Re: [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY
From: Bart Van Assche <hidden>
Date: 2016-08-29 17:00:33
Possibly related (same subject, not in this thread)
- 2016-09-05 · Re: [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY · Max Gurtovoy <hidden>
- 2016-09-05 · [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY · Christoph Hellwig <hidden>
- 2016-09-05 · Re: [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY · Christoph Hellwig <hidden>
- 2016-08-31 · Re: [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY · Max Gurtovoy <hidden>
- 2016-08-30 · Re: [PATCH 4/6] IB/srp: use IB_PD_UNSAFE_GLOBAL_RKEY · Sagi Grimberg <hidden>
On 08/29/2016 03:02 AM, Christoph Hellwig wrote:
quoted hunk
diff --git a/drivers/infiniband/ulp/srp/ib_srp.h b/drivers/infiniband/ulp/srp/ib_srp.h index 26bb9b0..446f559 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.h +++ b/drivers/infiniband/ulp/srp/ib_srp.h@@ -90,7 +90,6 @@ struct srp_device { struct list_head dev_list; struct ib_device *dev; struct ib_pd *pd; - struct ib_mr *global_mr; u64 mr_page_mask; int mr_page_size; int mr_max_size;@@ -179,7 +178,6 @@ struct srp_target_port { spinlock_t lock; /* read only in the hot path */ - struct ib_mr *global_mr; struct srp_rdma_ch *ch; u32 ch_count; u32 lkey;
Hello Christoph, Was it really necessary to remove these two member variables? If the per-PD global MR is ever removed all the code that uses pd->unsafe_global_rkey will have to be touched again. Keeping the two global_mr variables would reduce the number of load instructions the CPU has to perform while mapping SRP requests. Thanks, Bart. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html