Re: sorry guys... slow on uptake today :)

2010-06-25 Thread James Mckenzie
Misha Koshelev misha...@gmail.com wrote:

So I've been working on bug 13891, but got sidetracked by what I thought
was a race condition as dlls/shell32/tests/shlexec.c was hanging.

Spent several hours tracing it to line 464 in dlls/ntdll/directory.c

if (stat( entry-mnt_dir, st ) == -1) continue;

(read from /etc/mtab)

Turned out it was my curlftpfs file system, which tends to disconnect
randomly and not unmount (doh!).

Anyway I hope to have perhaps some tests for this bug before I leave to
Russia (on Sat morning), but unfortunately I might not be able to get
this bug fixed until post-code freeze. Sorry :(

If the fix is large and ugly, it will not be looked at until after the code 
freeze is lifted anyway.

Have a safe travel.

James Mckenzie





Re: sorry guys... slow on uptake today :)

2010-06-25 Thread Misha Koshelev
On Fri, 2010-06-25 at 07:31 -0700, James Mckenzie wrote:
 Misha Koshelev misha...@gmail.com wrote:
 
 So I've been working on bug 13891, but got sidetracked by what I thought
 was a race condition as dlls/shell32/tests/shlexec.c was hanging.
 
 Spent several hours tracing it to line 464 in dlls/ntdll/directory.c
 
 if (stat( entry-mnt_dir, st ) == -1) continue;
 
 (read from /etc/mtab)
 
 Turned out it was my curlftpfs file system, which tends to disconnect
 randomly and not unmount (doh!).
 
 Anyway I hope to have perhaps some tests for this bug before I leave to
 Russia (on Sat morning), but unfortunately I might not be able to get
 this bug fixed until post-code freeze. Sorry :(
 
 If the fix is large and ugly, it will not be looked at until after the code 
 freeze is lifted anyway.
Ah I see :)

 
 Have a safe travel.
Thank you. Much appreciated.

 
 James Mckenzie
 






sorry guys... slow on uptake today :)

2010-06-24 Thread Misha Koshelev
So I've been working on bug 13891, but got sidetracked by what I thought
was a race condition as dlls/shell32/tests/shlexec.c was hanging.

Spent several hours tracing it to line 464 in dlls/ntdll/directory.c

if (stat( entry-mnt_dir, st ) == -1) continue;

(read from /etc/mtab)

Turned out it was my curlftpfs file system, which tends to disconnect
randomly and not unmount (doh!).

Anyway I hope to have perhaps some tests for this bug before I leave to
Russia (on Sat morning), but unfortunately I might not be able to get
this bug fixed until post-code freeze. Sorry :(

Misha