm-svn: 209576
After the load/store refactoring, we were sometimes trying to feed a
GPR64 into a 32-bit register offset operand. This failed in
copyPhysReg.
llvm-svn: 209566
This matches both what we do for the non-thread case and what gcc does.
With this patch clang would match gcc's behaviour in
static __thread int a = 42;
extern __thread int b __attribute__((alias("a")));
int *f(void) { return &a; }
int *g(void) { return &b; }
if not for pr19843. Manually writing the IL does produce the same access modes.
It is also a step in the direction of fixing pr19844.
llvm-svn: 209543
Summary:
Instead the system is required to provide some means of handling unaligned
load/store without special instructions. Options include full hardware
support, full trap-and-emulate, and hybrids such as hardware support within
a cache line and trap-and-emulate for multi-line accesses.
MipsSETargetLowering::allowsUnalignedMemoryAccesses() has been configured to
assume that unaligned accesses are 'fast' on the basis that I expect few
hardware implementations will opt for pure-software handling of unaligned
accesses. The ones that do handle it purely in software can override this.
mips64-load-store-left-right.ll has been merged into load-store-left-right.ll
The stricter testing revealed a Bits!=Bytes bug in passByValArg(). This has
been fixed and the variables renamed to clarify the units they hold.
Reviewers: zoran.jovanovic, jkolek, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3872
llvm-svn: 209512
This patch teaches the x86 backend how to efficiently lower ISD::BITCAST dag
nodes from MVT::f64 to MVT::v4i16 (and vice versa), and from MVT::f64 to
MVT::v8i8 (and vice versa).
This patch extends the logic from revision 208107 to also handle MVT::v4i16
and MVT::v8i8. Also, this patch correctly propagates Undef values when
performing the widening of a vector (example: when widening from v2i32 to
v4i32, the upper 64bits of the resulting vector are 'undef').
llvm-svn: 209451
Summary:
* Split into two functions, one to test each struct.
* R0 and R2 must be defined by an lw with a %got reference to the correct
symbol.
* Test for $4 (first argument) where appropriate instead of accepting any
register.
* Test that the two lbu's are correctly combined into $4
Depends on D3844
Reviewers: jkolek, zoran.jovanovic, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3845
llvm-svn: 209424
Summary:
lwl and lwr are not available in MIPS32r6/MIPS64r6. The purpose of the test
is to check that the '$1' expands to '0($x)' rather than to test something related
to the lwl or lwr instructions so we can simply switch to lw.
Depends on D3842
Reviewers: jkolek, zoran.jovanovic, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3844
llvm-svn: 209423
Summary:
This patch is necessary so that they do not fail on MIPS32r6/MIPS64r6 when
-integrated-as is enabled by default and we correctly detect the host CPU.
No functional change since these tests are testing the behaviour of the
constraint used for the third operand rather than the mnemonic.
Depends on D3842
Reviewers: zoran.jovanovic, jkolek, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3843
llvm-svn: 209421
This intrinsic permits the emission of platform specific undefined sequences.
ARM has reserved the 0xde opcode which takes a single integer parameter (ignored
by the CPU). This permits the operating system to implement custom behaviour on
this trap. The llvm.arm.undefined intrinsic is meant to provide a means for
generating the target specific behaviour from the frontend. This is
particularly useful for Windows on ARM which has made use of a series of these
special opcodes.
llvm-svn: 209390
Committed in r209178 then reverted in r209251 due to LTO breakage,
here's a proper fix for the case of the missing subprogram DIE. The DIEs
were there, just in other compile units. Using the SPMap we can find the
right compile unit to search for and produce cross-unit references to
describe this kind of inlining.
One existing test case needed to be updated because it had a function
that wasn't in the CU's subprogram list, so it didn't appear in the
SPMap.
llvm-svn: 209335
ISD::VSELECT mask uses 1 to identify the first argument and 0 to identify the
second argument.
On the other hand, BLENDI uses 0 to identify the first argument and 1 to
identify the second argument.
Fix the generation of the blend mask to account for this difference.
The bug did not show up with r209043, because we were not checking for the
actual arguments of the blend instruction!
This commit also fixes the test cases.
Note: The same mask works for the BLENDr variant because the arguments are
swapped during instruction selection (see the BLENDXXrr patterns).
