Re: [PATCH v3] sparc: use generic dma_noncoherent_ops

2018-08-21 Thread Christoph Hellwig
On Tue, Aug 21, 2018 at 10:57:37PM -0700, David Miller wrote: > From: Christoph Hellwig > Date: Wed, 22 Aug 2018 07:36:13 +0200 > > > I fear you have already pushed out your tree, but otherwise it would > > be better do merge it through the dma-mapping tree. > > I did and asked Linus to pull as

Re: [PATCH v3] sparc: use generic dma_noncoherent_ops

2018-08-21 Thread David Miller
From: Christoph Hellwig Date: Wed, 22 Aug 2018 07:36:13 +0200 > I fear you have already pushed out your tree, but otherwise it would > be better do merge it through the dma-mapping tree. I did and asked Linus to pull as well. Sorry :-/ ___ iommu maili

Re: [PATCH v3] sparc: use generic dma_noncoherent_ops

2018-08-21 Thread Christoph Hellwig
On Tue, Aug 21, 2018 at 12:38:00PM -0700, David Miller wrote: > From: Christoph Hellwig > Date: Fri, 3 Aug 2018 09:48:40 +0200 > > > any chance you would consider reviewing and applying this patch, > > which already has an ACK from Sam? It would really help me with > > some core DMA API projects

Re: [PATCH v3] sparc: use generic dma_noncoherent_ops

2018-08-21 Thread David Miller
From: Christoph Hellwig Date: Fri, 3 Aug 2018 09:48:40 +0200 > any chance you would consider reviewing and applying this patch, > which already has an ACK from Sam? It would really help me with > some core DMA API projects planned for the next merge window. Applied, thanks Chrisoph. ___

Re: [PATCH v3] sparc: use generic dma_noncoherent_ops

2018-08-03 Thread Christoph Hellwig
Dave, any chance you would consider reviewing and applying this patch, which already has an ACK from Sam? It would really help me with some core DMA API projects planned for the next merge window. On Tue, Jul 31, 2018 at 10:02:04AM +0200, Christoph Hellwig wrote: > Switch to the generic noncoher

[PATCH v3] sparc: use generic dma_noncoherent_ops

2018-07-31 Thread Christoph Hellwig
Switch to the generic noncoherent direct mapping implementation. This removes the previous sync_single_for_device implementation, which looks bogus given that no syncing is happening in the similar but more important map_single case. Signed-off-by: Christoph Hellwig Acked-by: Sam Ravnborg ---