Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1345210
| From | "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [net-next][PATCH v2 11/13] RDS: IB: add Fastreg MR (FRMR) detection support |
| Date | 2016-02-28 11:30 +0100 |
| Message-ID | <r76LU-Ry-19@gated-at.bofh.it> (permalink) |
| References | <r6Zhn-3WL-3@gated-at.bofh.it> <r6Zho-3WL-23@gated-at.bofh.it> <r75wt-5M-9@gated-at.bofh.it> |
| Organization | Oracle Corporation |
On 2/28/16 1:08 AM, Christoph Hellwig wrote: > On Sat, Feb 27, 2016 at 06:19:48PM -0800, Santosh Shilimkar wrote: >> Discovere Fast Memmory Registration support using IB device >> IB_DEVICE_MEM_MGT_EXTENSIONS. Certain HCA might support just FRMR >> or FMR or both FMR and FRWR. In case both mr type are supported, >> default FMR is used. >> >> Default MR is still kept as FMR against what everyone else >> is following. Default will be changed to FRMR once the >> RDS performance with FRMR is comparable with FMR. The >> work is in progress for the same. >> >> Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org> >> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> >> --- >> v2: Dropped the module parameter as suggested by David Miller > > This means we only use the safer method if the HCA doesn't support > the other one. All other RDMA ULP that support both methods have > a module_param so the veto from Dave is a bit unfortunate. Anyway, > let's get the code in for now and figure out what to use later. > Indeed. It wasn't really deal breaker for RDS so I agreed to drop it. > Just curious: where / how do you see worse peformance using FRs? > I wouldn't call it worse but its not comparable. Use case is multi-threaded RDS RDMA perf test(s). Hopefully the follow up series which am working on should minimise the gap. I am leaving the details for later, but one of the main issue I saw was contention on driver post_send() lock from send, MR reg and MR invalidation. Regards, Santosh
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[net-next][PATCH v2 00/13] RDS: Major clean-up with couple of new features for 4.6 Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 12/13] RDS: IB: allocate extra space on queues for FRMR support Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 09/13] RDS: IB: handle the RDMA CM time wait event Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 04/13] RDS: IB: Remove the RDS_IB_SEND_OP dependency Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 10/13] RDS: IB: add mr reused stats Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 11/13] RDS: IB: add Fastreg MR (FRMR) detection support Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
Re: [net-next][PATCH v2 11/13] RDS: IB: add Fastreg MR (FRMR) detection support Christoph Hellwig <hch@infradead.org> - 2016-02-28 10:10 +0100
Re: [net-next][PATCH v2 11/13] RDS: IB: add Fastreg MR (FRMR) detection support "santosh.shilimkar@oracle.com" <santosh.shilimkar@oracle.com> - 2016-02-28 11:30 +0100
[net-next][PATCH v2 02/13] RDS: Add support for SO_TIMESTAMP for incoming messages Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 05/13] RDS: IB: Re-organise ibmr code Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 03/13] MAINTAINERS: update RDS entry Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 08/13] RDS: IB: add connection info to ibmr Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
[net-next][PATCH v2 13/13] RDS: IB: Support Fastreg MR (FRMR) memory registration mode Santosh Shilimkar <santosh.shilimkar@oracle.com> - 2016-02-28 03:30 +0100
Re: [net-next][PATCH v2 00/13] RDS: Major clean-up with couple of new features for 4.6 David Miller <davem@davemloft.net> - 2016-03-01 23:40 +0100
Re: [net-next][PATCH v2 00/13] RDS: Major clean-up with couple of new features for 4.6 santosh shilimkar <santosh.shilimkar@oracle.com> - 2016-03-02 00:10 +0100
csiph-web