This moves the MemoryAccess interface out of the ExecutorProcessControl
class and splits implementation classes InProcessMemoryManager and
SelfExecutorProcessControl out of ExecutorProcessControl.h and into
their own headers.
This refactor was motivated by two bugs identified in out-of-tree
builds:
1. Some implementations of the VisitMembersFunction type (often used to
implement special loading semantics, e.g. -all_load or -ObjC) were assuming
that buffers for archive members were null-terminated, which they are not in
general. This was triggering occasional assertions.
2. Archives may include multiple members with the same file name, e.g.
when constructed by appending files with the same name:
% llvm-ar crs libfoo.a foo.o
% llvm-ar q libfoo.a foo.o
% llvm-ar t libfoo.a foo.o
foo.o
While confusing, these members may be safe to link (provided that they're
individually valid and don't define duplicate symbols). In ORC however, the
archive member name may be used to construct an ORC initializer symbol,
which must also be unique. In that case the duplicate member names lead to a
duplicate definition error even if the members define unrelated symbols.
In addition to these bugs, StaticLibraryDefinitionGenerator had grown a
collection of all member buffers (ObjectFilesMap), a BumpPtrAllocator
that was redundantly storing synthesized archive member names (these are
copied into the MemoryBuffers created for each Object, but were never
freed in the allocator), and a set of COFF-specific import files.
To fix the bugs above and simplify StaticLibraryDefinitionGenerator this
patch makes the following changes:
1. StaticLibraryDefinitionGenerator::VisitMembersFunction is generalized
to take a reference to the containing archive, and the index of the
member within the archive. It now returns an Expected<bool> indicating
whether the member visited should be treated as loadable, not loadable,
or as invalidating the entire archive.
2. A static StaticLibraryDefinitionGenerator::createMemberBuffer method
is added which creates MemoryBuffers with unique names of the form
`<archive-name>[<index>](<member-name>)`. This defers construction of
member names until they're loaded, allowing the BumpPtrAllocator (with
its redundant name storage) to be removed.
3. The ObjectFilesMap (symbol name -> memory-buffer-ref) is replaced
with a SymbolToMemberIndexMap (symbol name -> index) which should be
smaller and faster to construct.
4. The 'loadability' result from VisitMemberFunctions is now taken into
consideration when building the SymbolToMemberIndexMap so that members
that have already been loaded / filtered out can be skipped, and do not
take up any ongoing space.
5. The COFF ImportedDynamicLibraries member is moved out into the
COFFImportFileScanner utility, which can be used as a
VisitMemberFunction.
This fixes the bugs described above; and should lower memory consumption
slightly, especially for archives with many files and / or symbol where
most files are eventually loaded.
This patch replaces:
llvm::copy(Src, std::back_inserter(Dst));
with:
llvm::append_range(Dst, Src);
for breavity.
One side benefit is that llvm::append_range eventually calls
llvm::SmallVector::reserve if Dst is of llvm::SmallVector.
Commit 253e11695ba added support for emulating weak-linking against dylibs
that are (under the emulation) absent at runtime. This commit extends emulated
weak linking support to allow a real dylib to supply the interface (i.e.
-weak-lx / -weak_library can be pointed at a dylib, in which case they should
be read as "weak-link against this dylib, behavining as if it weren't actually
present at runtime").
Linking libraries in ld64 with -weak-lx / -weak_library causes all references
to symbols in those libraries to be made weak, allowing the librarie to be
missing at runtime.
This patch extends EPCDynamicLibrarySearchGenerator with support for emulating
this behavior: If an instance is constructed with an Allow predicate but no
dylib handle then all symbols matching the predicate are immediately resolved
to null.
The llvm-jitlink tool is updated with -weak-lx / -weak_library options for
testing. Unlike their ld64 counterparts these options take a TBD file as input,
and always resolve all exports in the TBD file to null.
While processing library link options that check search paths (-lx, -hidden-lx,
etc.) we shouldn't generate candidate paths with extensions that are invalid
for the option being visited (e.g. -hidden-lx only applies to archives, so we
shouldn't generate candidates with `.so` extensions).
Note: Candidate extensions should probably be further filtered based on the OS
of the executing process. This patch is a step in the right direction though.
We had been abusing the setOverrideObjectFlagsWithResponsibilityFlags method to
do this. Handling it explicitly ensures that flags are only modified on the
intended files, and not accedintally modified elsewhere.
