This is the mail archive of the
mailing list for the GDB project.
Re: [RFC] win32-nat.c: Improve handling of 'set shell on'
- From: Christopher Faylor <cgf-use-the-mailinglist-please at sourceware dot org>
- To: gdb-patches at sourceware dot org, Pierre Muller <muller at ics dot u-strasbg dot fr>, Eli Zaretskii <eliz at gnu dot org>
- Date: Wed, 9 Jan 2008 20:52:43 -0500
- Subject: Re: [RFC] win32-nat.c: Improve handling of 'set shell on'
- References: <firstname.lastname@example.org> <email@example.com>
On Wed, Jan 09, 2008 at 08:54:51PM +0200, Eli Zaretskii wrote:
>> From: "Pierre Muller" <firstname.lastname@example.org>
>> Date: Wed, 9 Jan 2008 10:04:35 +0100
>> I recently discovered that if you use
>> 'set shell on' on cygwin native, there are
>> some specific problems that appear.
>> I ran the testsuite with a single line
>> -static int useshell = 0; /* use shell for subprocesses */
>> +static int useshell = 1; /* use shell for subprocesses */
>> Which made the use of shell to start the subprocess
>> the default.
>Is this really appropriate for the native Windows port of GDB as well
>as for Cygwin?
No. In fact, this really doesn't work all that well at all, AFAIK. The
current code represents my initial attempt to get it working but there
were problems which led me to believe that I needed to add some help in
Cygwin itself. I think that the main reason for wanting to use the
shell would be to interpret redirection and quoting characters and, last
I checked, that was not right. That's why it defaults to false.