I must say that 2.1.33 is looking very, very nice. =] I built the
latest modules with it and there are no undefined symbols in any of the
modules I built (which, at first pass, was almost everything). I did hand
apply the little patch to parport.c to get it to compile.
Now for the ever-present bug reports. I'm sorry that I don't have
fixes for this stuff; I'm only a novice when it comes to kernel hacking.
=]
Using a Mitsumi FX400 4x IDE CDROM (master, secondary IDE channel with
Intel Triton board and kernel support), built as a module, loads through
kerneld correctly and can be used to view data discs normally. However,
leaving the CD mounted (even if the data on it is not being explicitly
accessed from user space), I see these until I umount:
Apr 11 02:19:30 earth kernel: _isofs_bmap: block>= EOF(2, 4096)
Apr 11 02:19:42 earth last message repeated 11 times
Apr 11 02:19:42 earth kernel: _isofs_bmap: block>= EOF(40, 81920)
Apr 11 02:19:42 earth kernel: _isofs_bmap: block>= EOF(40, 81920)
Apr 11 02:19:43 earth kernel: _isofs_bmap: block>= EOF(2, 4096)
Apr 11 02:19:43 earth last message repeated 9 times
Apr 11 02:19:44 earth kernel: _isofs_bmap: block>= EOF(32, 65536)
Apr 11 02:19:44 earth kernel: _isofs_bmap: block>= EOF(32, 65536)
Apr 11 02:19:44 earth kernel: _isofs_bmap: block>= EOF(26, 53248)
Apr 11 02:19:44 earth kernel: _isofs_bmap: block>= EOF(26, 53248)
Apr 11 02:19:44 earth kernel: _isofs_bmap: block>= EOF(2, 4096)
Apr 11 02:19:45 earth last message repeated 13 times
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(4, 8192)
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(4, 8192)
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(2, 4096)
Apr 11 02:19:45 earth last message repeated 3 times
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(4, 8192)
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(4, 8192)
Apr 11 02:19:45 earth kernel: _isofs_bmap: block>= EOF(2, 4096)
Apr 11 02:19:45 earth last message repeated 5 times
The second thing I'm seeing is one I've been waiting on for a long, long
time. It's about SMB and mount. The documentation on mount would make it
very believable that it can mount SMB volumes. In fact, mount (I'm using
2.6b) recognizes '-t smbfs' as a valid fs type. kerneld will even load
the smbfs.o module. But the mount always fails with the following error:
Apr 11 02:16:55 earth kernel: smb_read_super: wrong data argument
Apr 11 02:17:16 earth kernel: smb_read_super: wrong data argument
That's it for now. Thanks for the great work,
-bp
-- # B. James Phillippe # Network/Sys Admin Terran.ORG # # <bryan@terran.org> # http://w3.terran.org/~bryan # # Finger for PGP key # Linux Convert - circa 1.1.59 # # Systems Administrator, Seattle Software Labs, Inc #