Date: Wed, 10 Jul 2002 09:59:13 -0400 (EDT) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Peter Wemm <peter@wemm.org> Cc: John Baldwin <jhb@FreeBSD.ORG>, alpha@FreeBSD.ORG Subject: Re: uac flags in the kernel Message-ID: <15660.15793.332512.618094@grasshopper.cs.duke.edu> In-Reply-To: <20020709224221.E98DB3808@overcee.wemm.org> References: <XFMail.20020709162339.jhb@FreeBSD.org> <20020709224221.E98DB3808@overcee.wemm.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Peter Wemm writes: > John Baldwin wrote: > > Ok, I have a couple of questions about the uac (unaligned access check) > > flags in the kernel. First, I think they should be a process-wide > > property, so we need to move them out of td_md.md_flags and into a > > p_md.md_flags or p_md.md_uac or some such. Secondly, why do the > > sysarch() syscalls get and set the uac of the parent process? > > uac works on the parent, because it would be a bit silly to work on > itself since it is just about to exit. > > Actually, I'd like it to behave more like env(1)/ktrace etc. > > uac ... - operates on parent > uac -p nnn ... - operates on process nnn > uac ... /bin/sh - execs /bin/sh with specified uac settings. All this sounds fine, except for the -p flag, which currently disables printing of unaligned access errors. FWIW, the uac flags were named so as to be compatable with the Tru64 uac routine for users migrating from Tru64. Drew To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15660.15793.332512.618094>