In message <[EMAIL PROTECTED]>
Philip Blundell <[EMAIL PROTECTED]> wrote:
: > This test is only run for swap files, not swap devices, and is some way
: > down the chain from a call in swapfile.c to page in the first page in the
: > swap file (containing the magic).
: > I'm investigating more fully. Anyone care to tell me what this bit is
: > meant to do (before I figure it out sooner or later)?
: You may be running into some issue to do with page sizes vs the NFS block
: size. It's possible you need to mount the filesystem with rsize and wsize
: both set to 32k.
You cannot do this with NFS 2. The protocol forbids blocksizes > 8K
(RFC1094, p21), and it certainly used (circa 1.2.13) to be an interesting way
of crashing one of the nfsd or the IP stack (yes, I managed this). NFS 3
allows it (though servers have to set a maximum, and are perfectly at liberty
to specify 8K), so you may be lucky.
--
Dickon Hood
Due to binaries posted to non-binary newsgroups, my .sig is
temporarily unavailable. Normal service will be resumed as soon as
possible. We apologise for the inconvenience in the mean time.
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