<rdar://problem/16975435>
llvm-svn: 209324
Although the previous code would construct a bundle and add the correct elements
to it, it would not finalise the bundle. This resulted in the InternalRead
markers not being added to the MachineOperands nor, more importantly, the
externally visible defs to the bundle itself. So, although the bundle was not
exposing the def, the generated code would be correct because there was no
optimisations being performed. When optimisations were enabled, the post
register allocator would kick in, and the hazard recognizer would reorder
operations around the load which would define the value being operated upon.
Rather than manually constructing the bundle, simply construct and finalise the
bundle via the finaliseBundle call after both MIs have been emitted. This
improves the code generation with optimisations where IMAGE_REL_ARM_MOV32T
relocations are emitted.
The changes to the other tests are the result of the bundle generation
preventing the scheduler from hoisting the moves across the loads. The net
effect of the generated code is equivalent, but, is much more identical to what
is actually being lowered.
llvm-svn: 209267
Povray and dealII currently assert with "Overran sorted position" in
AssignTopologicalOrder. The problem is that performPostLD1Combine can
introduce cycles.
Consider:
(insert_vector_elt (INSERT_SUBREG undef,
(load (add %vreg0, Constant<8>), undef), <= A
TargetConstant<2>),
(load %vreg0, undef), <= B
Constant<1>)
This is turned into a LD1LANEpost node. However the address in A is not a
valid user of the post-incremented address of B in LD1LANEpost.
llvm-svn: 209242
make the functions to set them non-static.
Move and rename the llvm specific backend options to avoid conflicting
with the clang option.
Paired with a backend commit to update.
llvm-svn: 209238
This commit introduces a canonical representation for the formulae.
Basically, as soon as a formula has more that one base register, the scaled
register field is used for one of them. The register put into the scaled
register is preferably a loop variant.
The commit refactors how the formulae are built in order to produce such
representation.
This yields a more accurate, but still perfectible, cost model.
<rdar://problem/16731508>
llvm-svn: 209230
The SplitIndexingFromLoad changes exposed a latent isel bug in the PowerPC64
backend. We matched an immediate offset with STWX8 even though it only
supports register offset.
The culprit is the complex-pattern predicate, SelectAddrIdx, which decides
that if the offset is not ISD::Constant it must be a register.
Many thanks to Bill Schmidt for testing this.
llvm-svn: 209219
Since we visit the whole list of subprograms for each CU at module
start, this is clearly true - don't test for the case, just assert it.
A few old test cases seemed to have incomplete subprogram lists, but any
attempt to reproduce them shows full subprogram lists that even include
entities that have been completely inlined and the out of line
definition removed.
llvm-svn: 209178
Instructions TZCNT (requires BMI1) and LZCNT (requires LZCNT), always
provide the operand size as output if the input operand is zero.
We can take advantage of this knowledge during instruction selection
stage in order to simplify a few corner case.
llvm-svn: 209159
Summary:
When inserting an element that's coming from a vector load or a broadcast
of a vector (or scalar) load, combine the load into the insertps
instruction.
Added PerformINSERTPSCombine for the case where we need to fix the load
(load of a vector + insertps with a non-zero CountS).
Added patterns for the broadcasts.
Also added tests for SSE4.1, AVX, and AVX2.
Reviewers: delena, nadav, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3581
llvm-svn: 209156
- On ARM/ARM64 we get a vrev because the shuffle matching code is really smart. We still unroll anything that's not v4i32 though.
- On X86 we get a pshufb with SSSE3. Required more cleverness in isShuffleMaskLegal.
- On PPC we get a vperm for v8i16 and v4i32. v2i64 is unrolled.
llvm-svn: 209123
Windows on ARM uses R11 for the frame pointer even though the environment is a
pure Thumb-2, thumb-only environment. Replicate this behaviour to improve
Windows ABI compatibility. This register is used for fast stack walking, and
thus is part of the Windows ABI.
llvm-svn: 209085
WoA uses COFF, not ELF. ARMISelLowering::createTLOF would previously return ELF
for any non-MachO platform. This was a missed site when the original change for
target format support for Windows on ARM was done.
llvm-svn: 209057
were added in SSE2, no SSSE3. Found this while auditing all uses of
SSSE3 in the X86 target. I don't actually expect this to make
a significant difference on anything and I don't have any detailed test
cases but I updated the existing test cases that already covered some of
this code path.
llvm-svn: 209056
vselects with constant masks, after legalization, will get turned into
specialized shuffle_vectors so they can be matched to blend+imm
instructions.
