NetBSD Problem Report #49457

From hauke@Espresso.Rhein-Neckar.DE  Sun Dec  7 18:17:41 2014
Return-Path: <hauke@Espresso.Rhein-Neckar.DE>
Received: from mail.netbsd.org (mail.netbsd.org [149.20.53.66])
	(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits))
	(Client CN "mail.netbsd.org", Issuer "Postmaster NetBSD.org" (verified OK))
	by mollari.NetBSD.org (Postfix) with ESMTPS id 991D0A5809
	for <gnats-bugs@gnats.NetBSD.org>; Sun,  7 Dec 2014 18:17:41 +0000 (UTC)
Message-Id: <201412071802.sB7I2MlW003157@pizza.causeuse.org>
Date: Sun, 7 Dec 2014 19:02:22 +0100 (CET)
From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Reply-To: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: gnats-bugs@NetBSD.org
Cc: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Subject: 'atactl sleep' loses drive
X-Send-Pr-Version: 3.95

>Number:         49457
>Category:       kern
>Synopsis:       'atactl sleep' loses drive
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Dec 07 18:20:00 +0000 2014
>Last-Modified:  Sat Aug 17 13:20:01 +0000 2019
>Originator:     Hauke Fath
>Release:        NetBSD 7.0_BETA
>Organization:
Falling Raindrops
>Environment:


System: NetBSD pizza.causeuse.org 7.0_BETA NetBSD 7.0_BETA (BLACKBOX-$Revision: 1.85 $) #0: Sun Dec 7 17:09:27 CET 2014 hauke@pizza.causeuse.org:/var/obj/netbsd-builds/7/amd64/sys/arch/amd64/compile/BLACKBOX amd64
Architecture: x86_64
Machine: amd64
>Description:

	The machine has a second sata disk that is only used as an
	Amanda spool disk, and unmounted otherwise. The backup scripts
	run 'atactl sleep' on the drive to power it down after
	work. 

	This worked without a problem on netbsd-6, but irreversibly
	puts the drive in a coma on netbsd-7.

>How-To-Repeat:

[hauke@pizza] ~ # atactl wd1 checkpower
Current power status: Standby mode
[hauke@pizza] ~ # atactl wd1 sleep
load: 0.06  cmd: atactl 4012 [ahcicmd] 0.00u 0.00s 0% 0k
[hauke@pizza] ~ # atactl wd1 checkpower
atactl: wd1: Input/output error
[hauke@pizza] ~ # 

The console logs

wd1: wd_flushcache: status=0x5128<TIMEOU>
wd1: IDENTIFY failed

At this point, only a reboot will bring the drive back.

>Fix:
	Yes, please.

>Audit-Trail:
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: 
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Mon, 8 Dec 2014 18:19:58 +0100

 On Sun, Dec 07, 2014 at 06:20:00PM +0000, Hauke Fath wrote:
 > 	The machine has a second sata disk that is only used as an
 > 	Amanda spool disk, and unmounted otherwise. The backup scripts
 > 	run 'atactl sleep' on the drive to power it down after
 > 	work. 
 > 
 > 	This worked without a problem on netbsd-6, but irreversibly
 > 	puts the drive in a coma on netbsd-7.
 > 
 > >How-To-Repeat:
 > 
 > [hauke@pizza] ~ # atactl wd1 checkpower
 > Current power status: Standby mode
 > [hauke@pizza] ~ # atactl wd1 sleep
 > load: 0.06  cmd: atactl 4012 [ahcicmd] 0.00u 0.00s 0% 0k
 > [hauke@pizza] ~ # atactl wd1 checkpower
 > atactl: wd1: Input/output error
 > [hauke@pizza] ~ # 
 > 
 > The console logs
 > 
 > wd1: wd_flushcache: status=0x5128<TIMEOU>
 > wd1: IDENTIFY failed
 > 
 > At this point, only a reboot will bring the drive back.

 that's strange, no more message from the kernel ?
 What would e.g. a disklabel -r show ?

 tested on a netbsd-7 host with ahcisata controller here:
 yquem# atactl wd0 checkpower
 Current power status: Active mode
 yquem# atactl wd0 sleep
 yquem# atactl wd0 checkpower
 ATA command timed out
 [this is expected, a drive in sleep doesn't reply to any ATA command - only
 a reset will wake it up]
 yquem# disklabel -r wd0
 [several seconds delay, I can hear the drive spin up]
 # /dev/rwd0d:
 type: unknown
 disk: ST500DM002-1BD1
 label: 
 [...]

 and the kernel said while the several seconds delay:
 wd0d: device timeout reading fsbn 0 (wd0 bn 0; cn 0 tn 0 sn 0), retrying
 wd0: soft error (corrected)

 The failed wd_flushcache in your case looks suspicious; if the partition
 is unmounted I'm not sure why a flushcache would be queued at this point.

