Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-10-04 Thread Konrad Rzeszutek Wilk
On Wed, Oct 02, 2013 at 06:31:57PM +0100, Stefano Stabellini wrote: > On Mon, 30 Sep 2013, Konrad Rzeszutek Wilk wrote: > > On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: > > > The swiotlb code has appropriate calls to dma_mark_clean in place for > > > buffers passed to

Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-10-04 Thread Konrad Rzeszutek Wilk
On Wed, Oct 02, 2013 at 06:31:57PM +0100, Stefano Stabellini wrote: On Mon, 30 Sep 2013, Konrad Rzeszutek Wilk wrote: On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: The swiotlb code has appropriate calls to dma_mark_clean in place for buffers passed to

Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-10-02 Thread Stefano Stabellini
On Mon, 30 Sep 2013, Konrad Rzeszutek Wilk wrote: > On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: > > The swiotlb code has appropriate calls to dma_mark_clean in place for > > buffers passed to swiotlb_map_page as an argument. However it assumes > > that the swiotlb bounce

Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-10-02 Thread Stefano Stabellini
On Mon, 30 Sep 2013, Konrad Rzeszutek Wilk wrote: On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: The swiotlb code has appropriate calls to dma_mark_clean in place for buffers passed to swiotlb_map_page as an argument. However it assumes that the swiotlb bounce buffer

Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-09-30 Thread Konrad Rzeszutek Wilk
On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: > The swiotlb code has appropriate calls to dma_mark_clean in place for > buffers passed to swiotlb_map_page as an argument. However it assumes > that the swiotlb bounce buffer (io_tlb_start-io_tlb_end) is already > coherent and

Re: [PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-09-30 Thread Konrad Rzeszutek Wilk
On Fri, Sep 27, 2013 at 05:10:00PM +0100, Stefano Stabellini wrote: The swiotlb code has appropriate calls to dma_mark_clean in place for buffers passed to swiotlb_map_page as an argument. However it assumes that the swiotlb bounce buffer (io_tlb_start-io_tlb_end) is already coherent and

[PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-09-27 Thread Stefano Stabellini
The swiotlb code has appropriate calls to dma_mark_clean in place for buffers passed to swiotlb_map_page as an argument. However it assumes that the swiotlb bounce buffer (io_tlb_start-io_tlb_end) is already coherent and doesn't need any calls to dma_mark_clean. On ARM the swiotlb bounce buffer

[PATCH v6 12/19] swiotlb: don't assume that io_tlb_start-io_tlb_end is coherent

2013-09-27 Thread Stefano Stabellini
The swiotlb code has appropriate calls to dma_mark_clean in place for buffers passed to swiotlb_map_page as an argument. However it assumes that the swiotlb bounce buffer (io_tlb_start-io_tlb_end) is already coherent and doesn't need any calls to dma_mark_clean. On ARM the swiotlb bounce buffer