Sanitize MCJIT remote execution

MCJIT remote execution (ChildTarget+RemoteTargetExternal) protocol was in
dire need of refactoring. It was fail-prone, had no error reporting and
implemented the same message logic on every single function.

This patch rectifies it, and makes it work on ARM, where it was randomly
failing. Other architectures shall profit from this change as well, making
their buildbots and releases more reliable.

llvm-svn: 199261

1 files changed, 5 insertions, 2 deletions

diff --git a/llvm/tools/lli/lli.cpp b/llvm/tools/lli/lli.cpp
index 3766f8c0caa..c6db51a2b07 100644
--- a/llvm/tools/lli/lli.cpp
+++ b/llvm/tools/lli/lli.cpp
@@ -685,7 +685,10 @@ int main(int argc, char **argv, char * const *envp) {
     MM->setRemoteTarget(Target.get());
 
     // Create the remote target.
-    Target->create();
+    if (!Target->create()) {
+      errs() << "ERROR: " << Target->getErrorMsg() << "\n";
+      return EXIT_FAILURE;
+    }
 
     // Since we're executing in a (at least simulated) remote address space,
     // we can't use the ExecutionEngine::runFunctionAsMain(). We have to
@@ -702,7 +705,7 @@ int main(int argc, char **argv, char * const *envp) {
     DEBUG(dbgs() << "Executing '" << EntryFn->getName() << "' at 0x"
                  << format("%llx", Entry) << "\n");
 
-    if (Target->executeCode(Entry, Result))
+    if (!Target->executeCode(Entry, Result))
       errs() << "ERROR: " << Target->getErrorMsg() << "\n";
 
     // Like static constructors, the remote target MCJIT support doesn't handle