The find-dynamic-unwind-info callback registration APIs in libunwind
limit the number of callbacks that can be registered. If we use multiple
UnwindInfoManager instances, each with their own own callback function
(as was the case prior to this patch) we can quickly exceed this limit
(see https://github.com/llvm/llvm-project/issues/126611).
This patch updates the UnwindInfoManager class to use a singleton
pattern, with the single instance shared between all LLVM JITs in the
process.
This change does _not_ apply to compact unwind info registered through
the ORC runtime (which currently installs its own callbacks).
As a bonus this change eliminates the need to load an IR "bouncer"
module to supply the unique callback for each instance, so support for
compact-unwind can be extended to the llvm-jitlink tools (which does not
support adding IR).
This reapplies 6d72bf47606, which was reverted in 57447d3ddf to investigate
build failures, e.g. https://lab.llvm.org/buildbot/#/builders/3/builds/10114.
The original patch contained an invalid unused friend declaration of
std::make_shared. This has been removed.
Also adds a new IdleTask type and updates DynamicThreadPoolTaskDispatcher to
schedule IdleTasks whenever the total number of threads running is less than
the maximum number of MaterializationThreads.
A SimpleLazyReexportsSpeculator instance maintains a list of speculation
suggestions ((JITDylib, Function) pairs) and registered lazy reexports. When
speculation opportunities are available (having been added via
addSpeculationSuggestions or when lazy reexports were created) it schedules
an IdleTask that triggers the next speculative lookup as soon as resources
are available. Speculation suggestions are processed first, followed by
lookups for lazy reexport bodies. A callback can be registered at object
construction time to record lazy reexport executions as they happen, and these
executions can be fed back into the speculator as suggestions on subsequent
executions.
The llvm-jitlink tool is updated to support speculation when lazy linking is
used via three new arguments:
-speculate=[none|simple] : When the 'simple' value is specified a
SimpleLazyReexportsSpeculator instances is used
for speculation.
-speculate-order <path> : Specifies a path to a CSV containing
(jit dylib name, function name) triples to use
as speculative suggestions in the current run.
-record-lazy-execs <path> : Specifies a path in which to record lazy function
executions as a CSV of (jit dylib name, function
name) pairs, suitable for use with
-speculate-order.
The same path can be passed to -speculate-order and -record-lazy-execs, in
which case the file will be overwritten at the end of the execution.
No testcase yet: Speculative linking is difficult to test (since by definition
execution behavior should be unaffected by speculation) and this is an new
prototype of the concept*. Tests will be added in the future once the interface
and behavior settle down.
* An earlier implementation of the speculation concept can be found in
llvm/include/llvm/ExecutionEngine/Orc/Speculation.h. Both systems have the
same goal (hiding compilation latency) but different mechanisms. This patch
relies entirely on information available in the controller, where the old
system could receive additional information from the JIT'd runtime via
callbacks. I aim to combine the two in the future, but want to gain more
practical experience with speculation first.
ORC and JITLink debugging output write the dbgs() raw_ostream, which isn't
thread-safe. Use -num-threads=0 to force single-threaded linking for tests that
produce debugging output.
The llvm-jitlink tool is updated to suggest -num-threads=0 when debugging
output is enabled.
Commit edca1d9bad2 enabled threaded linking by default in llvm-jitlink, but we
need to handle the case where LLVM is built with -DLLVM_ENABLE_THREADS=Off.
This patch updates the llvm-jitlink tool to switch back to materialization on
the main thread (equivalent to llvm-jitlink -num-threads=0 ...) when LLVM is
built without thread support.
This reapplies edca1d9bad2 which was reverted in 7ec139ad4bc due to bot
failures.
LocalDependencyPropagation.s is updated to use -num-threads=0 in order to
avoid interleaving debugging output.
ELFNixPlatform.h is updated to protect the deferred runtime function calls
map during bootstrap.
Reapplies commit edca1d9bad2 which was reverted in 34531cff638 while I
investigated bot failures, (e.g.
https://lab.llvm.org/buildbot/#/builders/137/builds/10791).
Commit 158a60051d2 should address the -check failures on the bots, which
were caused by checks running earlier under the concurrent linking
scheme before all files referenced by the checks had been fully linked.
