On 6/30/07, Kris Maglione <[EMAIL PROTECTED]> wrote:
On Sat, Jun 30, 2007 at 07:39:12PM +0200, Enrico Weigelt wrote: >I'm not sure if spfs sends this zero or libixp doesn't decode it. >With npfs I've got the same problem.spfs sends it. styxmon has nothing to do with libixp. >But it's interesting that the Linux 9p driver doesn't have any >problem w/ spfs or npfs. No idea if it does an similar hack as mine. lib9pclient ignores the iounit entirely. npfs does this, which is nowhere in the spec, and thus confuses me: fid->iounit = rc->iounit; if (!fid->iounit || fid->iounit>fid->fsys->msize-IOHDRSZ) fid->iounit = fid->fsys->msize-IOHDRSZ;
If the client gets iounit equal to zero, it has to pick a greater-than-zero number. msize - IOHDRSZ is the maximum number of bytes that can be read/written with a single message. Lucho