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Mon, 8 Dec 2014 21:59:03 +0100

 On Mon,  8 Dec 2014 17:25:00 +0000 (UTC), Manuel Bouyer wrote:
 > tested on a netbsd-7 host with ahcisata controller here:
 >  yquem# atactl wd0 checkpower
 >  Current power status: Active mode
 >  yquem# atactl wd0 sleep
 >  yquem# atactl wd0 checkpower
 >  ATA command timed out
 >  [this is expected, a drive in sleep doesn't reply to any ATA command - only
 >  a reset will wake it up]
 >  yquem# disklabel -r wd0
 >  [several seconds delay, I can hear the drive spin up]
 >  # /dev/rwd0d:
 >  type: unknown
 >  disk: ST500DM002-1BD1
 >  label: 
 >  [...]

 For the same procedure, I get

 [hauke@pizza] ~ # atactl wd1 checkpower
 Current power status: Active mode
 [hauke@pizza] ~ # atactl wd1 sleep
 [hauke@pizza] ~ # atactl wd1 checkpower
 atactl: wd1: Input/output error
 [hauke@pizza] ~ # disklabel -r wd1
 disklabel: /dev/rwd1d: Input/output error
 [hauke@pizza] ~ # 

 and dmesg says

 [...]
 wd1: wd_flushcache: status=0x5128<TIMEOU>
 wd1: IDENTIFY failed
 wd1: IDENTIFY failed

 I remember the "device timeout" line you quoted from netbsd-6, but have 
 not seen them on netbsd-7.

From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org,
        Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Tue, 9 Dec 2014 11:06:53 +0100

 On Mon, Dec 08, 2014 at 09:20:00PM +0000, Hauke Fath wrote:
 > The following reply was made to PR kern/49457; it has been noted by GNATS.
 > 
 > From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
 > To: gnats-bugs@NetBSD.org
 > Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org
 > Subject: Re: kern/49457: 'atactl sleep' loses drive
 > Date: Mon, 8 Dec 2014 21:59:03 +0100
 > 
 >  On Mon,  8 Dec 2014 17:25:00 +0000 (UTC), Manuel Bouyer wrote:
 >  > tested on a netbsd-7 host with ahcisata controller here:
 >  >  yquem# atactl wd0 checkpower
 >  >  Current power status: Active mode
 >  >  yquem# atactl wd0 sleep
 >  >  yquem# atactl wd0 checkpower
 >  >  ATA command timed out
 >  >  [this is expected, a drive in sleep doesn't reply to any ATA command - only
 >  >  a reset will wake it up]
 >  >  yquem# disklabel -r wd0
 >  >  [several seconds delay, I can hear the drive spin up]
 >  >  # /dev/rwd0d:
 >  >  type: unknown
 >  >  disk: ST500DM002-1BD1
 >  >  label: 
 >  >  [...]
 >  
 >  For the same procedure, I get
 >  
 >  [hauke@pizza] ~ # atactl wd1 checkpower
 >  Current power status: Active mode
 >  [hauke@pizza] ~ # atactl wd1 sleep
 >  [hauke@pizza] ~ # atactl wd1 checkpower
 >  atactl: wd1: Input/output error
 >  [hauke@pizza] ~ # disklabel -r wd1
 >  disklabel: /dev/rwd1d: Input/output error
 >  [hauke@pizza] ~ # 
 >  
 >  and dmesg says
 >  
 >  [...]
 >  wd1: wd_flushcache: status=0x5128<TIMEOU>
 >  wd1: IDENTIFY failed
 >  wd1: IDENTIFY failed
 >  
 >  I remember the "device timeout" line you quoted from netbsd-6, but have 
 >  not seen them on netbsd-7.

 Still not sure why the wd_flushcache() is there.
 What command above triggers which message ?

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

From: Hauke Fath <hf@spg.tu-darmstadt.de>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org,
        Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Tue, 9 Dec 2014 16:30:20 +0100

 On Tue,  9 Dec 2014 10:10:06 +0000 (UTC), Manuel Bouyer wrote:
 >  >  wd1: wd_flushcache: status=0x5128<TIMEOU>
 >  >  wd1: IDENTIFY failed
 >  >  wd1: IDENTIFY failed
 >  >  
 >  >  I remember the "device timeout" line you quoted from netbsd-6, but have 
 >  >  not seen them on netbsd-7.
 >  
 >  Still not sure why the wd_flushcache() is there.
 >  What command above triggers which message?

