On Sat, 4 Aug 2018, Matwey V. Kornilov wrote:

2018-07-30 18:35 GMT+03:00 Laurent Pinchart [off-list ref]:

quoted

Hi Matwey,

On Tuesday, 24 July 2018 21:56:09 EEST Matwey V. Kornilov wrote:

quoted

2018-07-23 21:57 GMT+03:00 Alan Stern:

quoted

On Mon, 23 Jul 2018, Matwey V. Kornilov wrote:

quoted

I've tried to strategies:

1) Use dma_unmap and dma_map inside the handler (I suppose this is
similar to how USB core does when there is no URB_NO_TRANSFER_DMA_MAP)

Yes.

quoted

2) Use sync_cpu and sync_device inside the handler (and dma_map only
once at memory allocation)

It is interesting that dma_unmap/dma_map pair leads to the lower
overhead (+1us) than sync_cpu/sync_device (+2us) at x86_64 platform.
At armv7l platform using dma_unmap/dma_map  leads to ~50 usec in the
handler, and sync_cpu/sync_device - ~65 usec.

However, I am not sure is it mandatory to call
dma_sync_single_for_device for FROM_DEVICE direction?

According to Documentation/DMA-API-HOWTO.txt, the CPU should not write
to a DMA_FROM_DEVICE-mapped area, so dma_sync_single_for_device() is
not needed.

Well, I measured the following at armv7l. The handler execution time
(URB_NO_TRANSFER_DMA_MAP is used for all cases):

1) coherent DMA: ~3000 usec (pwc is not functional)
2) explicit dma_unmap and dma_map in the handler: ~52 usec
3) explicit dma_sync_single_for_cpu (no dma_sync_single_for_device): ~56
usec

I really don't understand why the sync option is slower. Could you please
investigate ? Before doing anything we need to make sure we have a full
understanding of the problem.

Hi,

I've found one drawback in my measurements. I forgot to fix CPU
frequency at lowest state 300MHz. Now, I remeasured

2) dma_unmap and dma_map in the handler:
2A) dma_unmap_single call: 28.8 +- 1.5 usec
2B) memcpy and the rest: 58 +- 6 usec
2C) dma_map_single call: 22 +- 2 usec
Total: 110 +- 7 usec

3) dma_sync_single_for_cpu
3A) dma_sync_single_for_cpu call: 29.4 +- 1.7 usec
3B) memcpy and the rest: 59 +- 6 usec
3C) noop (trace events overhead): 5 +- 2 usec
Total: 93 +- 7 usec

So, now we see that 2A and 3A (as well as 2B and 3B) agree good within
error ranges.

Taken together, those measurements look like a pretty good argument for 
always using dma_sync_single_for_cpu in the driver.  Provided results 
on other platforms aren't too far out of line with these results.

Alan Stern