On Wed, Sep 21, 2022 at 04:28:32PM -0700, Bart Van Assche wrote:
> Hi Jens, Christoph and Ming,
>
> As we know the Linux kernel block layer does not support DMA segment sizes
> that are smaller than the page size. From block/blk-settings.c:
>
> void blk_queue_max_segment_size(struct request_queue *q,
> unsigned int max_size)
> {
> if (max_size < PAGE_SIZE) {
> max_size = PAGE_SIZE;
> printk(KERN_INFO "%s: set to minimum %d\n",
> __func__, max_size);
> }
>
> /* see blk_queue_virt_boundary() for the explanation */
> WARN_ON_ONCE(q->limits.virt_boundary_mask);
>
> q->limits.max_segment_size = max_size;
> }
>
> I have been asked to add support for DMA segment sizes that are smaller than
> the page size to support a UFS controller with a maximum DMA segment size of
> 4 KiB and a page size > 4 KiB (my understanding of the JEDEC UFS host
> controller specification is that UFS host controllers should support a
> maximum DMA segment size of 256 KiB). Does anyone want to comment on this
> before I start working on a patch series?
If the hardware's DMA segment is smaller than a page, why doesn't the driver
just split a kernel's larger segment into whatever representation the hardware
wants? We do that in nvme, at least.
