This generalizes the code to eliminate extra truncs/exts around i1 bit
operations to also do the same on PPC64 for i32 bit operations. This eliminates
a fairly prevalent code wart:
int foo(int a) {
return a == 5 ? 7 : 8;
}
On PPC64, because of the extension implied by the ABI, this would generate:
cmplwi 0, 3, 5
li 12, 8
li 4, 7
isel 3, 4, 12, 2
rldicl 3, 3, 0, 32
blr
where the 'rldicl 3, 3, 0, 32', the extension, is completely unnecessary. At
least for the single-BB case (which is all that the DAG combine mechanism can
handle), this unnecessary extension is no longer generated.
llvm-svn: 202600
X86Operand is extracted into individual header, because it allows to create an
arbitrary memory operand and append it to MCInst. It'll be reused in X86 inline
assembly instrumentation.
Patch by Yuri Gorshenin.
llvm-svn: 202496
* Align targets of indirect jumps to instruction bundle boundaries (in MI layer).
* Add masking instructions before indirect jumps (in MC layer).
Differential Revision: http://llvm-reviews.chandlerc.com/D2847
llvm-svn: 202479
The PPC isel instruction can fold 0 into the first operand (thus eliminating
the need to materialize a zero-containing register when the 'true' result of
the isel is 0). When the isel is fed by a bit register operation that we can
invert, do so as part of the bit-register-operation peephole routine.
llvm-svn: 202469
The CR bit tracking code broke PPC/Darwin; trying to get it working again...
(the darwin11 builder, which defaults to the darwin ABI when running PPC tests,
asserted when running test/CodeGen/PowerPC/inverted-bool-compares.ll)
llvm-svn: 202459
This change enables tracking i1 values in the PowerPC backend using the
condition register bits. These bits can be treated on PowerPC as separate
registers; individual bit operations (and, or, xor, etc.) are supported.
Tracking booleans in CR bits has several advantages:
- Reduction in register pressure (because we no longer need GPRs to store
boolean values).
- Logical operations on booleans can be handled more efficiently; we used to
have to move all results from comparisons into GPRs, perform promoted
logical operations in GPRs, and then move the result back into condition
register bits to be used by conditional branches. This can be very
inefficient, because the throughput of these CR <-> GPR moves have high
latency and low throughput (especially when other associated instructions
are accounted for).
- On the POWER7 and similar cores, we can increase total throughput by using
the CR bits. CR bit operations have a dedicated functional unit.
Most of this is more-or-less mechanical: Adjustments were needed in the
calling-convention code, support was added for spilling/restoring individual
condition-register bits, and conditional branch instruction definitions taking
specific CR bits were added (plus patterns and code for generating bit-level
operations).
This is enabled by default when running at -O2 and higher. For -O0 and -O1,
where the ability to debug is more important, this feature is disabled by
default. Individual CR bits do not have assigned DWARF register numbers,
and storing values in CR bits makes them invisible to the debugger.
It is critical, however, that we don't move i1 values that have been promoted
to larger values (such as those passed as function arguments) into bit
registers only to quickly turn around and move the values back into GPRs (such
as happens when values are returned by functions). A pair of target-specific
DAG combines are added to remove the trunc/extends in:
trunc(binary-ops(binary-ops(zext(x), zext(y)), ...)
and:
zext(binary-ops(binary-ops(trunc(x), trunc(y)), ...)
In short, we only want to use CR bits where some of the i1 values come from
comparisons or are used by conditional branches or selects. To put it another
way, if we can do the entire i1 computation in GPRs, then we probably should
(on the POWER7, the GPR-operation throughput is higher, and for all cores, the
CR <-> GPR moves are expensive).
