Right now we turn off IPC HTTP networking by default for the necko xpcshell tests in the 'unit_ipc' directory. Now that we have a working basic HTTP test, we should turn on NECKO_E10S_HTTP=1 for those xpcshell tests that support it, so we can catch any regressions. I'd like to land bug 526335 (full set of necko xpcshell tests work on child process) first, and then come up with a scheme to allow NECKO_E10S_HTTP=1 to be turned on a per-test basis.
Created attachment 434185 [details] [diff] [review] Turns on e10s HTTP for xpcshell tests in unit_ipc directory For now, just turn on e10s HTTP for all necko xpcshell tests in unit_ipc. (There's only one test there, and it works with e10s HTTP). I'll figure out a plan for the rest of the existing necko tests later--right now I just want to get e10s HTTP tested on tinderbox, so we'll be alerted if it breaks.
Assignee: nobody → jduell.mcbugs
Status: NEW → ASSIGNED
Attachment #434185 - Flags: review?(dwitte)
Comment on attachment 434185 [details] [diff] [review] Turns on e10s HTTP for xpcshell tests in unit_ipc directory r=dwitte. But don't we also need a unit_ipc bit here? http://mxr.mozilla.org/mozilla-central/source/netwerk/test/Makefile.in#97
Attachment #434185 - Flags: review?(dwitte) → review+
http://hg.mozilla.org/projects/electrolysis/rev/a01250f927d0 We don't need the unit_ipc stuff mentioned in comment 2 since we're not checking this into m-c, just e10s.
Status: ASSIGNED → RESOLVED
Last Resolved: 8 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.