Re: port-alpha/33073 (isp0: unable to load dma (22))

[Jason Thorpe <thorpej%me.com@localhost>]

The following reply was made to PR port-alpha/33073; it has been noted by GNATS.

From: Jason Thorpe <thorpej%me.com@localhost>
To: Michael Haardt <michael%moria.de@localhost>
Cc: gnats-bugs%netbsd.org@localhost,
 Jason Thorpe <thorpej%netbsd.org@localhost>,
 gnats-admin%netbsd.org@localhost,
 netbsd-bugs%netbsd.org@localhost
Subject: Re: port-alpha/33073 (isp0: unable to load dma (22))
Date: Wed, 21 Jul 2021 11:27:09 -0700

 > On Jul 21, 2021, at 11:14 AM, Michael Haardt <michael%moria.de@localhost> wrote:
 >=20
 > I see nothing odd during booting, but wonder if "auto" should have =
 been
 > propagated from nsphy0 to tlp0:
 >=20
 > [   1.0000000] cia0 at mainbus0: DECchip 2117x Core Logic Chipset =
 (Pyxis), pass 2
 > [   1.0000000] cia0: extended capabilities: 0x1<BWEN>
 > [   1.0000000] cia0: using BWX for PCI config access
 > [   1.0000000] pci0 at cia0 bus 0
 > [   1.0000000] pci0: i/o space, memory space enabled, rd/line, =
 rd/mult, wr/inv ok
 > [   1.0000000] tlp0 at pci0 dev 3 function 0: DECchip 21143 Ethernet, =
 pass 3.0
 > [   1.0000000] tlp0: interrupting at dec 550 irq 0
 > [   1.0000000] tlp0: DEC, Ethernet address 00:00:f8:76:0e:39
 > [   1.0000000] nsphy0 at tlp0 phy 5: DP83840 10/100 media interface, =
 rev. 1
 > [   1.0000000] nsphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, =
 auto
 > [   1.0000000] tlp0: 10baseT, 10baseT-FDX, 10base2, 10base5
 
 Huh... can you provide the tlp-related boot messages from 9.2 as well?  =
 This bug is almost certainly my fault, but seeing the previous output =
 would be helpful in figuring out how I broke it :-)
 
 If you could gather that up and then file a new bug report for the tlp =
 issue, I would really appreciate it!
 
 > During unpacking, I see a flood of "Warning: received processor
 > correctable error."  Is there any additional information which =
 component
 > of the system may be causing that?
 
 I guess this is probably due to a flaky memory module.  We would need to =
 decode the machine check logout frame.  I'll investigate this.
 
 > Anyway: The DMA problems are gone and unpacking the sets to the disk
 > finishes.  I guess none of the above is related to port-alpha/33073,
 > but I wanted to mention it nevertheless.  Thanks for fixing this one!
 
 Ok, good news, thanks for confirming!
 
 -- thorpej