This patch also fixes the -threads option failure by renaming the option
to -num-threads to avoid clashing with the ThreadCount cl::opt variable
defined in ThinLTOCodeGenerator.cpp.
ORC dependence tracking is fine-grained (i.e. per-symbol), however when running
-check mode we want to wait for all links triggered by the entry point lookup
to complete, regardless of whether the code / data in them is actually
reachable from the entry point. This simplifies test-cases, since authors don't
need to reason about per-symbol dependencies to know that additional files will
be linked (if referenced transitively in any way from the test-case).
The new Session::waitForFilesLinkedFromEntryPointFile utility does _not_ wait
for lazily linked (-lazy) files.
This will be used to fix buildbot errors caused by edca1d9bad2.
Adds a -threads option to llvm-jitlink. By default llvm-jitlink will now use a
DynamicThreadPoolTaskDispatcher with the number of materialization threads set
to whatever is returned by std::hardware_concurrency(). This brings the default
in-place linking behavior in line with the concurrent linking that is used for
-oop-executor and -oop-executor-connect mode.
In-place linking on the main thread can be forced by passing -threads=0.
This re-applies 570ecdcf8b4, which was reverted in 74e8a37ff32 due to bot
failures. This commit renames sysv_resolve.cpp to resolve.cpp, which was the
cause of the config errors.
This reapplies 570ecdcf8b4, which was reverted in 6073dd923b8 due to bot
failures.
The test failures on Linux were fixed by:
1. Removing an overly restrictive assertion (query dependence on a symbol no
longer implies a MaterializingInfo for that symbol)
2. Adding reentry and resolver files to the ORC runtime CMakeLists.txt for
Linux.
3. Adding the __orc_rt_reentry -> __orc_rt_sysv_reentry alias to ELFNixPlatform.
…d reentry.
These utilities provide new, more generic and easier to use support for
lazy compilation in ORC.
LazyReexportsManager is an alternative to LazyCallThroughManager. It
takes requests for lazy re-entry points in the form of an alias map:
lazy-reexports = {
( <entry point symbol #1>, <implementation symbol #1> ),
( <entry point symbol #2>, <implementation symbol #2> ),
...
( <entry point symbol #n>, <implementation symbol #n> )
}
LazyReexportsManager then:
1. binds the entry points to the implementation names in an internal
table.
2. creates a JIT re-entry trampoline for each entry point.
3. creates a redirectable symbol for each of the entry point name and
binds redirectable symbol to the corresponding reentry trampoline.
When an entry point symbol is first called at runtime (which may be on
any thread of the JIT'd program) it will re-enter the JIT via the
trampoline and trigger a lookup for the implementation symbol stored in
LazyReexportsManager's internal table. When the lookup completes the
entry point symbol will be updated (via the RedirectableSymbolManager)
to point at the implementation symbol, and execution will proceed to the
implementation symbol.
Actual construction of the re-entry trampolines and redirectable symbols
is delegated to an EmitTrampolines functor and the
RedirectableSymbolsManager respectively.
JITLinkReentryTrampolines.h provides a JITLink-based implementation of
the EmitTrampolines functor. (AArch64 only in this patch, but other
architectures will be added in the near future).
Register state save and reentry functionality is added to the ORC
runtime in the __orc_rt_sysv_resolve and __orc_rt_resolve_implementation
functions (the latter is generic, the former will need custom
implementations for each ABI and architecture to be supported, however
this should be much less effort than the existing OrcABISupport
approach, since the ORC runtime allows this code to be written as native
assembly).
The resulting system:
1. Works equally well for in-process and out-of-process JIT'd code.
2. Requires less boilerplate to set up.
Given an ObjectLinkingLayer and PlatformJD (JITDylib containing the ORC
runtime), setup is just:
```c++
auto RSMgr = JITLinkRedirectableSymbolManager::Create(OLL);
if (!RSMgr)
return RSMgr.takeError();
auto LRMgr = createJITLinkLazyReexportsManager(OLL, **RSMgr, PlatformJD);
if (!LRMgr)
return LRMgr.takeError();
```
after which lazy reexports can be introduced with:
```c++
JD.define(lazyReexports(LRMgr, <alias map>));
```
LazyObectLinkingLayer is updated to use this new method, but the LLVM-IR
level CompileOnDemandLayer will continue to use LazyCallThroughManager
and OrcABISupport until the new system supports a wider range of
architectures and ABIs.
The llvm-jitlink utility's -lazy option now uses the new scheme. Since
it depends on the ORC runtime, the lazy-link.ll testcase and associated
helpers are moved to the ORC runtime.
Use SymbolStringPtr for Symbol names in LinkGraph. This reduces string interning
on the boundary between JITLink and ORC, and allows pointer comparisons (rather
than string comparisons) between Symbol names. This should improve the
performance and readability of code that bridges between JITLink and ORC (e.g.
ObjectLinkingLayer and ObjectLinkingLayer::Plugins).
To enable use of SymbolStringPtr a std::shared_ptr<SymbolStringPool> is added to
LinkGraph and threaded through to its construction sites in LLVM and Bolt. All
LinkGraphs that are to have symbol names compared by pointer equality must point
to the same SymbolStringPool instance, which in ORC sessions should be the pool
attached to the ExecutionSession.
---------
Co-authored-by: Lang Hames <lhames@gmail.com>
LazyObjectLinkingLayer can be used to add object files that will not be linked
into the executor unless some function that they define is called at runtime.
(References to data members defined by these objects will still trigger
immediate linking)
To implement lazy linking, LazyObjectLinkingLayer uses the lazyReexports
utility to construct stubs for each function in a given object file, and an
ObjectLinkingLayer::Plugin to rename the function bodies at link-time. (Data
symbols are not renamed)
The llvm-jitlink utility is extended with a -lazy option that can be
passed before input files or archives to add them using the lazy linking
layer rather than the base ObjectLinkingLayer.
This patch adds support for forced loading of archive members, similar to the
behavior of the -all_load and -ObjC options in ld64. To enable this, the
StaticLibraryDefinitionGenerator class constructors are extended with a
VisitMember callback that is called on each member file in the archive at
generator construction time. This callback can be used to unconditionally add
the member file to a JITDylib at that point.
To test this the llvm-jitlink utility is extended with -all_load (all platforms)
and -ObjC (darwin only) options. Since we can't refer to symbols in the test
objects directly (these would always cause the member to be linked in, even
without the new flags) we instead test side-effects of force loading: execution
of constructors and registration of Objective-C metadata.
rdar://134446111
We can get a reference to the `ExecutionSession` from the
`ObjectLinkingLayer` argument, so there's no need to pass it in
separately.
This mirrors recent changes to `ElfNixPlatform` and `MachOPlatform` by
@lhames in
3dba4ca155
and
cc20dd285a.
This allows us to rewrite part of StaticLibraryDefinitionGenerator in terms of
loadLinkableFile.
It's also useful for clients who may not know (either from file extensions or
context) whether a given path will be an object file, an archive, or a
universal binary.
rdar://134638070
ORC supports loading relocatable object files into a JIT'd process. The
raw "add object file" API (ObjectLayer::add) accepts plain relocatable
object files as llvm::MemoryBuffers only and does not check that the
object file's format or architecture are compatible with the process
that it will be linked in to. This API is flexible, but places the
burden of error checking and universal binary support on clients.
This commit introduces a new utility, loadRelocatableObject, that takes
a path to load and a target triple and then:
1. If the path does not exist, returns a FileError containing the
invalid path.
2. If the path points to a MachO universal binary, identifies and
returns MemoryBuffer covering the slice that matches the given triple
(checking that the slice really does contains a valid MachO relocatable
object with a compatible arch).
3. If the path points to a regular relocatable object file, verifies
that the format and architecture are compatible with the triple.
Clients can use loadRelocatableObject in the common case of loading
object files from disk to simplify their code.
Note: Error checking for ELF and COFF is left as a FIXME.
rdar://133653290
This re-applies 6094b3b7db7, which was reverted in e7efd37c229 (and before that
in 1effa19de24) due to bot failures.
The test failures were fixed by having SelfExecutorProcessControl use an
InPlaceTaskDispatcher by default, rather than a DynamicThreadPoolTaskDispatcher.
This shouldn't be necessary (and indicates a concurrency issue elsewhere), but
InPlaceTaskDispatcher is a less surprising default, and better matches the
existing behavior (compilation on current thread by default), so the change
seems reasonable. I've filed https://github.com/llvm/llvm-project/issues/89870
to investigate the concurrency issue as a follow-up.
Coding my way home: 6.25133S 127.94177W
