Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-29 Thread Thomas Zimmermann
Am 29.09.20 um 13:09 schrieb Christian König: > Am 29.09.20 um 11:14 schrieb Daniel Vetter: >> On Mon, Sep 28, 2020 at 01:22:13PM +0200, Christian König wrote: >>> Am 28.09.20 um 09:37 schrieb Thomas Zimmermann: Hi Am 28.09.20 um 08:50 schrieb Christian König: > Am 27.09.20 um

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-29 Thread Christian König
Am 29.09.20 um 11:14 schrieb Daniel Vetter: On Mon, Sep 28, 2020 at 01:22:13PM +0200, Christian König wrote: Am 28.09.20 um 09:37 schrieb Thomas Zimmermann: Hi Am 28.09.20 um 08:50 schrieb Christian König: Am 27.09.20 um 21:16 schrieb Sam Ravnborg: Hi Thomas. struct simap {    

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-29 Thread Daniel Vetter
On Mon, Sep 28, 2020 at 01:22:13PM +0200, Christian König wrote: > Am 28.09.20 um 09:37 schrieb Thomas Zimmermann: > > Hi > > > > Am 28.09.20 um 08:50 schrieb Christian König: > > > Am 27.09.20 um 21:16 schrieb Sam Ravnborg: > > > > Hi Thomas. > > > > > > > > > > struct simap { > > > > > >  

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-28 Thread Christian König
Am 28.09.20 um 09:37 schrieb Thomas Zimmermann: Hi Am 28.09.20 um 08:50 schrieb Christian König: Am 27.09.20 um 21:16 schrieb Sam Ravnborg: Hi Thomas. struct simap {     union {     void __iomem *vaddr_iomem;     void *vaddr;     };     bool

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-28 Thread Thomas Zimmermann
Hi Am 28.09.20 um 08:50 schrieb Christian König: > Am 27.09.20 um 21:16 schrieb Sam Ravnborg: >> Hi Thomas. >> struct simap {     union {     void __iomem *vaddr_iomem;     void *vaddr;     };     bool is_iomem; };

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-28 Thread Christian König
Am 27.09.20 um 21:16 schrieb Sam Ravnborg: Hi Thomas. struct simap { union { void __iomem *vaddr_iomem; void *vaddr; }; bool is_iomem; }; Where simap is a shorthand for system_iomem_map And it could al be stuffed into a

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-27 Thread Sam Ravnborg
Hi Thomas. > > > > struct simap { > >union { > >void __iomem *vaddr_iomem; > >void *vaddr; > >}; > >bool is_iomem; > > }; > > > > Where simap is a shorthand for system_iomem_map > > And it could al be stuffed into a include/linux/simap.h

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-27 Thread Thomas Zimmermann
Hi Sam Am 26.09.20 um 09:13 schrieb Sam Ravnborg: > Hi Thomas. > > Sorry for chiming in late here, have been offline for a while. > > On Fri, Sep 25, 2020 at 01:55:57PM +0200, Thomas Zimmermann wrote: >> Dma-buf provides vmap() and vunmap() for retriving and releasing mappings >> of dma-buf

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-26 Thread Sam Ravnborg
Hi Thomas. Sorry for chiming in late here, have been offline for a while. On Fri, Sep 25, 2020 at 01:55:57PM +0200, Thomas Zimmermann wrote: > Dma-buf provides vmap() and vunmap() for retriving and releasing mappings > of dma-buf memory in kernel address space. The functions operate with plain >

Re: [Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-25 Thread Tomasz Figa
Hi Thomas, On Fri, Sep 25, 2020 at 01:55:57PM +0200, Thomas Zimmermann wrote: > Dma-buf provides vmap() and vunmap() for retriving and releasing mappings > of dma-buf memory in kernel address space. The functions operate with plain > addresses and the assumption is that the memory can be accessed

[Intel-gfx] [PATCH v3 0/4] dma-buf: Flag vmap'ed memory as system or I/O memory

2020-09-25 Thread Thomas Zimmermann
Dma-buf provides vmap() and vunmap() for retriving and releasing mappings of dma-buf memory in kernel address space. The functions operate with plain addresses and the assumption is that the memory can be accessed with load and store operations. This is not the case on some architectures (e.g.,