Fixed some tests.
llvm-svn: 209044
LowerVSELECT will, if possible, generate a X86ISD::BLENDI DAG node if the
condition is constant and we can emit that instruction, given the
subtarget.
This is not enough for all cases. An additional SELECTCombine optimization
will be committed.
Fixed tests that were expecting variable blends but where a blend+imm can
be generated.
Added test where we can't emit blend+immediate.
Added avx2 blend+imm tests.
llvm-svn: 209043
This patch changes the design of GlobalAlias so that it doesn't take a
ConstantExpr anymore. It now points directly to a GlobalObject, but its type is
independent of the aliasee type.
To avoid changing all alias related tests in this patches, I kept the common
syntax
@foo = alias i32* @bar
to mean the same as now. The cases that used to use cast now use the more
general syntax
@foo = alias i16, i32* @bar.
Note that GlobalAlias now behaves a bit more like GlobalVariable. We
know that its type is always a pointer, so we omit the '*'.
For the bitcode, a nice surprise is that we were writing both identical types
already, so the format change is minimal. Auto upgrade is handled by looking
through the casts and no new fields are needed for now. New bitcode will
simply have different types for Alias and Aliasee.
One last interesting point in the patch is that replaceAllUsesWith becomes
smart enough to avoid putting a ConstantExpr in the aliasee. This seems better
than checking and updating every caller.
A followup patch will delete getAliasedGlobal now that it is redundant. Another
patch will add support for an explicit offset.
llvm-svn: 209007
DIBuilder maintains this invariant and the current DwarfDebug code could
end up doing weird things if it contained declarations (such as putting
the definition DIE inside a CU that contained the declaration - this
doesn't seem like a good idea, so rather than adding logic to handle
this case we'll just ban in for now & cross that bridge if we come to
it later).
llvm-svn: 209004
This reverts commit r208934.
The patch depends on aliases to GEPs with non zero offsets. That is not
supported and fairly broken.
The good news is that GlobalAlias is being redesigned and will have support
for offsets, so this patch should be a nice match for it.
llvm-svn: 208978
TableGen has a fairly dubious heuristic to decide whether an alias should be
printed: does the alias have lest operands than the real instruction. This is
bad enough (particularly with no way to override it), but it should at least be
calculated consistently for both strings.
This patch implements that logic: first get the *correct* string for the
variant, in the same way as the Matcher, without guessing; then count the
number of whitespace chars.
There are basically 4 changes this brings about after the previous
commits; all of these appear to be good, so I have changed the tests:
+ ARM64: we print "neg X, Y" instead of "sub X, xzr, Y".
+ ARM64: we skip implicit "uxtx" and "uxtw" modifiers.
+ Sparc: we print "mov A, B" instead of "or %g0, A, B".
+ Sparc: we print "fcmpX A, B" instead of "fcmpX %fcc0, A, B"
llvm-svn: 208969
Add some Windows on ARM specific library calls. These are provided by msvcrt,
and can be used to perform integer to floating-point conversions (and
vice-versa) mirroring similar functions in the RTABI.
llvm-svn: 208949
This commit implements two command line switches -global-merge-on-external
and -global-merge-aligned, and both of them are false by default, so this
optimization is disabled by default for all targets.
For ARM64, some back-end behaviors need to be tuned to get this optimization
further enabled.
llvm-svn: 208934
Added target specific combine rules to fold blend intrinsics according
to the following rules:
1) fold(blend A, A, Mask) -> A;
2) fold(blend A, B, <allZeros>) -> A;
3) fold(blend A, B, <allOnes>) -> B.
Added two new tests to verify that the new folding rules work for all
the optimized blend intrinsics.
llvm-svn: 208895
In all cases, if a "mov" alias exists, it is the canonical form of the
instruction. Now that TableGen can support aliases containing syntax variants,
we can enable them and improve the quality of the asm output.
llvm-svn: 208874
Many old tests using prior schemas still had some brokenness here (both
indirect arrays and arrays with single bogus elements). Fixed those up
so they don't hit the new assertions.