POWER7 test-suite performance results (from 10 runs in each configuration):
SingleSource/Benchmarks/Misc/mandel-2: 35% speedup
MultiSource/Benchmarks/Prolangs-C++/city/city: 21% speedup
MultiSource/Benchmarks/MiBench/automotive-susan: 23% speedup
SingleSource/Benchmarks/CoyoteBench/huffbench: 13% speedup
SingleSource/Benchmarks/Misc-C++/Large/sphereflake: 13% speedup
SingleSource/Benchmarks/Misc-C++/mandel-text: 10% speedup
SingleSource/Benchmarks/Misc-C++-EH/spirit: 10% slowdown
MultiSource/Applications/lemon/lemon: 8% slowdown
llvm-svn: 202451
This is a temporary workaround for native arm linux builds:
PR18996: Changing regalloc order breaks "lencod" on native arm linux builds.
llvm-svn: 202433
scan the register file for sub- and super-registers.
No functionality change intended.
(Tests are updated because the comments in the assembler output are
different.)
llvm-svn: 202416
If a function returns a large struct by value return the first 4 words
in registers and the rest on the stack in a location reserved by the
caller. This is needed to support the xC language which supports
functions returning an arbitrary number of return values. This is
r202397 reapplied with a fix to avoid an uninitialized read of a member.
llvm-svn: 202414
Summary:
If a function returns a large struct by value return the first 4 words
in registers and the rest on the stack in a location reserved by the
caller. This is needed to support the xC language which supports
functions returning an arbitrary number of return values.
Reviewers: robertlytton
Reviewed By: robertlytton
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2889
llvm-svn: 202397
Summary:
If the src, dst and size of a memcpy are known to be 4 byte aligned we
can call __memcpy_4() instead of memcpy().
Reviewers: robertlytton
Reviewed By: robertlytton
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2871
llvm-svn: 202395
If the SI_KILL operand is constant, we can either clear the exec mask if
the operand is negative, or do nothing otherwise.
Reviewed-by: Tom Stellard <thomas.stellard@amd.com>
llvm-svn: 202337
The current approach to lower a vsetult is to flip the sign bit of the
operands, swap the operands and then use a (signed) pcmpgt. psubus (unsigned
saturating subtract) can be used to emulate a vsetult more efficiently:
+ case ISD::SETULT: {
+ // If the comparison is against a constant we can turn this into a
+ // setule. With psubus, setule does not require a swap. This is
+ // beneficial because the constant in the register is no longer
+ // destructed as the destination so it can be hoisted out of a loop.
I also enable lowering via psubus in a few other cases where it's clearly
beneficial: setule and setuge if minu/maxu cannot be used.
rdar://problem/14338765
Patch by Adam Nemet <anemet@apple.com>.
llvm-svn: 202301
Summary:
This should fix the MCJIT unit tests that were broken by r201792 on the MIPS buildbot.
MIPS currently uses the default implementation of sys::getHostCPUName() which
always returns "generic". For now, we will accept "generic" and coerce it to
"mips32" or "mips64" depending on the target architecture like we do for empty
CPU names.
Reviewers: jacksprat, matheusalmeida
Reviewed By: jacksprat
Differential Revision: http://llvm-reviews.chandlerc.com/D2878
llvm-svn: 202253
Eventually DataLayoutPass should go away, but for now that is the only easy
way to get a DataLayout in some APIs. This patch only changes the ones that
have easy access to a Module.
One interesting issue with sometimes using DataLayoutPass and sometimes
fetching it from the Module is that we have to make sure they are equivalent.
We can get most of the way there by always constructing the pass with a Module.
In fact, the pass could be changed to point to an external DataLayout instead
of owning one to make this stricter.
Unfortunately, the C api passes a DataLayout, so it has to be up to the caller
to make sure the pass and the module are in sync.
llvm-svn: 202204
We need to abort the formation of counter-register-based loops where there are
128-bit integer operations that might become function calls.
llvm-svn: 202192
The behaviour of the XCore's instruction buffer means that the performance
of the same code sequence can differ depending on whether it starts at a 4
byte aligned address or not. Since we don't model the instruction buffer
in the backend we have no way of knowing for sure if it is beneficial to
word align a specific function. However, in the absence of precise
modelling, it is better on balance to word align functions because:
* It makes a fetch-nop while executing the prologue slightly less likely.
