Date: Sat, 6 Nov 2010 23:19:33 -0700 From: Pyun YongHyeon <pyunyh@gmail.com> To: =?ISO-8859-1?Q?Ulrich_Sp=F6rlein?= <uqs@spoerlein.net> Cc: stable@freebsd.org Subject: Re: Abysmal re(4) performance under 8.1-STABLE (mid-August) Message-ID: <AANLkTinqcv0M_CR9uHWtjOeHNHt4QGjhS_wNNOjJinu_@mail.gmail.com> In-Reply-To: <20101106093700.GW85693@acme.spoerlein.net> References: <20101106093700.GW85693@acme.spoerlein.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Nov 6, 2010 at 2:37 AM, Ulrich Sp=F6rlein <uqs@spoerlein.net> wrote= : > Hello Pyun, > > On this new server, I cannot get more than ~280kByte/s up/downstream out = of > re(4) without any tweaking. > > re0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 15= 00 > =A0 =A0 =A0 =A0options=3D389b<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,VLAN_= HWCSUM,WOL_UCAST,WOL_MCAST,WOL_MAGIC> > =A0 =A0 =A0 =A0ether 00:21:85:63:74:34 > =A0 =A0 =A0 =A0inet6 fe80::221:85ff:fe63:7434%re0 prefixlen 64 scopeid 0x= 1 > =A0 =A0 =A0 =A0inet 46.4.12.147 netmask 0xffffffc0 broadcast 46.4.12.191 > =A0 =A0 =A0 =A0nd6 options=3D3<PERFORMNUD,ACCEPT_RTADV> > =A0 =A0 =A0 =A0media: Ethernet autoselect (100baseTX <half-duplex>) > =A0 =A0 =A0 =A0status: active > It seems the link was resolved to half-duplex. Does link partner also agree on the resolved speed/duplex? > re0@pci0:2:0:0: class=3D0x020000 card=3D0x368c1462 chip=3D0x816810ec rev= =3D0x01 hdr=3D0x00 > =A0 =A0vendor =A0 =A0 =3D 'Realtek Semiconductor' > =A0 =A0device =A0 =A0 =3D 'Gigabit Ethernet NIC(NDIS 6.0) (RTL8168/8111/8= 111c)' > =A0 =A0class =A0 =A0 =A0=3D network > =A0 =A0subclass =A0 =3D ethernet > > re0: <RealTek 8168/8111 B/C/CP/D/DP/E PCIe Gigabit Ethernet> port 0xd800-= 0xd8ff mem 0xfeaff000-0xfeafffff irq 19 at device 0.0 on pci2 > re0: Using 1 MSI messages > re0: Chip rev. 0x38000000 > re0: MAC rev. 0x00000000 > miibus0: <MII bus> on re0 > rgephy0: <RTL8169S/8110S/8211B media interface> PHY 1 on miibus0 > rgephy0: =A010baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 10= 00baseT-FDX, auto > re0: Ethernet address: 00:21:85:63:74:34 > re0: [FILTER] > > > It's interesting to note, that re0 only negotiates half-duplex, where > linux will negotiate full-duplex (and gets ~10MByte/s as expected). > > Next, I disabled almost all options, except that I cannot disable > VLAN_MTU, VLAN_HWCSUM. I also forced the adapter into full-duplex. > > # ifconfig re0 -vlanmtu > # ifconfig re0 -vlanhwcsum > ifconfig: -vlanhwcsum: bad value I'm sure this has nothing to do that this issue. If you want to disable checksum offloading of VLAN interface, use vlan interface instead of parent interface of the VLAN interface(i.e. ifconfig vlan0 -txcsum -rxcsum). And you can't disable VLAN_MTU on re(4). There is no reason to disable supporting VLAN oversized frames. > # ifconfig re0 > re0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 15= 00 > =A0 =A0 =A0 =A0options=3D88<VLAN_MTU,VLAN_HWCSUM> > =A0 =A0 =A0 =A0ether 00:21:85:63:74:34 > =A0 =A0 =A0 =A0inet6 fe80::221:85ff:fe63:7434%re0 prefixlen 64 scopeid 0x= 1 > =A0 =A0 =A0 =A0inet 46.4.12.147 netmask 0xffffffc0 broadcast 46.4.12.191 > =A0 =A0 =A0 =A0nd6 options=3D3<PERFORMNUD,ACCEPT_RTADV> > =A0 =A0 =A0 =A0media: Ethernet 100baseTX <full-duplex> (100baseTX <half-d= uplex>) > =A0 =A0 =A0 =A0status: active > This time, it seems you used forced media configuration instead of auto. It still shows duplex mismatch so it's normal to see poor performance. What makes me wonder is why you have duplex mismatch? Did you use forced media configuration on link partner? What happens when you use different switch? > If I then immediately start the test-download, I get a ~2MByte/s spike, > which quickly returns to something around 250kByte/s. > > Booting with > hw.pci.enable_msix=3D0 > hw.pci.enable_msi=3D0 > > I can get almost up to 400kByte/s, but this may be coincidence. > > So this is usually as far as it gets: > > re0 =A0in =A0 =A0190.504 KB/s =A0 =A0 =A0 =A0246.136 KB/s =A0 =A0 =A0 =A0= =A0 66.709 MB > =A0 =A0 out =A0 =A010.290 KB/s =A0 =A0 =A0 =A0 12.985 KB/s =A0 =A0 =A0 = =A0 =A0 =A06.076 MB > > But then I ran tcpdump in another session, and it looks like the ssh traf= fic on > the upstream of the interface will make the downloads running in another = window > go faster: > > re0 =A0in =A0 =A0805.961 KB/s =A0 =A0 =A0 =A0 =A01.577 MB/s =A0 =A0 =A0 = =A0 =A0116.523 MB > =A0 =A0 out =A0 222.940 KB/s =A0 =A0 =A0 =A0502.045 KB/s =A0 =A0 =A0 =A0 = =A0 19.267 MB > > Any ideas? > > Uli >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTinqcv0M_CR9uHWtjOeHNHt4QGjhS_wNNOjJinu_>