For targeting pecoff, ".def foo" appears before ".short 32".
.def foo;
...
.LCPI0_0:
.short 32
foo:
CHECK-LABEL seeks not from ".short 32" but from the top of the input.
llvm-svn: 201931
This works by asking cmake to use the "install rpath", but setting that rpath
to be relative.
Thanks a lot to Brad King for the help with CMake!
llvm-svn: 201921
The LLVM diagnostic are now wired-up in clang (since r200931),
thus the user experience will not be impacted by this change
anymore.
Related to <rdar://problem/15886697>
llvm-svn: 201915
CodeGenPrepare uses extensively TargetLowering which is part of libLLVMCodeGen.
This is a layer violation which would introduce eventually a dependence on
CodeGen in ScalarOpts.
Move CodeGenPrepare into libLLVMCodeGen to avoid that.
Follow-up of <rdar://problem/15519855>
llvm-svn: 201912
shifted mask rather than masking and shifting separately.
The patch adds this transformation to the DAGCombiner:
(shl (and (setcc:i8v16 ...) N01C) N1C) -> (and (setcc:i8v16 ...) N01C<<N1C)
<rdar://problem/16054492>
Patch by Adam Nemet <anemet@apple.com>
llvm-svn: 201906
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
This interface allows IRObjectFile to be implemented without having dummy
methods for all section and segment related methods.
Both llvm-ar and llvm-nm are changed to use it. Unfortunately the mangler is
still not plugged in since it requires some refactoring to make a Module hold
a DataLayout.
llvm-svn: 201881
We were just emitting a label for this section for no real reason - this
caused us to emit the section even though we never put anything in it.
Not bothering with a test (though not adamantly anti-test) because it
seems somewhat arbitrary to test for the absence of this section anymore
than the absence of any other section.
llvm-svn: 201876
in the dependence test, we used to discard some information that the
delinearization provides: the size of the innermost dimension of an array,
i.e., the size of scalars stored in the array, and the remainder of the
delinearization that provides the offset from which the array reads start,
i.e., the base address of the array.
To avoid losing this data in the rest of the data dependence analysis, the fix
is to multiply the access function in the last delinearized dimension by its
size, effectively making the size of the last dimension to always be in bytes,
and then add the remainder of delinearization to the last subscript,
effectively making the last subscript start at the base address of the array.
llvm-svn: 201867
Because the delinearization is not a global analysis pass, it will compute the
delinearization independently of knowledge about the way the delinearization
happened for other data accesses to the same array: the dependence analysis will
only trigger the delinearization on a tuple of access functions, and thus
delinearization may compute different subscripts sizes for a same array. When
that happens the safest is to discard the delinearized information.
llvm-svn: 201866
llvm_add_library(foo SHARED STATIC
DEPENDS <dependent targets...>
LINK_LIBS <required libraries...>
)
It generates both foo (foo.so) and foo_static(foo.a) and both of them depend on DEPENDS and LINK_LIBS.
Then, also obj.foo is generated. obj.foo depends on DEPENDS, but doesn't depend on LINK_LIBS.
llvm-svn: 201854
The module still needs to collect the list of all available libraries
in order to satisfy the 'all' component. Provide this in the package
configuration file, 'LLVMConfig.cmake', as a LLVM_AVAILABLE_LIBS
variable. (A variable is scoped better than a global property.)
Since this won't be set for our own build, fall back to looking up the
LLVM_LIBS property to get the value when it is not set.
Contributed by Brad King.
llvm-svn: 201853
LLVM library names are now available as logical CMake targets both
to our own build and to application CMake code. Replace use of
'list(FIND)' with a simple 'if(TARGET)' to determine whether a
library is available.
Contributed by Brad King.
llvm-svn: 201852
This replaces the old NoIntegratedAssembler with at TargetOption. This is
more flexible and will be used to forward clang's -no-integrated-as option.
llvm-svn: 201836
I am really sorry for the noise, but the current state where some parts of the
code use TD (from the old name: TargetData) and other parts use DL makes it
hard to write a patch that changes where those variables come from and how
they are passed along.
llvm-svn: 201827
should not be marked nounwind.
Marking them nounwind caused crashes in the WebKit FTL JIT, because if we enable
sufficient optimizations, LLVM starts eliding compact_unwind sections (or any unwind
data for that matter), making deoptimization via stackmaps impossible.
This changes the stackmap intrinsic to be may-throw, adds a test for exactly the
sympton that WebKit saw, and fixes TableGen to handle un-attributed intrinsics.
Thanks to atrick and philipreames for reviewing this.
llvm-svn: 201826
The SuppressWarnings flag, unfortunately, isn't very useful for custom tools
that want to use the LLVM module linker. So I'm changing it to a parameter of
the Linker, and the flag itself moves to the llvm-link tool.
For the time being as SuppressWarnings is pretty much the only "option" it
seems reasonable to propagate it to Linker objects. If we end up with more
options in the future, some sort of "struct collecting options" may be a
better idea.
llvm-svn: 201819
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
passing down an AsmPrinter instance so we could compute the size of
the block which could be target specific. All of the test cases in
the unittest don't have any target specific data so we can use a NULL
AsmPrinter there. This also depends upon block data being added as
integers.
