Date: Thu, 1 Sep 2005 11:56:06 +0300 (EEST) From: Dmitry Pryanishnikov <dmitry@atlantis.dp.ua> To: freebsd-arch@freebsd.org Subject: kern/85503: panic: wrong dirclust using msdosfs in RELENG_6 Message-ID: <20050901113819.F95708@atlantis.atlantis.dp.ua>
next in thread | raw e-mail | index | archive | help
Hello!
During the hunting the bug kern/85503 (panic: wrong dirclust in msdosfs)
I've tried to think about the solution, but it seems to be
architecture-related. The problem is: msdosfs uses pseudo-inodes (that is,
the offset from the start of the partition to the start of directory entry
in bytes) which must therefore have off_t bitness (at least 64 bits). I've
found the primary error (lack of casts leaded to 32-bit result), but then
we should transfer this 64-bit "inode" number to vfs_hash_get(). Oops,
it also limited to u_int (32 bits on i386). Finally, I see that the
primary shortcoming here: in sys/vnode.h we have
/*
* vfs_hash: (mount + inode) -> vnode hash.
*/
LIST_ENTRY(vnode) v_hashlist;
u_int v_hash;
I think it's feasible and useful to upgrade type of v_hash to at least off_t.
In our days of large media we will face filesystems with more than 4 billions
files (and thus at least 64-bit inode numbers) quite often. Am I right?
Sincerely, Dmitry
--
Atlantis ISP, System Administrator
e-mail: dmitry@atlantis.dp.ua
nic-hdl: LYNX-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050901113819.F95708>