signsign.cpp uses a static variable si_UserHasBeenSelected to solve certain problems regarding the list of multiple users that is sometimes displayed when more than one login is saved for the same URL. This could result in a race condition if two pages are loading simultaneously in two separate windows and both pages have login forms and there are more than one saved logins for each of the two pages. It's probably very unlikely that this will ever happen, and certainly very difficult to give a set of instructions that will actually demonstrate the problem.
Status: NEW → ASSIGNED
Summary: Possible race condition regarding static variable in singsign.cpp → [z]Possible race condition regarding static variable in singsign.cpp
Target Milestone: --- → Future
How do you know a bad race is unlikely? A frameset with two frames, each containing text widgets, where the content of the two frames results in the right combination of TCP segment boundaries, or just tag open/close relations, to sink content in an interleaved fashion, could do it. Cc'ing layout peers. Does the [z] cybercrud mean "I want to mark this bug WONTFIX but I'm scared of being criticized"? Maybe the right answer is to use the Priority field as it was intended to be used. /be
Summary: [z]Possible race condition regarding static variable in singsign.cpp → Possible race condition regarding static variable in singsign.cpp
Netscape nav triage team: based on Steve Morse's pretriage recommendation, this is not a beta stopper.
Reassigning to new module owner.
Assignee: morse → dveditz
Status: ASSIGNED → NEW
Priority: P3 → --
Target Milestone: Future → ---
Obsoleted by Bug 390025 (Move to LoginManager and remove wallet from SeaMonkey) Not using said file any more.
Status: NEW → RESOLVED
Last Resolved: 10 years ago
Resolution: --- → WONTFIX
You need to log in before you can comment on or make changes to this bug.