maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/unittests/ExecutionEngine/Orc/LazyCallThroughAndReexports...

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

75 lines
2.5 KiB
C++
Raw Normal View History

[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
#include "OrcTestCommon.h"
[ORC] Extract and tidy up JITTargetMachineBuilder, add unit test. (1) Adds comments for the API. (2) Removes the setArch method: This is redundant: the setArchStr method on the triple should be used instead. (3) Turns EmulatedTLS on by default. This matches EngineBuilder's behavior. llvm-svn: 343423
2018-10-01 03:12:23 +08:00
#include "llvm/ExecutionEngine/Orc/JITTargetMachineBuilder.h"
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
#include "llvm/ExecutionEngine/Orc/LazyReexports.h"
#include "gtest/gtest.h"
using namespace llvm;
using namespace llvm::orc;
class LazyReexportsTest : public CoreAPIsBasedStandardTest {};
static int dummyTarget() { return 42; }
TEST_F(LazyReexportsTest, BasicLocalCallThroughManagerOperation) {
// Create a callthrough manager for the host (if possible) and verify that
// a call to the lazy call-through:
// (1) Materializes the MU. This verifies that the symbol was looked up, and
// that we didn't arrive at the target via some other path
// (2) Returns the expected value (which we take as proof that the call
// reached the target).
auto JTMB = JITTargetMachineBuilder::detectHost();
// Bail out if we can not detect the host.
if (!JTMB) {
consumeError(JTMB.takeError());
return;
}
// Bail out if we can not build a local call-through manager.
auto LCTM = createLocalLazyCallThroughManager(JTMB->getTargetTriple(), ES, 0);
if (!LCTM) {
consumeError(LCTM.takeError());
return;
}
[ORC] Add an 'intern' method to ExecutionEngine for interning symbol names. This cuts down on boilerplate by reducing 'ES.getSymbolStringPool().intern(...)' to 'ES.intern(...)'. llvm-svn: 343427
2018-10-01 07:18:24 +08:00
auto DummyTarget = ES.intern("DummyTarget");
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
bool DummyTargetMaterialized = false;
[llvm] Migrate llvm::make_unique to std::make_unique Now that we've moved to C++14, we no longer need the llvm::make_unique implementation from STLExtras.h. This patch is a mechanical replacement of (hopefully) all the llvm::make_unique instances across the monorepo. llvm-svn: 369013
2019-08-15 23:54:37 +08:00
cantFail(JD.define(std::make_unique<SimpleMaterializationUnit>(
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
SymbolFlagsMap({{DummyTarget, JITSymbolFlags::Exported}}),
Re-apply "[ORC] Make MaterializationResponsibility immovable..." with fixes. Re-applies c74900ca672 with fixes for the ThinLtoJIT example.
2020-09-12 00:23:14 +08:00
[&](std::unique_ptr<MaterializationResponsibility> R) {
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
DummyTargetMaterialized = true;
[ORC] Fix a FIXME: Propagate errors to dependencies. When symbols are failed (via MaterializationResponsibility::failMaterialization) any symbols depending on them will now be moved to an error state. Attempting to resolve or emit a symbol in the error state (via the notifyResolved or notifyEmitted methods on MaterializationResponsibility) will result in an error. If notifyResolved or notifyEmitted return an error due to failure of a dependence then the caller should log or discard the error and call failMaterialization to propagate the failure to any queries waiting on the symbols being resolved/emitted (plus their dependencies). llvm-svn: 369808
2019-08-24 04:37:31 +08:00
// No dependencies registered, can't fail.
Re-apply "[ORC] Make MaterializationResponsibility immovable..." with fixes. Re-applies c74900ca672 with fixes for the ThinLtoJIT example.
2020-09-12 00:23:14 +08:00
cantFail(R->notifyResolved(
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
{{DummyTarget,
JITEvaluatedSymbol(static_cast<JITTargetAddress>(
reinterpret_cast<uintptr_t>(&dummyTarget)),
[ORC] Fix a FIXME: Propagate errors to dependencies. When symbols are failed (via MaterializationResponsibility::failMaterialization) any symbols depending on them will now be moved to an error state. Attempting to resolve or emit a symbol in the error state (via the notifyResolved or notifyEmitted methods on MaterializationResponsibility) will result in an error. If notifyResolved or notifyEmitted return an error due to failure of a dependence then the caller should log or discard the error and call failMaterialization to propagate the failure to any queries waiting on the symbols being resolved/emitted (plus their dependencies). llvm-svn: 369808
2019-08-24 04:37:31 +08:00
JITSymbolFlags::Exported)}}));
Re-apply "[ORC] Make MaterializationResponsibility immovable..." with fixes. Re-applies c74900ca672 with fixes for the ThinLtoJIT example.
2020-09-12 00:23:14 +08:00
cantFail(R->notifyEmitted());
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
})));
unsigned NotifyResolvedCount = 0;
[ORC] Update lazyReexports to support aliases with different symbol names. A bug in the existing implementation meant that lazyReexports would not work if the aliased name differed from the alias's name, i.e. all lazy reexports had to be of the form (lib1, name) -> (lib2, name). This patch fixes the issue by capturing the alias's name in the NotifyResolved callback. To simplify this capture, and the LazyCallThroughManager code in general, the NotifyResolved callback is updated to use llvm::unique_function rather than a custom class. No test case yet: This can only be tested at runtime, and the only in-tree client (lli) always uses aliases with matching names. I will add a new LLJIT example shortly that will directly test the lazyReexports API and the non-trivial alias use case.
2020-01-15 09:09:02 +08:00
auto NotifyResolved = [&](JITTargetAddress ResolvedAddr) {
++NotifyResolvedCount;
return Error::success();
};
[ORC] Add a "lazy call-through" utility based on the same underlying trampoline implementation as lazy compile callbacks, and a "lazy re-exports" utility that builds lazy call-throughs. Lazy call-throughs are similar to lazy compile callbacks (and are based on the same underlying state saving/restoring trampolines) but resolve their targets by performing a standard ORC lookup rather than invoking a user supplied compiler callback. This allows them to inherit the thread-safety of ORC lookups while blocking only the calling thread (whereas compile callbacks also block one compile thread). Lazy re-exports provide a simple way of building lazy call-throughs. Unlike a regular re-export, a lazy re-export generates a new address (a stub entry point) that will act like the re-exported symbol when called. The first call via a lazy re-export will trigger compilation of the re-exported symbol before calling through to it. llvm-svn: 343061
2018-09-26 12:18:30 +08:00
auto CallThroughTrampoline = cantFail((*LCTM)->getCallThroughTrampoline(
JD, DummyTarget, std::move(NotifyResolved)));
auto CTTPtr = reinterpret_cast<int (*)()>(
static_cast<uintptr_t>(CallThroughTrampoline));
// Call twice to verify nothing unexpected happens on redundant calls.
auto Result = CTTPtr();
(void)CTTPtr();
EXPECT_TRUE(DummyTargetMaterialized)
<< "CallThrough did not materialize target";
EXPECT_EQ(NotifyResolvedCount, 1U)
<< "CallThrough should have generated exactly one 'NotifyResolved' call";
EXPECT_EQ(Result, 42) << "Failed to call through to target";
}