Re: PR/52039 CVS commit: src/sys/dev/pci

[Frank Kardel <kardel%netbsd.org@localhost>]

The following reply was made to PR kern/52039; it has been noted by GNATS.

From: Frank Kardel <kardel%netbsd.org@localhost>
To: 
Cc: gnats-bugs%netbsd.org@localhost
Subject: Re: PR/52039 CVS commit: src/sys/dev/pci
Date: Wed, 08 Mar 2017 09:31:17 +0100

 Hi,
 
 you are right that that is strange. At boot things look like this:
 
 wm0 at pci0 dev 25 function 0: PCH LAN (82577LM) Controller (rev. 0x06)
 wm0: interrupting at msi2 vec 0
 wm0: PCI-Express bus
 wm0: 2048 words FLASH
 wm0: Ethernet address f0:de:f1:3b:df:f8
 ihphy0 at wm0 phy 2: i82577 10/100/1000 media interface, rev. 3
 ihphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 
 1000baseT-FDX, auto
 
 So there is a PHY attached. But at shutdown the code trips over the NULL 
 pointer,
 Could it be the the PHY is detached before the interface is shut-down?
 
 Thanks for having an eye on that. With the fix the kernel at least 
 manages to
 shut down cleanly without triggering a fsck at every boot - WOL might 
 not work though with the missing PHY. - I haven't checked that.
 
 Frank
 
 
 On 03/08/17 09:20, Masanobu SAITOH wrote:
 > On 2017/03/08 17:05, Frank Kardel wrote:
 >> The following reply was made to PR kern/52039; it has been noted by
 >> GNATS.
 >>
 >> From: "Frank Kardel" <kardel%netbsd.org@localhost>
 >> To: gnats-bugs%gnats.NetBSD.org@localhost
 >> Cc:
 >> Subject: PR/52039 CVS commit: src/sys/dev/pci
 >> Date: Wed, 8 Mar 2017 08:00:09 +0000
 >>
 >>  Module Name:    src
 >>  Committed By:    kardel
 >>  Date:        Wed Mar  8 08:00:09 UTC 2017
 >>
 >>  Modified Files:
 >>      src/sys/dev/pci: if_wm.c
 >>
 >>  Log Message:
 >>  PR kern/52039: use same safeguard as for the 82578
 >
 >  Your change is correct.
 > In your PR:
 >
 >> wm0 at pci0 dev 25 function 0: PCH LAN (82577LM) Controller (rev. 0x06)
 >> wm0: interrupting at msi2 vec 0
 >> wm0: PCI-Express bus
 >> wm0: 2048 words FLASH
 >> wm0: Ethernet address  xx::xx
 >
 > Isn't any PHY attached? Could you show me the dmesg of PHY device?
 > If it doesn't exist, it's strange.
 >
 >>
 >>  To generate a diff of this commit:
 >>  cvs rdiff -u -r1.496 -r1.497 src/sys/dev/pci/if_wm.c
 >>
 >>  Please note that diffs are not public domain; they are subject to the
 >>  copyright notices on the relevant files.
 >
 >
 >