Also reduced nesting in some places, etc.
llvm-svn: 208817
If the function has the landingpad instruction, then the
handlerdata should be emitted even if the function has
nouwnind attribute. Otherwise, following code will not
work:
void test1() noexcept {
try {
throw_exception();
} catch (...) {
log_unexpected_exception();
}
}
Since the cantunwind was incorrectly emitted and the
LSDA is not available.
llvm-svn: 208791
The commit r208166 will cause some regression on ARM EHABI.
This fix has been committed in r208715, and an assertion failure
test case has been committed in r208770.
This commit further extends the unittest so that the actual
value in the handlerdata will be checked.
llvm-svn: 208790
For example
tzcntl %edi, %ebx
testl %edi, %edi
je .label
can be rewritten into
tzcntl %edi, %ebx
jb .label
A minor complication is that tzcnt sets CF instead of ZF when the input
is zero, we have to rewrite users of the flags from ZF to CF. Currently
we recognize patterns using lzcnt, tzcnt and popcnt.
Differential Revision: http://reviews.llvm.org/D3454
llvm-svn: 208788
This commit was already commited as revision rL208689 and discussd in
phabricator revision D3704.
But the test file was crashing on OS X and windows.
I fixed the test file in the same way as in rL208340.
llvm-svn: 208711
compared to 'AddrMode.BaseReg'. In the case that 'AddrMode.BaseReg' is
nullptr, 'Result' will also be nullptr, so the cast causes an assertion. We
should use dyn_cast_or_null here to check 'Result' is not null and it is an
instruction.
Bug found by Mats Petersson, and I reduced his IR to get a test case.
llvm-svn: 208705
Normally, patterns like (add x, (setcc cc ...)) will be folded into
(csel x, x+1, not cc). However, if there is a ZEXT after SETCC, they
won't be folded. This patch recognizes the ZEXT and allows the
generation of CSINC.
This patch fixes bug 19680.
llvm-svn: 208660
Right now the load may not get DCE'd because of the side-effect of updating
the base pointer.
This can happen if we lower a read-modify-write of an illegal larger type
(e.g. i48) such that the modification only affects one of the subparts (the
lower i32 part but not the higher i16 part). See the testcase.
In order to spot the dead load we need to revisit it when SimplifyDemandedBits
decided that the value of the load is masked off. This is the
CommitTargetLoweringOpt piece.
I checked compile time with ARM64 by sending SPEC bitcode files through llc.
No measurable change.
Fixes <rdar://problem/16031651>
llvm-svn: 208640
r208453 added support for having sret on the second parameter. In that
change, the code for copying sret into a virtual register was hoisted
into the loop that lowers formal parameters. This caused a "Wrong
topological sorting" assertion failure during scheduling when a
parameter is passed in memory. This change undoes that by creating a
second loop that deals with sret.
I'm worried that this fix is incomplete. I don't fully understand the
dependence issues. However, with this change we produce the same DAGs
we used to produce, so if they are broken, they are just as broken as
they have always been.
llvm-svn: 208637
Tested by comparing make check VERBOSE=1 before and after to make sure
no tests are missed. (VERBOSE=1 prints the list of tests.)
Only one test :( remains where .cpp is required:
tools/llvm-cov/range_based_for.cpp:// RUN: llvm-cov range_based_for.cpp | FileCheck %s --check-prefix=STDOUT
The topic was discussed in this thread:
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20140428/214905.html
llvm-svn: 208621
The current patterns for REV16 misses mostn __builtin_bswap16() due to
legalization promoting the operands to from load/stores toi32s and then
truncing/extending them. This patch adds new patterns that catch the resultant
DAGs and codegens them to rev16 instructions. Tests included.
rdar://15353652
llvm-svn: 208620
1) Changed gather and scatter intrinsics. Now they are aligned with GCC built-ins. There is no more non-masked form. Masked intrinsic receives -1 if all lanes are executed.
2) I changed the function that works with intrinsics inside X86ISelLowering.cpp. I put all intrinsics in one table. I did it for INTRINSICS_W_CHAIN and plan to put all intrinsics from WO_CHAIN set to the same table in order to avoid the long-long "switch". (I wanted to use static map initialization that allowed by C++11 but I wasn't able to compile it on VS2012).
3) I added gather/scatter prefetch intrinsics.
