This is the mail archive of the
cygwin
mailing list for the Cygwin project.
Re: on the road to 1.5.19 - snapshot testing needed
On Fri, Sep 30, 2005 at 07:34:58PM -0400, Christopher Faylor wrote:
>On Fri, Sep 30, 2005 at 05:08:10PM -0500, Brian Ford wrote:
>>On Fri, 30 Sep 2005, Brian Ford wrote:
>>> On Thu, 29 Sep 2005, Christopher Faylor wrote:
>>> > So, once again, I'd like to finally release this thing and so I'm asking
>>> > for testing help. You all should know the drill by now but, just in case
>>> > you don't:
>>>
>>> This isn't a recent regression, and it may not even be supported, but I
>>> thought I'd report it. I'm also working on a STC, but just in case the
>>> description is good enough to get it fixed...
>>>
>>> When doing a fork/exec of a non-cygwin process and subsequently sending it
>>> a signal (I know, probably not supported) the cygwin stub process receives
>>> the signal (confirmed by strace), but it tries to call the handler
>>> previously installed by the parent process (if there was one, and in my
>>> case there was) instead of doing the SIG_DFL action (in this case,
>>> exiting). The result is that the process does not exit and hangs (both
>>> the stub and the non-cygwin process).
>>>
>>> If this isn't enough, please respond with "not enough info" and I'll
>>> supply the STC, cygcheck, etc.
>>
>>Here ya go. Self explanatory STC and cygcheck attached.
>
>As you noted, this is not a regression from 1.5.18. It is not going to
>be fixed for 1.5.19.
Actually, nevermind. A fairly non-intrusive way to deal with this
occurred to me so it will be fixed in the next snapshot.
There are still other, somewhat similar issues involving the execing of
non-cygwin processes with things like threads still sitting around and
alarms potentially causing problems. These will not be fixed for
1.5.19.
Thanks for the test case.
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/