Date: Mon, 18 Feb 2008 17:20:14 +0300 From: Yar Tikhiy <yar@comp.chem.msu.su> To: Craig Rodrigues <rodrigc@crodrigues.org> Cc: cvs-src@FreeBSD.org, src-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sbin/mount_nfs mount_nfs.c Message-ID: <20080218142013.GH30694@comp.chem.msu.su> In-Reply-To: <20080218104846.GA81672@crodrigues.org> References: <200802181024.m1IAOmoF089632@repoman.freebsd.org> <20080218104846.GA81672@crodrigues.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Feb 18, 2008 at 10:48:46AM +0000, Craig Rodrigues wrote: > On Mon, Feb 18, 2008 at 10:24:48AM +0000, Yar Tikhiy wrote: > > yar 2008-02-18 10:24:48 UTC > > > > FreeBSD src repository > > > > Modified files: > > sbin/mount_nfs mount_nfs.c > > Log: > > Add a work-around to make it possible again to remount > > NFS root r/w. > > This is really a horrible hack. If this hack is necessary > due to your 1.274 change to vfs_mount.c, can you back this This horrible hack is necessary because Someone(tm) started converting our NFS client to the new nmount(2) semantics a year ago but he has never finished the job. Now the NFS client just uses old mount(2) semantics over nmount(2) by stuffing all NFS-related options in "nfs_args" and passing the standard options as bit flags. At the same time, the kernel part of the NFS client already relies on string options instead of bit flags -- ever tried to remount a r/w NFS mount to read-only since nfs_vfsops.c #1.176? Hence the hack. > change and 1.274 to vfs_mount.c out, until we can analyze > the full implications of your change, and properly fix the > necessary infrastructure, instead of committing hack after hack? Which else _hacks_ did I commit? -- Yar
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080218142013.GH30694>