Re: [PATCH v4 0/7] ARM: Fix dma_alloc_coherent() and friends for NOMMU

2017-05-24 Thread Vladimir Murzin
On 23/05/17 21:04, Russell King - ARM Linux wrote: > On Mon, May 15, 2017 at 09:52:58AM +0100, Vladimir Murzin wrote: >> Ping again... > > Apart from the one comment, the ARM bits look fine to me - but I'm > not saying anything about the lib/dma-noop.c or the drivers/base > changes. > > What's th

Re: [PATCH v4 0/7] ARM: Fix dma_alloc_coherent() and friends for NOMMU

2017-05-23 Thread Russell King - ARM Linux
On Mon, May 15, 2017 at 09:52:58AM +0100, Vladimir Murzin wrote: > Ping again... Apart from the one comment, the ARM bits look fine to me - but I'm not saying anything about the lib/dma-noop.c or the drivers/base changes. What's the dependency between the ARM bits and those bits? > On 02/05/17 0

Re: [PATCH v4 0/7] ARM: Fix dma_alloc_coherent() and friends for NOMMU

2017-05-15 Thread Vladimir Murzin
Ping again... On 02/05/17 09:32, Vladimir Murzin wrote: > Gentle ping! > > On 24/04/17 11:16, Vladimir Murzin wrote: >> It seem that addition of cache support for M-class CPUs uncovered >> latent bug in DMA usage. NOMMU memory model has been treated as being >> always consistent; however, for R/M

Re: [PATCH v4 0/7] ARM: Fix dma_alloc_coherent() and friends for NOMMU

2017-05-02 Thread Vladimir Murzin
Gentle ping! On 24/04/17 11:16, Vladimir Murzin wrote: > It seem that addition of cache support for M-class CPUs uncovered > latent bug in DMA usage. NOMMU memory model has been treated as being > always consistent; however, for R/M CPU classes memory can be covered > by MPU which in turn might co