RE: omap3 gpmc: irq-20 could not claim: err -2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Hi Peter,

On Thu, Oct 11, 2012 at 15:05:43, Peter Meerwald wrote:

> I'm getting the kernel log messages early in the bootup sequence (kernel 
> 3.6.1 mainline)
> [    0.121337] GPMC revision 5.0
> [    0.121582] gpmc: irq-20 could not claim: err -22

This issue has been present for quite some time, now as
certain gpmc patches got merged recently (during the
ongoing merge window), this should be gone.

> NAND still works since it is operated in NAND_OMAP_PREFETCH_POLLED mode; 
> however, I'd like to try NAND_OMAP_PREFETCH_IRQ to see if it brings any 
> performance gains

I vaguely remember that once while checking irq mode (not mainline,
on am335x-evm), speed was less compared to default, so I doubt whether
you get any performance gains, anyway it would be good to know
performance results in your case.

> just passing NAND_OMAP_PREFETCH_IRQ via gmpc_nand_init() hangs the kernel 
> (I guess since the request_irq() failed earlier)

With mainline HEAD, you should be able to use irq.

> another question is dev_ready which also seems to depend on IRQ also

Now it is no more.

> can you make any recommendation w.r.t NAND performance on OMAP?

Once DMA was checked, but found it slower than other two, not sure
whether I was doing anything wrong. Right now don't have any
suggestions, if I come across any will let you know.

Regards
Afzal
��.n��������+%������w��{.n�����{�������ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f



[Index of Archives]     [Linux Arm (vger)]     [ARM Kernel]     [ARM MSM]     [Linux Tegra]     [Linux WPAN Networking]     [Linux Wireless Networking]     [Maemo Users]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux