230456 - potential invalid memory access [@nsFontMetricsGTK::TryNodes]

Status: RESOLVED FIXED

(Core Graveyard :: GFX: Gtk, defect)

Description

[Darin Fisher]

potential invalid memory access [@nsFontMetricsGTK::TryNodes]

the following code is invalid:

  const char *FFREName = PromiseFlatCString(aFFREName).get();
  nsCStringKey key(FFREName);

FFREName could point to junk once key is constructed.

Comment 1

[Darin Fisher]

Attachment: v1 patch

Comment 2

[Darin Fisher]

Attachment: v1.1 patch

slightly better version.

Comment 3

[Darin Fisher]

fixed-on-trunk

Comment 4

[Roland Mainz]

Reopeing for Xlib port...

Comment 5

[Roland Mainz]

Suffering is all mine...

Comment 6

[Roland Mainz]

Attachment: Xlib port for v1.1 patch (attachment 138673), patch for 2004-02-01-16-trunk

Comment 7

[Darin Fisher]

didn't dbaron fix this already?  (i.e., last week sometime)

Comment 8

[Roland Mainz]

Darin Fisher wrote:
> didn't dbaron fix this already?  (i.e., last week sometime)

Per
http://bonsai.mozilla.org/cvsquery.cgi?branch=HEAD&file=mozilla/gfx/&date=month
- no.

The patch was made from yesterdays trunk CVS, I simply copied the code from GTK+
to Xlib land and tested the Xprint module.

Comment 9

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

please don't reopen and reassign fixed bugs because you want to fix a different bug.

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I'll fix the Xlib issue as part of bug 232339.

Comment 11

[Roland Mainz]

David Baron wrote:
> please don't reopen and reassign fixed bugs because you want to fix a 
> different bug.

This is not a different bug. This is the same bug. The GTK+ and Xlib code is
nearly identical in this area.

Comment 12

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

I don't mind you attaching the patch to this bug.  However, the bug report as
described, is fixed, so it shouldn't be reopened, and it should remain assigned
to the person who actually did the work of fixing it.