Re: port-i386/42966: kernel built from today's netbsd-5 crashes

[Anne Bennett <anne%porcupine.montreal.qc.ca@localhost>]

The following reply was made to PR port-i386/42966; it has been noted by GNATS.

From: Anne Bennett <anne%porcupine.montreal.qc.ca@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: port-i386-maintainer%NetBSD.org@localhost, gnats-admin%NetBSD.org@localhost,
    netbsd-bugs%NetBSD.org@localhost
Subject: Re: port-i386/42966: kernel built from today's netbsd-5 crashes
Date: Thu, 18 Mar 2010 15:30:47 -0400

 Okay, the options suggested to me so far have been:
 
 1) David H. Gutteridge: disable option DIAGNOSTIC
 
 2) matthew green #1: disable i915drm
 
 3) matthew green #2:
      change drm_pci_alloc(dev, PAGE_SIZE, PAGE_SIZE, 0xffffffff);
      to     drm_pci_alloc(dev, PAGE_SIZE, PAGE_SIZE, 0x0);
      or     drm_pci_alloc(dev, PAGE_SIZE, PAGE_SIZE, 0x80000000);
      in /usr/src/sys/external/bsd/drm/dist/shared-core/i915_dma.c
 
 4) matthew green #3: Try supplied patches to
      /src/sys/external/bsd/drm/dist/bsd-core/drmP.h    (1.9.10.3)
      /src/sys/external/bsd/drm/dist/bsd-core/drm_pci.c (1.4.10.1)
 
 5) Christos Zoulas: see revision 1.2 of
      /usr/src/sys/external/bsd/drm/dist/shared-core/i915_dma.c
      to find out what's going on (and revert to that version???)
 
 
 I selected option 4, and applied the two patches, which seem to
 fix data types and an off-by-one error.  I booted successfully into
 the new kernel, and here I am on the machine, which seems to be
 behaving mostly normally.  Yay!  :-)   Thank you.
 
 The only abnormal thing I notice is more disk activity than I expect
 at the moment: "top" shows "[system]" using about 35% of CPU, and 
 RAID parity status is "dirty" and currently being reconstructed.
 Since my shutdown was clean, this is puzzling.
 
 I'm comparing the old boot messages with the new ones, and, setting
 aside expected changes (kernel id lines), meaningless changes (minor
 differences in reported memory, CPU speed, and time frequency), the
 fact that some messages which used to take two lines are now
 compressed onto one, and USB devices "moving around", I see (not
 in order, grouped by type):
 
     < i915drm0 at vga1: Intel i945G (unit 0)
     > i915drm0 at vga1: Intel i945G
 
     < i915drm0: Initialized i915 1.5.0 20060119
     > i915drm0: Initialized i915 1.6.0 20080730
 
     < audio0 at azalia0: full duplex, independent
     > audio0 at azalia0: full duplex, playback, capture, independent
 
     < audio1 at pad0: half duplex
     > audio1 at pad0: half duplex, playback, capture
 
     < wd1: quirks 2<FORCE_LBA48>
 
     > raid0: cooldown 1168701999 out of range
 
 Most of those look like the result of improved reporting or version
 changes.  The last two puzzle me, especially "cooldown out of range",
 and especially since my RAID parity is currently being rebuilt for
 (as far as I know) no good reason.  How worried should I be?
 
 
 Anne.
 -- 
 Ms. Anne Bennett, as a private citizen:  
anne%porcupine.montreal.qc.ca@localhost
 Also reachable more officially at work:  anne%encs.concordia.ca@localhost