Linux2.0.31+Adaptec2840VLB?

[Christian Brandt]

Hallo dear readers,

 I updated my Linux-Kernel from 2.0.30 to 2.0.31 for getting most of my
hardware running without relying on inofficial patches, eg hisax, smp,
3c900 and so on.

 But to my surprise the initialization of my Adaptec VLB2840 (yes, the
VLB-Version, not the PCI-Version) fails in most obscure ways. As old
versions like 2.0.14/29/30 worked good for me, I guess something got
broken in the updating process (the size of aic7xxx.c increased quite
notable, much chance for a nice bug :-)

 But before bothering the developers, I´ll take a small survey if
someone else can reproduce the following errors:

 My System: AMD5x86-486dx4-160Mhz (i know, adaptec states 33mhz busclock
are maximum, but as said, NT, W95 and old linux never failed).
 Soyo-SiS471-Chipset, VLB284x with disabled onboard-fdc as I have to use
the fdc of my $5-ide/fd-controller, Revsd
 Adaptec 2840 identifing itself within 2.0.30 as follows:

aic7xxx: Reading SEEPROM...done.
aic7xxx: Extended translation enabled.
aic7xxx: AHA-2840 Rev E and subsequent.
aic7xxx: Memory check yields 4 SCBs, paging not enabled.
aic7xxx: Using level sensitive interrupts.
AHA-2840 AT VLB SLOT 1:
    irq 15
    bus release time 40 bclks
    data fifo threshold 100%
    SCSI CHANNEL A:
        scsi id 7
        scsi selection timeout 256 ms
        scsi bus reset at power-on enabled
        scsi bus parity enabled
        scsi bus termination (low byte) enabled
aic7xxx: Downloading sequencer code...done.
scsi0 : Adaptec AHA274x/284x/294x (EISA/VLB/PCI-Fast SCSI) 4.0/3.2/4.0
scsi : 1 host.
scsi0: Scanning channel A for devices.
Started kswapd v 1.4.2.2 

 and from 2.0.31 on it crashes most times at initializing with lots of
crap appering on the screen, sometimes it hangs with an wrong
irq-level-message, regardless if I set any at the lilo-prompt. Before
2.0.31 I never stumbled about the irq-level at all.
 Update: In some rare, not reproducable cases the system boots nice and
runs stable after it got through the scsi-init. I can just guess, but
its reappering more frequently if I hard-reset the system and disable
most of the aic-features while configuring the kernel.

 Just want to mention: I ran the tests with a very simplified and also
with a full-featured kernel, reduced the busclock from 40 to 33mhz and
took all unimportant boards out of my system. Didn´t change anythink :-(

-- 
Christian Brandt

 "Live is short and in most cases it ends with death." Sir Sinclair