Re: [PATCH V3 2/3] PCI: rcar: Do not abort on too many inbound dma-ranges

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

 



On 11/18/19 7:42 PM, Robin Murphy wrote:
[...]

>>> Despite the good intentions here, it doesn't seem like dma-ranges is
>>> designed for this purpose and as the hardware has limited ranges it will
>>> only be best-effort.
>> So what other options do we have ?
> 
> If you really want to sacrifice DMA efficiency for a perceived increase
> in theoretical robustness by setting very conservative windows,

That really means configuring the hardware correctly.

> then
> ultimately it's your choice, go ahead. It's just that you *need* to make
> that choice in the bootloader, not in Linux. If Linux gets passed
> dma-ranges that aren't actually reflected by the hardware, such that
> this patch is needed, then it *will* go wrong eventually, and you'll
> only get an "I told you so" from me.
> 
> The bootloader knows what platform it's running on, so it has no excuse
> for emitting more ranges than there are available windows on that platform.

So basically the conclusion is to limit the amount of DMA ranges added
into the DT and be done with it ?

-- 
Best regards,
Marek Vasut



[Index of Archives]     [DMA Engine]     [Linux Coverity]     [Linux USB]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [Greybus]

  Powered by Linux