 The 'atactl sleep' (which takes quite a while to return) triggers the 
 TIMEOU; the subsequent 'atactl checkpower' and 'disklabel' commands 
 trigger the 'IDENTIFY failed' messages.

From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Mon, 1 Apr 2019 20:59:18 +0200

 At 18:20 Uhr +0000 07.12.2014, Hauke Fath wrote:
 >[hauke@pizza] ~ # atactl wd1 checkpower
 >Current power status: Standby mode
 >[hauke@pizza] ~ # atactl wd1 sleep
 >load: 0.06  cmd: atactl 4012 [ahcicmd] 0.00u 0.00s 0% 0k
 >[hauke@pizza] ~ # atactl wd1 checkpower
 >atactl: wd1: Input/output error
 >[hauke@pizza] ~ #
 >
 >The console logs
 >
 >wd1: wd_flushcache: status=0x5128<TIMEOU>
 >wd1: IDENTIFY failed
 >
 >At this point, only a reboot will bring the drive back.

 FTR, the issue is unchanged in netbsd-8.


 --
 "It's never straight up and down"     (DEVO)


From: mlelstv@serpens.de (Michael van Elst)
To: gnats-bugs@netbsd.org
Cc: 
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Mon, 1 Apr 2019 22:17:39 -0000 (UTC)

 hauke@Espresso.Rhein-Neckar.DE (Hauke Fath) writes:

 > >[hauke@pizza] ~ # atactl wd1 sleep
 > >load: 0.06  cmd: atactl 4012 [ahcicmd] 0.00u 0.00s 0% 0k
 > >wd1: wd_flushcache: status=0x5128<TIMEOU>
 > >wd1: IDENTIFY failed
 > >
 > >At this point, only a reboot will bring the drive back.
 > FTR, the issue is unchanged in netbsd-8.

 If only a reboot brings it back, it's probably an issue
 with the drive.

 Here I see timeouts when accessing a sleeping drive, but the
 drive wakes up after several seconds. Not perfect either.
 Maybe a sucessful sleep command should configure larger timeouts.

 -- 
 -- 
                                 Michael van Elst
 Internet: mlelstv@serpens.de
                                 "A potential Snark may lurk in every tree."

From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Tue, 2 Apr 2019 07:36:47 +0200

 At 22:20 Uhr +0000 01.04.2019, Michael van Elst wrote:
 > If only a reboot brings it back, it's probably an issue
 > with the drive.

 The drive is in daily use as an Amanda spool, and it has never given me any
 problems. Plus, as mentioned further up in the PR, the setup worked  with
 netbsd-6, and broke after an upgrade to -7. So - not impossible, but
 unlikely.  ;)

 Cheerio,
 Hauke



 --
 "It's never straight up and down"     (DEVO)


From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@netbsd.org
Cc: kern-bug-people@netbsd.org, gnats-admin@netbsd.org, netbsd-bugs@netbsd.org,
        Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Tue, 2 Apr 2019 12:28:44 +0200

 On Mon, Apr 01, 2019 at 10:20:01PM +0000, Michael van Elst wrote:
 >  > >[hauke@pizza] ~ # atactl wd1 sleep
 >  > >load: 0.06  cmd: atactl 4012 [ahcicmd] 0.00u 0.00s 0% 0k
 >  > >wd1: wd_flushcache: status=0x5128<TIMEOU>
 >  > >wd1: IDENTIFY failed
 >  > >
 >  > >At this point, only a reboot will bring the drive back.
 >  > FTR, the issue is unchanged in netbsd-8.
 >  
 >  If only a reboot brings it back, it's probably an issue
 >  with the drive.
 >  
 >  Here I see timeouts when accessing a sleeping drive, but the
 >  drive wakes up after several seconds. Not perfect either.
 >  Maybe a sucessful sleep command should configure larger timeouts.

 No, the reset is the only way to wake up a sleeping drive.
 The sleep command could tag the drive as sleeping, and so the next command
 could trigger the reset without the timeout. But that's not perfect either,
 because the drive may go to sleep on its own.

