Re: [PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-28 Thread Moritz Fischer
Hi Jason, On Fri, Jan 27, 2017 at 10:58 PM, Jason Gunthorpe wrote: > On Mon, Jan 09, 2017 at 04:13:38PM -0600, Alan Tull wrote: >> On Mon, Jan 9, 2017 at 10:12 AM, Jason Gunthorpe >> wrote: >> > On Mon, Jan 09, 2017 at 10:04:36AM -0600, Alan Tull wrote: >> > >> >> > diff --git a/drivers/fpga/fpg

Re: [PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-27 Thread Jason Gunthorpe
On Mon, Jan 09, 2017 at 04:13:38PM -0600, Alan Tull wrote: > On Mon, Jan 9, 2017 at 10:12 AM, Jason Gunthorpe > wrote: > > On Mon, Jan 09, 2017 at 10:04:36AM -0600, Alan Tull wrote: > > > >> > diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c > >> > index f0a69d3e60a584..30f9778d0632d

Re: [PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-09 Thread Alan Tull
On Mon, Jan 9, 2017 at 10:12 AM, Jason Gunthorpe wrote: > On Mon, Jan 09, 2017 at 10:04:36AM -0600, Alan Tull wrote: > >> > diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c >> > index f0a69d3e60a584..30f9778d0632d2 100644 >> > +++ b/drivers/fpga/fpga-mgr.c >> > @@ -1,4 +1,4 @@ >> > -

Re: [PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-09 Thread Jason Gunthorpe
On Mon, Jan 09, 2017 at 10:04:36AM -0600, Alan Tull wrote: > > diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c > > index f0a69d3e60a584..30f9778d0632d2 100644 > > +++ b/drivers/fpga/fpga-mgr.c > > @@ -1,4 +1,4 @@ > > -/* > > + /* > > Hi Jason, > > Need to take these added 2 space

Re: [PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-09 Thread Alan Tull
On Fri, 6 Jan 2017, Jason Gunthorpe wrote: > Requiring contiguous kernel memory is not a good idea, this is a limited > resource and allocation can fail under normal work loads. > > This introduces a .write_sg op that supporting drivers can provide > to DMA directly from dis-contiguous memory and

[PATCH v3 fpga 3/4] fpga: Add scatterlist based programming

2017-01-06 Thread Jason Gunthorpe
Requiring contiguous kernel memory is not a good idea, this is a limited resource and allocation can fail under normal work loads. This introduces a .write_sg op that supporting drivers can provide to DMA directly from dis-contiguous memory and a new entry point fpga_mgr_buf_load_sg that users can