There already have two "dead" functions, initialize{IPO|IPA}, defined for
similar purpose. I decide not to call these two functions for two reasons:
o. they don't cover all LTO passes (which will soon be separated into IPO
and post-IPO passes)
o. We have not yet figured out the right passes and the ordering for IPO
and post-IPO stages, meaning this change is only for the time being.
Since LTO passes are registered, we are now able to print IR before and
after particular point.
For OSX users:
--------------
"...-Wl,-mllvm -Wl,-print-after=<pass-name>" will print IR after the
specified pass.
For Other UNIX with GNU gold linker:
------------------------------------
"-Wl,-plugin-opt=-print-after=<pass-name>" should work.
(NOTE: no need for "-Wl,-mllvm")
Strip "-Wl," if flags are fed directly to linker instead of clang/clang++.
llvm-svn: 186853
This matches gnu archive behavior and since archive member order can change
which member is used, not changing the order on replacement looks like the
right thing to do.
This patch also refactors the logic for which archive member to keep and
whether to move it to a helper function (computeInsertAction). The
nesting in computeNewArchiveMembers was getting a bit confusing.
llvm-svn: 186829
The original change was rolled back in r186627 because of test
failures on the big endian machine. I believe I fixed the issue
so re-submitting.
llvm-svn: 186734
This allows "llvm-mc -disassemble" to accept two new features:
+ Using comma as a byte separator
+ Grouping bytes with '[' and ']' pairs.
The behaviour outside a [...] group is unchanged. But within the group once
llvm-mc encounters a true error, it stops rather than trying to resynchronise
the stream at the next byte. This is more useful for disassembly tests, where
we have an almost-instruction in mind and don't care what the misaligned
interpretation would be. Particularly if it means llvm-mc won't actually see
the next intended almost-instruction.
As a side effect, this means llvm-mc can disassemble its own -show-encoding
output if copy-pasted.
llvm-svn: 186661
Summary:
Dump optional data directory entries in the PE/COFF header, so that
we can test the output of LLD linker. This patch updates the test binary
file, but the source of the binary is the same. I just re-linked the file.
I don't know how the previous file was linked, but the previous file did
not have any data directory entries for some reason.
Reviewers: rafael
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D1148
llvm-svn: 186623
This is a micro optimization. Instead of going char*->StringRef->Twine->char*,
go char*->Twine->char* and avoid having to copy the filename on the stack.
llvm-svn: 186380
This fixes two bugs is lib/Object that the use in llvm-ar found:
* In OS X created archives, the name can be padded with nulls. Strip them.
* In the constructor, remember the first non special member and use that in
begin_children. This makes sure we skip all special members, not just the
first one.
The change to llvm-ar itself consist of
* Using lib/Object for reading archives instead of ArchiveReader.cpp.
* Writing the modified archive directly, instead of creating an in memory
representation.
The old Archive library was way more general than what is needed, as can
be seen by the diffstat of this patch.
Having llvm-ar using lib/Object now opens the way for creating regular symbol
tables for both native objects and bitcode files so that we can use those
archives for LTO.
llvm-svn: 186197
* All systems we support have some form of long name support.
* The options has different names and semantics in different implementations
('f' on gnu, 'T' on OS X), which makes it unlikely it is normally used on
build systems.
* It was completely untested.
llvm-svn: 186078
The problem with running internalize before we're ready to output an object file
is that it may change a 'weak' symbol into an internal one, but that symbol
could be needed by an external object file --- e.g. with arclite.
<rdar://problem/14334895>
llvm-svn: 185882
This fixes a regression introduced by r185726: the new call to get
a unique file does not prepend the system temporary directory, so
we need to anchor on the file that the temporary file gets moved
to to ensure we're on the same file system.
llvm-svn: 185825
This function is complementary to createTemporaryFile. It handles the case were
the unique file is *not* temporary: we will rename it in the end. Since we
will rename it, the file has to be in the same filesystem as the final
destination and we don't prepend the system temporary directory.
