Date: Tue, 15 Jul 1997 11:51:28 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: chuckr@glue.umd.edu (Chuck Robey) Cc: joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.ORG Subject: Re: no SYSVSHM in GENERIC now.. Message-ID: <199707151851.LAA03875@phaeton.artisoft.com> In-Reply-To: <Pine.BSF.3.96.970715142504.15399C-100000@Journey2.mat.net> from "Chuck Robey" at Jul 15, 97 02:27:23 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > Sure. But why not adding a small `patch' that catches SIGSYS, and > > exits cleanly telling the user to include SYSVSHM into her kernel? > > (That's the same way the XFree86 servers use to disable MIT-SHM.) > > I think that's great, but it was the XFree86 folks who did that, not the > FreeBSD porter. If FreeBSD porters had to patch the software to eliminate > bugs, there'd be nearly no ports. Such things should be forwarded to the > software authors. The software authors, not unreasonably, expect the system calls required for the software to function, to exist. If a particular port of the software can potentially be run on a platform where the calls may not exist, then it is the job of that port to resolve the missing calls. Preferably, the OS will demand-load the facilities backing the call, via a demand-loaded LKM. Alternately, the porter should patch the software to run in the new environment, and hope the originators of the code adopt his patch. Either way, while the originators have not adopted the patch, it's the porters responsibility to ensure that it is part of the port. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199707151851.LAA03875>