Re: [PATCH] dmatest: flush and invalidate destination bufferbefore DMA

[Atsushi Nemoto]

On Mon, 5 Jan 2009 11:31:57 -0700, "Dan Williams" <dan.j.williams@xxxxxxxxx> wrote:
> > 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.
> 
> I am tempted to say this is a bug in the MIPS dma_map_single
> implementation.  It is at least a difference in behavior from ARM
> where partial cachelines are cleaned+invalidated in the
> DMA_FROM_DEVICE case [1].

Yes, MIPS and ARM do different thing on partial cache line.  But I
suppose this belongs to "implementation dependent" area of DMA API so
users of the API should not depend on it.  (Well, maybe I'm biased to
MIPS ;-))

In general, drivers must not put normal data and DMA buffer on same
cacheline anyway to avoid unexpected writeback and data loss.  So this
ambiguity is not a problem.  IMHO writeback of the partial line for
DMA_FROM_DEVICE just _hides_ abusing of the DMA API and potential data
loss.

---
Atsushi Nemoto
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/