We can now hash the entire fission-cu.ll compile unit so turn the
flag on there with the hash value.
llvm-svn: 201752
Change parameter names exposed in headers to avoid collisions with Objective-C++
keywords.
Contributed-by: Graham Lee <graham@iamleeg.com>
llvm-svn: 201727
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
The same code (~20 lines) for initializing a TargetOptions object from CodeGen
cmdline flags is duplicated 4 times in 4 different tools. This patch moves it
into a utility function.
Since the CodeGen/CommandFlags.h file defines cl::opt flags in a header, it's
a bit of a touchy situation because we should only link them into tools. So this
patch puts the init function in the header.
llvm-svn: 201699
According to http://gcc.gnu.org/projects/cxx0x.html,
override and final keyword was added in gcc 4.7. Thus,
we should not use these keywords in gcc 4.6 even when
__GXX_EXPERIMENTAL_CXX0X__ is available.
llvm-svn: 201679
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
Load Configuration Table may contain a pointer to SEH table. This patch is to
print the offset to the table. Printing SEH table contents is a TODO.
The layout of Layout Configuration Table is described in Microsoft PE/COFF
Object File Format Spec, but the table's offset/size descriptions seems to be
totally wrong, at least in revision 8.3 of the spec. I believe the table in
this patch is the correct one.
llvm-svn: 201638
This enhances the macro parser to parse and handle parameter qualifications,
which is needed to support required formal parameters in macro definitions. A
required parameter may not be defaulted (though providing a default value is
accepted with a warning). This improves GAS compatibility.
Partially addresses PR9248.
llvm-svn: 201630
Rather than using std::pair, create a structure to represent the type. This is
a preliminary refactoring to enable required parameter handling. Additional
state is needed to indicate required parameters. This has a minor side effect
of improving readability by providing more accurate names compared to first and
second.
llvm-svn: 201629
When outputting an object we check its section to find its name, but when
looking for the section with -ffunction-section we look for the symbol name.
Break the loop by requesting a name with the private prefix when constructing
the section name. This matches the behavior before r201608.
llvm-svn: 201622
Some references to llvm-gcc were so crusty that I wasn't sure how to
proceed and so I've left them intact.
I also slipped in a quick peephole fix to use a :doc: link instead of
raw HTML link.
llvm-svn: 201619
From a cursory look it seems like all the described commandline options
and such apply to clang just fine, but I'd appreciate a second opinion.
llvm-svn: 201616
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
findOrEmitSection).
Vaidas Gasiunas's patch, r201259, fixed one instance where we were always
allocating sections as text. This patch fixes the remaining buggy call sites.
No test case: This isn't breaking anything that I know of, it's just
inconsistent.
<rdar://problem/15943542>
llvm-svn: 201605
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
Modifying build_llvm to handle SDKROOT being the name of an SDK rather than a
path. This will still work if SDKROOT is a path.
rdar://problem/15162322
llvm-svn: 201560
In gcov, the -o flag can accept either a directory or a file name.
When given a directory, the gcda and gcno files are expected to be in
that directory. When given a file, the gcda and gcno files are
expected to be named based on the stem of that file. Non-existent
paths are treated as files.
This implements compatible behaviour.
llvm-svn: 201555
Introducing llvm-profdata, a tool for merging profile data generated by
PGO instrumentation in clang.
- The name indicates a file extension of <name>.profdata. Eventually
profile data output by clang should be changed to that extension.
- llvm-profdata merges two profiles. However, the name is more general,
since it will likely pick up more tasks (such as summarizing a single
profile).
- llvm-profdata parses the current text-based format, but will be
updated once we settle on a binary format.
<rdar://problem/15949645>
llvm-svn: 201535
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
This is implemented by handling assignments to the '.' pseudo symbol
as ".org" directives.
Differential Revision: http://llvm-reviews.chandlerc.com/D2625
llvm-svn: 201530
Until this point only macro definition with named parameters were parsed but the
names were ignored. This adds support for using that information for named
parameter instantiation.
In order to support the full semantics of the keyword arguments, the arguments
are no longer lazily initialised since the keyword arguments can be specified
out of order and partially if they are defaulted. Prepopulate the arguments
with the default value for any defaulted parameters, and then parse the
specified arguments.
This simplies some of the handling of the arguments in the inner loop since
empty arguments simply increment the parameter index and move on.
Note that keyword and positional arguments cannot be mixed.
llvm-svn: 201499
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
During LSR of one loop we can run into a situation where we have to expand the
start of a recurrence of a loop induction variable in this loop. This start
value is a value derived of the induction variable of a preceeding loop. SCEV
has cannonicalized this value to a different recurrence than the recurrence of
the preceeding loop's induction variable (the type and/or step direction) has
changed). When we come to instantiate this SCEV we created a second induction
variable in this preceeding loop. This patch tries to base such derived
induction variables of the preceeding loop's induction variable.
This helps twolf on arm and seems to help scimark2 on x86.
Reapply with a fix for the case of a value derived from a pointer.
radar://15970709
llvm-svn: 201496
alongside DIEBlock and replace uses accordingly. Use DW_FORM_exprloc
in DWARF4 and later code. Update testcases.
