thr3ads.net - freebsd stable - Heads UP - MFC for em coming shortly [Oct 2007]

If this information is useful, please help other people find it:
Share via:

Jack Vogel

2007-Oct-05 13:28 UTC

Heads UP - MFC for em coming shortly

I am preparing to update the em driver to the equivalent of my
6.6.6 driver. Just doing some last minute sanity checking, I
hope to the checkin before end of day.

This will provide support for devices up to our latest, the
next stage I'm planning will incorporate multiqueue and
multiple locks that are already in my Oplin driver (ixgbe)
but this will take a bit longer.

This driver has the support for TSO in it but its not on
by default, as soon as Andre is able to get that MFC'd
it can be enabled via a compile option.

Jack

Artem Kuchin

2007-Oct-05 13:43 UTC

head link

Heads UP - MFC for em coming shortly

Jack Vogel wrote:> I am preparing to update the em driver to the equivalent of my
> 6.6.6 driver. Just doing some last minute sanity checking, I
> hope to the checkin before end of day.
> 
> This will provide support for devices up to our latest, the
> next stage I'm planning will incorporate multiqueue and
> multiple locks that are already in my Oplin driver (ixgbe)
> but this will take a bit longer.
> 
> This driver has the support for TSO in it but its not on
> by default, as soon as Andre is able to get that MFC'd
> it can be enabled via a compile option.
Hello, Jack!

Remember, i wrote about problems with TSO (dropping
send perfomace). I just posted news about it. Switching
to SCHED_ULE solved to problem or.. i could be switching
from HZ 1000 to HZ 100. I need to check into it.
I hope it helps in some way.

--
Regards,
Artem

Eugene Grosbein

2007-Oct-06 12:02 UTC

head link

Heads UP - MFC for em coming shortly

On Fri, Oct 05, 2007 at 01:28:20PM -0700, Jack Vogel wrote:
> I am preparing to update the em driver to the equivalent of my
> 6.6.6 driver. Just doing some last minute sanity checking, I
> hope to the checkin before end of day.
The MFC broke compilation of kernel containing both device em
and options DEVICE_POLLING:

/home/obj/src/sys/dev/em/if_em.c: In function `em_allocate_intr':
/home/obj/src/sys/dev/em/if_em.c:2647: warning: passing arg 6 of
`bus_setup_intr' from incompatible pointer type
/home/obj/src/sys/dev/em/if_em.c:2647: error: too many arguments to function
`bus_setup_intr'

Eugene Grosbein

Eugene Grosbein

2007-Oct-07 07:04 UTC

head link

Heads UP - MFC for em coming shortly

On Fri, Oct 05, 2007 at 01:28:20PM -0700, Jack Vogel wrote:
> I am preparing to update the em driver to the equivalent of my
> 6.6.6 driver. Just doing some last minute sanity checking, I
> hope to the checkin before end of day.
Latest fix works for me (c)  :-)
Thank you, Jack, for your patience and cooperation.

Eugene Grosbein
_______________________________________________
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to
"freebsd-stable-unsubscribe@freebsd.org"

Mike Tancsa

2007-Oct-08 13:09 UTC

head link

Heads UP - MFC for em coming shortly

At 04:28 PM 10/5/2007, Jack Vogel wrote:>I am preparing to update the em driver to the equivalent of my
>6.6.6 driver. Just doing some last minute sanity checking, I
>hope to the checkin before end of day.
Hi,
         thanks for fixing the compile issue, but I have another 
possible problem.  Do you know if there were any performance 
regressions with this rev? On one firewall / router, I am seeing 
dropped packets.  The PPS rate is only about 30k which should not be that much


Oct  8 02:15:02 c1 kernel: em2: Excessive collisions = 0
Oct  8 02:15:02 c1 kernel: em2: Sequence errors = 0
Oct  8 02:15:02 c1 kernel: em2: Defer count = 0
Oct  8 02:15:02 c1 kernel: em2: Missed Packets = 2068
Oct  8 02:15:02 c1 kernel: em2: Receive No Buffers = 280
Oct  8 02:15:02 c1 kernel: em2: Receive Length Errors = 0
Oct  8 02:15:02 c1 kernel: em2: Receive errors = 3
Oct  8 02:15:02 c1 kernel: em2: Crc errors = 3
Oct  8 02:15:02 c1 kernel: em2: Alignment errors = 0
Oct  8 02:15:02 c1 kernel: em2: Collision/Carrier extension errors = 0
Oct  8 02:15:02 c1 kernel: em2: RX overruns = 17
Oct  8 02:15:02 c1 kernel: em2: watchdog timeouts = 0
Oct  8 02:15:02 c1 kernel: em2: XON Rcvd = 0
Oct  8 02:15:02 c1 kernel: em2: XON Xmtd = 0
Oct  8 02:15:02 c1 kernel: em2: XOFF Rcvd = 0
Oct  8 02:15:02 c1 kernel: em2: XOFF Xmtd = 0
Oct  8 02:15:02 c1 kernel: em2: Good Packets Rcvd = 146043044
Oct  8 02:15:02 c1 kernel: em2: Good Packets Xmtd = 75481090
Oct  8 02:15:02 c1 kernel: em2: TSO Contexts Xmtd = 0
Oct  8 02:15:02 c1 kernel: em2: TSO Contexts Failed = 0


% vmstat -i
interrupt                          total       rate
irq1: atkbd0                           5          0
irq4: sio0                         10094          0
irq16: em0                     291014020       5831
irq17: em1 em2 em3             298574806       5983
irq19: atapci1                     53969          1
cpu0: timer                     99802510       1999
cpu1: timer                     99791176       1999
Total                          789246580      15815


Load avg is quite low, although top doesnt show any interrupt 
activity for some reason.

