[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZdRqcIRmDD-70ap3@infradead.org>
Date: Tue, 20 Feb 2024 01:01:36 -0800
From: Christoph Hellwig <hch@...radead.org>
To: Maxime Coquelin <maxime.coquelin@...hat.com>
Cc: mst@...hat.com, jasowang@...hat.com, xuanzhuo@...ux.alibaba.com,
xieyongji@...edance.com, axboe@...nel.dk,
gregkh@...uxfoundation.org, brauner@...nel.org, lstoakes@...il.com,
virtualization@...ts.linux.dev, linux-kernel@...r.kernel.org,
netdev@...r.kernel.org, david.marchand@...hat.com
Subject: Re: [PATCH] vduse: implement DMA sync callbacks
On Mon, Feb 19, 2024 at 06:06:06PM +0100, Maxime Coquelin wrote:
> Since commit 295525e29a5b ("virtio_net: merge dma
> operations when filling mergeable buffers"), VDUSE device
> require support for DMA's .sync_single_for_cpu() operation
> as the memory is non-coherent between the device and CPU
> because of the use of a bounce buffer.
>
> This patch implements both .sync_single_for_cpu() and
> sync_single_for_device() callbacks, and also skip bounce
> buffer copies during DMA map and unmap operations if the
> DMA_ATTR_SKIP_CPU_SYNC attribute is set to avoid extra
> copies of the same buffer.
vduse really needs to get out of implementing fake DMA operations for
something that is not DMA.
Powered by blists - more mailing lists