Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH] drm/xen-front: Make shmem backed display buffer coherent
To:     Christoph Hellwig <hch@infradead.org>,
        =?UTF-8?Q?Noralf_Tr=c3=b8nnes?= <noralf@tronnes.org>
Cc:     xen-devel@lists.xenproject.org, linux-kernel@vger.kernel.org,
        dri-devel@lists.freedesktop.org, daniel.vetter@intel.com,
        jgross@suse.com, boris.ostrovsky@oracle.com,
        Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com>
References: <20181127103252.20994-1-andr2000@gmail.com>
 <17640791-5306-f7e4-8588-dd39c14e975b@tronnes.org>
 <20181220153646.GA22997@infradead.org>
From:   Oleksandr Andrushchenko <andr2000@gmail.com>
Message-ID: <f2cfadc7-2e0b-b63d-d094-d5339b0c85fe@gmail.com>
Date:   Thu, 20 Dec 2018 17:49:39 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.2.1
MIME-Version: 1.0
In-Reply-To: <20181220153646.GA22997@infradead.org>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 8bit
Content-Language: en-US
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 12/20/18 5:36 PM, Christoph Hellwig wrote:
> On Tue, Dec 18, 2018 at 08:20:22PM +0100, Noralf Trønnes wrote:
>>> +	if (!dma_map_sg(dev->dev, xen_obj->sgt->sgl, xen_obj->sgt->nents,
>>> +			DMA_BIDIRECTIONAL)) {
>>
>> Are you using the DMA streaming API as a way to flush the caches?
> This looks rather broken.  Please send the whole patch series to
> the iommu list for proper review.
This is the only patch [1], no series. And at the moment I think
there is nothing to review as I am not sure how to deal with those
shmem pages: this patch is rather to start a discussion on how shmem
pages can be flushed on ARM (the only workaround I have so far is
in this patch which uses DMA API). This is where I am looking for
some advice, so I can implement the patch the right way.
>> Does this mean that GFP_USER isn't making the buffer coherent?
> How could GFP_USER make memory coherent in any way?
I am no way an expert here, but other DRM drivers allocate buffers
from shmem and then use DMA API [2], for example [3]

[1] https://patchwork.kernel.org/patch/10700089/
[2] https://elixir.bootlin.com/linux/v4.20-rc7/ident/drm_gem_get_pages
[3] 
https://elixir.bootlin.com/linux/v4.20-rc7/source/drivers/gpu/drm/omapdrm/omap_gem.c#L248