This has a small semantic difference from unique_file: the default mode is 0666.
This matches the behavior of most unix tools. For example, with this change
lld now produces files with the same permissions as ld. I will add a test
of this change when I port clang over to createUniqueFile (next commit).
llvm-svn: 185726
This is dead code since PIC16 was removed in 2010. The result was an odd mix,
where some parts would carefully pass it along and others would assert it was
zero (most of the object streamer for example).
llvm-svn: 185436
This kind of simplification is sometimes useful, but in general it's not correct.
As GNU/kFreeBSD is an hybrid system, for kernel-related issues we want to match the
build definitions used for FreeBSD, whereas for userland-related issues we want to
match the definitions used for other systems with Glibc.
The current modification adjusts the build system so that they can be distinguished,
and explicitly adds GNU/kFreeBSD to the build checks in which it belongs.
Fixes bug #16444.
Patch by Robert Millan in the context of Debian.
llvm-svn: 185311
- warn users when -debug-ir is used with old JIT engine (only partial debug
info is available)
For example, to debug an IR file with GDB (that supports JIT registration), do:
$ gdb --args lli -use-mcjit -debug-ir testcase.ll
(gdb) break main
(gdb) run
<Process continues to lli main>
(gdb) continue
<Process continues to testcase.ll main()
(gdb) step
<Now stepping through the LLVM IR in testcase.ll>
llvm-svn: 185197
sys::fs::unique_file will now loop infinitely if provided with a file name
without '%' characters and the input file already exists. As a result, bugpoint
cannot use a fixed file name for the execution output (including the reference
output).
llvm-svn: 185166
No functionality change.
It should suffice to check the type of a debug info metadata, instead of
calling Verify. For cases where we know the type of a DI metadata, use
assert.
Also update testing cases to make them conform to the format of DI classes.
llvm-svn: 185135
Where a source tree is complete with lld, lldb and polly, it may not be possible to use cmake to configure build scripts if the host compiler it not capable of compiling these sub-projects. This change makes it possible to first build a bootstrap clang compiler when can then be used to build a complete llvm toolchain. An example bootstrap build sequence could be as follows:
$ mkdir bootstrap
$ cd bootstrap
$ cmake -G 'Unix Makefiles'
-DCMAKE_BUILD_TYPE:STRING=Release
-DCMAKE_PREFIX_PATH:STRING=$(pwd)
-DLLVM_TARGETS_TO_BUILD:STRING=host
-DLLVM_INCLUDE_TOOLS:STRING=bootstrap-only
../source
$ make clang # build clang only for host
$ cd ..
$ export CC=$(realpath bootstrap/bin)/clang
$ export CXX=$(realpath bootstrap/bin)/clang++
$ mkdir final
$ cd final
$ cmake -G 'Unix Makefiles' ../source
$ make all check-all
llvm-svn: 184924
Although in reality the symbol table in ELF resides in a section, the
standard requires that there be no more than one SHT_SYMTAB. To enforce
this constraint, it is cleaner to group all the symbols under a
top-level `Symbols` key on the object file.
llvm-svn: 184627
The improperly aligned section content in the output was causing
buildbot failures. This should fix them.
Incidentally, this results in a simpler and more robust API for
ContiguousBlobAccumulator.
llvm-svn: 184621
Previously we unconditionally enforced that section references in
symbols in the YAML had a name that was a section name present in the
object, and linked the references to that section. Now, permit empty
section names (already the default, if the `Section` key is not
provided) to indicate SHN_UNDEF.
llvm-svn: 184513
Instead, just have 3 sub-lists, one for each of
{STB_LOCAL,STB_GLOBAL,STB_WEAK}.
This allows us to be a lot more explicit w.r.t. the symbol ordering in
the object file, because if we allowed explicitly setting the STB_*
`Binding` key for the symbol, then we might have ended up having to
shuffle STB_LOCAL symbols to the front of the list, which is likely to
cause confusion and potential for error.
