Re: [PATCH rdma-next 0/3] Dump and fill MKEY

2018-07-04 Thread Jason Gunthorpe
On Wed, Jul 04, 2018 at 09:54:59PM +0300, Leon Romanovsky wrote: > On Wed, Jul 04, 2018 at 11:47:39AM -0600, Jason Gunthorpe wrote: > > On Tue, Jun 19, 2018 at 08:47:21AM +0300, Leon Romanovsky wrote: > > > From: Leon Romanovsky > > > > > > MLX5 IB HCA offers the memory key, dump_fill_mkey to

Re: [PATCH rdma-next 0/3] Dump and fill MKEY

2018-07-04 Thread Leon Romanovsky
On Wed, Jul 04, 2018 at 11:47:39AM -0600, Jason Gunthorpe wrote: > On Tue, Jun 19, 2018 at 08:47:21AM +0300, Leon Romanovsky wrote: > > From: Leon Romanovsky > > > > MLX5 IB HCA offers the memory key, dump_fill_mkey to increase > > performance, when used in a send or receive operations. > > > >

Re: [PATCH rdma-next 0/3] Dump and fill MKEY

2018-07-04 Thread Jason Gunthorpe
On Tue, Jun 19, 2018 at 08:47:21AM +0300, Leon Romanovsky wrote: > From: Leon Romanovsky > > MLX5 IB HCA offers the memory key, dump_fill_mkey to increase > performance, when used in a send or receive operations. > > It is used to force local HCA operations to skip the PCI bus access, > while

Re: [PATCH rdma-next 0/3] Dump and fill MKEY

2018-06-19 Thread Leon Romanovsky
On Tue, Jun 19, 2018 at 08:21:06AM -0600, Jason Gunthorpe wrote: > On Tue, Jun 19, 2018 at 08:47:21AM +0300, Leon Romanovsky wrote: > > From: Leon Romanovsky > > > > MLX5 IB HCA offers the memory key, dump_fill_mkey to increase > > performance, when used in a send or receive operations. > > > >

Re: [PATCH rdma-next 0/3] Dump and fill MKEY

2018-06-19 Thread Jason Gunthorpe
On Tue, Jun 19, 2018 at 08:47:21AM +0300, Leon Romanovsky wrote: > From: Leon Romanovsky > > MLX5 IB HCA offers the memory key, dump_fill_mkey to increase > performance, when used in a send or receive operations. > > It is used to force local HCA operations to skip the PCI bus access, > while

[PATCH rdma-next 0/3] Dump and fill MKEY

2018-06-18 Thread Leon Romanovsky
From: Leon Romanovsky MLX5 IB HCA offers the memory key, dump_fill_mkey to increase performance, when used in a send or receive operations. It is used to force local HCA operations to skip the PCI bus access, while keeping track of the processed length in the ibv_sge handling. In this three