Adding a DIELoc instead of using extra forms inside DIEBlock so
that we can keep location expressions separate from other uses. No
direct use at the moment, however, it's not a lot of code and
using a separately named class keeps it somewhat more obvious
what's going on in various locations.
llvm-svn: 201481
The Linux kernel defines empty macros for compatibility with ARM UAL syntax.
The comma after the name is optional, and if present can be safely lexed. This
improves compatibility with the GNU assembler.
llvm-svn: 201474
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
This broke in r185459 while TLS support was being generalized to handle
non-symbol TLS representations.
I thought about/tried having an enum rather than a bool to track the
TLS-ness of the address table entry, but namespaces and naming seemed
more hassle than it was worth for only one caller that needed to specify
this.
llvm-svn: 201469
During LSR of one loop we can run into a situation where we have to expand the
start of a recurrence of a loop induction variable in this loop. This start
value is a value derived of the induction variable of a preceeding loop. SCEV
has cannonicalized this value to a different recurrence than the recurrence of
the preceeding loop's induction variable (the type and/or step direction) has
changed). When we come to instantiate this SCEV we created a second induction
variable in this preceeding loop. This patch tries to base such derived
induction variables of the preceeding loop's induction variable.
This helps twolf on arm and seems to help scimark2 on x86.
radar://15970709
llvm-svn: 201465
Type units will share the statement list of their defining compile unit.
This is a tradeoff that reduces .o debug info size at the cost of some
linked debug info size (since the contents of those string tables won't
be deduplicated along with the type unit) which seems right for now.
llvm-svn: 201445
These types have an out of line virtual function each (emitHeader at
least) so they won't have weak vtables - no need for more than that.
llvm-svn: 201444
This probably also addresses the FIXME in the fission case regarding
multiple compile units, though I haven't tested that.
This code still confuses me (the literal zero offset makes little sense,
the limitations surrounding asm output I'm not sure about either - but
perhaps we should just always emit one line table? Or should we not rely
on .loc/.file even in assembly so we can produce the same output between
asm and object output?) but this maintains the existing functionality.
llvm-svn: 201441
This should be a small build time improvement in general and fixes
the build on OS X with -DBUILD_SHARED_LIBS=ON.
The issue is that not all users are including GenericDomTreeConstruction.h,
causing undefined references when ld64 managed to hide the
linkonce_odr symbols.
llvm-svn: 201440
Recommitting r201380 (reverted in r201389)
Recommitting r201351 and r201355 (reverted in r201351 and r201355)
We weren't emitting the an empty (header only) line table when the line
table was empty - this made the DWARF invalid (the compile unit would
point to the zero-size debug_lines section where there should've been an
empty line table but there was nothing at all). Fix that, and as a
consequence this works around/addresses PR18809.
Also, we emit a non-empty line table to workaround a darwin linker bug,
so XFAILing on darwin too.
Also, mark the test as 'REQUIRES: object-emission' because it does.
llvm-svn: 201429
Summary:
This adds support for emitting DWARF path discriminator values in
the object streamer. It also changes the DWARF dumper to show
discriminator values in the line table output.
Reviewers: echristo
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2794
llvm-svn: 201427
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
Recommitting r201351 and r201355 (reverted in r201351 and r201355)
We weren't emitting the an empty (header only) line table when the line
table was empty - this made the DWARF invalid (the compile unit would
point to the zero-size debug_lines section where there should've been an
empty line table but there was nothing at all). Fix that, and as a
consequence this works around/addresses PR18809.
llvm-svn: 201380
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:
There should be a space before each of these two keywords to avoid
generating invalid assembly files.
NOTE: I could not find an obvious maintainers in CODE_OWNERS.TXT, but
this seems related to debug info.
Reviewers: echristo
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2791
llvm-svn: 201359
In theory, Clang should figure out how to parse this correctly without
typename, but since this is the last TU that Clang falls back on in the
self-host, I'm going to compromise and check for __clang__.
And now Clang can self-host on -win32 without fallback! The 'check' and
'check-clang' targets both pass.
llvm-svn: 201358
This is an optimistic version of create_diretories: it tries to create the
directory first and looks at the parent only if that fails.
Running strace on "mkdir -p" shows that it is pessimistic, calling mkdir on
every element of the path. We could implement that if needed.
In any case, with both strategies there is no reason to call stat, just check
the return of mkdir.
llvm-svn: 201347
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
On unsupported platforms, llvm_add_library(MODULE) doesn't create any targets.
Caller may be responsible to check and add extra target properties.
llvm-svn: 201320
FIXME: llvm/test may be aware of LLVM_PLUGIN_EXT, like as clang/test does.
FIXME: CMAKE_*_SUFFIX may be set in HandleLLVMOptions if those variables could be writable, rather than to set one as target properties.
llvm-svn: 201316
This fix checks the original LLVM IR node to identify opaque constants by
looking for the bitcast-constant pattern. Originally we looked at the generated
SDNode, but this might lead to incorrect results. The SDNode could have been
generated by an constant expression that was folded to a constant.
