Date: Mon, 3 Apr 2000 00:18:45 -0500 (CDT) From: Mike Isely <isely@enteract.com> To: aic7xxx@freebsd.org Subject: AHA2842 controller doesn't work anymore? Message-ID: <Pine.NEB.3.96.1000403001506.22732B-100000@shell-1.enteract.com> In-Reply-To: <200003310450.NAA24304@smtp0.dti.ne.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
Am I the only one left who's actually still using an Adaptec 2842? It
appears that 2842 support in the current version of the aic7xxx driver
may be broken. Allow me to explain:
I have one old 486/66MHz 32MB system with an Adaptec AHA2842
controller in it. It uses a Micropolis 1.7GB 2217 SCSI2-Fast
harddrive, and also has an old Toshiba 3401 2X CDROM. This system was
assembled out of old spare parts over a year ago. It runs only Linux
and has been stable the entire time. Its hardware configuration
hasn't changed for almost a year. (Why have it? Because it makes a
great X station.)
I tend to run very late but stable-series kernels on it. It's been
running kernel 2.2.15pre10 (with the IDE patch applied - this kernel
is used on several different machines). This kernel has version
5.1.23 of the aic7xxx driver. This works fine.
Tonight I tried out 2.2.15pre17. Same exact kernel configuration
(same .config file). This has version 5.1.28 of the aic7xxx driver.
It failed to boot, panicing inside of the aic7xxx driver. Here's the
relevant text (hand-transposed from the panic screen, hopefully no
typos):
|
| Detected scsi disk sda at scsi0, channel 0, id 0, lun0
| (scsi0) BRKADRINT error(0x2):
| Illegal Sequencer Address referenced
| (scsi0) SEQADDR=0x1
| Kernel panic: aic7xxx: unrecoverable BRKADRINT.
|
It appears that something broke AHA2842 support between version 5.1.23
and version 5.1.28. I'm trying to narrow that down now. The first
thing I tried was to patch the 2.2.15pre10 kernel forward, starting
with the 5.1.24 aic7xxx patch. However the patch failed to apply
correctly, with rejects inside of aic7xxx.c. No idea why except that
perhaps the "version" of 5.1.23 inside of aic7xxx.c is different than
what the 5.1.24 patch was generated from. So for now I'm compiling
different 2.2.15preXX "stock" kernels (i.e. no extra patches like the
IDE patch) in order to see where the breakage occurs. What I've got
so far is that that 2.2.15pre13 is ok and the driver panics in
2.2.15pre15. Unfortunately that doesn't say much because 2.2.15pre13
uses 5.1.23 and 2.2.15pre15 uses 5.1.28. Argh!
I'll try 2.2.15pre14 and see if that works. Then I'll take the last
working kernel version and try to patch the aic7xxx driver forward in
order to see where the breakage happens.
Is this ringing any bells with anyone? Is there anything more
specific I can provide (like my .config) that can help uncover what is
going on? As I said, I **really** don't see this as being a hardware
problem, given that the hardware in question hasn't changed for a long
time and has been running Linux with older versions of the driver.
I've had this controller for years, and have used the aic7xxx driver
(or an ancestor of it) since back when the current Linux kernel was
1.1.49.
Is this just a case of the AHA2842 not being verified for a while? I
can be a test bed for this if needed. The machine in question is not
critical for me at the moment.
-Mike
| Mike Isely | PGP fingerprint
POSITIVELY NO | | 03 54 43 4D 75 E5 CC 92
UNSOLICITED JUNK MAIL! | isely @ pobox (dot) com | 71 16 01 E2 B5 F5 C1 E8
| (spam-foiling address) |
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe aic7xxx" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96.1000403001506.22732B-100000>