* If we don't word align functions then a small perturbation in one
function can have a dramatic knock on effect. If the size of the function
changes it might change the alignment and therefore the performance of
all the functions that happen to follow it in the binary. This butterfly
effect makes it harder to reason about and measure the performance of
code.
llvm-svn: 202163
The patch defines new or refines existing generic scheduling classes to match
the behavior of the SSE instructions.
It also maps those scheduling classes on the related SSE instructions.
<rdar://problem/15607571>
llvm-svn: 202065
After this I will set the default back to F_None. The advantage is that
before this patch forgetting to set F_Binary would corrupt a file on windows.
Forgetting to set F_Text produces one that cannot be read in notepad, which
is a better failure mode :-)
llvm-svn: 202052
.align is handled specially on certain targets. .align without any parameters
on ARM indicates a default alignment (4). Handle the special case in the target
parser, but fall back to the generic parser for the normal version.
llvm-svn: 201988
This adds support for the .short and its alias .hword for adding literal values
into the object file. This is similar to the .word directive, however, rather
than inserting a value of 4 bytes, adds a 2-byte value.
llvm-svn: 201968
This commit moves getSLEB128Size() and getULEB128Size() from
MCAsmInfo to LEB128.h and removes some copy-and-paste code.
Besides, this commit also adds some unit tests for the LEB128
functions.
llvm-svn: 201937
The lowering of the frame index for stackmaps and patchpoints requires some
target-specific magic and should therefore be handled in the target-specific
eliminateFrameIndex method.
This is related to <rdar://problem/16106219>
llvm-svn: 201904
The va_start macro for AArch64 must set va_list.__stack to the address
following the last named argument on the stack, rounded up to an alignment
of 8 bytes.
llvm-svn: 201797
Summary:
This removes the need to coerce UnknownABI to the default ABI (O32 for
MIPS32, N64 for MIPS64 [*]) in both MipsSubtarget and MipsAsmParser.
Clang has been updated to disable both possible default ABI's before enabling
the ABI it intends to use.
[*] N64 being the default for MIPS64 is not actually correct.
However N32 is not fully implemented/tested yet.
Depends on: D2830
Reviewers: jacksprat, matheusalmeida
Reviewed By: matheusalmeida
Differential Revision: http://llvm-reviews.chandlerc.com/D2832
Differential Revision: http://llvm-reviews.chandlerc.com/D2846
llvm-svn: 201792
Summary:
This is consistent with the integrated assembler.
All mips64 codegen tests previously passed -mcpu. Removed -mcpu from
blez_bgez.ll and const-mult.ll to cover the default case.
Ideally, the two implementations of selectMipsCPU() will be merged but it's
proven difficult to find a home for the function that doesn't cause link errors.
For now, we'll hoist the common functionality into a function and mark it with
FIXME's.
Reviewers: jacksprat, matheusalmeida
Reviewed By: matheusalmeida
Differential Revision: http://llvm-reviews.chandlerc.com/D2830
llvm-svn: 201782
TargetLoweringBase is implemented in CodeGen, so before this patch we had
a dependency fom Target to CodeGen. This would show up as a link failure of
llvm-stress when building with -DBUILD_SHARED_LIBS=ON.
This fixes pr18900.
llvm-svn: 201711
r201608 made llvm corretly handle private globals with MachO. r201622 fixed
a bug in it and r201624 and r201625 were changes for using private linkage,
assuming that llvm would do the right thing.
They all got reverted because r201608 introduced a crash in LTO. This patch
includes a fix for that. The issue was that TargetLoweringObjectFile now has
to be initialized before we can mangle names of private globals. This is
trivially true during the normal codegen pipeline (the asm printer does it),
but LTO has to do it manually.
llvm-svn: 201700
On x86, shifting a vector by a scalar is significantly cheaper than shifting a
vector by another fully general vector. Unfortunately, because SelectionDAG
operates on just one basic block at a time, the shufflevector instruction that
reveals whether the right-hand side of a shift *is* really a scalar is often
not visible to CodeGen when it's needed.
