(This is a mass-update to all current FC6 kernel bugs in NEW state)

Hello,

I'm reviewing this bug list as part of the kernel bug triage project, an attempt
to isolate current bugs in the Fedora kernel.

http://fedoraproject.org/wiki/KernelBugTriage

I am CC'ing myself to this bug, however this version of Fedora is no longer
maintained.

Please attempt to reproduce this bug with a current version of Fedora (presently
Fedora 8). If the bug no longer exists, please close the bug or I'll do so in a
few days if there is no further information lodged.

Thanks for using Fedora!

The problem still exists in FC8:

Linux version 2.6.23.12-99.fc8 (mockbuild.redhat.com) (gcc
version 4.1.2 20070925 (Red Hat 4.1.2-33)) #1 SMP Wed Dec 19 17:20:50 EST 2007
...
...
...
pccard: PCMCIA card inserted into slot 1
cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff
pcmcia: registering new device pcmcia1.0
smc91c92_cs: Unable to find hardware address.
unregister_netdevice: device eth%d/cff31800 never was registered
WARNING: at net/core/dev.c:3769 unregister_netdevice() (Not tainted)
 [<c05bc678>] unregister_netdev+0xf/0x15
 [<d4b58a9c>] smc91c92_probe+0xb46/0xb9a [smc91c92_cs]
 [<c056ea60>] pccard_get_tuple_data+0x65/0x6e
 [<c056ef6a>] pccard_read_tuple+0x7e/0x88
 [<c0570a08>] pcmcia_device_probe+0xaa/0xfa
 [<c0566fd3>] driver_probe_device+0xea/0x168
 [<c04f3589>] kobject_uevent_env+0x37a/0x3a1
 [<c05670f7>] __driver_attach+0x0/0xa5
 [<c0567167>] __driver_attach+0x70/0xa5
 [<c056655f>] bus_for_each_dev+0x39/0x5b
 [<c0566e2e>] driver_attach+0x16/0x18
 [<c05670f7>] __driver_attach+0x0/0xa5
 [<c0566843>] bus_add_driver+0x6d/0x16d
 [<c05712af>] pcmcia_register_driver+0xd1/0x101
 [<c044931b>] __link_module+0x0/0x10
 [<c044ab9e>] sys_init_module+0x150d/0x1651
 [<c06206ea>] do_page_fault+0x2c0/0x5ef
 [<c0570845>] pcmcia_dev_present+0x0/0x33
 [<c040518a>] syscall_call+0x7/0xb
 [<c0610000>] xfrm_tmpl_sort+0x36/0x5a
 =======================
1.0: ttyS3 at I/O 0x2e8 (irq = 4) is a 16550A

Miklos,

I'm not sure a driver exists for your hardware in linux. However would you mind
running the following and attach it as type text/plain to this bug:

pccardctl ident > pccardctl.out
# lspci -vvxxx > lspci.out

We may then be able to add the hardware ident to /etc/pcmcia/config.opts

Cheers
Chris

Created attachment 293878 [details]
lspci output file

Created attachment 293879 [details]
pccardctl output

Hi Chris,
I was using this adapter so this driver did work correctly in the earlier
releases of Fedora and RHx linuxes. I was examining the source code of the
driver both of the working and the not working versions (release nums are above)
with the Meld diff viewer as I recognized the problem. I have searched for the
output message in the sources and found the place where the code was changed
which caused the error-exit. As I remember it was somewhere around the
initialization routines. Although I have more than 10 yrs programming experience
[but rather in Bill's op system :-( not in linux driver programming] I am still
a hobby linux user and fun so I could'nt do anything except reporting the problem.
Thanks for help!
Miklos

If you have the diff that would be appreciated...

Created attachment 294330 [details]
source files

Guys,
unfortunately I couldn't provide the driver source files of the same kernel
versions as given in the first report entry. But I attach two other versions of
the source files: 2.6.15 which was working, and 2.6.22 which shows up some
changes that made the EM1144T not working.
If you compare these files, there will be a lot of changes. But I think the
important difference for us is in the 'mhz_setup' routine (line 600-609 in the
old file). There is a comment there: "Ugh -- the EM1144 card has two VERS_1
tuples!?!" I don't know what does that mean, but I feel that there was some
hacking there which is missing from the new source. Therefore the new setup
routine returns with nonzero so the "smc91c92_cs: Unable to find hardware
address." is printed out.

Created attachment 294348 [details]
Diff of changes between 2.6.15 and 2.6.22

Created attachment 294350 [details]
Diff between 2.6.19 and 2.6.20

We can try reverting this commit and see if it resolves the issue for you as
you mentioned the 2.6.19 > 2.6.20 upgrade broke things for you.

We would only need to revert this part... and it does look like that would fix it:

--- a/drivers/net/pcmcia/smc91c92_cs.c	Tue Dec 05 06:09:20 2006 +0500
+++ b/drivers/net/pcmcia/smc91c92_cs.c	Tue Dec 05 06:09:21 2006 +0500
@@ -560,16 +560,8 @@ static int mhz_setup(struct pcmcia_devic
 
     /* Read the station address from the CIS.  It is stored as the last
        (fourth) string in the Version 1 Version/ID tuple. */
-    tuple->DesiredTuple = CISTPL_VERS_1;
-    if (first_tuple(link, tuple, parse) != CS_SUCCESS) {
-	rc = -1;
-	goto free_cfg_mem;
-    }
-    /* Ugh -- the EM1144 card has two VERS_1 tuples!?! */
-    if (next_tuple(link, tuple, parse) != CS_SUCCESS)
-	first_tuple(link, tuple, parse);
-    if (parse->version_1.ns > 3) {
-	station_addr = parse->version_1.str + parse->version_1.ofs[3];
+    if (link->prod_id[3]) {
+	station_addr = link->prod_id[3];
 	if (cvt_ascii_address(dev, station_addr) == 0) {
 		rc = 0;
 		goto free_cfg_mem;

Fixed in 2.6.23.15-136

kernel-2.6.23.15-137.fc8 has been submitted as an update for Fedora 8

kernel-2.6.23.15-137.fc8 has been pushed to the Fedora 8 stable repository.  If problems still persist, please make note of it in this bug report.

Hi,
this modification solved the problem. Here is the dmesg output:
----------------------------------------
pccard: PCMCIA card inserted into slot 1
cs: memory probe 0xa0000000-0xa0ffffff: excluding 0xa0000000-0xa00fffff
pcmcia: registering new device pcmcia1.0
eth1: smc91c92 rev 3: io 0x310, irq 4, hw_addr xx:xx:xx:xx:xx:xx
  4608 byte buffer, auto xcvr
pcmcia: registering new device pcmcia1.1
1.1: ttyS2 at I/O 0x3e8 (irq = 4) is a 16550A
eth1: flipped to 10baseT
eth1: no IPv6 routers present
----------------------------------------

Thanks!
m