Mail Archives: cygwin-developers/1998/11/27/16:41:44
On Sat, Nov 28, 1998 at 12:59:06AM +0100, Corinna Vinschen wrote:
>Christopher Faylor wrote:
>>
>> On Fri, Nov 27, 1998 at 10:05:28AM -0800, Earnie Boyd wrote:
>> >> Ok. I've tweaked a couple of things in the signal code. There is a
>> >> new snapshot and a new binary at:
>> >>
>> >> ftp://cambridge.cygnus.com/pub/cygwin/
>> >> --
>> >> cgf AT cygnus DOT com
>> >> http://www.cygnus.com/
>> >>
>> >
>> >With this snapshot in NT 3.51 I occasionally get a pause after which
>> >the message "proc_terminate: WFSO(hwait_subproc) timed out" is
>> >displayed. This happens with a repetitive null command (hold the
>> >ENTER button down). Most often the repetitive ENTER will "freeze" the
>> >window and will not break so I must kill the process.
>>
>> If you can duplicate this with an strace log it should be easy to track
>> down.
>>
>> >I rebuilt the snapshot with no optimization switch. I still get the
>> >afore mentioned error but haven't been able to get it to "freeze".
>>
>> There is no evidence that optimization has anything to do with this,
>> AFAIK.
>>
>> cgf
>
>NT40SP3, CYGWIN="binmode ntea notty"
>
>bash:
> works as before as expected.
>
>tcsh:
> On command execution, sometimes approx. 10 seconds 'til prompt reappears
> but mostly indeterminant freeze after command execution.
>
>pdksh:
> On command execution, approx. 10 seconds 'til same message as
> described in my previous mail:
>
> ****** exceptions.cc:419 __get_signal_mutex failed, res 258, error 2
> 0) function exceptions.cc, line 365
>
> Very frequent indeterminant freeze after the above message.
>
>I have attached an STRACE=0xfff output of a command `/usr/bin/echo', started
>from pdksh, which leads to a frozen shell.
I missed that this was strace output from echo. This doesn't provide
enough context to be useful. If you can get this to fail with strace=1
prior to running the shell then that should provide some more useful
information.
cgf
- Raw text -