On 2019-10-24 16:51, Will Deacon wrote:
On Thu, Oct 24, 2019 at 04:41:04PM +0530, Sai Prakash Ranjan wrote:
On 2019-10-24 16:21, Will Deacon wrote:
> In commit 90ec7a76cc4b ("iommu/io-pgtable-arm: Add support to use system
> cache") we added support for IOMMU_QCOM_SYS_CACHE which was merged into
> 5.3.
> This allows non-coherent devices to request an outer cacheable memory
> type.... except that nobody actually does this in mainline. I remember
> there
> being a potential DRM user but I don't know what happened to it.
>
> Given that this isn't actually exposed in the DMA API, I worry that
> we're
> just carrying part of an out-of-tree hack here and propose that we drop
> the flag altogether unless we get an upstream user, preferably by
> plumbing
> this into the DMA API via a new attribute.
>
> Thoughts?
>

There is definitely a user of this coming soon atleast for SC7180 SoC once
we have support for this SoC upstream.

Ok, I'm keen to see how you end up using this. How soon is soon?


We have already started upstreaming for SC7180, so this should also come pretty soon. Sorry, I cannot tell the exact date but can make sure that your message reaches to appropriate team.

Thanks,
Sai

--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu

Reply via email to