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

From: Manuel Bouyer <bouyer@antioche.eu.org>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org, netbsd-bugs@NetBSD.org,
        Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Tue, 2 Apr 2019 17:33:42 +0200

 --Qxx1br4bt0+wmkIi
 Content-Type: text/plain; charset=us-ascii
 Content-Disposition: inline

 On Tue, Dec 09, 2014 at 04:40:00PM +0000, Hauke Fath wrote:
 > The following reply was made to PR kern/49457; it has been noted by GNATS.
 > 
 > From: Hauke Fath <hf@spg.tu-darmstadt.de>
 > To: gnats-bugs@NetBSD.org
 > Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org,
 >         Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
 > Subject: Re: kern/49457: 'atactl sleep' loses drive
 > Date: Tue, 9 Dec 2014 16:30:20 +0100
 > 
 >  On Tue,  9 Dec 2014 10:10:06 +0000 (UTC), Manuel Bouyer wrote:
 >  >  >  wd1: wd_flushcache: status=0x5128<TIMEOU>
 >  >  >  wd1: IDENTIFY failed
 >  >  >  wd1: IDENTIFY failed
 >  >  >  
 >  >  >  I remember the "device timeout" line you quoted from netbsd-6, but have 
 >  >  >  not seen them on netbsd-7.
 >  >  
 >  >  Still not sure why the wd_flushcache() is there.
 >  >  What command above triggers which message?
 >  
 >  The 'atactl sleep' (which takes quite a while to return) triggers the 
 >  TIMEOU; the subsequent 'atactl checkpower' and 'disklabel' commands 
 >  trigger the 'IDENTIFY failed' messages.

 The fact that the flushcache command times out is suspect by itself. This
 command is issued before the device goes to sleep.

 It seems that to reproduce it, the disk needs to be unmounted when put
 to sleep. In this case, the WDF_LOADED is cleared on device close,
 and any attempt to open it afterward issues a wd_get_params() command.
 But this one doens't reset and retry in case of timeout, and the open fails.
 So there's no way to wake up the disk.

 While there I also implemented the DIRTY flag (copied from sd(4))
 so avoid flushing the cache if there has been no write.
 As a side effect, the atactl command on an unmounted disk doesn't trigger
 a cache flush, and so avoids the timeout.

 The attached patch fixes it for me on netbsd-7; it should also work for
 netbsd-8. Can you test ?

 -- 
 Manuel Bouyer <bouyer@antioche.eu.org>
      NetBSD: 26 ans d'experience feront toujours la difference
 --

 --Qxx1br4bt0+wmkIi
 Content-Type: text/plain; charset=us-ascii
 Content-Disposition: attachment; filename=diff

 Index: sys/dev/ata/wd.c
 ===================================================================
 RCS file: /cvsroot/src/sys/dev/ata/wd.c,v
 retrieving revision 1.412.2.2
 diff -u -r1.412.2.2 wd.c
 --- sys/dev/ata/wd.c	5 Jul 2016 19:09:17 -0000	1.412.2.2
 +++ sys/dev/ata/wd.c	2 Apr 2019 15:16:11 -0000
 @@ -752,8 +752,12 @@
  		wd->sc_wdc_bio.flags |= ATA_LBA48;
  	if (wd->sc_flags & WDF_LBA)
  		wd->sc_wdc_bio.flags |= ATA_LBA;
 -	if (bp->b_flags & B_READ)
 +	if (bp->b_flags & B_READ) {
  		wd->sc_wdc_bio.flags |= ATA_READ;
 +	} else {
 +		/* it's a write */
 +		wd->sc_flags |= WDF_DIRTY;
 +	}
  	/* Instrumentation. */
  	disk_busy(&wd->sc_dk);
  	switch (wd->atabus->ata_bio(wd->drvp, &wd->sc_wdc_bio)) {
 @@ -961,7 +965,6 @@
  		}
  	} else {
  		if ((wd->sc_flags & WDF_LOADED) == 0) {
 -
  			/* Load the physical device parameters. */
  			if (wd_get_params(wd, AT_WAIT, &wd->sc_params) != 0) {
  				aprint_error_dev(wd->sc_dev,
 @@ -1014,7 +1017,8 @@
  {
  	struct wd_softc *wd = device_private(self);

 -	wd_flushcache(wd, AT_WAIT);
 +	if (wd->sc_flags & WDF_DIRTY)
 +		wd_flushcache(wd, AT_WAIT);

  	if (! (wd->sc_flags & WDF_KLABEL))
  		wd->sc_flags &= ~WDF_LOADED;
 @@ -1811,11 +1815,20 @@
  int
  wd_get_params(struct wd_softc *wd, u_int8_t flags, struct ataparams *params)
  {
 +	int retry = 0;

 +again:
  	switch (wd->atabus->ata_get_params(wd->drvp, flags, params)) {
  	case CMD_AGAIN:
  		return 1;
  	case CMD_ERR:
 +		if (retry == 0) {
 +			retry++;
 +			(*wd->atabus->ata_reset_drive)(wd->drvp,
 +			    flags | AT_RST_NOCMD, NULL);
 +			goto again;
 +		}
 +
  		if (wd->drvp->drive_type != ATA_DRIVET_OLD)
  			return 1;
  		/*
 @@ -1975,6 +1988,7 @@
  		    sbuf);
  		return EIO;
  	}
 +	wd->sc_flags &= ~WDF_DIRTY;
  	return 0;
  }

