This is the mail archive of the
gdb-patches@sourceware.cygnus.com
mailing list for the GDB project.
Re: [RFA] Don't talk about ptrace when there's none
- To: dan at cgsoftware dot com (Daniel Berlin+list.gdb-patches)
- Subject: Re: [RFA] Don't talk about ptrace when there's none
- From: Eli Zaretskii <eliz at delorie dot com>
- Date: Thu, 6 Apr 2000 10:19:16 -0400 (EDT)
- CC: gdb-patches at sourceware dot cygnus dot com
- References: <200004061216.IAA21692@mescaline.gnu.org><ya6r8kms.fsf@dan.resnet.rochester.edu>
- Reply-to: Eli Zaretskii <eliz at is dot elta dot co dot il>
> Isn't there a test for PTRACE in the configuration script?
No, not as far as I could see. AFAIK, the target-specific files are
expected to take care of that.
> Can't we use that somehow, rather than __DJGPP__?
I'm open to suggestions. I looked but couldn't find anything
appropriate.
Ideally, HAVE_PTRACE and CANT_FORK would be enough in this case. But
the former doesn't seem to exist, while the latter is used
sporadically in a couple of source files, but doesn't appear to be a
legit global symbol that we could use.
> maybe HAVE_PTRACE exists in config.h?
I cannot find it there.
> I ask because we don't have ptrace on BeOS either, and i don't think
> we want to have more than one ifdef.
We could have some OS-specific ifdefs at the top of infrun.c, if no
other good options exist, but that's ugly.