This adds another handler to CodeGenPrepare, to sink any useful shufflevector
instructions down to the basic block where they're used, predicated on a target
hook (since on other architectures, doing so will often just introduce extra
real work).
rdar://problem/16063505
llvm-svn: 201655
The IR
@foo = private constant i32 42
is valid, but before this patch we would produce an invalid MachO from it. It
was invalid because it would use an L label in a section where the liker needs
the labels in order to atomize it.
One way of fixing it would be to just reject this IR in the backend, but that
would not be very front end friendly.
What this patch does is use an 'l' prefix in sections that we know the linker
requires symbols for atomizing them. This allows frontends to just use
private and not worry about which sections they go to or how the linker handles
them.
One small issue with this strategy is that now a symbol name depends on the
section, which is not available before codegen. This is not a problem in
practice. The reason is that it only happens with private linkage, which will
be ignored by the non codegen users (llvm-nm and llvm-ar).
llvm-svn: 201608
BuildMI instructions were not including MachineMemOperand information.
This was discovered by 'SingleSource/Benchmarks/Stanford/Oscar' failing
due to a FrameIndex load incorrectly being hoisted by postra-machine-licm.
No other tests have been found to fail.
llvm-svn: 201562
ldrd r6, r7 [r2, #15]
simply gives an error and does not triggers an assertion.
As Jim points out, the diagnostic is really strange here,
but fixing that would be more complicated. The missing
comma results in the parser expecting a construct like r2[2],
which is the vector index thing the error message is talking
about. That's not what the user intended, though, and there's
nothing else in the instruction that looks at all like a vector.
Yet more fallout from not having a real parser here and trying
to do context-free generic matching for addressing modes.
rdar://15097243
llvm-svn: 201531
NaCl's ARM ABI uses 16 byte stack alignment, so set that in
ARMSubtarget.cpp.
Using 16 byte alignment exposes an issue in code generation in which a
varargs function leaves a 4 byte gap between the values of r1-r3 saved
to the stack and the following arguments that were passed on the
stack. (Previously, this code only needed to support 4 byte and 8
byte alignment.)
With this issue, llc generated:
varargs_func:
sub sp, sp, #16
push {lr}
sub sp, sp, #12
add r0, sp, #16 // Should be 20
stm r0, {r1, r2, r3}
ldr r0, .LCPI0_0 // Address of va_list
add r1, sp, #16
str r1, [r0]
bl external_func
Fix the bug by checking for "Align > 4". Also simplify the code by
using OffsetToAlignment(), and update comments.
Differential Revision: http://llvm-reviews.chandlerc.com/D2677
llvm-svn: 201497
This adds a partial implementation of the .arch_extension directive to the
integrated ARM assembler. There are a number of limitations to this
implementation arising from the target backend support rather than the
implementation itself. Namely, iWMMXT (v1 and v2), Maverick, and XScale support
is not present in the ARM backend. Currently, there is no check for A-class
only (needed for virt), and no ARMv6k detection (needed for os and sec). The
remainder of the extensions are fully supported.
llvm-svn: 201471
1) Fix a specific bug when certain conversion functions are called in a program compiled as mips16 with hard float and
the program is linked as c++. There are two libraries that are reversed in the link order with gcc/g++ and clang/clang++ for
mips16 in this case and the proper stubs will then not be called. These stubs are normally handled in the Mips16HardFloat pass
but in this case we don't know at that time that we need to generate the stubs. This must all be handled later in code generation
and we have moved this functionality to MipsAsmPrinter. When linked as C (gcc or clang) the proper stubs are linked in from libc.