 Index: sys/dev/ata/wdvar.h
 ===================================================================
 RCS file: /cvsroot/src/sys/dev/ata/wdvar.h,v
 retrieving revision 1.40
 diff -u -r1.40 wdvar.h
 --- sys/dev/ata/wdvar.h	2 Feb 2012 19:43:02 -0000	1.40
 +++ sys/dev/ata/wdvar.h	2 Apr 2019 15:16:11 -0000
 @@ -59,6 +59,7 @@
  #define WDF_LBA		0x040 /* using LBA mode */
  #define WDF_KLABEL	0x080 /* retain label after 'full' close */
  #define WDF_LBA48	0x100 /* using 48-bit LBA mode */
 +#define	WDF_DIRTY	0x200 /* disk cache dirty */
  	u_int64_t sc_capacity; /* full capacity of the device */
  	u_int32_t sc_capacity28; /* capacity accessible with LBA28 commands */


 --Qxx1br4bt0+wmkIi--

From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Wed, 3 Apr 2019 22:31:00 +0200

 At 17:33 Uhr +0200 02.04.2019, Manuel Bouyer wrote:
 >The attached patch fixes it for me on netbsd-7; it should also work for
 >netbsd-8. Can you test ?

 This works fine, thanks! The patch to sys/dev/ata/wdvar.h was rejected, but
 trivially to apply by hand.

 Thanks a lot!

 Cheerio,
 hauke

 --
 "It's never straight up and down"     (DEVO)


From: "Manuel Bouyer" <bouyer@netbsd.org>
To: gnats-bugs@gnats.NetBSD.org
Cc: 
Subject: PR/49457 CVS commit: src/sys/dev/ata
Date: Fri, 5 Apr 2019 18:23:45 +0000

 Module Name:	src
 Committed By:	bouyer
 Date:		Fri Apr  5 18:23:45 UTC 2019

 Modified Files:
 	src/sys/dev/ata: wd.c wdvar.h

 Log Message:
 Implement a DIRTY flag (copied from sd(4)) so avoid flushing the cache if
 there has been no write. This avoids a (long) timeout on the flush cache
 command triggered by atactl sleep, when the device is open only by the atactl
 command itself.
 If a drive has no partition open and goes to sleep, the WDF_LOADED
 flag is clear, and the next open will issue  wd_get_params() command.
 But to wake up the drive a reset is required, and wd_get_params() doens't
 issue a reset on timeout, so there's no way to wake up the disk.
 Add a retry after reset to wd_get_params().

 Tested by Hauke Fath; fixes PR kern/49457


 To generate a diff of this commit:
 cvs rdiff -u -r1.446 -r1.447 src/sys/dev/ata/wd.c
 cvs rdiff -u -r1.48 -r1.49 src/sys/dev/ata/wdvar.h

 Please note that diffs are not public domain; they are subject to the
 copyright notices on the relevant files.

From: Hauke Fath <hauke@Espresso.Rhein-Neckar.DE>
To: gnats-bugs@NetBSD.org
Cc: kern-bug-people@NetBSD.org, gnats-admin@NetBSD.org,
        Manuel Bouyer <bouyer@antioche.eu.org>
Subject: Re: kern/49457: 'atactl sleep' loses drive
Date: Sat, 17 Aug 2019 15:09:52 +0200

 At 15:35 Uhr +0000 02.04.2019, Manuel Bouyer wrote:
 > The attached patch fixes it for me on netbsd-7; it should also work for
 > netbsd-8. Can you test ?

 I carried your patch locally in the netbsd-8 tree, where it worked fine. A
 pull-up was never requested, and -current is probably too far away from -8
 to make your commit viable.

 After upgrading the machine to -9 though, the problem is back in its
 original shape; the only difference to your patch I could find in wd.c was
 the demise of AT_RST_NOCMD.

 Can you see any reason why that would be?

 Cheerio,
 Hauke




 --
 "It's never straight up and down"     (DEVO)


>Unformatted:

NetBSD Home
NetBSD PR Database Search

(Contact us) $NetBSD: query-full-pr,v 1.43 2018/01/16 07:36:43 maya Exp $
$NetBSD: gnats_config.sh,v 1.9 2014/08/02 14:16:04 spz Exp $
Copyright © 1994-2017 The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.