Re: Bug#47512: zgv (or svgalib, or whoever) disturbs SCSI by PCI probi

Search this archive.

From: Kai Henningsen (kaih@khms.westfalen.de)
Date: Sat 16 Oct 1999 - 17:05:00 IST


andy.mortimer@zetnet.co.uk (Andy Mortimer)  wrote on 16.10.99 in <87r9ivy42n.fsf@zetnet.co.uk>:

> Thanks for your report.  I'm afraid this is a little out of my area,
> and I don't have SCSI installed so I can try it.  It's almost
> certainly an SVGAlib problem, so I've reassigned the bug anyway, but
> if you do have another svgalib program installed (and it's not a fatal
> error), I should be interested to know if starting another SVGAlib
> program gives the same results.  I'll forward it on to the SVGAlib
> mailing list, to see if they have any ideas.
>
> Also, is this new with this version of zgv/SVGAlib, or did you just
> not use them before this?

Well, zgv is just about the only one I actually use, but a short try with  
maelstrom gave this:

kai@khms tty10 ~ $ maelstrom
sym53c875-0: SCSI parity error detected: SCR1=132 DBC=50000000 SBCL=0
ncr53c810-0: SCSI parity error detected: SCR1=65 DBC=50000000 SSTAT1=f
svgalib: Using S3 driver (Trio64, 2048K).
svgalib: s3: chipsets newer than S3-864 is not supported well yet.
svgalib: RAMDAC: Trio64: MCLK = 59.957 MHz
scsi : aborting command due to timeout : pid 1026935, scsi1, channel 0, id 5, lu
n 0 Read (10) 00 00 4d be fe 00 00 80 00
ncr53c8xx_abort: pid=1026935 serial_number=1026953 serial_number_at_timeout=1026
953
ncr53c810-0: abort ccb=cffac820 (cancel)
SCSI host 1 abort (pid 1026935) timed out - resetting
SCSI bus is being reset for host 1 channel 0.
ncr53c8xx_reset: pid=1026935 reset_flags=2 serial_number=1026953 serial_number_a
t_timeout=1026953
Warning: you have not yet configured your mouse type. If you have no mouse,
setting the type to none' in /etc/vga/libvga.config will get rid of this
annoying message.
svgalib: Failed to initialize mouse.
Killed by signal 11 (SIGSEGV)
kai@khms tty10 ~ $

As for being fatal, I suppose _usually_ the SCSI layer will recover (after  
blocking everything for like 20 seconds), but I sure wouldn't want to rely  
on that!

Re: version, it's recent, but I couldn't say since which version exactly.


MfG Kai


This archive was generated by hypermail 2.1.4 : Wed 21 Jan 2004 - 22:10:22 IST