ib_read_lat question

DISCUSSION Steve Wise 3 participants 4 messages Feb 21, 2013 – Feb 27, 2013 3 messages not in the archive InfiniBand

Share & tools

Steve Wise

author

2013-02-21 17:50 UTC #0 lore raw

Hey Ido,

ib_read_lat displays RTT latency, while the other lat programs display RTT/2. Is this intended? Or is read_lat.c/print_report() missing a divide-by-two?

Thanks,

Steve.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Hefty, Sean

2013-02-21 18:27 UTC #2 lore raw

In reply to a message not in the archive

Quoted 3 lines

ib_read_lat displays RTT latency, while the other lat programs display RTT/2. Is this intended? Or is read_lat.c/print_report() missing a divide-by-two?

The tests should measure how long it takes the data to reach the destination. For writes/sends, that would be RTT/2. For reads, the destination is local, so you should want RTT. --
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Steve Wise

author

2013-02-21 18:34 UTC #4 lore raw

In reply to a message not in the archive

Quoted 4 lines · Sean

Quoted 3 lines
ib_read_lat displays RTT latency, while the other lat programs display RTT/2. Is this intended? Or is read_lat.c/print_report() missing a divide-by-two?
The tests should measure how long it takes the data to reach the destination. For writes/sends, that would be RTT/2. For reads, the destination is local, so you should want RTT.

Guess that makes sense.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html

Ido Shamai

2013-02-27 08:46 UTC #6 lore raw

In reply to a message not in the archive

Quoted 8 lines · Steve Wise

Quoted 4 lines · Sean
Quoted 3 lines
ib_read_lat displays RTT latency, while the other lat programs display RTT/2. Is this intended? Or is read_lat.c/print_report() missing a divide-by-two?
The tests should measure how long it takes the data to reach the destination. For writes/sends, that would be RTT/2. For reads, the destination is local, so you should want RTT.
Guess that makes sense.

Yes, to me too.
Thanks Sean
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html