re(4) problem

[Pyun YongHyeon]

On Tue, Mar 11, 2008 at 11:34:12AM -0700, Abdullah Ibn Hamad Al-Marri wrote:
 > ----- Original Message ----
 > > From: Pyun YongHyeon <pyunyh at gmail.com>
 > > To: Daniel Gerzo <danger at FreeBSD.org>
 > > Cc: current at FreeBSD.org; yongari at FreeBSD.org
 > > Sent: Tuesday, March 11, 2008 5:24:41 AM
 > > Subject: Re: re(4) problem
 > > 
 > > On Mon, Mar 10, 2008 at 09:38:32PM +0100, Daniel Gerzo wrote:
 > >  > Hello Daniel,
 > >  > 
 > >  > Monday, March 10, 2008, 4:26:49 PM, you wrote:
 > >  > 
 > >  > > Hello Pyun,
 > >  > 
 > >  > >> Since you're using MSI, would you try attached patch?
 > >  > >> Sorry, I don't have PCIe based hardware and RealTek's datasheet for
 > >  > >> PCIe variants is too poor to get useful information.
 > >  > >> 
 > >  > >> -- 
 > >  > >> Regards,
 > >  > >> Pyun YongHyeon
 > >  > 
 > >  > > Thank you for your quick reply, I will test your patch ASAP, but I have one
 > >  > > simple question:
 > >  > 
 > >  > > Should I apply this patch on top of
 > >  > > http://people.freebsd.org/~yongari/re/re.HEAD.patch
 > >  > > or what is this patch aginst?
 > >  > 
 > >  > > Thank you.
 > >  > 
 > >  > Or better, please send me (or place somewhere on your home page) the
 > >  > complete if_re.c and if_rlreg.h which should adress my problems, so
 > >  > that I can test it easily.
 > >  > 
 > > 
 > > Done.
 > > http://people.freebsd.org/~yongari/re/if_re.c
 > > http://people.freebsd.org/~yongari/re/if_rlreg.h
 > > The above files would build on CURRENT/RELENG_7/7.0-RELEASE.
 > > 
 > > I'm not sure whether it helps or not.
 > > 
 > >  > Your efforst are greatly appreciated!
 > >  > 
 > >  > -- 
 > >  > Best regards,
 > >  >  Daniel                            mailto:danger at FreeBSD.org
 > >  > 
 > > 
 > > -- 
 > > Regards,
 > > Pyun YongHyeon
 > 
 > Ok, here is the ping to 2 servers using same nic to another server using fxp in same lan.
 > 
 > ping -c 100 -s 5000 after using 
 > http://people.freebsd.org/~yongari/re/if_re.c
 > http://people.freebsd.org/~yongari/re/if_rlreg.h
 > 
 > re0: <RealTek 8168/8111B PCIe Gigabit Ethernet> port 0xc000-0xc0ff mem 0xd0020000-0xd0020fff irq 17 at device 0.0 on pci3
 > re0: Using 2 MSI messages
 > miibus0: <MII bus> on re0
 > rgephy0: <RTL8169S/8110S/8211B media interface> PHY 1 on miibus0
 > rgephy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto
 > re0: Ethernet address: 00:19:d1:a7:a4:72
 > re0: [FILTER]
 > re0: [FILTER]
 > 
 > 100 packets transmitted, 100 packets received, 0.0% packet loss
 > round-trip min/avg/max/stddev = 2.410/2.429/2.587/0.019 ms
 > 
 > Here without.
 > 
 > re0: <RealTek 8168/8111B PCIe Gigabit Ethernet> port 0xd000-0xd0ff mem 0xf2000000-0xf2000fff irq 17 at device 0.0 on pci3
 > re0: Using 2 MSI messages
 > miibus0: <MII bus> on re0
 > rgephy0: <RTL8169S/8110S/8211B media interface> PHY 1 on miibus0
 > rgephy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto
 > re0: Ethernet address: 00:1a:4d:4f:76:07
 > re0: [FILTER]
 > re0: [FILTER]
 > 
 > 100 packets transmitted, 99 packets received, 1.0% packet loss
 > round-trip min/avg/max/stddev = 632.148/633.103/640.812/0.872 ms
 > 
 > So as I could tell the ping latency got fixed by using latest changes.
 > 

Sorry, I don't know what exact issues you had. Just pinging wouldn't
prove anything. Did you encounter random network loss without any
relevant messages in console prior to trying latest patch?
If you know a way that can reliably reproduce re(4) issue would you
try to reproduce it again?

Thanks.
-- 
Regards,
Pyun YongHyeon