Date: Wed, 15 Mar 2006 21:57:23 +0100 From: "OxY" <oxy@field.hu> To: <freebsd-net@freebsd.org>, <pyunyh@gmail.com> Subject: Re: problem with Marwell gigabit performance Message-ID: <000401c64873$0f4273d0$0201a8c0@oxy> References: <001101c64754$cbbd65d0$0201a8c0@oxy><20060315001806.GA52826@cdnetworks.co.kr><000f01c64846$c7148840$0201a8c0@oxy> <000a01c64848$41d6f710$0201a8c0@oxy>
next in thread | previous in thread | raw e-mail | index | archive | help
after further testing and transferred for a couple hours i realized that the performance is ok, not changed dramatically (the server was slow when i measured 9mb/s..) but the load! when i used ftp before the load increased to ~2.0, but now it's 4.5! changed back to the original driver and got 2.0 again.. would you guys give me an advice what to buy? 3com or intel? (i heard em(4) driver is not too good..) thx! ----- Original Message ----- From: "OxY" <oxy@field.hu> To: <freebsd-net@freebsd.org>; <pyunyh@gmail.com> Sent: Wednesday, March 15, 2006 4:51 PM Subject: Re: problem with Marwell gigabit performance >i forgot to mention the load is twice, than before and the transfer is >speed is > decreased from 20mbyte/s to 9mbyte/s > > ----- Original Message ----- > From: "OxY" <oxy@field.hu> > To: <freebsd-net@freebsd.org> > Sent: Wednesday, March 15, 2006 4:40 PM > Subject: Re: problem with Marwell gigabit performance > > >> with the modified sk(4) driver i got the following: >> >> sk0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500 >> options=2b<RXCSUM,TXCSUM,VLAN_MTU,JUMBO_MTU> >> >> the performance didn't changed, packet drops as usual... >> when the system is 100% idle, packet drop is around 1%, >> when i got ~0.80 load on the other (fxp0) interface, packet drop is 8-11% >> may i change the card to intel or 3com, or what to do? >> >> ----- Original Message ----- >> From: "Pyun YongHyeon" <pyunyh@gmail.com> >> To: "OxY" <oxy@field.hu> >> Cc: <freebsd-net@freebsd.org> >> Sent: Wednesday, March 15, 2006 1:18 AM >> Subject: Re: problem with Marwell gigabit performance >> >> >>> On Tue, Mar 14, 2006 at 11:48:14AM +0100, OxY wrote: >>> > hi! >>> > >>> > i have a Marwell (SMC) gigabit ethernet card (sk0) and have serious >>> > problems with performance.. >>> > the machine is a pc, amd 2000+ xp, 512mb ram. >>> > >>> > tested with iperf (bidirectional test, udp transfer, not stream) >>> > and got 8-15% packet drop when the system was idle. >>> > >>> >>> There is a modified sk(4) driver at >>> http://people.freebsd.org/~yongari/sk/sk_test/if_sk.c >>> http://people.freebsd.org/~yongari/sk/sk_test/if_skreg.h >>> >>> Would you please give it try? >>> I can't sure but you would get better results as the new driver >>> uses one lock for standard MTU size and supports Tx TCP checksum >>> offload and Rx IP checksum offload. >>> I could push 32bit PCI NIC(DGE-530T) to the PCI bus limit on sparc64. >>> >>> > then tuned the sysctl with these settings: >>> > >>> > kern.ipc.maxsockbuf=81920000 >>> > kern.ipc.nmbclusters=5000000000 >>> > kern.ipc.somaxconn=8192 >>> > net.inet.tcp.inflight.enable=1 >>> > net.inet.raw.recvspace=4096 >>> > net.inet.tcp.recvspace=131072 >>> > kern.ipc.shmmax=409600000 >>> > >>> > with these i got around 3-7% packet drop, >>> > but it's very high compared to zero :) >>> > >>> > my question is where/what should i optimize to not >>> > have packet drop at all, or at least reduce it as much as >>> > possible. >>> > >>> > thanks for your help! >>> > >>> > Csaba Banhalmi, HU >>> -- >>> Regards, >>> Pyun YongHyeon >> _______________________________________________ >> freebsd-net@freebsd.org mailing list >> http://lists.freebsd.org/mailman/listinfo/freebsd-net >> To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org" > > _______________________________________________ > freebsd-net@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-net > To unsubscribe, send any mail to "freebsd-net-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?000401c64873$0f4273d0$0201a8c0>