Yes- this means the request queue is overflowing pretty regularly. This
seems odd for your setup, though. You have a 1020 here-it can take 256
request queue entries (circular queue). This also includes s/g entries
for long transfers. I'm a bit surprised you're hitting this.

I haven't been paying attention much to NetBSD lately- I haven't seen
this stuff, for instance, on a NetBSD-alpha 1.6 PC164 which is pretty
close to this, nor on a i386 larger system.

In any case, this stuff gets retried, so it's not really an error.

I turned this into a LOGERR instead of LOG_DEBUG0 at one point- that was
probably a mistake. You can change it back if it bugs you (it's in
isp_pci.c). I still would like to know what you're running that's
overflowing the queue though. You don't have boatloads of buffer memory
nor have a really fast disk that can take a lot.

Did you up the count of simultaneous disk operations you can try?

-matt


On Mon, 14 Jul 2003, Sean Davis wrote:

> Has anybody else seen this? I was running bonnie on an AlphaStation 600
> 5/266, 128MB RAM. The hard drive is a western digital Ultra2/LVD connected
> via a sca -> 68pin adapter to the fast/wide bus.
>
> Jul 14 14:42:29 jane.endersgame.net /netbsd: isp0: Request Queue Overflow++
> Jul 14 14:42:29 jane.endersgame.net /netbsd: sd0(isp0:0:0:0): adapter
> resource shortage
> Jul 14 14:42:30 jane.endersgame.net /netbsd: isp0: Request Queue Overflow++
> Jul 14 14:42:30 jane.endersgame.net /netbsd: sd0(isp0:0:0:0): adapter
> resource shortage
> Jul 14 14:42:31 jane.endersgame.net /netbsd: sd0(isp0:0:0:0): adapter
> resource shortage
> Jul 14 14:42:33 jane.endersgame.net /netbsd: isp0: Request Queue Overflow++
> Jul 14 14:42:33 jane.endersgame.net /netbsd: sd0(isp0:0:0:0): adapter
> resource shortage
> Jul 14 14:42:36 jane.endersgame.net last message repeated 2 times
> Jul 14 14:42:37 jane.endersgame.net /netbsd: isp0: Request Queue Overflow++
> Jul 14 14:42:37 jane.endersgame.net /netbsd: sd0(isp0:0:0:0): adapter
> resource shortage
>
> Is this a problem, or can it be safely ignored? If the former, is it fixable
> by some kernel option or sysctl, or do I just need a new scsi card?
>
> The partition I was running bonnie on is mounted with softdep, if that makes
> a difference.
>
> -Sean
>
> dmesg follows:
> Copyright (c) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003
>     The NetBSD Foundation, Inc.  All rights reserved.
> Copyright (c) 1982, 1986, 1989, 1991, 1993
>     The Regents of the University of California.  All rights reserved.
>
> NetBSD 1.6U (JANE) #4: Sun Jul 13 12:14:43 EDT 2003
> 	dive@jane.endersgame.net:/sys/arch/alpha/compile/JANE
> Digital AlphaStation 600 5/266, 266MHz, s/n
> 8192 byte page size, 1 processor.
> total memory = 128 MB
> (2304 KB reserved for PROM, 125 MB used by NetBSD)
> avail memory = 114 MB
> using 817 buffers containing 6536 KB of memory
> mainbus0 (root)
> cpu0 at mainbus0: ID 0 (primary), 21164-0
> cia0 at mainbus0: DECchip 2117x Core Logic Chipset (ALCOR/ALCOR2), pass 2
> pci0 at cia0 bus 0
> pci0: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
> tlp0 at pci0 dev 9 function 0: ADMtek AN985 Ethernet, pass 1.1
> tlp0: interrupting at kn20aa irq 12
> tlp0: Ethernet address 00:04:5a:67:5e:a7
> acphy0 at tlp0 phy 1: ACXXX 10/100 media interface, rev. 0
> acphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
> pceb0 at pci0 dev 10 function 0: vendor 0x8086 product 0x0482 (rev. 0x05)
> ppb0 at pci0 dev 11 function 0: vendor 0x1011 product 0x0001 (rev. 0x02)
> pci1 at ppb0 bus 1
> pci1: memory space enabled, rd/line, wr/inv ok
> tlp1 at pci1 dev 0 function 0: DECchip 21040 Ethernet, pass 2.3
> tlp1: interrupting at kn20aa irq 0
> tlp1: Ethernet address 08:00:2b:e7:94:72
> tlp1: 10baseT, 10baseT-FDX, 10base5, manual
> isp0 at pci1 dev 1 function 0: QLogic 1020 Ultra Wide SCSI HBA
> isp0: interrupting at kn20aa irq 1
> scsibus0 at isp0: 16 targets, 8 luns per target
> isp1 at pci1 dev 2 function 0: QLogic 1020 Ultra Wide SCSI HBA
> isp1: interrupting at kn20aa irq 2
> scsibus1 at isp1: 16 targets, 8 luns per target
> vga0 at pci0 dev 12 function 0: vendor 0x1011 product 0x0017 (rev. 0x01)
> wsdisplay0 at vga0 (kbdmux ignored): console (80x25, vt100 emulation)
> eisa0 at pceb0
> isa0 at pceb0
> com0 at isa0 port 0x3f8-0x3ff irq 4: ns16550a, working fifo
> pckbc0 at isa0 port 0x60-0x64
> pckbd0 at pckbc0 (kbd slot)
> pckbc0: using irq 1 for kbd slot
> wskbd0 at pckbd0 (mux ignored): console keyboard, using wsdisplay0
> pcppi0 at isa0 port 0x61
> spkr0 at pcppi0
> isabeep0 at pcppi0
> mcclock0 at isa0 port 0x70-0x71: mc146818 or compatible
> scsibus0: waiting 2 seconds for devices to settle...
> scsibus1: waiting 2 seconds for devices to settle...
> sd0 at scsibus0 target 0 lun 0: <WDIGTL, WDE4550 ULTRA2, 1.22> disk fixed
> sd0: 4340 MB, 6932 cyl, 6 head, 213 sec, 512 bytes/sect x 8890000 sectors
> sd0: sync (100.00ns offset 12), 16-bit (20.000MB/s) transfers, tagged queueing
> cd0 at scsibus0 target 3 lun 0: <PLEXTOR, CD-ROM PX-40TW, 1.04> cdrom removable
> cd0: sync (100.00ns offset 12), 16-bit (20.000MB/s) transfers
> root on sd0a dumps on sd0b
>
> --
> /~\ The ASCII
> \ / Ribbon Campaign                   Sean Davis
>  X  Against HTML                       aka dive
> / \ Email!
>