lkml.org 
[lkml]   [2004]   [Mar]   [16]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
SubjectRe: [3C509] Fix sysfs leak.
From
Date
>>>>> "Dave" == Dave Jones <davej@redhat.com> writes:

Dave> no it doesn't, which is the whole purpose of the patch I sent.
Dave> try it..

Dave> modprobe 3c509
Dave> lsmod | grep 3c509 # module didnt stay around
Dave> find /sys | grep 3c509 # oh look, it left crap in sysfs

Real problem, wrong fix. You just killed 3c509 EISA support
altogether. Could you please test the following patch (against latest
bk) :

===== drivers/net/3c509.c 1.49 vs edited =====
--- 1.49/drivers/net/3c509.c Mon Mar 15 22:24:30 2004
+++ edited/drivers/net/3c509.c Tue Mar 16 16:44:24 2004
@@ -1655,14 +1655,14 @@
}

#ifdef CONFIG_EISA
- if (eisa_driver_register (&el3_eisa_driver) <= 0) {
+ if (eisa_driver_register (&el3_eisa_driver) < 0) {
eisa_driver_unregister (&el3_eisa_driver);
}
#endif
#ifdef CONFIG_MCA
mca_register_driver(&el3_mca_driver);
#endif
- return el3_cards ? 0 : -ENODEV;
+ return 0;
}

static void __exit el3_cleanup_module(void)
This is not pretty either, but 3c579 probing will work, and in your
case it won't leave a dangling directory in sysfs.

Dave> Why is this even an issue so late on? Bus probing should have
Dave> been done as part of bootup. By the time I get to modprobing
Dave> device drivers, it should have been determined already.

Modprobing is perfectly OK, and indeed everything has been probed at
this stage. But having built-in drivers raises a few different
problems (the driver may be initialized before all busses are probed).

Dave> Your argument seems to be "probing is hard, so we don't do it",
Dave> which is just *wrong*.

If you want to get back to the 2.4 state, where every single EISA
driver reinvents EISA probing, fair enough. I think 2.6 has it mostly
right (at least, it respects the bus hierarchy, which is necessary to
set futile things like dma_mask correctly). Drivers may be broken
(does hp100 rings a bell ?), but the framework looks sane to me.

Again, if you have something better to offer, I'm all ears.

Regards,

M.
--
Places change, faces change. Life is so very strange.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 14:01    [W:0.357 / U:0.904 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site