4) I fixed MRMm encoding for masked instructions.
llvm-svn: 208522
Support for the intrinsics that read from and write to global named registers
is added for r1, r2 and r13 (depending on the subtarget).
llvm-svn: 208509
The counter-loops formation pass needs to know what operations might be
function calls (because they can't appear in counter-based loops). On PPC32,
128-bit shifts might be runtime calls (even though you can't use __int128 on
PPC32, it seems that SROA might form them).
Fixes PR19709.
llvm-svn: 208501
When lowering build_vector to an insertps, we would still lower it, even
if the source vectors weren't v4x32. This would break on avx if the source
was a v8x32. We now check the type of the source vectors.
llvm-svn: 208487
We were swapping the true & false results while testing for FMAX/FMIN,
but not putting them back to the original state if the later checks
failed.
Should fix PR19700.
llvm-svn: 208469
This reverts commit r200561.
This calling convention was an attempt to match the MSVC C++ ABI for
methods that return structures by value. This solution didn't scale,
because it would have required splitting every CC available on Windows
into two: one for methods and one for free functions.
Now that we can put sret on the second arg (r208453), and Clang does
that (r208458), revert this hack.
llvm-svn: 208459
MSVC always places the implicit sret parameter after the implicit this
parameter of instance methods. We used to handle this for
x86_thiscallcc by allocating the sret parameter on the stack and leaving
the this pointer in ecx, but that doesn't handle alternative calling
conventions like cdecl, stdcall, fastcall, or the win64 convention.
Instead, change the verifier to allow sret on the second parameter.
This also requires changing the Mips and X86 backends to return the
argument with the sret parameter, instead of assuming that the sret
parameter comes first.
The Sparc backend also returns sret parameters in a register, but I
wasn't able to update it to handle secondary sret parameters. It
currently calls report_fatal_error if you feed it an sret in the second
parameter.
Reviewers: rafael.espindola, majnemer
Differential Revision: http://reviews.llvm.org/D3617
llvm-svn: 208453
This patch adds support to ARM for custom lowering of the
llvm.{u|s}add.with.overflow.i32 intrinsics for i32/i64. This is particularly useful
for handling idiomatic saturating math functions as generated by
InstCombineCompare.
Test cases included.
rdar://14853450
llvm-svn: 208435
When using the ARM AAPCS, HFAs (Homogeneous Floating-point Aggregates) must
be passed in a block of consecutive floating-point registers, or on the stack.
This means that unused floating-point registers cannot be back-filled with
part of an HFA, however this can currently happen. This patch, along with the
corresponding clang patch (http://reviews.llvm.org/D3083) prevents this.
llvm-svn: 208413
Summary:
Adds MIPS32r6/MIPS64r6 and checks the compatibility requirements for these
processors.
I've also included comments to describe removed and re-encoded instructions,
along with placeholder def's for the new instructions but there are no
functional changes to codegen at this point.
Reviewers: jkolek, vmedic
Reviewed By: vmedic
Differential Revision: http://reviews.llvm.org/D3622
llvm-svn: 208399
Handle lowering of global addresses for PIC mode compilation on Windows. Always
use the movw/movt load to load the address as Windows on ARM requires ARMv7+ and
is a pure Thumb environment.
llvm-svn: 208385
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper, delena
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 208372
This patch teaches the backend how to combine packed SSE2/AVX2 arithmetic shift
intrinsics.
The rules are:
- Always fold a packed arithmetic shift by zero to its first operand;
- Convert a packed arithmetic shift intrinsic dag node into a ISD::SRA only if
the shift count is known to be smaller than the vector element size.
This patch also teaches to function 'getTargetVShiftByConstNode' how fold
target specific vector shifts by zero.
Added two new tests to verify that the DAGCombiner is able to fold
sequences of SSE2/AVX2 packed arithmetic shift calls.
llvm-svn: 208342
When building on Windows, the default target is Windows. Windows on ARM does
not support ARM mode compilation, resulting in test failures. Simply specify a
triple to ensure that we are testing the correct behaviour.
llvm-svn: 208340
Summary:
Vectors built with zeros and elements in the same order as another
(source) vector are optimized to be built using a single insertps
instruction.
Also optimize when we move one element in a vector to a different place
in that vector while zeroing out some of the other elements.
Further optimizations are possible, described in TODO comments.
I will be implementing at least some of them in the near future.
