Date: Sun, 11 Jan 1998 15:23:13 +1000 From: Carl.Makin@aipo.gov.au To: freebsd-stable@freebsd.org, freebsd-scsi@freebsd.org Subject: seagate.c kernel compile error and sea device problems. Message-ID: <4A256589.001DDBEF.00@notes.aipo.gov.au>
next in thread | raw e-mail | index | archive | help
Under FreeBSD 2.2.5-RELEASE, I get the following error from seagate.c when I compile a kernel. root@brain:/sys/compile/BRAIN$ make cc -c -O -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -nostdinc -I- -I. -I../.. -I/usr/include -DAPM_BROKEN_STATCLOCK -DIPDIVERT -DNETATALK -DFAILSAFE -DCOMPAT_43 -DMSDOSFS -DNFS -DFFS -DINET -DKERNEL ../../i386/isa/seagate.c ../../i386/isa/seagate.c:581: warning: no previous prototype for `seaintr' ../../i386/isa/seagate.c: In function `sea_data_output': ../../i386/isa/seagate.c:1111: warning: assignment discards `volatile' from pointer target type ../../i386/isa/seagate.c: In function `sea_data_input': ../../i386/isa/seagate.c:1181: warning: assignment discards `volatile' from pointer target type sh ../../conf/newvers.sh BRAIN -DAPM_BROKEN_STATCLOCK -DIPDIVERT -DNETATALK -DFAILSAFE -DCOMPAT_43 -DMSDOSFS -DNFS -DFFS -DINET cc -O -Wreturn-type -Wcomment -Wredundant-decls -Wimplicit -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -nostdinc -I- -I. -I../.. -I/usr/include -DAPM_BROKEN_STATCLOCK -DIPDIVERT -DNETATALK -DFAILSAFE -DCOMPAT_43 -DMSDOSFS -DNFS -DFFS -DINET -DKERNEL -c vers.c loading kernel rearranging symbols text data bss dec hex 1011712 81920 89084 1182716 120bfc root@brain:/sys/compile/BRAIN$ I noticed this as I've been having a few problems with the sea driver driving a TMC-850MEX (Future Domain TMC-950 chip) scsi card and talking to a DEC TZK10 tape drive (QIC-525). I've been getting; Jan 10 21:20:33 brain /kernel: sea0/5/0 data length underflow and the write to tape continues however reading back from the tape often (ie not always but often) fails at the same point as the above error occured. I modified /sys/i386/isa/seagate.c and commented out "BLINDTRANSFERS". It had no noticible effect so then I changed the kernel config file to remove the IRQ and memory address information (it then autodetects memory and uses no IRQ). This seems to have stopped the "data length underflow" errors (limited testing so far). I uncommented the "BLINDTRANSFERS" option and tried again with the autodetect sea0 config line and the errors started again. I'm not overly worried as it seems to be working now and the only thing I have on the sea0 adaptor is the tape drive however I'm not a kernel hacking wizard and I'd like to bring this to the attention to someone who might be! ;) please cc me on any replies to this mail (I'm not on freebsd-scsi). Carl.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4A256589.001DDBEF.00>