Re: kern/54289: wd1 fails to be identified properly

["John D. Baker" <jdbaker%consolidated.net@localhost>]

The following reply was made to PR kern/54289; it has been noted by GNATS.

From: "John D. Baker" <jdbaker%consolidated.net@localhost>
To: gnats-bugs%netbsd.org@localhost
Cc: 
Subject: Re: kern/54289: wd1 fails to be identified properly
Date: Thu, 5 Dec 2019 11:16:37 -0600 (CST)

 On Fri, 25 Oct 2019, John D. Baker wrote:
 
 > Booting a 9.99.17 kernel (the first I've built since the switch to GCC-8)
 > seems to detect wd0 (the first in the RAID set), but still complains
 > about IDENTIFY FAILED and can't find identify wd1-wd3 (more RAID disks)
 > at all.  These are all on the siisata card.  The system disk is wd4 on
 > 
 > [   1.0254336] siisata0 at pci5 dev 4 function 0: CMD Technology SiI3124 SATALink (rev. 0x02)
 > [   1.0254336] siisata0: interrupting at msi1 vec 0
 > [   1.0254336] siisata0: SiI3124, 3.0Gb/s
 > [   1.0254336] siisata0: 32-bit 66MHz PCI
 > [   1.0254336] atabus0 at siisata0 channel 0
 > [   1.0254336] atabus1 at siisata0 channel 1
 > [   1.0254336] atabus2 at siisata0 channel 2
 > [   1.0254336] atabus3 at siisata0 channel 3
 > 
 > [   1.7984930] siisata0 port 0: device present, speed: 3.0Gb/s
 > [   1.7984930] siisata0 port 1: device present, speed: 3.0Gb/s
 > [   1.7984930] siisata0 port 2: device present, speed: 3.0Gb/s
 > [   1.7984930] siisata0 port 3: device present, speed: 3.0Gb/s
 > 
 > [   1.8084982] wd0 at atabus0 drive 0
 > [   1.8084982] wd0: <Hitachi HDS721010CLA332>
 > [   1.8084982] wd0: drive supports 16-sector PIO transfers, LBA48 addressing
 > [   1.8084982] wd0: 931 GB, 1938021 cyl, 16 head, 63 sec, 512 bytes/sect x 1953525168 sectors
 > 
 > [   7.8311162] wd0: autoconfiguration error: IDENTIFY failed
 > [   7.8311162] autoconfiguration error: wd0: unable to open device, error = 5
 > [  10.8383800] wd0(siisata0:0:0): using PIO mode 0
 > [  10.8383800] wd1 at atabus1 drive 0
 > [  16.8629319] wd1: autoconfiguration error: IDENTIFY failed
 > [  16.8629319] wd1: fixing 0 sector size
 > [  16.8629319] wd1: secperunit and ncylinders are zero
 > [  19.8701959] wd1(siisata0:1:0): using PIO mode 0
 > [  19.8701959] wd2 at atabus2 drive 0
 > [  25.8947475] wd2: autoconfiguration error: IDENTIFY failed
 > [  25.8947475] wd2: fixing 0 sector size
 > [  25.8947475] wd2: secperunit and ncylinders are zero
 > [  28.9020114] wd2(siisata0:2:0): using PIO mode 0
 > [  28.9020114] wd3 at atabus3 drive 0
 > [  34.9265633] wd3: autoconfiguration error: IDENTIFY failed
 > [  34.9265633] wd3: fixing 0 sector size
 > [  34.9265633] wd3: secperunit and ncylinders are zero
 > [  37.9338271] wd3(siisata0:3:0): using PIO mode 0
 > 
 > [  44.3801110] wd0: autoconfiguration error: IDENTIFY failed
 > [  44.3801110] opendisk: can't open dev wd0 (5)
 > [  50.4448822] wd0: autoconfiguration error: IDENTIFY failed
 > [  50.4448822] opendisk: can't open dev wd0 (5)
 > [  56.5498116] wd0: autoconfiguration error: IDENTIFY failed
 > [  56.5498116] WARNING: 9 errors while detecting hardware; check system log.
 
 All previous work on this PR has been directed at ahcisata(4).  On this
 machine, there is no issue with ahcisata(4).  Rather, the problem
 manifests only on with siisata(4) and the disks attached to it.
 
 I don't know if or how much siisata(4) shares with ahcisata(4), but it
 would seem that siisata(4) needs to be investigated to advance the state
 of this PR.
 
 Although there has been posted one example where an siisata(4) card is
 working properly, it does not do so in my case.  Alas, it is the only
 one I have and is mission-critical, so I cannot test it in other host
 systems.
 
 -- 
 |/"\ John D. Baker, KN5UKS               NetBSD     Darwin/MacOS X
 |\ / jdbaker[snail]consolidated[flyspeck]net  OpenBSD            FreeBSD
 | X  No HTML/proprietary data in email.   BSD just sits there and works!
 |/ \ GPGkeyID:  D703 4A7E 479F 63F8 D3F4  BD99 9572 8F23 E4AD 1645