This repository has been archived by the owner on Mar 24, 2022. It is now read-only.
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Outdated suggestions cannot be applied.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.
Over in #429 we adjusted
KillSwitch
to be useful in that a timeout before a guest begins is upheld, rather than ignored. As it turns out, our tests didn't cover one of the expectations we have around instance lifetimes: after a fault, instances can be run without reset. Consequently, users of lucet-runtime that ran instances post-fault without reset (such aslucet-spectest
) caught a bug: rerunning after a fault could cause us to panic in a way that aborts the process.In normal guest exits we disable termination on the instance's
KillState
, then reset it to a new one so newKillSwitch
can cancel execution for some eventual function call. To avoid the badKillState
state, we just needed to do the same when exiting through a signal handler, too.The first commit here introduces a test that fails like
(
SIGABRT
because we panic where there isn't full unwind information, and libunwind falls over)where the second commit fixes this issue, and gets us back to passing all but five spectests:
data
,elem
,globals
,imports
,linking
!