Added some tests for different cases where this optimization triggers.
Reviewers: nadav, delena, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3521
llvm-svn: 208271
Prior to r208252, the FMA 231 family was marked as isCommutable. However the
memory variants of this family are not commutable. Therefore, we did not
implemented the findCommutedOpIndices for those variants and missed that
the default implementation (more or less: commute indices 1 and 2) was
firing behind our back.
As a result, as demonstrated in the test case before the fix, we were
transforming a = b * c + a into a = a * c + b.
I.e., before r208252 we were generating for this test case:
vmovaps %xmm0, %xmm1
vmoss (%rsi), %xmm0
vfmadd231ss (%rdi), %xmm1, %xmm0
Instead of:
vmoss (%rsi), %xmm1
vfmadd231ss (%rdi), %xmm1, %xmm0
<rdar://problem/16800495>
llvm-svn: 208260
this patch disables the dead register elimination pass and the load/store pair
optimization pass at -O0. The ILP optimizations don't require the optimization
level to be checked because the call to addILPOpts is predicated with the
necessary check. The AdvSIMDScalar pass is disabled by default at all
optimization levels. This patch leaves that pass disabled by default.
Also, move command-line options into ARM64TargetMachine.cpp and add a few
additional flags to aid in debugging. This fixes an issue with the
-debug-pass=Structure flag where passes were printed, but not actually run
(i.e., AdvSIMDScalar pass).
llvm-svn: 208223
When performing a scalar comparison that feeds into a vector select,
it's actually better to do the comparison on the vector side: the
scalar route would be "CMP -> CSEL -> DUP", the vector is "CM -> DUP"
since the vector comparisons are all mask based.
llvm-svn: 208210
The AAPCS states that values passed in registers must have a value as though
they had been loaded with "LDR". LDR is equivalent to "LD1.64 vX.1D" - that is,
loading scalars to vector registers and loading 1-element vectors is equivalent.
The logic implemented here is to ensure that at all call boundaries and during
formal argument lowering all vectors are treated as their bitwidth-based floating
point scalar counterpart, which is always one of f64 or f128 (v2i32 -> f64,
v4i32 -> f128 etc). A BITCAST is inserted so that the appropriate REV will be
generated during code generation.
llvm-svn: 208198
Because we've canonicalised on using LD1/ST1, every time we do a bitcast
between vector types we must do an equivalent lane reversal.
Consider a simple memory load followed by a bitconvert then a store.
v0 = load v2i32
v1 = BITCAST v2i32 v0 to v4i16
store v4i16 v2
In big endian mode every memory access has an implicit byte swap. LDR and
STR do a 64-bit byte swap, whereas LD1/ST1 do a byte swap per lane - that
is, they treat the vector as a sequence of elements to be byte-swapped.
The two pairs of instructions are fundamentally incompatible. We've decided
to use LD1/ST1 only to simplify compiler implementation.
LD1/ST1 perform the equivalent of a sequence of LDR/STR + REV. This makes
the original code sequence: v0 = load v2i32
v1 = REV v2i32 (implicit)
v2 = BITCAST v2i32 v1 to v4i16
v3 = REV v4i16 v2 (implicit)
store v4i16 v3
But this is now broken - the value stored is different to the value loaded
due to lane reordering. To fix this, on every BITCAST we must perform two
other REVs:
v0 = load v2i32
v1 = REV v2i32 (implicit)
v2 = REV v2i32
v3 = BITCAST v2i32 v2 to v4i16
v4 = REV v4i16
v5 = REV v4i16 v4 (implicit)
store v4i16 v5
This means an extra two instructions, but actually in most cases the two REV
instructions can be combined into one. For example:
(REV64_2s (REV64_4h X)) === (REV32_4h X)
There is also no 128-bit REV instruction. This must be synthesized with an
EXT instruction.
Most bitconverts require some sort of conversion. The only exceptions are:
a) Identity conversions - vNfX <-> vNiX
b) Single-lane-to-scalar - v1fX <-> fX or v1iX <-> iX
Even though there are hundreds of changed lines, I have a fairly high confidence
that they are somewhat correct. The changes to add two REV instructions per
bitcast were pretty mechanical, and once I'd done that I threw the resulting
.td at a script I wrote which combined the two REVs together (and added
an EXT instruction, for f128) based on an instruction description I gave it.