Also, this new approach is simpler ;)
llvm-svn: 184506
After this patch, the ELF file produced by
`yaml2obj-elf-symbol-basic.yaml`, when linked and executed on x86_64
(under SysV ABI, obviously; I tested on Linux), produces a working
executable that goes into an infinite loop!
llvm-svn: 184469
One of the key things that the YAML format abstracts over is the use of
section numbers for referencing sections. Instead, textual section names
are used, which yaml2obj then translates into appropriate section
numbers. (Technically ELF doesn't care about section names (only section
numbers), but since this is a testing tool, readability counts).
This simplifies using section names as symbolic references in various
parts of the code. An upcoming commit will use this to allow symbols to
reference sections.
llvm-svn: 184467
This commit completely removes what is left of the simplify-libcalls
pass. All of the functionality has now been migrated to the instcombine
and functionattrs passes. The following C API functions are now NOPs:
1. LLVMAddSimplifyLibCallsPass
2. LLVMPassManagerBuilderSetDisableSimplifyLibCalls
llvm-svn: 184459
This is a basic implementation - we still don't have any support (that I
know of) for dumping DWARF expressions in a meaningful way, so the
location information itself is just printed as a sequence of bytes as we
do elsewhere.
llvm-svn: 184361
This matches GNU ar behavior. Also remove the now unused getFileStatus method.
Not sure how to add a test, it would have to run ls -l or something like that.
llvm-svn: 184337
Previously, we would monkeypatch the vector of YAML::Section's in order
to ensure that the SHT_NULL entry is present. Now we just add it
unconditionally.
The proliferation of small numerical adjustments is beginning to
frighten me, but I can't think of a way having a single point of truth
for them without introducing a whole new layer of data structures (i.e.
lots of code and complexity) between the YAML and binary ELF formats.
llvm-svn: 184260
A bug in libObject will cause it to assert() if a symbol table's string
table and the section header string table are the same section, so we
need to ensure that we emit two different string tables (among other
things). The problematic code is the hardcoded usage of ".strtab"
(`dot_strtab_sec`) for looking up symbol names in
ELFObjectFile<ELFT>::getSymbolName.
I discussed this with Michael, and he has some local improvements to the
ELF code in libObject that, among other things, should fix our handling
of this scenario.
llvm-svn: 184161
I was spotting garbage in the output. I'd like to just zero the entire
ELFYAML::Section to be sure, but it contains non-POD types. (I'm also
trying to avoid bloating the ELFYAML::Foo classes with a bunch of
constructor code).
No test, since this is by its very nature unpredictable. I'm pretty sure
that one of the sanitizers would catch it immediately though.
llvm-svn: 184160
The error message was:
/home/espindola/llvm/llvm/tools/gold/gold-plugin.cpp: In function ‘ld_plugin_status cleanup_hook()’:
/home/espindola/llvm/llvm/tools/gold/gold-plugin.cpp:461:30: error: cannot pass objects of non-trivially-copyable type ‘std::string {aka class std::basic_string<char>}’ through ‘...’
I will check if this was a clang or gcc issue.
llvm-svn: 184138
Archive files (.a) can have a symbol table indicating which object
files in them define which symbols. The purpose of this symbol table
is to speed up linking by allowing the linker the read only the .o
files it is actually going to use instead of having to parse every
object's symbol table.
LLVM's archive library currently supports a LLVM specific format for
such table. It is hard to see any value in that now that llvm-ld is
gone:
* System linkers don't use it: GNU ar uses the same plugin as the
linker to create archive files with a regular index. The OS X ar
creates no symbol table for IL files, I assume the linker just parses
all IL files.
* It doesn't interact well with archives having both IL and native objects.
* We probably don't want to be responsible for yet another archive
format variant.
This patch then:
* Removes support for creating and reading such index from lib/Archive.
* Remove llvm-ranlib, since there is nothing left for it to do.
