Thread (111 messages) 111 messages, 5 authors, 2018-02-06

Re: [PATCH v6 16/36] nds32: DMA mapping API

From: Arnd Bergmann <hidden>
Date: 2018-01-24 11:36:17
Also in: linux-arch, linux-devicetree, linux-serial, lkml

On Tue, Jan 23, 2018 at 12:52 PM, Greentime Hu [off-list ref] wrote:
Hi, Arnd:

2018-01-23 16:23 GMT+08:00 Greentime Hu [off-list ref]:
quoted
Hi, Arnd:

2018-01-18 18:26 GMT+08:00 Arnd Bergmann [off-list ref]:
quoted
On Mon, Jan 15, 2018 at 6:53 AM, Greentime Hu [off-list ref] wrote:
quoted
From: Greentime Hu <greentime-MUIXKm3Oiri1Z/+hSey0Gg@public.gmane.org>

This patch adds support for the DMA mapping API. It uses dma_map_ops for
flexibility.

Signed-off-by: Vincent Chen <vincentc-MUIXKm3Oiri1Z/+hSey0Gg@public.gmane.org>
Signed-off-by: Greentime Hu <greentime-MUIXKm3Oiri1Z/+hSey0Gg@public.gmane.org>
I'm still unhappy about the way the cache flushes are done here as discussed
before. It's not a show-stopped, but no Ack from me.
How about this implementation?
I am not sure if I understand it correctly.
I list all the combinations.

RAM to DEVICE
    before DMA => writeback cache
    after DMA => nop

DEVICE to RAM
    before DMA => nop
    after DMA => invalidate cache

static void consistent_sync(void *vaddr, size_t size, int direction, int master)
{
        unsigned long start = (unsigned long)vaddr;
        unsigned long end = start + size;

        if (master == FOR_CPU) {
                switch (direction) {
                case DMA_TO_DEVICE:
                        break;
                case DMA_FROM_DEVICE:
                case DMA_BIDIRECTIONAL:
                        cpu_dma_inval_range(start, end);
                        break;
                default:
                        BUG();
                }
        } else {
                /* FOR_DEVICE */
                switch (direction) {
                case DMA_FROM_DEVICE:
                        break;
                case DMA_TO_DEVICE:
                case DMA_BIDIRECTIONAL:
                        cpu_dma_wb_range(start, end);
                        break;
                default:
                        BUG();
                }
        }
}
That looks reasonable enough, but it does depend on a number of factors,
and the dma-mapping.h implementation is not just about cache flushes.

As I don't know the microarchitecture, can you answer these questions:

- are caches always write-back, or could they be write-through?
- can the cache be shared with another CPU or a device?
- if the cache is shared, is it always coherent, never coherent, or
either of them?
- could the same memory be visible at different physical addresses
  and have conflicting caches?
- is the CPU physical address always the same as the address visible to the
  device?
- are there devices that can only see a subset of the physical memory?
- can there be an IOMMU?
- are there write-buffers in the CPU that might need to get flushed before
  flushing the cache?
- could cache lines be loaded speculatively or with read-ahead while
  a buffer is owned by a device?

        Arnd
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help