The settings are all default

% sysctl -a dev.em.2
dev.em.2.%desc: Intel(R) PRO/1000 Network Connection Version - 6.6.6
dev.em.2.%driver: em
dev.em.2.%location: slot=0 function=0
dev.em.2.%pnpinfo: vendor=0x8086 device=0x108c subvendor=0x8086 
subdevice=0x348f class=0x020000
dev.em.2.%parent: pci3
dev.em.2.debug_info: -1
dev.em.2.stats: -1
dev.em.2.rx_int_delay: 0
dev.em.2.tx_int_delay: 66
dev.em.2.rx_abs_int_delay: 66
dev.em.2.tx_abs_int_delay: 66
dev.em.2.rx_processing_limit: 100


em0: <Intel(R) PRO/1000 Network Connection Version - 6.6.6> port 
0x3020-0x303f mem 0x88260000-0x8827ffff,0x88240000-0x8825ffff
irq 16 at device 0.0 on pci2
em0: Ethernet address: 00:15:17:0b:70:98
em0: [FAST]
em1: <Intel(R) PRO/1000 Network Connection Version - 6.6.6> port 
0x3000-0x301f mem 0x88220000-0x8823ffff,0x88200000-0x8821ffff
irq 17 at device 0.1 on pci2
em1: Ethernet address: 00:15:17:0b:70:99
em1: [FAST]
pcib3: <ACPI PCI-PCI bridge> at device 28.5 on pci0
pci3: <ACPI PCI bus> on pcib3
em2: <Intel(R) PRO/1000 Network Connection Version - 6.6.6> port 
0x2000-0x201f mem 0x88180000-0x8819ffff,0x88100000-0x8817ffff irq 17 
at device 0.0 on pci3
em2: Ethernet address: 00:15:17:1e:94:0b
em2: [FAST]
pci3: <simple comms, UART> at device 0.3 (no driver attached)
pci3: <serial bus> at device 0.4 (no driver attached)
pcib4: <ACPI PCI-PCI bridge> at device 30.0 on pci0
pci4: <ACPI PCI bus> on pcib4
pci4: <display, VGA> at device 4.0 (no driver attached)
em3: <Intel(R) PRO/1000 Network Connection Version - 6.6.6> port 
0x1100-0x113f mem 0x88020000-0x8803ffff,0x88000000-0x8801ffff irq 17 
at device 5.0 on pci4
em3: Ethernet address: 00:15:17:1e:94:0c
em3: [FAST]

em0@pci2:0:0:   class=0x020000 card=0x115e8086 chip=0x105e8086 
rev=0x06 hdr=0x00
     vendor     = 'Intel Corporation'
     device     = 'PRO/1000 PT'
     class      = network
     subclass   = ethernet
     cap 01[c8] = powerspec 2  supports D0 D3  current D0
     cap 05[d0] = MSI supports 1 message, 64 bit
     cap 10[e0] = PCI-Express 1 endpoint
em1@pci2:0:1:   class=0x020000 card=0x115e8086 chip=0x105e8086 
rev=0x06 hdr=0x00
     vendor     = 'Intel Corporation'
     device     = 'PRO/1000 PT'
     class      = network
     subclass   = ethernet
     cap 01[c8] = powerspec 2  supports D0 D3  current D0
     cap 05[d0] = MSI supports 1 message, 64 bit
     cap 10[e0] = PCI-Express 1 endpoint
em2@pci3:0:0:   class=0x020000 card=0x348f8086 chip=0x108c8086 
rev=0x03 hdr=0x00
     vendor     = 'Intel Corporation'
     device     = 'PRO/1000 PM'
     class      = network
     subclass   = ethernet
     cap 01[c8] = powerspec 2  supports D0 D3  current D0
     cap 05[d0] = MSI supports 1 message, 64 bit
     cap 10[e0] = PCI-Express 1 endpoint
em3@pci4:5:0:   class=0x020000 card=0x348f8086 chip=0x10768086 
rev=0x05 hdr=0x00
     vendor     = 'Intel Corporation'
     device     = '82547EI Gigabit Ethernet Controller'
     class      = network
     subclass   = ethernet
     cap 01[dc] = powerspec 2  supports D0 D3  current D0
     cap 07[e4] = PCI-X supports 2048 burst read, 1 split transaction


         ---Mike
>This will provide support for devices up to our latest, the
>next stage I'm planning will incorporate multiqueue and
>multiple locks that are already in my Oplin driver (ixgbe)
>but this will take a bit longer.
>
>This driver has the support for TSO in it but its not on
>by default, as soon as Andre is able to get that MFC'd
>it can be enabled via a compile option.
>
>Jack
>_______________________________________________
>freebsd-stable@freebsd.org mailing list
>http://lists.freebsd.org/mailman/listinfo/freebsd-stable
>To unsubscribe, send any mail to
"freebsd-stable-unsubscribe@freebsd.org"

freebsd stable - Oct 2007 - Heads UP - MFC for em coming shortly

Heads UP - MFC for em coming shortly

Heads UP - MFC for em coming shortly

Heads UP - MFC for em coming shortly

Heads UP - MFC for em coming shortly

Heads UP - MFC for em coming shortly