[PATCH v4] dma: Add Xilinx AXI Direct Memory Access Engine driver support

[Vinod Koul]

On Fri, Feb 20, 2015 at 11:42:31AM +0100, Michal Simek wrote:
> Hi Vinod,
> 
> On 02/20/2015 07:29 AM, Vinod Koul wrote:
> > On Sat, Feb 14, 2015 at 02:21:52PM +0000, Appana Durga Kedareswara Rao wrote:
> >> Hi Vinod,
> >>
> >>
> >>> -----Original Message-----
> >>> From: Vinod Koul [mailto:vinod.koul at intel.com]
> >>> Sent: Thursday, February 12, 2015 3:05 PM
> >>> To: Appana Durga Kedareswara Rao
> >>> Cc: Arnd Bergmann; linux-arm-kernel at lists.infradead.org;
> >>> devicetree at vger.kernel.org; Srikanth Vemula; linux-
> >>> kernel at vger.kernel.org; Srikanth Thokala; dmaengine at vger.kernel.org;
> >>> robh+dt at kernel.org; Michal Simek; Anirudha Sarangi;
> >>> grant.likely at linaro.org; dan.j.williams at intel.com
> >>> Subject: Re: [PATCH v4] dma: Add Xilinx AXI Direct Memory Access Engine
> >>> driver support
> >>>
> >>> On Thu, Jan 29, 2015 at 11:24:41AM +0000, Appana Durga Kedareswara Rao
> >>> wrote:
> >>>>>  Earlier in the driver we are embedding dma_slave_config into custom
> >>>>> structs But Lars suggested to modify it. please refer to the below
> >>>>> thread for more details.
> >>>>>  http://www.spinics.net/lists/dmaengine/msg00010.html
> >>>>>
> >>>>>  And later we thought of creating a separate private member in
> >>>>> dma_slave_config for sharing additional configuration between slave
> >>>>> device and dma engine Or a new dma_ctrl_cmd like
> >>>>>
> >>> FSLDMA_EXTERNAL_START(http://www.kernelhub.org/?msg=405535&p=2 )
> >>>>>
> >>>>>  Finally Vinod was ok with the existing suggestion so we dropped the
> >>>>> above thought.
> >>>>>
> >>>>>  @Vinod : Could you please comment on this.
> >>> FSLDMA_EXTERNAL_START has been removed, and you cna use the removal
> >>> as a guidance of how to solve your issue
> >>>
> >>
> > Please limit you mailer to set wrap at 80 chars, otherwise it becomes
> > unreadable
> > 
> >> The current VDMA driver that got merged in the open source and the
> >> current patch (dma: xilinx dma v4) Srikanth T submitted both contains the
> >> Similar implementation that you pointed as above(I mean similar to the
> >> free scale dma driver).
> >>
> >> The discussion starts here is why we need the header file(Xilinx_dma.h) in
> >> the include/linux/amba/Xilinx_dma.h  folder.
> >>
> >> Like for FSL DMA it is got defined (fsl_dma_external_start) as a function
> >> prototype in include/linux/fsldma.h And it is being getting used in the
> >> drivers/misc/carma/carma-fpga-program.c
> >>
> >> Similarly in the Xilinx VDMA driver(xilinx_vdma_channel_set_config )
> >> defined as a function prototype in the include/linux/amba/xilinx_dma.h And
> >> is being used by Xilinx videos  drivers.  Those video driver need to use
> >> the (xilinx_vdma_channel_set_config) which requires this global header
> >> file.
> >>
> >> Could you please suggest us whether that header file (Xilinx_dma.h) should
> >> be there  in include/linux/amba or we should move it to include/linux?
> >> And any functional comments for this v4 patch?
> > Let me ask you a question, what is the relation of xilinx_dma with amba?
> > 
> > If none then only include/linux/ seems to be the right choice. If not then
> > amba maybe okay but you need to provide justfication why it should be in amba?
> 
> I see your point. I think there is real reason to have this header in amba.
> The reason for that was that xilinx_dma.h was added there.
and what would be that?
> 
> Anyway there is include/linux/dma/ folder. Is it ok to move xilinx_dma.h there
> too? The same with this dma one.
Yes that is fine too... As i said this or include/linux should be default
location..

-- 
~Vinod