Date: Thu, 2 Sep 2010 15:43:09 -0400 From: Ed Maste <emaste@freebsd.org> To: Ryan Stone <rysto32@gmail.com> Cc: Attilio Rao <attilio@freebsd.org>, svn-src-projects@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r212142 - in projects/sv/sys: amd64/amd64 arm/arm geom i386/i386 ia64/ia64 net powerpc/powerpc sparc64/sparc64 sun4v/sun4v sys Message-ID: <20100902194309.GA36288@sandvine.com> In-Reply-To: <AANLkTi=%2Bh-Pm%2BncXekd4-B6sc5nzKy9Hf8p1pY2RmOCZ@mail.gmail.com> References: <201009020933.o829XmS9098339@svn.freebsd.org> <AANLkTi=%2Bh-Pm%2BncXekd4-B6sc5nzKy9Hf8p1pY2RmOCZ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Sep 02, 2010 at 10:35:13AM -0400, Ryan Stone wrote: > On Thu, Sep 2, 2010 at 5:33 AM, Attilio Rao <attilio@freebsd.org> wrote: > > Log: > > Introduce a flag field for the dumpers and use it immediately for the > > DIF_NET flag. > > This flag is responsible for toggling a netdump and having a correct set > > of dumplo in the dumping routines. > > > > To be honest, I'd be happier if the DIF_NET flag disappeared entirely. > It seems to me that dumpsys shouldn't need knowledge of disk geometry > -- especially now when there may not be a disk at all. It seems to me > that the correct interface would be for dumpsys to pass the offset > into the core file to the dumper, and the dumper should convert that > to a value appropriate for the dump device. I agree; in fact, on my ToDo list I've got: 14. dump_write API change to take offset (and add di->mediaoffset itself) * makes netdumping slightly cleaner If this is the only use of DIF_NET flag then we certainly should be able to get rid of it. - Ed
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100902194309.GA36288>