This fixes <rdar://problem/16050719>
llvm-svn: 201291
As defined in LangRef, aliases do not have sections. However, LLVM's
GlobalAlias class inherits from GlobalValue, which means we can read and
set its section. We should probably ban that as a separate change,
since it doesn't make much sense for an alias to have a section that
differs from its aliasee.
Fixes PR18757, where the section was being lost on the global in code
from Clang like:
extern "C" {
__attribute__((used, section("CUSTOM"))) static int in_custom_section;
}
Reviewers: rafael.espindola
Differential Revision: http://llvm-reviews.chandlerc.com/D2758
llvm-svn: 201286
I was insightless then about unknown optional parameters.
(Consider that LINK_LIBS foo bar ADDITIONAL_HEADERS qux quux)
Suggested by Michael Kruse. Thanks!
llvm-svn: 201283
logical operations on the i1's driving them. This is a bad idea for every
target I can think of (confirmed with micro tests on all of: x86-64, ARM,
AArch64, Mips, and PowerPC) because it forces the i1 to be materialized into
a general purpose register, whereas consuming it directly into a select generally
allows it to exist only transiently in a predicate or flags register.
Chandler ran a set of performance tests with this change, and reported no
measurable change on x86-64.
llvm-svn: 201275
'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
This does exactly the same thing as for Win32, except it passes -m64 to
the compiler and the files go in a different directory.
Differential Revision: http://llvm-reviews.chandlerc.com/D2749
llvm-svn: 201269
We are now no longer relying on the target-specific call lowering implementation
to lower a stackmap intrinsic call. Instead we perform the call lowering in a
target-independent way directly in the stackmap lowering code. This simplifies
the code and removes the need to fixup the code after the target-specific call
lowering.
llvm-svn: 201263
The ID type for the stackmap and patchpoint intrinsics are in both cases i64.
This fixes an zero extend in the SelectionDAGBuilder that still used i32. This
also updates the target independent instructions STACKMAP and PATCHPOINT to use
the correct type.
llvm-svn: 201262
required for all sections in a module. This can be useful when targets or
code-models place strict requirements on how sections must be laid out
in memory.
If RTDyldMemoryManger::needsToReserveAllocationSpace() is overridden to return
true then the JIT will call the following method on the memory manager, which
can be used to preallocate the necessary memory.
void RTDyldMemoryManager::reserveAllocationSpace(uintptr_t CodeSize,
uintptr_t DataSizeRO,
uintptr_t DataSizeRW)
Patch by Vaidas Gasiunas. Thanks very much Viadas!
llvm-svn: 201259
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.
Reviewers: rafael
Reviewed By: rafael
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D2686
llvm-svn: 201237
I found that swapping the order of some header files helped fix a
build issue that we're seeing on mingw32. Without the swap, windows.h
was being included before _WIN32_WINNT was being defined and the
CreateHardLinkW function was #ifdef'd out.
It looks like the header is mainly used to get the SHGetFolderPathW
function, so I don't think that there'll be much fallout from the
switch.
Suggested by Alex Crichton. Thanks!
llvm-svn: 201230
This macro depends on several variables to be set in the calling
context. Check them and report an error if they are not set.
Without this, custom commands may be silently specified that
will fail at build time.
Patch by Brad King.
llvm-svn: 201229
There's still one piece missing here, which is adding the
DW_AT_stmt_list to the type unit that refer's to the compile unit's line
table. Working on that.
llvm-svn: 201198
This is preliminary work to fix type unit file strings so they appear in
their originating CU's line table - but it's also just good/simple
cleanup, so I'm committing it ahead of time.
llvm-svn: 201195
We used to be pretty vague about what debug entities were what, with
many conditionals to silently drop/skip/accept things. These don't seem
to be relevant anymore.
llvm-svn: 201194
* CPRCs may be allocated to co-processor registers or the stack – they may never be allocated to core registers
* When a CPRC is allocated to the stack, all other VFP registers should be marked as unavailable
The difference is only noticeable in rare cases where there are a large number of floating point arguments (e.g.
7 doubles + additional float, double arguments). Although it's probably still better to avoid vmov as it can cause
stalls in some older ARM cores. The other, more subtle benefit, is to minimize difference between the various
calling conventions.
rdar://16039676
llvm-svn: 201193
Debug info: Emit values in subregisters that do not have a separate
DWARF register number by emitting a super-register + DW_OP_bit_piece.
This is necessary because on x86_64, there are no DWARF register numbers
for i386-style subregisters.
Fixes a bunch of FIXMEs.
rdar://problem/16015314
llvm-svn: 201190
This comes up in empty files or files containing #file directives that
never reference the actual source file name. Came up in a small test of
line tables I was playing with.
llvm-svn: 201187
These tests were unnecessarily sensitive to the presence and ordering of
elements in the line table file_names list which will break on a future
change I'm working on.
llvm-svn: 201185
DWARF register number by emitting a super-register + DW_OP_bit_piece.
This is necessary because on x86_64, there are no DWARF register numbers
for i386-style subregisters.
