Closed Bug 543763 Opened 14 years ago Closed 14 years ago

AbcEnv::~AbcEnv touchs AvmCore which could be gone

Categories

(Tamarin Graveyard :: Garbage Collection (mmGC), defect, P1)

x86
macOS
defect

Tracking

(Not tracked)

VERIFIED FIXED
flash10.1

People

(Reporter: treilly, Assigned: treilly)

Details

Attachments

(1 file)

Yes the player will delete the AvmCore and the GC will finalize the AbcEnv later, chaos ensues.
Assignee: nobody → treilly
Severity: normal → blocker
Status: NEW → ASSIGNED
Priority: -- → P1
Target Milestone: --- → flash10.1
Attachment #424795 - Flags: superreview?(edwsmith)
Attachment #424795 - Flags: review?(stejohns)
Comment on attachment 424795 [details] [diff] [review]
prevent said touching

nice find. bad injection on my part.

r+ on the condition that we add a comment indicating that m_core can be stale and referencing this bug id.
Attachment #424795 - Flags: review?(stejohns) → review+
do we typically put bug id's in the code?   I'd prefer to let the hg message contain the bug id
protocol for "fixme" comments is:
// fixme: bug ####: description

(or todo, etc).  elaboration and status can go in the bug (or code, if appropriate).  duplication in the hg comments is fine too.
Comment on attachment 424795 [details] [diff] [review]
prevent said touching

The only way this is safe is if core being null implies caches don't need flushing.  i think that's valid, second opinion?
Attachment #424795 - Flags: superreview?(edwsmith) → superreview+
(In reply to comment #5)
> (From update of attachment 424795 [details] [diff] [review])
> The only way this is safe is if core being null implies caches don't need
> flushing.  i think that's valid, second opinion?

correct
tr: 3710
tr-argo: 3646
Status: ASSIGNED → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
Engineering work item.  Marking verified fixed.
Status: RESOLVED → VERIFIED
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: