On Sat, Sep 13, 2003 at 10:30:55PM -0400, Christopher Faylor wrote:
> On Sat, Sep 13, 2003 at 10:07:42PM -0400, Pierre A. Humblet wrote:
> >This is the second and final part of the pinfo security patch.
>
> Looks like a Corinna yea or nay on this one.
The changes look good. Please apply, Pierre
Christopher Faylor wrote:
Thomas, I made the change below to stop a SEGV on thread exit as evinced
by the threadidafterfork test in the testsuite.
The problem is that this code overwrites impure_ptr with the contents of
the thread which called fork, which is not the correct thing to do since
_impur
At 10:03 AM 9/15/2003 +0200, you wrote:
>On Sat, Sep 13, 2003 at 10:30:55PM -0400, Christopher Faylor wrote:
>> On Sat, Sep 13, 2003 at 10:07:42PM -0400, Pierre A. Humblet wrote:
>> >This is the second and final part of the pinfo security patch.
>>
>> Looks like a Corinna yea or nay on this one.
Cygwin uses a "delete queue" in a shared file mapping to hold
the names of files that could not be deleted on unlink, usually
because they were still opened. The queue is scanned by all
processes so that the files eventually get deleted after they
are closed.
Because Everyone has write access to
You are a *God*, Pierre. ;-)
--
Gary R. Van Sickle
Brewer. Patriot.
> Cygwin uses a "delete queue" in a shared file mapping to hold
> the names of files that could not be deleted on unlink, usually
> because they were still opened. The queue is scanned by all
> processes so that the files eve