We should in the future add support for regular indexes to llvm-ar for
both native and IL objects. When we do that, llvm-ranlib should be
reimplemented as a symlink to llvm-ar, as it is equivalent to "ar s".
llvm-svn: 184019
For consistency, change the address in the test case from 0xDEADBEEF to
0xCAFEBABE since 0xCAFEBABE that actually has a 2-byte alignment.
llvm-svn: 183962
It was only used to implement ExecuteAndWait and ExecuteNoWait. Expose just
those two functions and make Execute and Wait implementations details.
llvm-svn: 183864
These records are mandatory for executables and are used by the loader.
Reviewers: rafael
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D939
llvm-svn: 183852
Currently, only emitting the ELF header is supported (no sections or
segments).
The ELFYAML code organization is broadly similar to the COFFYAML code.
llvm-svn: 183711
from the LC_DATA_IN_CODE load command. And when disassembling print
the data in code formatted for the kind of data it and not disassemble those
bytes.
I added the format specific functionality to the derived class MachOObjectFile
since these tables only appears in Mach-O object files. This is my first
attempt to modify the libObject stuff so if folks have better suggestions
how to fit this in or suggestions on the implementation please let me know.
rdar://11791371
llvm-svn: 183424
Previously, yaml2coff.cpp had a writeHexData static helper function to
do this, but it is generally useful functionality.
Also, validate hex strings up-front to avoid running having to handle
errors "deep inside" the yaml2obj code (it also gives better diagnostics
than it used to).
llvm-svn: 183345
See the comment in yaml2obj.cpp for why this is currently needed.
Eventually we can get rid of this, but for now it is needed in order to
make forward progress with adding ELF support, and should be
straightforward to remove later.
Also, preserve the default of COFF, to avoid breaking existing tests.
This policy can easily be changed later though.
llvm-svn: 183332
In ELF (as in MachO), not all relocations point to symbols. Represent this
properly by using a symbol_iterator instead of a SymbolRef. Update llvm-readobj
ELF's dumper to handle relocatios without symbols.
llvm-svn: 183284
Specifying the load address for Darwin i386 dylibs was a performance
optimization for dyld that is not relevant for x86_64 or arm. We can just
remove this now.
llvm-svn: 183230
This patch builds on some existing code to do CFG reconstruction from
a disassembled binary:
- MCModule represents the binary, and has a list of MCAtoms.
- MCAtom represents either disassembled instructions (MCTextAtom), or
contiguous data (MCDataAtom), and covers a specific range of addresses.
- MCBasicBlock and MCFunction form the reconstructed CFG. An MCBB is
backed by an MCTextAtom, and has the usual successors/predecessors.
- MCObjectDisassembler creates a module from an ObjectFile using a
disassembler. It first builds an atom for each section. It can also
construct the CFG, and this splits the text atoms into basic blocks.
MCModule and MCAtom were only sketched out; MCFunction and MCBB were
implemented under the experimental "-cfg" llvm-objdump -macho option.
This cleans them up for further use; llvm-objdump -d -cfg now generates
graphviz files for each function found in the binary.
In the future, MCObjectDisassembler may be the right place to do
"intelligent" disassembly: for example, handling constant islands is just
a matter of splitting the atom, using information that may be available
in the ObjectFile. Also, better initial atom formation than just using
sections is possible using symbols (and things like Mach-O's
function_starts load command).
This brings two minor regressions in llvm-objdump -macho -cfg:
- The printing of a relocation's referenced symbol.
- An annotation on loop BBs, i.e., which are their own successor.
Relocation printing is replaced by the MCSymbolizer; the basic CFG
annotation will be superseded by more related functionality.
llvm-svn: 182628
This is a basic first step towards symbolization of disassembled
instructions. This used to be done using externally provided (C API)
callbacks. This patch introduces:
- the MCSymbolizer class, that mimics the same functions that were used
in the X86 and ARM disassemblers to symbolize immediate operands and
to annotate loads based off PC (for things like c string literals).
