Re: NCR5380 stuff (and Oops)

Olivier Tharan (olive@laria.u-picardie.fr)
Thu, 1 Apr 1999 15:35:25 +0200


Le 31 mars 1999, Alan Cox écrivait :
> Ok how does this look. I think this correctly interprets the 2,0 drivers
> intention unlike 2.2 which as you pointed out got the one case of

Apart from the obvious typo, it correctly detects the SCSI card (and the
scanner attached to it), but still the driver is subject to Oopses. Here
is the output of ksymoops against stock 2.2.5 (and your patch), it seems
to occur in a call to sg_write.

I include also the output of the ksyms command because I loaded
g_NCR5380 as a module (modprobe g_NCR5380 ncr_addr=0x240 dtc_3181e=1).

--- cut oops result ---

Apr 1 16:02:06 vega kernel: Oops: 0000
Apr 1 16:02:06 vega kernel: CPU: 0
Apr 1 16:02:06 vega kernel: EIP: 0010:[sg_close+31/132]
Apr 1 16:02:06 vega kernel: EFLAGS: 00010246
Apr 1 16:02:06 vega kernel: eax: 00000000 ebx: c13c5d20 ecx: 000000d8 edx: c0007800
Apr 1 16:02:06 vega kernel: esi: c113add0 edi: c10de760 ebp: c1828b40 esp: c0e9bebc
Apr 1 16:02:06 vega kernel: ds: 0018 es: 0018 ss: 0018
Apr 1 16:02:06 vega kernel: Process xscanimage (pid: 295, process nr: 33, stackpage=c0e9b000)
Apr 1 16:02:06 vega kernel: Stack: c13c5d20 c13c5d20 00000000 c10de760 c0122f67 c13c5d20 c13c5d20 c13c5d20
Apr 1 16:02:06 vega kernel: c0121ece c13c5d20 00000009 00000001 c1828b40 00000001 c0114cfc c13c5d20
Apr 1 16:02:06 vega kernel: c1828b40 00000009 c0e9a000 00000009 c0e9bfc4 00000001 c0e9a000 c0107963
Apr 1 16:02:06 vega kernel: Call Trace: [fput+23/72] [filp_close+90/100] [do_exit+304/568] [do_signal+619/708] [sg_write+166/720] [sg_write+0/720] [do_page_fault+267/792]
Apr 1 16:02:06 vega kernel: Code: 8b 40 0c 8b 40 24 8b 50 04 85 d2 74 1b 8d 42 10 ff 4a 10 a1
Warning: trailing garbage ignored on Code: line
Text: 'Code: 8b 40 0c 8b 40 24 8b 50 04 85 d2 74 1b 8d 42 10 ff 4a 10 a1 '
Garbage: ' '

Code: 00000000 Before first symbol 00000000 <_IP>: <===
Code: 00000000 Before first symbol 0: 8b 40 0c movl 0xc(%eax),%eax <===
Code: 00000003 Before first symbol 3: 8b 40 24 movl 0x24(%eax),%eax
Code: 00000006 Before first symbol 6: 8b 50 04 movl 0x4(%eax),%edx
Code: 00000009 Before first symbol 9: 85 d2 testl %edx,%edx
Code: 0000000b Before first symbol b: 74 1b je 00000028 Before first symbol
Code: 0000000d Before first symbol d: 8d 42 10 leal 0x10(%edx),%eax
Code: 00000010 Before first symbol 10: ff 4a 10 decl 0x10(%edx)
Code: 00000013 Before first symbol 13: a1 00 00 00 00 movl 0x0,%eax

2 warnings issued. Results may not be reliable.

**** Result of `ksyms`:

Address Symbol Defined by
c201451c usleep_timer [g_NCR5380]
c20104e0 generic_NCR5380_info [g_NCR5380]
c2012d7c generic_NCR5380_proc_info [g_NCR5380]
c2011130 generic_NCR5380_queue_command [g_NCR5380]
c2010514 generic_NCR5380_biosparam [g_NCR5380]
c2012c4c generic_NCR5380_reset [g_NCR5380]
c2010128 generic_NCR53C400A_setup [g_NCR5380]
c2012aec generic_NCR5380_abort [g_NCR5380]
c2010158 generic_NCR5380_detect [g_NCR5380]
c2010140 generic_DTC3181E_setup [g_NCR5380]
c20100f8 generic_NCR5380_setup [g_NCR5380]
c2010b30 notyet_generic_proc_info [g_NCR5380]
c20108f0 NCR5380_timer_fn [g_NCR5380]
c20104e8 generic_NCR5380_release_resources [g_NCR5380]
c201412c driver_template [g_NCR5380]
c2010110 generic_NCR53C400_setup [g_NCR5380]
c2013f8c proc_scsi_g_ncr5380 [g_NCR5380]
--- cut oops result ---

olive

-- 
		Olivier Tharan, <olive@laria.u-picardie.fr>

While preceding your entrance with a grenade is a good tactic in Quake, it can lead to problems if attempted at work.

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/