2012-11-04 13:20:05

by Geert Uytterhoeven

Subject: Re: [v2 PATCH 1/2] Add new uio device for dynamic memory allocation

On Tue, Sep 25, 2012 at 8:09 AM, Damian Hobson-Garcia
<[email protected]> wrote:
> are holding the device file open, the address returned to userspace is
> DMA_ERROR_CODE.

Only a small subset of the architectures
(arm/ia64/microblaze/openrisc/powerpc/sparc/x86)
seem to define DMA_ERROR_CODE, causing everywhere else:

drivers/uio/uio_dmem_genirq.c: In function ‘uio_dmem_genirq_release’:
drivers/uio/uio_dmem_genirq.c:95: error: ‘DMA_ERROR_CODE’ undeclared
(first use in this function)
drivers/uio/uio_dmem_genirq.c:95: error: (Each undeclared identifier
is reported only once
drivers/uio/uio_dmem_genirq.c:95: error: for each function it appears in.)
drivers/uio/uio_dmem_genirq.c: In function ‘uio_dmem_genirq_probe’:
drivers/uio/uio_dmem_genirq.c:238: error: ‘DMA_ERROR_CODE’ undeclared
(first use in this function)
make[1]: *** [drivers/uio/uio_dmem_genirq.o] Error 1
make: *** [drivers/uio/] Error 2

(e.g. http://kisskb.ellerman.id.au/kisskb/buildresult/7462173/)

Furthermore, none of them define it in <uapi/...>, so DMA_ERROR_CODE is not
part of the userspace API (yet), while it is architecture-specific (some use 0,
others ~0).

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

2012-11-05 02:34:58

by Damian Hobson-Garcia

Subject: Re: [v2 PATCH 1/2] Add new uio device for dynamic memory allocation

Hi Geert,

On 2012/11/04 22:20, Geert Uytterhoeven wrote:
> On Tue, Sep 25, 2012 at 8:09 AM, Damian Hobson-Garcia
> <[email protected]> wrote:
>> are holding the device file open, the address returned to userspace is
>> DMA_ERROR_CODE.
>
> Only a small subset of the architectures
> (arm/ia64/microblaze/openrisc/powerpc/sparc/x86)
> seem to define DMA_ERROR_CODE, causing everywhere else:
>
> drivers/uio/uio_dmem_genirq.c: In function ‘uio_dmem_genirq_release’:
> drivers/uio/uio_dmem_genirq.c:95: error: ‘DMA_ERROR_CODE’ undeclared
> (first use in this function)
> drivers/uio/uio_dmem_genirq.c:95: error: (Each undeclared identifier
> is reported only once
> drivers/uio/uio_dmem_genirq.c:95: error: for each function it appears in.)
> drivers/uio/uio_dmem_genirq.c: In function ‘uio_dmem_genirq_probe’:
> drivers/uio/uio_dmem_genirq.c:238: error: ‘DMA_ERROR_CODE’ undeclared
> (first use in this function)
> make[1]: *** [drivers/uio/uio_dmem_genirq.o] Error 1
> make: *** [drivers/uio/] Error 2
>
> (e.g. http://kisskb.ellerman.id.au/kisskb/buildresult/7462173/)
>
> Furthermore, none of them define it in <uapi/...>, so DMA_ERROR_CODE is not
> part of the userspace API (yet), while it is architecture-specific (some use 0,
> others ~0).
>
Good points, thank you. In that case, does fixing the value to
something like ~0 sound like a better (more portable) solution?

Damian

--
Damian Hobson-Garcia
IGEL Co.,Ltd
http://www.igel.co.jp