[Webkit-unassigned] [Bug 154169] REGRESSION(r196384): [GTK] ASSERTION FAILED: SessionTracker::storageSession(parameters.sessionID)

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=154169

--- Comment #2 from Adrien Plazas <aplazas at igalia.com> ---
Created attachment 272013
  --> https://bugs.webkit.org/attachment.cgi?id=272013&action=review
Add some prints to stderr

This patch add traces (prints to stderr) to accessors of the session map.

When running the http/tests/security/appcache-in-private-browsing.html test with this patch, it produces this trace before the usual stacktrace:
SessionTracker::storageSessionMap()
SessionTracker::storageSessionMap()
SessionTracker::storageSessionMap()
SessionTracker::destroySession(2)
SessionTracker::storageSession(1) default case: 0x7fe2397fb480
SessionTracker::storageSession(1) default case: 0x7fe2397fb480
SessionTracker::storageSession(1) default case: 0x7fe2397fb480
SessionTracker::storageSession(2) null case: nullptr
SessionTracker::setSession(2, 0x7fff56df8620)
SessionTracker::setSession(2, 0x7fff56df8620) : 0x7fad345b3f80
SessionTracker::storageSession(2) session case: 0x7fad345b3f80
SessionTracker::storageSession(2) null case: nullptr
ASSERTION FAILED: SessionTracker::storageSession(parameters.sessionID)
../../Source/WebKit2/NetworkProcess/NetworkLoad.cpp(63) : WebKit::NetworkLoad::NetworkLoad(WebKit::NetworkLoadClient&, const WebKit::NetworkLoadParameters&)

I checked the first calls to SessionTracker::storageSessionMap as they give direct access to the session map put I found nothing suspicious with the code there.

Though what I find suspicious is the end of the trace:
SessionTracker::storageSession(2) session case: 0x7fad345b3f80
SessionTracker::storageSession(2) null case: nullptr
ASSERTION FAILED: SessionTracker::storageSession(parameters.sessionID)

When looking two times for the same element without editing the map, the first time it is found but not the second time...

So far I have no way to explain this, does anybody have a clue?

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20160223/548b2c79/attachment.html>