This is the mail archive of the gdb-patches@sources.redhat.com mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[rfa] Always use O_NOCTTY when opening tty


Based on a tip from MarkK, I tried removing the conditionalization
on USE_O_NOCTTY in inflow.c when opening a tty:

        http://sources.redhat.com/ml/gdb-patches/2004-08/msg00244.html

So I tested a simplified[1] version of Mark's patch on the following
systems: - x86-linux
         - sparc-solaris 2.5 and 2.8
         - hppa-hpux 11.00
         - alpha-osf 4.0f and 5.1a
         - ppc-aix 4.3.2 and 5.1
         - mips-irix 6.5

No regression.

So I would like Mark's patch to be committed asap, so we can rid
ourselves of the last xm file immediately after that (actually, there
is the ptrace patch too, but this one is close to being committed,
I think). Mark's patch copied here, for your convenience:

2004-08-10 Mark Kettenis  <kettenis@gnu.org>
 
        * inflow.c (O_NOCTTY): Define to zero if not already defined.
        (new_tty): Use O_NOCTTY unconditionally.

Once Mark has committed this patch, I will remove all traces of
USE_NOCTTY (including in the doc).

Thanks,
-- 
Joel

[1]: Unfortunately, I tested the same patch as Mark's, except that
     I forgot the following hunk:

        +#ifndef O_NOCTTY
        +#define O_NOCTTY 0
        +#endif

     So technically, I didn't test the same patch. But I did test
     the principle (of using O_NOCTTY regardless of the host).

Attachment: inflow.c.diff
Description: Text document


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]