Closed
Bug 246329
Opened 20 years ago
Closed 20 years ago
SchemeIs failure leaks nsHttpChannel
Categories
(Core :: Networking: HTTP, defect)
Tracking
()
RESOLVED
FIXED
mozilla1.8alpha2
People
(Reporter: Biesinger, Assigned: darin.moz)
References
()
Details
(Keywords: fixed-aviary1.0, fixed1.7.5, memory-leak)
Attachments
(1 file)
1.43 KB,
patch
|
Biesinger
:
review+
sspitzer
:
superreview+
sspitzer
:
approval1.7.5+
|
Details | Diff | Splinter Review |
in nsHttpHandler::NewProxiedChannel: 1346 NS_NEWXPCOM(httpChannel, nsHttpChannel); 1347 if (!httpChannel) 1348 return NS_ERROR_OUT_OF_MEMORY; 1349 NS_ADDREF(httpChannel); 1350 1351 nsresult rv; 1352 1353 PRBool https; 1354 rv = uri->SchemeIs("https", &https); 1355 if (NS_FAILED(rv)) return rv; so if SchemeIs fails, httpChannel will never get deleted. I'd suggest moving the schemeIs call above NS_NEWXPCOM.
Assignee | ||
Updated•20 years ago
|
Status: NEW → ASSIGNED
Target Milestone: --- → mozilla1.8alpha2
Assignee | ||
Comment 1•20 years ago
|
||
Assignee | ||
Updated•20 years ago
|
Attachment #150524 -
Flags: review?(cbiesinger)
Reporter | ||
Updated•20 years ago
|
Attachment #150524 -
Flags: review?(cbiesinger) → review+
Assignee | ||
Comment 2•20 years ago
|
||
fixed-on-trunk
Status: ASSIGNED → RESOLVED
Closed: 20 years ago
Resolution: --- → FIXED
Comment 3•20 years ago
|
||
It is a fixed memory leak bug, could it be considered for 1.7.1 ?
Flags: blocking1.7.1?
Assignee | ||
Updated•20 years ago
|
Attachment #150524 -
Flags: approval1.7.1?
Comment 4•20 years ago
|
||
Comment on attachment 150524 [details] [diff] [review] v1 patch sr/a=sspitzer for 1.7.1. curious, did someone actually see this leak, or did someone find it while looking at the code? my guess is that if they saw it in action, we passed in null for the first arg, which sounds like broken code.
Attachment #150524 -
Flags: superreview+
Attachment #150524 -
Flags: approval1.7.1?
Attachment #150524 -
Flags: approval1.7.1+
Reporter | ||
Comment 6•20 years ago
|
||
someone found this while looking at the code.
> my guess is that if they saw it in action, we passed in null for the first arg,
actually no, since _that_ would've lead to a crash. well... I guess you could
call this a shutdown leak of this channel. ;)
Updated•19 years ago
|
Keywords: fixed-aviary1.0
Whiteboard: fixed-aviary1.0
You need to log in
before you can comment on or make changes to this bug.
Description
•