Re: New bug?

[Alexander Shiyan <shc_work@xxxxxxx>]

Среда, 19 февраля 2014, 8:45 +01:00 от Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>:
> On Wed, Feb 19, 2014 at 8:32 AM, Alexander Shiyan <shc_work@xxxxxxx> wrote:
> > During testing functions transfer_one(), stumbled upon an error.
> > I'm using kernel 3.13, so I had to apply a patch from upstream
> > (spi: core: Fix transfer failure when master->transfer_one returns positive value).
> > After applying all becomes normal with probing, but where did the error
> > "Unable to handle kernel NULL pointer dereference" ?
> > Did someone have similar bugs?
> >
> > Thanks.
> >
> > spi_master spi0: failed to transfer one message from queue: 1
> > 74x164 spi0.1: Failed writing: 1
> > 74x164: probe of spi0.1 failed with error 1
> > Unable to handle kernel NULL pointer dereference at virtual address 00000020
> > CAN device driver interface
> > pgd = c0024000
> > [00000020] *pgd=00000000
> > Internal error: Oops: 63c11805 [#1] PREEMPT ARM
> > CPU: 0 PID: 22 Comm: spi0 Not tainted 3.13.3 #1
> > task: c396b260 ti: c396e000 task.ti: c396e000
> > PC is at spi_pump_messages+0x150/0x194
> 
> Have you applied other patches from upstream?
> 
> E.g. e120cc0dcf2880a4c5c0a6cb27b655600a1cfa1d introduced a NULL
> pointer dereference, with got fixed by
> 1f802f8249a0da536877842c43c7204064c4de8b.

It helped me. Thanks.

Mark, Is any chances to put these two patches to 3.13-stable?
(spi: core: Fix transfer failure when master->transfer_one returns positive value),
(spi: Fix crash with double message finalisation on error handling)

---
��.n��������+%������w��{.n�����{����)��jg��������ݢj����G�������j:+v���w�m������w�������h�����٥