Date: Fri, 3 Nov 1995 01:40:02 -0800 From: Bruce Evans <bde@zeta.org.au> To: freebsd-bugs Subject: Re: kern/810: Some I/O attempts to CD-ROMs hang at isoilk FDIV038 Message-ID: <199511030940.BAA01604@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
The following reply was made to PR kern/810; it has been noted by GNATS. From: Bruce Evans <bde@zeta.org.au> To: FreeBSD-gnats-submit@freebsd.org, uhclem%nemesis@fw.ast.com Cc: Subject: Re: kern/810: Some I/O attempts to CD-ROMs hang at isoilk FDIV038 Date: Fri, 3 Nov 1995 20:31:24 +1100 >Now, login on a second terminal as a plain user: > login: user1 > Password: > % pwd > /usr/user1 > % gunzip /mnt/packages/All/lynx-2.3.7.tgz > (escapes may be used for filename completion) > You get the non-helpful message: > /mnt/packages/All/lynz-2.3.7.tgz: Operation not supported gzip should print which operation it tried. > (Mental note: We need to add error codes for "Write Protected" and > one for "Not Ready", instead of using "Not Configured" or "Not > Supported" for these legitimate events.) cd9660 uses cd9660_enotsupp() for all operations that it doesn't support, including for link and rename, which must be cleaned up by calling VOP_ABORTOP() and vput(). See msdosfs_abortop() for correct handling. ENODEV might be a better errno. EOPNOTSUPP only applies to networking according to the comment in errno.h :-). EROFS isn't quite right for file systems on devices that are inherently readonly. OTOH, cd9660 isn't inherently readonly. The current implementation just doesn't suppport writing, so EOPNOTSUPP is reasonable. If writing was supported then EROFS would be the natural errno. Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199511030940.BAA01604>