Fixes a bunch of FIXMEs.
rdar://problem/16015314
llvm-svn: 201180
sections. The call to data.getUnsigned(&Offset, AddressSize) only
increments Offset if the read succeeds, which will result in an infinite
loop.
llvm-svn: 201179
I noticed this convention from the commit logs. It seems like it
would be useful to document it, to encourage other committers to link
back to code reviews in their commits.
Differential Revision: http://llvm-reviews.chandlerc.com/D2678
llvm-svn: 201160
BUILD_VECTOR nodes, e.g.:
(concat_vectors (BUILD_VECTOR a1, a2, a3, a4), (BUILD_VECTOR b1, b2, b3, b4))
->
(BUILD_VECTOR a1, a2, a3, a4, b1, b2, b3, b4)
This fixes an issue with AVX, where a sequence was not recognized as a 256-bit
vbroadcast due to the concat_vectors.
llvm-svn: 201158
Fixes PR18753 and PR18782.
This is necessary for LICM to preserve LCSSA correctly and efficiently.
There is still some active discussion about whether we should be using
LCSSA, but we can't just immediately stop using it and we *need* LICM to
preserve it while we are using it. We can restore the old SSAUpdater
driven code if and when there is a serious effort to remove the reliance
on LCSSA from all of the loop passes.
However, this also serves as a great example of why LCSSA is very nice
to have. This change significantly simplifies the process of sinking
instructions for LICM, and makes it quite a bit less expensive.
It wouldn't even be as complex as it is except that I had to start the
process of removing the big recursive LCSSA formation hammer in order to
switch even this much of the re-forming code to asserting that LCSSA was
preserved. I'll fully remove that next just to tidy things up until the
LCSSA debate settles one way or the other.
llvm-svn: 201148
Xcore target ABI requires const data that is externally visible
to be handled differently if it has C-language linkage rather than
C++ language linkage.
Clang now emits ".cp.rodata" section information.
All other externally visible constant data will be placed in the DP section.
llvm-svn: 201144
profitability check due to some other checks in the addressing
mode matcher. I.e., test case for commit r201121.
<rdar://problem/16020230>
llvm-svn: 201132
uintptr_t. An unsigned could overflow for large sections.
No test case - anything big enough to overflow an unsigned is going to take an
appreciable time to zero when the test passes.
The choice of uintptr_t was made to match the RTDyldMemoryManager APIs, but
these should probably be hardcoded to uint64_ts: It is legitimate to JIT for
64-bit targets from a 32-bit host/compiler.
llvm-svn: 201127
The addressing mode matcher checks at some point the profitability of folding an
instruction into the addressing mode. When the instruction to be folded has
several uses, it checks that the instruction can be folded in each use.
To do so, it creates a new matcher for each use and check if the instruction is
in the list of the matched instructions of this new matcher.
The new matchers may promote some instructions and this has to be undone to keep
the state of the original matcher consistent.
A test case will follow.
<rdar://problem/16020230>
llvm-svn: 201121
These are self-contained in functionality so it makes sense to separate them,
as opt.cpp has grown quite big already.
Following Eric's suggestions, if this code is ever deemed useful outside of
tools/opt, it will make sense to move it to one of the LLVM libraries like IR.
llvm-svn: 201116
This function adds an extra path argument to lto_module_create_from_memory.
The path argument will be passed to makeBuffer to make sure the MemoryBuffer
has a name and the created module has a module identifier.
This is mainly for emitting warning messages from the linker. When we emit
warning message on a module, we can use the module identifier.
rdar://15985737
llvm-svn: 201114
A const ObjectFile needs to be able to provide its name. For an IRObjectFile,
that means being able to call the mangler. Since each IRObjectFile can have
a different mangling, it is natural for them to contain a Mangler which is
therefore also const.
llvm-svn: 201113
The crux of the issue is that LCSSA doesn't preserve stateful alias
analyses. Before r200067, LICM didn't cause LCSSA to run in the LTO pass
manager, where LICM runs essentially without any of the other loop
passes. As a consequence the globalmodref-aa pass run before that loop
pass manager was able to survive the loop pass manager and be used by
DSE to eliminate stores in the function called from the loop body in
Adobe-C++/loop_unroll (and similar patterns in other benchmarks).
When LICM was taught to preserve LCSSA it had to require it as well.
This caused it to be run in the loop pass manager and because it did not
preserve AA, the stateful AA was lost. Most of LLVM's AA isn't stateful
and so this didn't manifest in most cases. Also, in most cases LCSSA was
already running, and so there was no interesting change.
The real kicker is that LCSSA by its definition (injecting PHI nodes
only) trivially preserves AA! All we need to do is mark it, and then
everything goes back to working as intended. It probably was blocking
some other weird cases of stateful AA but the only one I have is
a 1000-line IR test case from loop_unroll, so I don't really have a good
test case here.
Hopefully this fixes the regressions on performance that have been seen
since that revision.
llvm-svn: 201104
DS instructions that access local memory can only uses addresses that
are less than or equal to the value of M0. When M0 is uninitialized,
then we experience undefined behavior.