- the MCExternalSymbolizer class, which implements the old C API.
- the MCRelocationInfo class, which provides a way for targets to
translate relocations (either object::RelocationRef, or disassembler
C API VariantKinds) to MCExprs.
- the MCObjectSymbolizer class, which does symbolization using what it
finds in an object::ObjectFile. This makes simple symbolization (with
no fancy relocation stuff) work for all object formats!
- x86-64 Mach-O and ELF MCRelocationInfos.
- A basic ARM Mach-O MCRelocationInfo, that provides just enough to
support the C API VariantKinds.
Most of what works in otool (the only user of the old symbolization API
that I know of) for x86-64 symbolic disassembly (-tvV) works, namely:
- symbol references: call _foo; jmp 15 <_foo+50>
- relocations: call _foo-_bar; call _foo-4
- __cf?string: leaq 193(%rip), %rax ## literal pool for "hello"
Stub support is the main missing part (because libObject doesn't know,
among other things, about mach-o indirect symbols).
As for the MCSymbolizer API, instead of relying on the disassemblers
to call the tryAdding* methods, maybe this could be done automagically
using InstrInfo? For instance, even though PC-relative LEAs are used
to get the address of string literals in a typical Mach-O file, a MOV
would be used in an ELF file. And right now, the explicit symbolization
only recognizes PC-relative LEAs. InstrInfo should have already have
most of what is needed to know what to symbolize, so this can
definitely be improved.
I'd also like to remove object::RelocationRef::getValueString (it seems
only used by relocation printing in objdump), as simply printing the
created MCExpr is definitely enough (and cleaner than string concats).
llvm-svn: 182625
Move the processing of the command line options to right before we create the
TargetMachine instead of after.
<rdar://problem/13468287>
llvm-svn: 182611
On 32-bit hosts %p can print garbage when given a uint64_t, we should
use %llx instead. This only affects the output of the debugging text
produced by lli.
llvm-svn: 182209
BitVector/SmallBitVector::reference::operator bool remain implicit since
they model more exactly a bool, rather than something else that can be
boolean tested.
The most common (non-buggy) case are where such objects are used as
return expressions in bool-returning functions or as boolean function
arguments. In those cases I've used (& added if necessary) a named
function to provide the equivalent (or sometimes negative, depending on
convenient wording) test.
One behavior change (YAMLParser) was made, though no test case is
included as I'm not sure how to reach that code path. Essentially any
comparison of llvm::yaml::document_iterators would be invalid if neither
iterator was at the end.
This helped uncover a couple of bugs in Clang - test cases provided for
those in a separate commit along with similar changes to `operator bool`
instances in Clang.
llvm-svn: 181868
EngineBuilder interface required a JITMemoryManager even if it was being used
to construct an MCJIT. But the MCJIT actually wants a RTDyldMemoryManager.
Consequently, the SectionMemoryManager, which is meant for MCJIT, derived
from the JITMemoryManager and then stubbed out a bunch of JITMemoryManager
methods that weren't relevant to the MCJIT.
This patch fixes the situation: it teaches the EngineBuilder that
RTDyldMemoryManager is a supertype of JITMemoryManager, and that it's
appropriate to pass a RTDyldMemoryManager instead of a JITMemoryManager if
we're using the MCJIT. This allows us to remove the stub methods from
SectionMemoryManager, and make SectionMemoryManager a direct subtype of
RTDyldMemoryManager.
llvm-svn: 181820
It was just a less powerful and more confusing version of
MCCFIInstruction. A side effect is that, since MCCFIInstruction uses
dwarf register numbers, calls to getDwarfRegNum are pushed out, which
should allow further simplifications.
