Re: earmhf issues on Beaglebone Black

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Wed, Jul 23, 2014 at 12:58:14AM +0200, Manuel Bouyer wrote:
> On Wed, Jul 23, 2014 at 12:22:03AM +0200, Manuel Bouyer wrote:
> > > > Sure, here it is:
> > > > data_abort_handler: data_aborts fsr=0x80e far=0x20000
> > > > dab_buserr far 0x20000 -> NULL pm->pm_l1[far >> L1_S_SHIFT] = 0x0
> > > > pm->pm_l1 0x9f036000 armreg_ttbr_read() 0x9f03601b
> > > 
> > > What happens if you change the [14] entry to NULL?
> > 
> > I'm now running with the attached patch, which should have the same
> > effect as setting data_aborts[fsr].func to NULL.
> > 
> > I got
> > fsr 0x182e, continuing anyway
> > fsr 0x180e, continuing anyway
> > fsr 0x182e, continuing anyway
> > 
> > but no error from make so far, it's now building the first package.
> 
> It has built libtool-base and pkg-config without problems so far, despite:
> fsr 0x182e, continuing anyway
> fsr 0x180e, continuing anyway
> fsr 0x182e, continuing anyway
> fsr 0x18be, continuing anyway
> fsr 0x184e, continuing anyway
> fsr 0x18ce, continuing anyway

While building packages, I got a variant of the bug:
data_abort_handler: data_aborts fsr=0xc06 far=0xc9202080
dab_buserr far 0xc9202080 -> NULL pm->pm_l1[far >> L1_S_SHIFT] = 0x0
pm->pm_l1 0x87f6e000 armreg_ttbr_read() 0x87f6e01b
data_abort_handler: data_aborts fsr=0xc06 far=0x97ec7a34
dab_buserr far 0x97ec7a34 -> NULL pm->pm_l1[far >> L1_S_SHIFT] = 0x0
pm->pm_l1 0x87f6e000 armreg_ttbr_read() 0x87f6e01b
copyin_vmspace: copyin 0x26000 0xc9202000 4728 return 14
uiomove: copyout_vmspace error 14

note that the fault type is 6, not 0xe ... but it's probably the
same underlying issue: we got a fault on an unmapped page,
but not the right fault.

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