Date: Fri, 10 Dec 2010 01:04:09 +0200 From: Andriy Gapon <avg@freebsd.org> To: Kostik Belousov <kostikbel@gmail.com> Cc: Bruce Cran <bruce@cran.org.uk>, src-committers@freebsd.org, svn-src-all@freebsd.org, Bruce Evans <brde@optusnet.com.au>, Bruce Cran <brucec@freebsd.org>, svn-src-head@freebsd.org, Matthew Jacob <mj@feral.com> Subject: Re: svn commit: r216269 - head/sys/geom/part Message-ID: <4D016069.7050505@freebsd.org> In-Reply-To: <20101209225555.GS33073@deviant.kiev.zoral.com.ua> References: <201012072046.oB7KkB4L079555@svn.freebsd.org> <4CFEAD09.30904@freebsd.org> <4CFEAFA6.4020103@feral.com> <4CFEB1AD.70906@freebsd.org> <20101208153857.H1428@besplex.bde.org> <20101208225235.501ced0e@core.draftnet> <20101209191657.B1400@besplex.bde.org> <20101209221458.42448075@core.draftnet> <4D015A09.7070608@freebsd.org> <20101209225555.GS33073@deviant.kiev.zoral.com.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
on 10/12/2010 00:55 Kostik Belousov said the following: > On Fri, Dec 10, 2010 at 12:36:57AM +0200, Andriy Gapon wrote: >> on 10/12/2010 00:14 Bruce Cran said the following: >>> Since it appears that disks are still using the CHS fields despite >>> having been obsolete since ATA-7 I guess it makes sense to continue >>> printing them. >> >> I apologize for repeating myself, that probably starts to be annoying... >> We read some reserved/unused/obsolete bytes and get some garbage in them. >> The fact that the garbage looks like "valid" (for some definition of "valid") >> CHS parameters doesn't mean that those are really valid, or CHS parameters, or >> we should use them, or we should even report them, even for debug. >> That's my personal position and I am sticking to it :-) > > I think that the drive manufacturers are trying to be compatible not only > with the latest ATA specifications, but also with the reasonable set of > previous revisions of the specs. At least they do this to allow older > BIOSes or DOS or whatever old software user has to continue to work. > > So, althought the fields are marked as reserved by newest spec, they > still filled with something semi-reasonable to satisfy (very) old > software. I think it is safe to read and interpret the values still. That could be true, but that's not a reason to keep using them. That's good for software that was written and compiled 10 years ago and now being run unchanged. But there is no reason to perpetuate that old stuff in modern and alive software that is being developed now. IMO, of course. In a way it's similar to how we kept trying to interpret value in RTC NVRAM/CMOS register E as BIOS diagnostics when BIOSes had already stopped using it. -- Andriy Gapon
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4D016069.7050505>