I left the MachineModuleInfo::addFrameMove interface unchanged since
this patch was already fairly big.
llvm-svn: 181680
- requires existing debug information to be present
- fixes up file name and line number information in metadata
- emits a "<orig_filename>-debug.ll" succinct IR file (without !dbg metadata
or debug intrinsics) that can be read by a debugger
- initialize pass in opt tool to enable the "-debug-ir" flag
- lit tests to follow
llvm-svn: 181467
The alignment is just a byte in the middle of Characteristics, not an
independent flag. Making it an independent field in the yaml
representation makes it more yamlio friendly.
llvm-svn: 181243
Update comments, fix * placement, fix method names that are not
used in clang, add a linkInModule that takes a Mode and put it
in Linker.cpp.
llvm-svn: 181099
Build attribute sections can now be read if they exist via ELFObjectFile, and
the llvm-readobj tool has been extended with an option to dump this information
if requested. Regression tests are also included which exercise these features.
Also update the docs with a fixed ARM ABI link and a new link to the Addenda
which provides the build attributes specification.
llvm-svn: 181009
For Mach-O there were 2 implementations for parsing object files. A
standalone llvm/Object/MachOObject.h and llvm/Object/MachO.h which
implements the generic interface in llvm/Object/ObjectFile.h.
This patch adds the missing features to MachO.h, moves macho-dump to
use MachO.h and removes ObjectFile.h.
In addition to making sure that check-all is clean, I checked that the
new version produces exactly the same output in all Mach-O files in a
llvm+clang build directory (including executables and shared
libraries).
To test the performance, I ran macho-dump over all the files in a
llvm+clang build directory again, but this time redirecting the output
to /dev/null. Both the old and new versions take about 4.6 seconds
(2.5 user) to finish.
llvm-svn: 180624
getRelocationAddress is for dynamic libraries and executables,
getRelocationOffset for relocatable objects.
Mark the getRelocationAddress of COFF and MachO as not implemented yet. Add a
test of ELF's. llvm-readobj -r now prints the same values as readelf -r.
llvm-svn: 180259
While here, don't report a dummy symbol for relocations that don't have symbols.
We used to says such relocations were for the first defined symbol, but now we
return end_symbols(). The llvm-readobj output change agrees with otool.
llvm-svn: 180214
LTO was always creating an empty llvm.compiler.used. With this patch we
now first check if there is anything to be added first.
Unfortunately, there is no good way to test libLTO in isolation as it needs gold
or ld64, but there are bots doing LTO builds that found this problem.
llvm-svn: 180202
The COFFParser now contains only a COFFYAML::Object and the string table
(which is recomputed, not serialized).
The structs in COFFParser now all begin with a Header field with what is
actually on the COFF object. The other fields are things that are semantically
part of the struct (relocations in a section for exmaple), but are not actually
represented that way in the object file.
llvm-svn: 180134
This is part of a future patch to use yamlio that incorrectly ended up in a
cleanup patch.
Thanks to Benjamin Kramer for reporting it.
llvm-svn: 179938
Instead, use MappingNormalization to directly parse COFF::header. Also change
the naming convention of the helper classes to be a bit shorter.
llvm-svn: 179917
Thanks to Evgeniy Stepanov for reporting this.
It might be a good idea to add a command iterator abstraction to MachO.h, but
this fixes the bug for now.
llvm-svn: 179848
This is a rework of the broken parts in r179373 which were subsequently reverted in r179374 due to incompatibility with C++98 compilers. This version should be ok under C++98.
llvm-svn: 179520
This option expands shown relocations from single line to a dictionary
format:
Relocation {
Offset: 0x4
Type: R_386_32 (1)
Symbol: sym
Info: 0x0
}
llvm-svn: 179359
Original message:
Print more information about relocations.
With this patch llvm-readobj now prints if a relocation is pcrel, its length,
if it is extern and if it is scattered.
It also refactors the code a bit to use bit fields instead of shifts and
masks all over the place.
llvm-svn: 179345
With this patch llvm-readobj now prints if a relocation is pcrel, its length,
if it is extern and if it is scattered.
It also refactors the code a bit to use bit fields instead of shifts and
masks all over the place.
llvm-svn: 179294
