> collected timestamps from the Pentium II tick counter. This has the CPU > clock frequency resolution (400 MHz in our case). Our systems use > NetGear FA310TX cards (some with DEC chips, most with Lite-On chips) and > the latest testing version 0.90Q of tulip.c driver. Ok. When you do the run are you tcpdumping the receiver side to see if there are lost frames ? > Finally, I found it rather curious that a stalled MPI code would > sometimes resume running if we sent a single "ping" to all hosts. Thats quite interesting, although it may be simply that it kicked in timewise while you were doing the ping. That can be hard to measure quantitively. Alan - To unsubscribe from this list: send the line "unsubscribe linux-net" in the body of a message to [EMAIL PROTECTED]
- Linux 2.2.2 TCP delays every 41st small packet by 10-20 m... Josip Loncaric
- Re: Linux 2.2.2 TCP delays every 41st small packet b... Alan Cox
- Re: Linux 2.2.2 TCP delays every 41st small packet b... Ben Gertzfield
- Re: Linux 2.2.2 TCP delays every 41st small pack... Donald Becker
- Re: Linux 2.2.2 TCP delays every 41st small packet b... Andrea Arcangeli
- Re: Linux 2.2.2 TCP delays every 41st small pack... David Miller
- Re: Linux 2.2.2 TCP delays every 41st small ... Rogier Wolff
- Re: Linux 2.2.2 TCP delays every 41st sm... Josip Loncaric
- Re: Linux 2.2.2 TCP delays every 41st small pack... Josip Loncaric
- Re: Linux 2.2.2 TCP delays every 41st small ... Donald Becker
- Re: Linux 2.2.2 TCP delays every 41st sm... Josip Loncaric
- Re: Linux 2.2.2 TCP delays every 41st sm... Josip Loncaric