This patch also changes the behavior to emit S_WQM_B64 on pixel shaders
no matter what kind of DS instruction is used.
llvm-svn: 201097
This doesn't change any functionality, since we only have two shader
types (compute and pixel) that use local memory. We're just changing
the logic to match the documentation.
llvm-svn: 201096
Similarly to the vshrn instructions, these are simple zext/sext + trunc
operations. Using normal LLVM IR should allow for better code, and more sharing
with the AArch64 backend.
llvm-svn: 201093
For A- and R-class processors, r12 is not normally callee-saved, but is for
interrupt handlers. See AAPCS, 5.3.1.1, "Use of IP by the linker".
llvm-svn: 201089
vshrn is just the combination of a right shift and a truncate (and the limits
on the immediate value actually mean the signedness of the shift doesn't
matter). Using that representation allows us to get rid of an ARM-specific
intrinsic, share more code with AArch64 and hopefully get better code out of
the mid-end optimisers.
llvm-svn: 201085
The CMake install(DIRECTORY) command documents that it sets permissions
on directories it is asked to install. Since the <prefix>/include
directory may not be exclusive to the LLVM installation, we should not
ask CMake to manage permissions of that directory for us. Instead, give
only our own include/llvm and include/llvm-c subdirectories to the
install(DIRECTORY) command.
Fixes PR4500. Patch by Brad King.
llvm-svn: 201075
- MODULE;SHARED;STATIC
STATIC by default w/o BUILD_SHARED_LIBS.
SHARED by default w/ BUILD_SHARED_LIBS.
- OUTPUT_NAME name
Corresponds to OUTPUT_NAME in target properties.
- DEPENDS targets...
Same semantics as add_dependencies().
- LINK_COMPONENTS components...
Same as the variable LLVM_LINK_COMPONENTS.
- LINK_LIBS lib_targets...
Same semantics as target_link_libraries().
- ADDITIONAL_HEADERS (implemented in LLVMProcessSources)
May specify header files for IDE generators.
I suggest llvm_add_library() may be used for inter-project add_library stuff
and also suggest add_***_library() may be used project-specific.
Please be patient that llvm_add_library might be ambiguous against add_llvm_library.
llvm-svn: 201072
Original commits messages:
Add MRMXr/MRMXm form to X86 for use by instructions which treat the 'reg' field of modrm byte as a don't care value. Will allow for simplification of disassembler code.
Simplify a bunch of code by removing the need for the x86 disassembler table builder to know about extended opcodes. The modrm forms are sufficient to convey the information.
llvm-svn: 201065
r201059 appears to cause a crash in a bootstrapped build of clang. Craig
isn't available to look at it right now, so I'm reverting it while he
investigates.
llvm-svn: 201064
I am sure it'd not be required any more.
In trunk, all of tablegen's users depend on ${TABLEGEN_OUTPUT} as not file dependency but inter-target dependency.
llvm-svn: 201063
CMake's target_link_libraries() will manage dependencies with Brad's LLVMConfig improvements.
Configuration time may be reduced by a few seconds.
llvm-svn: 201062
Some of the more complex directive and macro handling for GAS compatibility
requires lookahead. Add a single token lookahead in the MCAsmLexer.
llvm-svn: 201058
Teach the Makefile build system to generate and install CMake modules
LLVMConfig.cmake and LLVMConfigVersion.cmake so that applications that
build with CMake can use 'find_package(LLVM)' even when LLVM is not
built with CMake. These modules tell such applications about available
LLVM libraries and their dependencies.
Run llvm-config to generate the list of libraries and use the results of
llvm-build to generate the library dependencies. Use sed to perform
substitutions in the LLVMConfig.cmake.in and LLVMConfigVersion.cmake.in
sources that our CMake build system uses.
Teach the Makefile build system to generate the LLVMExports.cmake file
with content similar to that produced by the CMake install(EXPORT)
command. Extend llvm-build with an option to generate the library
dependencies fragment for this file.
Contributed by Brad King.
llvm-svn: 201053
Teach autoconf/configure.ac to AC_SUBST several additional values in
Makefile.config to make them available to Makefile code. These will
be useful to generate CMake package modules from the Makefile build.
Contributed by Brad King.
llvm-svn: 201052
Teach each package configuration file to load the LLVMExports file for
its corresponding tree. This will allow application CMake code to use
logical library and executable target names from LLVM as if they were in
our own build process (e.g. LLVMSupport). CMake will have enough
information to propagate LLVM library link dependencies automatically
while configuring applications.
Contributed by Brad King.
llvm-svn: 201051
Record every logical target that we install with install(TARGETS) in a
global LLVM_EXPORTS property. Then use the export(TARGETS) command to
provide a "LLVMExports.cmake" file that exports logical targets for
import into applications directly from our build tree.
The "LLVMExports.cmake" file is not meant for direct inclusion by
application code but should be included by "LLVMConfig.cmake" in a
future change.
Contributed by Brad King.
llvm-svn: 201050
Use the install(TARGETS) command EXPORT option for every library and
executable that we install with LLVM. Then use the install(EXPORT)
command to provide a "LLVMExports.cmake" file that exports logical
targets for import into applications from our install tree.
The "LLVMExports.cmake" file is not meant for direct inclusion by
application code but should be included by "LLVMConfig.cmake" in a
future change.
