Re: pccardc dumpcis - 0 slots found

Patrick Bowen wrote:

> M. Warner Losh wrote:
>
>> In message: <432F870D.2080302_at_fastmail.fm>
>>            Patrick Bowen <pbowen_at_fastmail.fm> writes:
>> : I'm running -current on a Dell C600 with a GENERIC kernel, and am 
>> having : no luck in dumping the CIS from the wi0 pccard (an 
>> SMC-2532W-B). I've : had the same problem with 5.4#6 and 6.0BETA4.
>> : : sg1# pccardc dumpcis
>> : 0 slots found
>>
>> This is expected.  pccardc dumpcis doesn't work, and likely will never
>> work again.  It is part of the old pccard support.
>>
>> If you have current as of today, however, you can run 'pccard
>> dumpcisfile /dev/pccard0.cis'.
>>
>> : When I remove the card I get the following;
>> : : taskqueue_drain with the following non-sleepable locks held:
>> : exclusive sleep mutex wi0 (network driver) r = 0 (0xc1c0bb68) 
>> locked _at_ : /usr/src/sys/dev/wi/if_wi.c:847
>> : KDB: stack backtrace:
>> : kdb_backtrace(1,c1b62650,c1b62400,c1ac1d00,d44dcc2c) at 
>> kdb_backtrace+0x29
>> : witness_warn(5,0,c0862050,c1b62400,c1b62400) at witness_warn+0x18e
>> : taskqueue_drain(c1ac1d00,c1b62650) at taskqueue_drain+0x1a
>> : if_detach(c1b62400,c1b62400) at if_detach+0x1a
>> : ether_ifdetach(c1b62400,0,c1c0b000,d44dcc94,c05e2590) at 
>> ether_ifdetach+0x3a
>> : ieee80211_ifdetach(c1c0b004,c1b62400,c1b62400,0,c1bec700) at : 
>> ieee80211_ifdetach+0x50
>> : wi_detach(c1bec700) at wi_detach+0x64
>> : device_detach(c1bec700) at device_detach+0x70
>> : pccard_detach_card(c1af9e00) at pccard_detach_card+0x41
>> : exca_removal(c1ac3004) at exca_removal+0x46
>> : cbb_removal(c1ac3000) at cbb_removal+0x2c
>> : cbb_event_thread(c1ac3000,d44dcd38,c1ac3000,c057cb24,0) at : 
>> cbb_event_thread+0x9a
>> : fork_exit(c057cb24,c1ac3000,d44dcd38) at fork_exit+0xa0
>> : fork_trampoline() at fork_trampoline+0x8
>> : --- trap 0x1, eip = 0, esp = 0xd44dcd6c, ebp = 0 ---
>> : wi0: detached
>> : : ...which doesn't look like a good thing.
>>
>> It is likely a warning that can be ignored.
>>
>> : I'm wondering if there is some place in pccardd that is not 
>> recognizing : the slots on this particular machine. If that were the 
>> case then it : wouldn't be called when the card was removed, 
>> accounting (I think) for : the lock error above.
>>
>> pccardd is deprecated and isn't needed.  It should just work.
>>
>> : Anyone have an idea of what's going on here? I'd be happy to try 
>> any : patches anyone may have, or to provide more information if 
>> necessary. In : the meantime, I'll keep looking into it.
>>
>> Other than not being able to dump the cis, what's the problem that you
>> are having?
>>
>> Warner
>>
>>  
>>
> That was actually the only problem. It came up when I replied to a 
> post from someone asking how to get an SMC card to work. You send a 
> msg back to me asking me to run the 'pccardc dumpcis' command and send 
> you the output. Since the 'pccardc ...' command didn't work, I got the 
> info using Slackware Linux, which I've got on the second slice, and 
> sent you that.
>
> In this thread I was essentially trying to provide info for the 
> 'pccardc dumpcis' command, since it seemed to be broken in some way or 
> another. But if it's deprecated, that's a whole 'nother story.
>
> My -current is about a week old, so I cvsup'ed another just now. I'll 
> rebuild everything and try the command you suggested.
>
> Thanks for your quick reply,
> Patrick
>
>
Warner;

Sorry about taking so long to get back to you. I've been on the road 
pretty heavily for the last few days.

Got a panic after I rebuilt everything...same panic that Munehiro
Matsuda got (re: Panic with recent PCcard changes). So I re-cvsuped
sources, re-build everything and tested just now. It all works fine.

You're a good man. Thanks for all your hard work.

Patrick