Re: [PATCH] vduse: implement DMA sync callbacks

2024-02-23 Thread Christoph Hellwig
On Thu, Feb 22, 2024 at 03:29:10PM -0500, Michael S. Tsirkin wrote: > In a sense ... but on the other hand, the "fake DMA" metaphor seems to > work surprisingly well, like in this instance - internal bounce buffer > looks a bit like non-coherent DMA. A way to make this all prettier > would I

Re: [PATCH] vduse: implement DMA sync callbacks

2024-02-22 Thread Michael S. Tsirkin
On Tue, Feb 20, 2024 at 01:01:36AM -0800, Christoph Hellwig wrote: > 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

Re: [PATCH] vduse: implement DMA sync callbacks

2024-02-21 Thread Maxime Coquelin
Hello Christoph, On 2/20/24 10:01, Christoph Hellwig wrote: 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

Re: [PATCH] vduse: implement DMA sync callbacks

2024-02-20 Thread Christoph Hellwig
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 >

[PATCH] vduse: implement DMA sync callbacks

2024-02-19 Thread Maxime Coquelin
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