567428 - Turn e10s/necko on by default.

Status: RESOLVED FIXED

(Core :: Networking: HTTP, defect)

Description

[Jason Duell]

Attachment: Turns on necko/e10s by default: Can set NECKO_SEPARATE_STACKS=1 to revert to previous, separate necko stack per process model.

Turns on necko/e10s by default:  Can set NECKO_SEPARATE_STACKS=1 to revert to previous, separate necko stack per process model.

Should only take a sec to review. Giving to jdm/dwitte for review--but only need 1 review, so whoever gets to it first, clear the other's review flag.

Comment 1

[[e-mail]]

Comment on attachment 446788 [details] [diff] [review]
Turns on necko/e10s by default:  Can set NECKO_SEPARATE_STACKS=1 to revert to previous, separate necko stack per process model.

>   if (!didCheck) {
>-    const char * e = PR_GetEnv("NECKO_E10S_HTTP");
>-    if (e && *e)
>+    // This allows independent necko-stacks (instead of single stack in chrome)
>+    // to still be run.  
>+    // TODO: Remove eventually.
>+    const char * e = PR_GetEnv("NECKO_SEPARATE_STACKS");
>+    if (e && *e) {
>+      amChild = false;

No need for this set, since you default it to false above. So only need the 'else' branch.

>+    } else {
>       amChild = (XRE_GetProcessType() == GeckoProcessType_Content);
>+    }

>diff --git a/netwerk/test/unit_ipc/head_ipc_setup.js b/netwerk/test/unit_ipc/head_ipc_setup.js

>+// If using NSPR logging, create child log as "${NSPR_LOG_FILE}.child" 

Do we need this for the cookie head_ipc_setup.js as well?

r=dwitte

Comment 2

[Jason Duell]

> Do we need this for the cookie head_ipc_setup.js as well?

It's already there.

http://hg.mozilla.org/projects/electrolysis/rev/5c29542bba4e