2) Set up the infrastructure to handle 90% of what is in the Mips16HardFloat pass in this new area of MipsAsmPrinter. This is a more
logical place to handle this and we have known for some time that we needed to move the code later and not implement it using
inline asm as we do now but it was not clear exactly where to do this and what mechanism should be used. Now it's clear to us
how to do this and this patch contains the infrastructure to move most of this to MipsAsmPrinter but the actual moving will be done
in a follow on patch. The same infrastructure is used to fix this current bug as described in #1. This change was requested by the list
during the original putback of the Mips16HardFloat pass but was not practical for us do at that time.
llvm-svn: 201426
As v1i1 is illegal, the type legalizer tries to scalarize such node. But if the type operands of SETCC is legal, the scalarization algorithm will cause an assertion failure.
llvm-svn: 201381
A simple register copy on X86 is just 3 bytes, whereas movabsq is a 10 byte
instruction. Marking movabsq as not beeing cheap will allow LICM to move it
out of the loop and it also prevents unnecessary rematerializations if the
value is needed in more than one register.
llvm-svn: 201377
Summary:
AsmPrinter::EmitInlineAsm() will no longer use the EmitRawText() call for
targets with mature MC support. Such targets will always parse the inline
assembly (even when emitting assembly). Targets without mature MC support
continue to use EmitRawText() for assembly output.
The hasRawTextSupport() check in AsmPrinter::EmitInlineAsm() has been replaced
with MCAsmInfo::UseIntegratedAs which when true, causes the integrated assembler
to parse inline assembly (even when emitting assembly output). UseIntegratedAs
is set to true for targets that consider any failure to parse valid assembly
to be a bug. Target specific subclasses generally enable the integrated
assembler in their constructor. The default value can be overridden with
-no-integrated-as.
All tests that rely on inline assembly supporting invalid assembly (for example,
those that use mnemonics such as 'foo' or 'hello world') have been updated to
disable the integrated assembler.
Changes since review (and last commit attempt):
- Fixed test failures that were missed due to configuration of local build.
(fixes crash.ll and a couple others).
- Fixed tests that happened to pass because the local build was on X86
(should fix 2007-12-17-InvokeAsm.ll)
- mature-mc-support.ll's should no longer require all targets to be compiled.
(should fix ARM and PPC buildbots)
- Object output (-filetype=obj and similar) now forces the integrated assembler
to be enabled regardless of default setting or -no-integrated-as.
(should fix SystemZ buildbots)
Reviewers: rafael
Reviewed By: rafael
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2686
llvm-svn: 201333
'OK_NonUniformConstValue' to identify operands which are constants but
not constant splats.
The cost model now allows returning 'OK_NonUniformConstValue'
for non splat operands that are instances of ConstantVector or
ConstantDataVector.
With this change, targets are now able to compute different costs
for instructions with non-uniform constant operands.
For example, On X86 the cost of a vector shift may vary depending on whether
the second operand is a uniform or non-uniform constant.
This patch applies the following changes:
- The cost model computation now takes into account non-uniform constants;
- The cost of vector shift instructions has been improved in
X86TargetTransformInfo analysis pass;
- BBVectorize, SLPVectorizer and LoopVectorize now know how to distinguish
between non-uniform and uniform constant operands.
Added a new test to verify that the output of opt
'-cost-model -analyze' is valid in the following configurations: SSE2,
SSE4.1, AVX, AVX2.
llvm-svn: 201272
Instead of expanding a packed shift into a sequence of scalar shifts,
the backend now tries (when possible) to convert the vector shift into a
vector multiply.
Before this change, a shift of a MVT::v8i16 vector by a
build_vector of constants was always scalarized into a long sequence of "vector
extracts + scalar shifts + vector insert".
With this change, if there is SSE2 support, we emit a single vector multiply.
This change also affects SSE4.1, AVX, AVX2 shifts:
- A shift of a MVT::v4i32 vector by a build_vector of non uniform constants
is now lowered when possible into a single SSE4.1 vector multiply.
- Packed v16i16 shift left by constant build_vector are now expanded when
possible into a single AVX2 vpmullw.
This change also improves the lowering of AVX512f vector shifts.
Added test CodeGen/X86/vec_shift6.ll with some code examples that are affected
by this change.
llvm-svn: 201271
