Path: ifm.liu.se!isy.liu.se!liuida!sunic!EU.net!howland.reston.ans.net!europa.eng.gtefsd.com!MathWorks.Com!noc.near.net!gradient.gradient.com!gradient!ddj From: ddj@gradient.gradient.com (Dave Johnson) Newsgroups: comp.unix.solaris,comp.sys.sun.hardware Subject: Re: Fujitsu M2694 drive on Solaris 2.3 Date: 18 Feb 1994 13:23:50 -0500 Organization: Gradient Technologies, Marlboro, MA Lines: 56 Distribution: inet Message-ID: References: <2k2ngs$17v@news.doit.wisc.edu> NNTP-Posting-Host: indy.gradient.com Xref: ifm.liu.se comp.unix.solaris:12198 comp.sys.sun.hardware:8599 orchard@eceserv0.ece.wisc.edu (Bruce Orchard) writes: >I have a Sparc Classic with an external Fujitsu M2694ESA disk drive >installed. I installed it while I was running Solaris 2.2 and it >worked. Then I upgraded to Solaris 2.3 and now get the following >message twice whenever I boot: > polled command timeout >esp: State=DATA_DONE Last State=DATA >esp: Latched stat=0x91 intr=0x10 fifo 0x80 >esp: last msg out: EXTENDED; last msg in: EXTENDED >esp: DMA csr=0xa6200310 >esp: addr=fc001038 dmacnt=8000 last=fc001028 last_cnt=30 >esp: Cmd dump for Target 1 Lun 0: >esp: cdblen=6, cdb=[ 0x12 0x0 0x0 0x0 0x30 0x0 ] >esp: pkt_state=0x7 pkt_flags=0xb pkt_statistics=0x0 >esp: cmd_flags=0x1422 cmd_timeout=60 >The active light stays on for about a minute before the message >comes out. >In addition, if I do a boot -r, the drive is polled a few more times, >resulting in a few more messages. The entries in the /devices >directory got created but not the entries in the /dev directory. >I tried doing an ln -s command to create the /dev entry, but >I still could not access the drive. >Is there a way to get around this problem? >Side note: I have a Sparcstation I with an internal 2694 running >Solaris 2.3 that works fine. >Bruce Orchard (orchard@engr.wisc.edu) You need to either patch your drive's firmware, or disable the scsi disk drive from using tagged command queueing in solaris 2.3. If the Fuji is not your boot disk, edit /etc/system and add the line: set scsi_options=0x178 Then reboot. Otherwise, boot from your cdrom and exit from the installation script. Mount your boot disk and perform the same edit. The scsi_options bits default to 1f8, and are documented in /usr/include/sys/scsi/conf/autonconf.h If you have a dos machine with an adaptec scsi controller 1542B, or controller 1542C with the aspi driver, you can get a copy of the program PB5E.EXE from the Fujitsu America BBS. Running this with your disk attached to the dos machine will update the firmware to fix this bug. The disk we have here is at firmware revision 811F, and it has this problem. The new revision is reported to be 812A. Dave Johnson Gradient Technologies, Inc. ddj@gradient.com Path: ifm.liu.se!isy.liu.se!liuida!sunic!EU.net!howland.reston.ans.net!math.ohio-state.edu!jussieu.fr!univ-lyon1.fr!swidir.switch.ch!news.univie.ac.at!fstgds15.tu-graz.ac.at!fsztmss01!schinagl From: schinagl@fsztmss01 (Hermann Schinagl) Newsgroups: comp.unix.solaris,comp.sys.sun.hardware Subject: Re: Fujitsu M2694 drive on Solaris 2.3 Followup-To: comp.unix.solaris,comp.sys.sun.hardware Date: 18 Feb 1994 20:45:18 GMT Organization: Graz University of Technology, Austria Lines: 26 Distribution: world Message-ID: <2k39cuINN9km@fstgds15.tu-graz.ac.at> References: <2k2ngs$17v@news.doit.wisc.edu> NNTP-Posting-Host: fsztmss01.tu-graz.ac.at X-Newsreader: TIN [version 1.2 PL2] Xref: ifm.liu.se comp.unix.solaris:12204 comp.sys.sun.hardware:8603 Bruce Orchard (orchard@eceserv0.ece.wisc.edu) wrote: : I have a Sparc Classic with an external Fujitsu M2694ESA disk drive : installed. I installed it while I was running Solaris 2.2 and it : worked. Then I upgraded to Solaris 2.3 and now get the following : message twice whenever I boot: : polled command timeout : esp: State=DATA_DONE Last State=DATA : esp: Latched stat=0x91 intr=0x10 fifo 0x80 : esp: last msg out: EXTENDED; last msg in: EXTENDED : esp: DMA csr=0xa6200310 : esp: addr=fc001038 dmacnt=8000 last=fc001028 last_cnt=30 : esp: Cmd dump for Target 1 Lun 0: : esp: cdblen=6, cdb=[ 0x12 0x0 0x0 0x0 0x30 0x0 ] : esp: pkt_state=0x7 pkt_flags=0xb pkt_statistics=0x0 : esp: cmd_flags=0x1422 cmd_timeout=60 I had the same problem, and applied patch 101318-32 and 101378-02. Then the problem never occoured ... Perfect software ;-( Ciao Hermann