Re: [PATCH] dmatest: flush and invalidate destination buffer before DMA

[Atsushi Nemoto <anemo@mba.ocn.ne.jp>]

On Sat, 27 Dec 2008 11:10:37 +0100, Haavard Skinnemoen <haavard.skinnemoen@atmel.com> wrote:
> Atsushi Nemoto wrote:
> > @@ -226,6 +227,12 @@ static int dmatest_func(void *data)
> >  
> >  		dmatest_init_srcbuf(thread->srcbuf, src_off, len);
> >  		dmatest_init_dstbuf(thread->dstbuf, dst_off, len);
> > +		/* flush and invalidate caches for whole dstbuf */
> > +		dma_dest = dma_map_single(chan->device->dev,
> > +				thread->dstbuf,
> > +				test_buf_size, DMA_BIDIRECTIONAL);
> > +		dma_unmap_single(chan->device->dev, dma_dest,
> > +				test_buf_size, DMA_BIDIRECTIONAL);
> 
> You're supposed to unmap after the DMA operation is done, not before
> it's submitted.
> 
> In this case, the DMA engine framework will do the unmapping for you
> (probably using the wrong primitive, but they're really all the same in
> practice, right?) so you can just drop the unmap call.

Well, let me explain more.

On nono-coherent MIPS platforms, dma_map_single() for DMA_TO_DEVICE
writeback the cache, dma_map_single() for DMA_FROM_DEVICE invalidated
(without writeback) the cache.  dma_unmap_simgle() is a nop.

If dst_off was not cacheline aligned, dma_map_single(...,
DMA_FROM_DEVICE) in dma_async_memcpy_buf_to_buf() invalidate whole
cachelines including dst_off.  So, for example, the initialized data
at dst_off - 1 will be just discarded.  This result mismatch error of
course.  Same error can be happen at end of the real DMA area.

I added dma_map_single/dma_unmap_single to just flush all dstbuf to
main memory.

> Now, I suspect the dw_dmac driver maps the buffer when it's not
> supposed to, masking this kind of error...should probably get that
> fixed too.

I don't think so.  But I'm not sure dma_map_sg() case in
dwc_prep_slave_sg().

---
Atsushi Nemoto