RE: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations

2010-08-10 Thread Walukiewicz, Miroslaw
: linux-rdma@vger.kernel.org Subject: Re: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations The proposed path optimization is removing of dynamic allocations by redefining a structure definition passed to kernel. To struct ibv_post_send { __u32 command

RE: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations

2010-08-10 Thread Walukiewicz, Miroslaw
...@vger.kernel.org [mailto:linux-rdma-ow...@vger.kernel.org] On Behalf Of Jason Gunthorpe Sent: Friday, August 06, 2010 6:33 PM To: Walukiewicz, Miroslaw Cc: Roland Dreier; linux-rdma@vger.kernel.org Subject: Re: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations On Fri, Aug 06, 2010

Re: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations

2010-08-06 Thread Roland Dreier
The proposed path optimization is removing of dynamic allocations by redefining a structure definition passed to kernel. To struct ibv_post_send { __u32 command; __u16 in_words; __u16 out_words; __u64 response; __u32 qp_handle;

Re: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations

2010-08-06 Thread Jason Gunthorpe
On Fri, Aug 06, 2010 at 11:03:36AM +0100, Walukiewicz, Miroslaw wrote: Currently the transmit/receive path works following way: User calls ibv_post_send() where vendor specific function is called. When the path should go through kernel the ibv_cmd_post_send() is called. The function creates

Re: {RFC] ibv_post_send()/ibv_post_recv() kernel path optimizations

2010-08-06 Thread Ralph Campbell
On Fri, 2010-08-06 at 03:03 -0700, Walukiewicz, Miroslaw wrote: Currently the ibv_post_send()/ibv_post_recv() path through kernel (using /dev/infiniband/rdmacm) could be optimized by removing dynamic memory allocations on the path. Currently the transmit/receive path works following way: