<p dir="ltr"></p>
<p dir="ltr">On Sun, Dec 25, 2016 at 2:37 PM, Greg KH <<a href="mailto:greg@kroah.com">greg@kroah.com</a>> wrote:<br>
> On Sat, Dec 24, 2016 at 05:47:19PM +0200, Ran Shalit wrote:<br>
>><br>
>><br>
>> On Tue, Dec 20, 2016 at 12:26 PM, Greg KH <<a href="mailto:greg@kroah.com">greg@kroah.com</a>> wrote:<br>
>><br>
>> On Tue, Dec 20, 2016 at 12:15:22PM +0200, Ran Shalit wrote:<br>
>> > On Tue, Dec 20, 2016 at 11:38 AM, Greg KH <<a href="mailto:greg@kroah.com">greg@kroah.com</a>> wrote:<br>
>> > > On Mon, Dec 19, 2016 at 06:08:47PM +0200, Ran Shalit wrote:<br>
>> > >> Hello,<br>
>> > >><br>
>> > >> I want to use DMA from userspace.<br>
>> > ><br>
>> > > Why?<br>
>> ><br>
>> > Hi Greg,<br>
>> ><br>
>> > We want that a userspace layer (a library) will do some HW related<br>
>> issues.<br>
>> > We have a memory mapped space (from FPGA), so we think it will be<br>
>> > easier, and I think also more correct way , that we create the driver,<br>
>> > and interact hadrware using the mapped memory space, and also do the<br>
>> > protocols in userspace. The only thing that is less easy in userspace<br>
>> > is using interrupt, and dma. but that is also possible if we just wrap<br>
>> > the dma, and interrupt in a character device (or use uio as you<br>
>> > suggested below).<br>
>> > ><br>
>> > >> I already use dma in kernel, and now I want can create a character<br>
>> > >> device which will be responsible for this.<br>
>> > ><br>
>> > > Why?<br>
>> > ><br>
>> > >> The only problem is that I want to use the same memory which was<br>
>> > >> allocated in kernel with dma_alloc_coherent.<br>
>> > ><br>
>> > > Why?<br>
>> ><br>
>> > in kernel we use dma_alloc_coherent, which returns contiguous memory.<br>
>> > As I understand, we can mmap in userspace the returned physical<br>
>> > address, and use the returned virtual address in userspace.<br>
>> ><br>
>> ><br>
>> > ><br>
>> > >> Is it correct to use mmap in order to use the phsyical memory which<br>
>> > >> was allocated with dma_alloc_coherent ?<br>
>> > >><br>
>> > >> If it's that simple it can be surely helpful, and the simple driver<br>
>> > >> which wraps dma_alloc_coherent can do the job for dmaing from<br>
>> > >> userspace.<br>
>> > ><br>
>> > > Have you looked at the uio driver interface?<br>
>> > ><br>
>> > > But again, why? What problem are you trying to solve here?<br>
>> ><br>
>> > We need to do some interaction with HW , but since most of the HW is<br>
>> > mapped to physical address (FPGA), it seem simpler to do that in<br>
>> > userspace (HW library), instead of doing this in kernel. What do you<br>
>> > think ?<br>
>><br>
>> I think you should use the UIO driver api, as that's exactly what it was<br>
>> written for. Have you looked at it yet? It handles your interrupt<br>
>> logic for you.<br>
>><br>
>><br>
>> Hello,<br>
>><br>
>> If I may please ask, I made some reading about uio, but didn't yet understand<br>
>> what's the benefit of using uio instead of creating a character device ?<br>
><br>
> It's a lot less work than writing a custom char driver that will not be<br>
> accepted upstream because you are not using the expected UIO api<br>
> interface :)<br>
><br>
> Writing a UIO driver should be very simple, and very small, all of the<br>
> framework is already done, in a correct way, why would you _not_ want to<br>
> use the UIO interface?<br>
><br>
Hi,</p>
<p dir="ltr">UIO drivers seems like a good choice in my case, I am familiar with generic-uio interface in devicetree.<br>
Just for my understanding, I am trying to understand the difference between writing a small character device which notifies the interrupt, to using uio interface.<br>
Is there any advantage of using uio over the the small chracter device?(I am sure there is. I just do not know it yet)<br>
Another question, in performance terms which is better: uio driver or kernel driver?</p>
<p dir="ltr">Thank you,<br>
Ran</p>
<p dir="ltr">> thanks,<br>
><br>
> greg k-h</p>