Re: 6.99.32: panic when starting X

[Thomas Klausner <wiz%NetBSD.org@localhost>]

On Fri, Feb 28, 2014 at 05:38:53PM +0100, Thomas Klausner wrote:
> On Fri, Feb 28, 2014 at 05:02:40PM +0100, Thomas Klausner wrote:
> > On Mon, Feb 24, 2014 at 08:01:58AM +0000, Nick Hudson wrote:
> > > Trying kmem_guard_depth might pick something up - see kmem(9)
> > 
> > I've tried a kernel with
> > options DEBUG
> > options KMEM_GUARD_DEPTH=30000
> > 
> > but that doesn't even finish boot. In 2/2 tries it stops with:
> > cd0 at atapibus0 drive 0: ...
> > uvm_fault(0xf..80fb9380, 0xffff8000054ad9000, 1) -> e
> > fatal page fault in supervisor mode
> > trap type 6 code 0 rip f...801517c1 cs8 rflags 10206 cr2 ffff800054ad9004 
> > ilevel 0 rsp f...e815150cb00
> > curlwp 0xf...e81510b8040 pid 0.114  lowest kstack 0xf...e81515092c0
> > kernel: page fault trap, code=0
> > Stopped in pid 0.114 (system) at netbsd:ahci_atapi_probe_device+0x91: 
> > m
> > ovzbl 4(%rcx,%rbx,1),%eax
> > db{0}>
> > 
> > Ideas?
> 
> # addr2line -e /netbsd.6.99.33.debug.gdb -a ffffffff801517c1
> 0xffffffff801517c1
> /usr/src/sys/arch/amd64/compile/KVOTHE.kmem/../../../../dev/ic/ahcisata_core.c:1784
> 
> Line 1784ff:
>          if (drvp->drive_type != ATA_DRIVET_ATAPI) {
>                  AHCIDEBUG_PRINT(("ahci_atapi_probe_device: drive %d "
>                      "not present\n", target), DEBUG_PROBE);
>                  return;
>          }

Backtrace:
ahci_atapi_probe_device() at netbsd:ahci_atapi_probe_device+0x91
atabibusattach() at netbsd:atapibusattach+0xd0
config_attach_loc() at netbsd:config_attach_loc+0x213
ahci_atapibus_attach() at netbsd:ahci_atapibus_attach+0xbb
atabusconfig_thread() at netbsd:atabusconfig_thread+0xd6

 Thomas