This was much less prone to error than doing it all manually, plus my brain
would not just have melted but would have vapourised.
llvm-svn: 208194
This completes the port of r204814 (cpirker "AArch64_BE function argument
passing for ARM ABI") from AArch64 to ARM64, and fixes a bunch of issues
found during later development along the way. The biggest of these was
that the alignment fixup logic wasn't replicated into all the places it
should have been.
llvm-svn: 208192
The ARM::BLX instruction is an ARM mode instruction. The Windows on ARM target
is limited to Thumb instructions. Correctly use the thumb mode tBLXr
instruction. This would manifest as an errant write into the object file as the
instruction is 4-bytes in length rather than 2. The result would be a corrupted
object file that would eventually result in an executable that would crash at
runtime.
llvm-svn: 208152
remove it from the list of unspilled registers. Otherwise the following
attempt to keep the stack aligned by picking an extra GPR register to
spill will not work as it picks up r11.
llvm-svn: 208129
Before this patch, the backend always emitted a store+load sequence to
bitconvert from f64 to i64 the input operand of a ISD::BITCAST dag node that
performed a bitconvert from type MVT::f64 to type MVT::v2i32. The resulting
i64 node was then used to build a v2i32 vector.
With this patch, the backend now produces a cheaper SCALAR_TO_VECTOR from
MVT::f64 to MVT::v2f64. That SCALAR_TO_VECTOR is then followed by a "free"
bitcast to type MVT::v4i32. The elements of the resulting
v4i32 are then extracted to build a v2i32 vector (which is illegal and
therefore promoted to MVT::v2i64).
This is in general cheaper than emitting a stack store+load sequence
to bitconvert the operand from type f64 to type i64.
llvm-svn: 208107
This patch implements the infrastructure to use named register constructs in
programs that need access to specific registers (bare metal, kernels, etc).
So far, only the stack pointer is supported as a technology preview, but as it
is, the intrinsic can already support all non-allocatable registers from any
architecture.
llvm-svn: 208104
An alias has the address of what it points to, so it also has the same
alignment.
This allows a few optimizations to see past aliases for free.
llvm-svn: 208103
The Win64 docs are very clear that anything larger than 8 bytes is
passed by reference, and GCC MinGW64 honors that for __modti3 and
friends.
Patch by Jameson Nash!
llvm-svn: 208029
Summary:
Also ran clang-format on the function. The code added is the last else
if block.
Reviewers: nadav, craig.topper
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D3518
llvm-svn: 207992
Windows on ARM does not conform to AEABI. However, memset would be emitted
using the AEABI signature, resulting in inverted parameters. Handle this
special case appropriately.
llvm-svn: 207943
Both MinGW and cygwin (i686) construct export directives without the global
leader prefix. This is mostly due to the fact that they use GNU ld which does
not correctly handle the export directive. This apparently has been been broken
for a while. However, this was recently reported as being broken by
mingwandroid and diorcety of the msys2 project.
Remove the global leader prefix if targeting MinGW or cygwin, otherwise, retain
the global leader prefix. Add an explicit test for cygwin's behaviour of export
directives.
llvm-svn: 207926
While post-indexed LD1/ST1 instructions do exist for vector loads,
this patch makes use of the more flexible addressing-modes in LDR/STR
instructions.
llvm-svn: 207838
Given the following C code llvm currently generates suboptimal code for
x86-64:
__m128 bss4( const __m128 *ptr, size_t i, size_t j )
{
float f = ptr[i][j];
return (__m128) { f, f, f, f };
}
=================================================
define <4 x float> @_Z4bss4PKDv4_fmm(<4 x float>* nocapture readonly %ptr, i64 %i, i64 %j) #0 {
%a1 = getelementptr inbounds <4 x float>* %ptr, i64 %i
%a2 = load <4 x float>* %a1, align 16, !tbaa !1
%a3 = trunc i64 %j to i32
%a4 = extractelement <4 x float> %a2, i32 %a3
%a5 = insertelement <4 x float> undef, float %a4, i32 0
%a6 = insertelement <4 x float> %a5, float %a4, i32 1
%a7 = insertelement <4 x float> %a6, float %a4, i32 2
%a8 = insertelement <4 x float> %a7, float %a4, i32 3
ret <4 x float> %a8
}
=================================================
shlq $4, %rsi
addq %rdi, %rsi
movslq %edx, %rax
vbroadcastss (%rsi,%rax,4), %xmm0
retq
=================================================
The movslq is uneeded, but is present because of the trunc to i32 and then
sext back to i64 that the backend adds for vbroadcastss.
