[perftests and rdma-core]: Location of upstream perftest repo and inclusion in rdma-core

[perftests and rdma-core]: Location of upstream perftest repo and inclusion in rdma-core

[Stephen Bates]

Hi All

Firstly, a big thanks to Parav for solving my “two RNICs in the same server” loopback woes [1]. The script you provided worked a treat with rdma_cm things like rping and NVMe over Fabrics working very nicely.

One thing I did notice is that for the perftests (e.g. ib_write_bw) there is no way to bind to an IP address when using the rdma_cm option –R. Does anyone know if this is possible with the existing code? If not I’d like to propose a patch for the perftests but I am not sure where the current upstream repo for perftests is? I also wanted to ask if this is something that could be considered for rdma-core as we have other user-space tools there?

A repo for perftests lives here [2] (which Logan forked a while back [3]) but it seems pretty inactive…

Cheers
 
Stephen

[1] https://marc.info/?l=linux-rdma&m=150575871330293&w=2
[2] git://git.openfabrics.org/~grockah/perftest.git
[3] https://github.com/lsgunth/perftest

Re: [perftests and rdma-core]: Location of upstream perftest repo and inclusion in rdma-core

[Leon Romanovsky]

[-- Attachment #1: Type: text/plain, Size: 1154 bytes --]

On Tue, Sep 26, 2017 at 09:01:35PM +0000, Stephen  Bates wrote:
> Hi All
>
> Firstly, a big thanks to Parav for solving my “two RNICs in the same server” loopback woes [1]. The script you provided worked a treat with rdma_cm things like rping and NVMe over Fabrics working very nicely.
>
> One thing I did notice is that for the perftests (e.g. ib_write_bw) there is no way to bind to an IP address when using the rdma_cm option –R. Does anyone know if this is possible with the existing code? If not I’d like to propose a patch for the perftests but I am not sure where the current upstream repo for perftests is? I also wanted to ask if this is something that could be considered for rdma-core as we have other user-space tools there?
>
> A repo for perftests lives here [2] (which Logan forked a while back [3]) but it seems pretty inactive…

Perftest official repo is on the github:
https://github.com/linux-rdma/perftest

Thanks

>
> Cheers
>
> Stephen
>
> [1] https://marc.info/?l=linux-rdma&m=150575871330293&w=2
> [2] git://git.openfabrics.org/~grockah/perftest.git
> [3] https://github.com/lsgunth/perftest
>
>

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [perftests and rdma-core]: Location of upstream perftest repo and inclusion in rdma-core

[Stephen Bates]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 916 bytes --]

>>
>> One thing I did notice is that for the perftests (e.g. ib_write_bw) there is no way to bind to an IP address when using the rdma_cm option –R. Does anyone know if this is possible >>with the existing code? If not I’d like to propose a patch for the perftests but I am not sure where the current upstream repo for perftests is? I also wanted to ask if this is something >> that could be considered for rdma-core as we have other user-space tools there?
>>
>> A repo for perftests lives here [2] (which Logan forked a while back [3]) but it seems pretty inactive…
>
>Perftest official repo is on the github:
>https://github.com/linux-rdma/perftest

Thanks Leon! I will see if a bind option of –R makes sense and submit a patch if I think it does.

Stephen


N‹§²æìr¸›yúèšØb²X¬¶Ç§vØ^–)Þº{.nÇ+‰·¥Š{±­ÙšŠ{ayº\x1dʇڙë,j\a­¢f£¢·hš‹»öì\x17/oSc¾™Ú³9˜uÀ¦æå‰È&jw¨®\x03(­éšŽŠÝ¢j"ú\x1a¶^[m§ÿïêäz¹Þ–Šàþf£¢·hšˆ§~ˆmš