Contributed by Brad King.
llvm-svn: 201049
Create separate package configuration files "LLVMConfig.cmake" for the
LLVM build and install trees so that each can have information specific
to its tree. Configure each with the corresponding include, lib, and
cmake directories. Include the "LLVM-Config" API modules directly from
the configured cmake modules directory.
In the install tree, compute the installation prefix relative to the
file location. In the build tree, provide information specific to the
build tree for use by tools like Clang that can build externally against
the LLVM build tree. Prefix such values in "LLVM_BUILD_" and comment
them as such.
Contributed by Brad King.
llvm-svn: 201048
Do not modify this value on the application's behalf and just ensure API
modules are always available next to the LLVMConfig module. This is
already the case in the install tree so use file(COPY) to make it so in
the build tree. Include the LLVM-Config API module from next to the
LLVMConfig location.
Contributed by Brad King.
llvm-svn: 201047
These methods normally call each other and it is really annoying if the
arguments are in different order. The more common rule was that the arguments
specific to call are first (GV, Encoding, Suffix) and the auxiliary objects
(Mang, TM) come after. This patch changes the exceptions.
llvm-svn: 201044
Fun fact: looking at the TableGen code (around TGParser.cpp:1166), the
only difference in handling is that adjacent regular string literals are
concatenated in the parser.
llvm-svn: 201035
They're called code fragments, but they are really multiline string
literals. Just spotted this usage in a patch by Aaron using "code
fragments" for holding documentation text. I remember someone bemoaning
the lack of multiline string literals in TableGen, so I'm explicitly
documenting that code fragments are multiline string literals.
Let it be known that any use case needing multiline string literals in
TableGen (such as descriptions of options, or whatnot) can use use
code fragments (instead of C-style string concatenation or exceedingly
long lines). E.g.
class Bar<int n>;
class Baz<int n>;
class Doc<string desc> {
string Desc = desc;
}
def Foo : Bar<1>, Baz<3>, Doc<[{
This Foo is a Bar, and also a Baz. It can take 3 values:
* Qux
* Quux
* Quuux
}]>;
llvm-svn: 201033
In some cases it is possible to have a personality 0 unwinding opcodes in the
extab (such as when .handlerdata is used in the assembly). Simply decode the 3
opcodes for that case.
llvm-svn: 201030
This makes the tests more readable by using the -arm-attributes decoding support
in llvm-readobj since that is now available. Change the invocation commands to
be similar to other test and use a more precise triple (the tests only require
ARM EABI support).
llvm-svn: 201029
Before conditional store vectorization/unrolling we had only one
vectorized/unrolled basic block. After adding support for conditional store
vectorization this will not only be one block but multiple basic blocks. The
last block would have the back-edge. I updated the code to use a vector of basic
blocks instead of a single basic block and fixed the users to use the last entry
in this vector. But, I forgot to add the basic blocks to this vector!
Fixes PR18724.
llvm-svn: 201028
The bitcast instruction during constant materialization was not placed correcly
in the presence of phi nodes. This commit fixes the insertion point to be in the
idom instead.
This fixes PR18768
llvm-svn: 201009
This fix first traverses the whole use list of the constant expression and
keeps track of the instructions that need to be updated. Then perform the
fixup afterwards.
llvm-svn: 201008
- Properly displaying non null terminated StringRef.
- Auto expanding pointer types.
- Displaying real type names for PointerUnions.
- Using "size" and "capacity" across all containers.
- Simplifying code where possible.
llvm-svn: 201004
An intermediate solution until the problems with analyzer plugins linking with
llvm/Support and causing assertions due to duplicate GeneralCategory are solved.
llvm-svn: 200981
According to the AAPCS, when a CPRC is allocated to the stack, all other
VFP registers should be marked as unavailable.
I have also modified the rules for allocating non-CPRCs to the stack, to make
it more explicit that all GPRs must be made unavailable. I cannot think of a
case where the old version would produce incorrect answers, so there is no test
for this.
llvm-svn: 200970
Generalize the AArch64 .td nodes for AssertZext and AssertSext. Use
them to match the relevant pextr store instructions.
The test widen_load-2.ll requires a slight change because with the
stores gone, the remaining instructions are scheduled in a different
order.
Add test cases for SSE4 and AVX variants.
Resolves rdar://13414672.
Patch by Adam Nemet <anemet@apple.com>.
llvm-svn: 200957
mode.
Basically the idea is to transform code like this:
%idx = add nsw i32 %a, 1
%sextidx = sext i32 %idx to i64
%gep = gep i8* %myArray, i64 %sextidx
load i8* %gep
Into:
%sexta = sext i32 %a to i64
%idx = add nsw i64 %sexta, 1
%gep = gep i8* %myArray, i64 %idx
load i8* %gep
That way the computation can be folded into the addressing mode.
This transformation is done as part of the addressing mode matcher.
If the matching fails (not profitable, addressing mode not legal, etc.), the
matcher will revert the related promotions.
<rdar://problem/15519855>
llvm-svn: 200947
This solves a problem where a def machine operand has no uses but has
not been marked dead. In this case, the initial RP analysis was being
extra precise and determining from LiveIntervals the the register was
actually dead. This caused us to omit the register from the RP
tracker's block live out. That's all good, but the per-instruction
summary still accounted for it as a valid def. This could cause an
assertion in the tracker later when we underflow pressure.
