Issue 284019: Speculative fix for a crash where RenderWidgetHostViewWin tries to access a d...

Keyboard Shortcuts

	File
u :	up to issue
j / k :	jump to file after / before current file
J / K :	jump to next file with a comment after / before current file
	Side-by-side diff
i :	toggle intra-line diffs
e :	expand all comments
c :	collapse all comments
s :	toggle showing all comments
n / p :	next / previous diff chunk or comment
N / P :	next / previous comment
<Up> / <Down> :	next / previous line

	Issue
u :	up to list of issues
j / k :	jump to patch after / before current patch
o / <Enter> :	open current patch in side-by-side view
i :	open current patch in unified diff view

	Issue List
j / k :	jump to issue after / before current issue
o / <Enter> :	open current issue

Issue 284019: Speculative fix for a crash where RenderWidgetHostViewWin tries to access a d... (Closed)

Created:
11 years, 2 months ago by eroman

Modified:
9 years, 6 months ago

Reviewers:
jcampan

CC:
chromium-reviews_googlegroups.com, brettw+cc_chromium.org, darin (slow to review), jam, ben+cc_chromium.org

Base URL:
svn://chrome-svn/chrome/trunk/src/

Visibility:
Public.

Description

Speculative fix for a crash where RenderWidgetHostViewWin tries to access a deleted RenderWidgetHost during OnFinalMessage(). BUG=24248 Committed in r29418, and subsequently reverted due to bug 25290.

Patch Set 1 #

Patch Set 2 : '' #

Created: 11 years, 2 months ago

Download [raw] [tar.bz2]

		Unified diffs	Side-by-side diffs	Delta from patch set	Stats (+8 lines, -1 line)			Patch
	M	chrome/browser/renderer_host/render_widget_host_view_win.cc	View	1	2 chunks	+8 lines, -1 line	0 comments	Download

Messages

Total messages: 2 (0 generated)

Expand Messages | Collapse Messages

eroman

I don't know anything about this code, so my fix could be total garbage... please ...

11 years, 2 months ago (2009-10-17 00:26:07 UTC) #1

jcampan

11 years, 2 months ago (2009-10-19 17:17:56 UTC) #2

My (limited) understanding of Windows and ATL is that you call Destroy on an
HWND and then, WM_DESTROY and then WM_NCDESTROY are sent to the HWND. The ATL
processing of WM_NCDESTROY calls  OnFinalMessage(). All that happens
synchronously.
So this stack is really puzzling.
I think this speculative patch is OK.
It seemed once we destroyed the HWND, we are not using render_widget_host_
anymore. And if we do, we'd crash hard so it'd be easy to figure-out.

LGTM

Expand Messages | Collapse Messages