new support for Raspberry Pi B+

[Bernd Walter]

On Sat, Nov 22, 2014 at 02:20:02PM -0700, Warner Losh wrote:
> 
> On Nov 22, 2014, at 2:05 PM, Ian Lepore <ian at FreeBSD.org> wrote:
> 
> > On Sat, 2014-11-22 at 13:56 -0700, Dan Raymond wrote:
> >> On 11/22/2014 1:09 PM, Bernd Walter wrote:
> >>> On Sat, Nov 22, 2014 at 12:46:34PM -0700, Dan Raymond wrote:
> >>>> Further testing with the 16GB Kingston reveals that it is unstable even
> >>>> at low speed.  I hit the following panic while executing "portsnap fetch
> >>>> update" on a r274416 build:
> >>>> 
> >>>> dev = mmcsd0s2a, ino = 155101, fs = /mnt/ufs
> >>>> panic: ffs_freefile: freeing free inode
> >>>> KDB: enter: panic
> >>>> [ thread pid 8 tid 100056 ]
> >>>> Stopped at      $d:     ldrb    r15, [r15, r15, ror r15]!
> >>>> db>
> >>> Maybe you've received a counterfeit.
> >>> Unfortunately there are a lot of them on the market.
> >>> Some of them are technically smaller than they claim and overwrite
> >>> other data if they run short.
> >> 
> >> I considered that originally which is why I bought 4 different cards of 
> >> different brands for testing (Transcend, SanDisk, Samsung, Kingston):
> >> 
> >> http://www.amazon.com/gp/product/B00CBD0XSI/ref=oh_aui_detailpage_o06_s00?ie=UTF8&psc=1
> >> http://www.amazon.com/gp/product/B00M55C0NS/ref=oh_aui_detailpage_o04_s00?ie=UTF8&psc=1
> >> http://www.amazon.com/gp/product/B00IVPU7DQ/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
> >> http://www.amazon.com/gp/product/B003WIRFD2/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1
> >> 
> >> None of them will boot FreeBSD at high speed and I have seen SD related 
> >> panics with 3 of them (so far) at low speed.

Well high speed seems to be a different issue.
But if it happens on multiple different cards, then the corruption is unlikely
card related.

> > I would actually dispute that this one is an sd-related panic, unless
> > you have messages about gvfs_iodone failures on the console before the
> > panic.  Otherwise this is more likely another memory corruption problem
> > similar to the panics being reported on wandboard.
> 
> The RPi is also sensitive to power. I?ve seen weird corruption like this when I tried to run
> my RPi off a battery and the battery was getting weak...

They completely redesigned the power design in the B+, including the
use of decent switching regulators.
Assuming the problem happens on a B+ as the subject claims.

-- 
B.Walter <bernd at bwct.de> http://www.bwct.de
Modbus/TCP Ethernet I/O Baugruppen, ARM basierte FreeBSD Rechner uvm.