We can't remove it because it changes the meaning. The IR that clang
generates is already suboptimal. What clang really should emit is:
%a4 = extractelement <4 x float> %a2, i64 %j
This patch makes that legal. A separate patch will teach clang to do it.
Differential Revision: http://reviews.llvm.org/D3519
llvm-svn: 207801
This creates a lot of core infrastructure in which to add, with little
effort, quite a bit more to mips fast-isel
Test Plan: simplestore.ll
Reviewers: dsanders
Reviewed By: dsanders
Differential Revision: http://reviews.llvm.org/D3527
llvm-svn: 207790
The canonical form of the BFM instruction is always one of the more explicit
extract or insert operations, which makes reading output much easier.
llvm-svn: 207752
For pattern like ((x >> C1) & Mask) << C2, DAG combiner may convert it
into (x >> (C1-C2)) & (Mask << C2), which makes pattern matching of ubfx
more difficult.
For example:
Given
%shr = lshr i64 %x, 4
%and = and i64 %shr, 15
%arrayidx = getelementptr inbounds [8 x [64 x i64]]* @arr, i64 0, %i64 2, i64 %and
%0 = load i64* %arrayidx
With current shift folding, it takes 3 instrs to compute base address:
lsr x8, x0, #1
and x8, x8, #0x78
add x8, x9, x8
If using ubfx, it only needs 2 instrs:
ubfx x8, x0, #4, #4
add x8, x9, x8, lsl #3
This fixes bug 19589
llvm-svn: 207702
There is no need to check if we want to hoist the immediate value of an
shift instruction. Simply return TCC_Free right away.
This change is like r206101, but for X86.
rdar://problem/16190769
llvm-svn: 207692
target cannot be determined accurately. This is the case for NaCl where the
sandboxing instructions are added in MC layer, after the MipsLongBranch pass.
It is also the case when the code has inline assembly. Instead of calculating
offset in the MipsLongBranch pass, use %hi(sym1 - sym2) and %lo(sym1 - sym2)
expressions that are resolved during the fixup.
This patch also deletes microMIPS test file test/CodeGen/Mips/micromips-long-branch.ll
and implements microMIPS CHECKs in a much simpler way in a file
test/CodeGen/Mips/longbranch.ll, together with MIPS32 and MIPS64.
llvm-svn: 207656
The canonical syntax for shifts by a variable amount does not end with 'v', but
that syntax should be supported as an alias (presumably for legacy reasons).
llvm-svn: 207649
On instructions using the NZCV register, a couple of conditions have dual
representations: HS/CS and LO/CC (meaning unsigned-higher-or-same/carry-set and
unsigned-lower/carry-clear). The first of these is more descriptive in most
circumstances, so we should print it.
llvm-svn: 207644
Summary:
This isn't supported directly so we rotate the vector by the desired number of
elements, insert to element zero, then rotate back.
The i64 case generates rather poor code on MIPS32. There is an obvious
optimisation to be made in future (do both insert.w's inside a shared
rotate/unrotate sequence) but for now it's sufficient to select valid code
instead of aborting.
Depends on D3536
Reviewers: matheusalmeida
Reviewed By: matheusalmeida
Differential Revision: http://reviews.llvm.org/D3537
llvm-svn: 207640
Since these are mostly used in "lsl #16", "lsl #32", "lsl #48" combinations to
piece together an immediate in 16-bit chunks, hex is probably the most
appropriate format.
llvm-svn: 207635
This is mostly aimed at the NEON logical operations and MOVI/MVNI (since they
accept weird shifts which are more naturally understandable in hex notation).
Also changes BRK/HINT etc, which is probably a neutral change, but easier than
the alternative.
llvm-svn: 207634
Since these instructions only accept a 12-bit immediate, possibly shifted left
by 12, the canonical syntax used by the architecture reference manual is "#N {,
lsl #12 }". We should accept an immediate that has already been shifted, (e.g.
Also, print a comment giving the full addend since it can be helpful.
llvm-svn: 207633
Hal Finkel