[openib-general][patch review] srp: fmr implementation,

[Helen Chen]

Hi Vu,

I am very excited to hear that you are archeiving such impressive SRP
performance.  We are evaluating the Engenio SRP and are having trouble 
with the SRP initiator using the srp stack from the openfabric-1.0rc4 
revision.  We are able to discover the SRP LUN's, but can't seem to 
complete mke2fs with the following dmesg.  Dave Ellis claims that you 
had successfully conducted the interoperability test with the Engenip 
storage a while ago, and may be able to lend us a hand in trouble shooting.  
Is this a configuration issue or protocol compatibility problem?  We'd 
really appreciate your taking time out of your busy schedule to help.

Sincerely,
Helen Chen   

........
printk: 97 messages suppressed.
ib_srp: Target has req_lim 0
printk: 147 messages suppressed.
ib_srp: Target has req_lim 0
printk: 104 messages suppressed.
ib_srp: Target has req_lim 0
printk: 103 messages suppressed.
ib_srp: Target has req_lim 0
printk: 119 messages suppressed.
ib_srp: Target has req_lim 0
printk: 146 messages suppressed.
ib_srp: Target has req_lim 0
printk: 65 messages suppressed.
ib_srp: Target has req_lim 0
printk: 90 messages suppressed.
ib_srp: Target has req_lim 0
printk: 139 messages suppressed.
ib_srp: Target has req_lim 0
printk: 116 messages suppressed.
ib_srp: Target has req_lim 0
printk: 133 messages suppressed.
ib_srp: Target has req_lim 0

-----Original Message-----
>From openib-general-bounces at openib.org Wed May 10 08:37:04 2006
>Roland Dreier wrote:
>> BTW, does Mellanox (or anyone else) have any numbers showing that
>> using FMRs makes any difference in performance on a semi-realistic benchmark?
>> 
>
>I'm using xdd to test the performance 
>www.ioperformance.com/products.htm
>
>The target is Mellanox srp target reference implemenation 
>with 14 SATA spindles
>
>I can get ~780 MB/s max without FMRs and ~920 MB/s with FMRs 
>(using 256 KB sequential read direct IO request)
>
>Vu
>_______________________________________________
>openib-general mailing list
>openib-general at openib.org
>http://openib.org/mailman/listinfo/openib-general
>
>tO UNSUBSCribe, please visit http://openib.org/mailman/listinfo/openib-general