Disable runtime assumption table on JITServer #9189
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
harryyu1994
force-pushed
the
disableRAT
branch
3 times, most recently
from
9d0508a
to
e89e865
Compare
|
|
harryyu1994
force-pushed
the
disableRAT
branch
3 times, most recently
from
67e71ba
to
6ba79f7
Compare
harryyu1994
changed the title
|
Ready for review. |
mpirvu
reviewed
Apr 9, 2020
| TR_RuntimeAssumptionTable * rat = compInfo->getPersistentInfo()->getRuntimeAssumptionTable(); | ||
| if (rat) | ||
| if (rat | ||
| #if defined(J9VM_OPT_JITSERVER) |
There was a problem hiding this comment.
If we take this path at the server something is wrong.
Thus, i would rather place an ASSERT_FATAL here.
There was a problem hiding this comment.
We do take this path at the server. Investigating what's wrong here.
| @@ -3550,9 +3555,6 @@ static bool updateCHTable(J9VMThread * vmThread, J9Class * cl) | |||
| TR_PersistentCHTable * table = 0; | |||
| if (TR::Options::getCmdLineOptions()->allowRecompilation() | |||
| && !TR::Options::getCmdLineOptions()->getOption(TR_DisableCHOpts) | |||
| #if defined(J9VM_OPT_JITSERVER) | |||
| && compInfo->getPersistentInfo()->getRemoteCompilationMode() != JITServer::SERVER | |||
There was a problem hiding this comment.
Why is this code removed. The idea is that, we don't want to update the CHTable at the server because it is not used for compilation anyway.
There was a problem hiding this comment.
I think persistentMemory->getPersistentInfo()->getRuntimeAssumptionTable()->init() is the only place where we call updateCHTable() at the server. Since that's disabled, updateCHTable() is no longer called at the server thus doesn't need a guard.
There was a problem hiding this comment.
To be sure let's add an ASSERT_FATAL
|
Path to |
|
So the hook was triggered by something from the "Unsafe" class from the sun.misc package (primarily designed for internal use within the core libraries). When the static VMINLINE void
putObject(J9VMThread *currentThread, MM_ObjectAccessBarrierAPI *objectAccessBarrier, j9object_t object, UDATA offset, bool isVolatile, j9object_t *value)
{
if (VM_VMHelpers::objectIsArray(currentThread, object)) {
UDATA index = convertOffsetToIndex(currentThread, offset, logFJ9ObjectSize(currentThread));
objectAccessBarrier->inlineIndexableObjectStoreObject(currentThread, object, index, *value, isVolatile);
} else if (offset & J9_SUN_STATIC_FIELD_OFFSET_TAG) {
/* Static field */
J9Class *fieldClass = J9VM_J9CLASS_FROM_HEAPCLASS(currentThread, object);
if (J9_ARE_ANY_BITS_SET(offset, J9_SUN_FINAL_FIELD_OFFSET_TAG)) {
VM_VMHelpers::reportFinalFieldModified(currentThread, fieldClass);
}
void *valueAddress = (void*)((UDATA)fieldClass->ramStatics + (offset & ~(UDATA)J9_SUN_FIELD_OFFSET_MASK));
objectAccessBarrier->inlineStaticStoreObject(currentThread, fieldClass, (j9object_t*)valueAddress, *value, isVolatile);
} else {
/* Instance field */
objectAccessBarrier->inlineMixedObjectStoreObject(currentThread, object, offset, *value, isVolatile);
}
} |
|
Ready for another review |
mpirvu
reviewed
Apr 13, 2020
| @@ -3136,6 +3136,15 @@ void jitMethodBreakpointed(J9VMThread * vmThread, J9Method *j9method) | |||
| */ | |||
| void jitIllegalFinalFieldModification(J9VMThread *currentThread, J9Class *fieldClass) | |||
| { | |||
| #if defined(J9VM_OPT_JITSERVER) | |||
There was a problem hiding this comment.
compInfo is not defined at this point
There was a problem hiding this comment.
We should have got the jitConfig like below:
J9JITConfig * jitConfig = currentThread->javaVM->jitConfig;
Maybe it compiles because there is a global somewhere.
There was a problem hiding this comment.
Good eye! I can't believe that compiled.
| @@ -3550,9 +3555,6 @@ static bool updateCHTable(J9VMThread * vmThread, J9Class * cl) | |||
| TR_PersistentCHTable * table = 0; | |||
| if (TR::Options::getCmdLineOptions()->allowRecompilation() | |||
| && !TR::Options::getCmdLineOptions()->getOption(TR_DisableCHOpts) | |||
| #if defined(J9VM_OPT_JITSERVER) | |||
| && compInfo->getPersistentInfo()->getRemoteCompilationMode() != JITServer::SERVER | |||
There was a problem hiding this comment.
To be sure let's add an ASSERT_FATAL
|
jenkins compile plinuxjit jdk11 |
JITServer should not create any runtime assumptions, instead it should buffer them and send them to the client. In order to catch all mistakes, it's better to disable RuntimeAssumptionTable on JITServer so any misuses will be caught by crashes. Issue: eclipse-openj9#8029 Signed-off-by: Harry Yu <harryyu1994@gmail.com>
|
jenkins test sanity plinuxjit,xlinuxjit jdk11 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
JITServer should not create any runtime assumptions, instead it
should buffer them and send them to the client. In order to catch
all mistakes, it's better to disable RuntimeAssumptionTable on JITServer
so any misuses will be caught by crashes.
Issue: #8029
Signed-off-by: Harry Yu harryyu1994@gmail.com