This is from a bug report on an out-of-tree target. It is not
reproducible on well-behaved targets. I'm just making an obvious fix
without unit test.
llvm-svn: 200941
There was a problem with the old pattern, so we were copying some
larger immediates into registers when we could have been encoding
them in the instruction.
llvm-svn: 200932
In a previous commit (r199818) we added a const_cast to an existing
subtarget info instead of creating a new one so that we could reuse
it when creating the TargetAsmParser for parsing inline assembly.
This cast was necessary because we needed to reuse the existing STI
to avoid generating incorrect code when the inline asm contained
mode-switching directives (e.g. .code 16).
The root cause of the failure was that there was an implicit sharing
of the STI between the parser and the MCCodeEmitter. To fix a
different but related issue, we now explicitly pass the STI to the
MCCodeEmitter (see commits r200345-r200351).
The const_cast is no longer necessary and we can now create a fresh
STI for the inline asm parser to use.
Differential Revision: http://llvm-reviews.chandlerc.com/D2709
llvm-svn: 200929
The most important part of this is probably adding any cost at all for
operations like zext <8 x i8> to <8 x i32>. Before they were being
recorded as extremely costly (24, I believe) which made LLVM fall back
on a 4-wide vectorisation of a loop.
It also rebalances the values for sext, zext and trunc. Lacking any
other sane metric that might work across CPU microarchitectures I went
for instructions. This seems to be in reasonable accord with the rest
of the table (sitofp, ...) though no doubt at least one value is
sub-optimal for some bizarre reason.
Finally, separate AVX and AVX2 values are provided where appropriate.
The CodeGen is quite different in many cases.
rdar://problem/15981990
llvm-svn: 200928
The aim in this patch is to reduce work that VirtRegRewriter needs to do when
telling MachineRegisterInfo which physregs are in use. Up until now
VirtRegRewriter::rewrite has been doing rewriting and populating def info and
then proceeding to set whether a physreg is used based this info for every
physreg that the target provides. This can be expensive when a target has an
unusually high number of supported physregs, and is a noticeable chunk of
compile time for small programs on such targets.
So to reduce compile time, this patch simply adds the use of a SparseSet to the
rewrite function that is used to flag each physreg that is encountered in a
MachineFunction. Afterward, rather than iterating over the set of all physregs
for a given target to set the physregs used in MachineRegisterInfo, the new way
is to iterate over the set of physregs that were actually encountered and set
in the SparseSet. This improves compile time because the existing rewrite
function was iterating over all MachineOperands already, and because the
iterations afterward to setPhysRegUsed is reduced by use of the SparseSet data.
llvm-svn: 200919
I believe VZEXT_MOVL means "zero all vector elements except the first" (and
should have identical input & output types) whereas VZEXT means "zero extend
each element of a vector (discarding higher elements if necessary)".
For example:
(v4i32 (vzext (v16i8 ...)))
should zero extend the low 4 bytes of the incoming vector to 32-bits,
discarding higher bytes.
However, somewhere in the past, these two concepts had become confused, even
leading to a nonsensical VSEXT_MOVL.
This re-merges the nodes where appropriate (all VSEXT_MOVL -> VSEXT, VZEXT_MOVL
-> VZEXT when it's an actual extension).
rdar://problem/15981990
llvm-svn: 200918
programs on targets with large register files. The root of the compile time
overhead was in the use of llvm::SmallVector to hold PhysRegEntries, which
resulted in slow-down from calling llvm::SmallVector::assign(N, 0). In contrast
std::vector uses the faster __platform_bzero to zero out primitive buffers when
assign is called, while SmallVector uses an iterator.
The fix for this was simply to replace the SmallVector with a dynamically
allocated buffer and to initialize or reinitialize the buffer based on the
total registers that the target architecture requires. The changes support
cases where a pass manager may be reused for different targets, and note that
the PhysRegEntries is allocated using calloc mainly for good for, and also to
quite tools like Valgrind (see comments for more info on this).
There is an rdar to track the fact that SmallVector doesn't have platform
specific speedup optimizations inside of it for things like this, and I'll
create a bugzilla entry at some point soon as well.
TL;DR: This fix replaces the expensive llvm::SmallVector<unsigned
char>::assign(N, 0) with a call to calloc for N bytes which is much faster
because SmallVector's assign uses iterators.
llvm-svn: 200917
large register files. The omission of Queries.clear() is perfectly safe because
LiveIntervalUnion::Query doesn't contain any data that needs freeing and
because LiveRegMatrix::runOnFunction happens to reset the OwningArrayPtr
holding Queries every time it is run, so there's no need to zero out the
queries either. Not having to do this for very large numbers of physregs
is a noticeable constant cost reduction in compilation of small programs.
llvm-svn: 200913
build but spectacularly changed behavior of the C++98 build. =]
This shows my one problem with not having unittests -- basic API
expectations aren't well exercised by the integration tests because they
*happen* to not come up, even though they might later. I'll probably add
a basic unittest to complement the integration testing later, but
I wanted to revive the bots.
llvm-svn: 200905
NAKAMURA Takumi