Messages in this thread | | | Date | Fri, 27 Jan 2006 19:37:48 +0100 | From | Pierre Ossman <> | Subject | Re: How to map high memory for block io |
| |
Pierre Ossman wrote: > > Doesn't seem like a generic solution is easily implemented. I'll start > hacking together some way of specifying that highmem isn't supported so > that mmc_block can indicate this to the block layer. >
Whilst doing this I discovered that the MMC layer already does some bounce buffer limiting. It defaults to BLK_BOUNCE_HIGH, but when dev->dma_mask is set it uses that. The problem is that the default value for PCI devices covers high mem transfers.
Russell, what was your plan here? Should MMC drivers set dma_mask to BLK_BOUNCE_HIGH when not doing DMA? Or perhaps 0?
Rgds Pierre
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |