[PATCH] S2io: Updated documentation
Hi, This patch provides updated documentation on the Neterion(S2io) driver. Please review the patch. Signed-off-by: Ravinandan Arakali <ravinandan.arakali@neterion.com> Signed-off-by: Jeff Garzik <jgarzik@pobox.com>hifive-unleashed-5.1
Ravinandan Arakali
Jeff Garzik
parent
da6971d8ec
commit
571de88b87
|
|
@ -1,48 +1,153 @@
|
||||||
S2IO Technologies XFrame 10 Gig adapter.
|
Release notes for Neterion's (Formerly S2io) Xframe I/II PCI-X 10GbE driver.
|
||||||
-------------------------------------------
|
|
||||||
|
|
||||||
I. Module loadable parameters.
|
Contents
|
||||||
When loaded as a module, the driver provides a host of Module loadable
|
=======
|
||||||
parameters, so the device can be tuned as per the users needs.
|
- 1. Introduction
|
||||||
A list of the Module params is given below.
|
- 2. Identifying the adapter/interface
|
||||||
(i) ring_num: This can be used to program the number of
|
- 3. Features supported
|
||||||
receive rings used in the driver.
|
- 4. Command line parameters
|
||||||
(ii) ring_len: This defines the number of descriptors each ring
|
- 5. Performance suggestions
|
||||||
can have. There can be a maximum of 8 rings.
|
- 6. Available Downloads
|
||||||
(iii) frame_len: This is an array of size 8. Using this we can
|
|
||||||
set the maximum size of the received frame that can
|
|
||||||
be steered into the corrsponding receive ring.
|
|
||||||
(iv) fifo_num: This defines the number of Tx FIFOs thats used in
|
|
||||||
the driver.
|
|
||||||
(v) fifo_len: Each element defines the number of
|
|
||||||
Tx descriptors that can be associated with each
|
|
||||||
corresponding FIFO. There are a maximum of 8 FIFOs.
|
|
||||||
(vi) tx_prio: This is a bool, if module is loaded with a non-zero
|
|
||||||
value for tx_prio multi FIFO scheme is activated.
|
|
||||||
(vii) rx_prio: This is a bool, if module is loaded with a non-zero
|
|
||||||
value for tx_prio multi RING scheme is activated.
|
|
||||||
(viii) latency_timer: The value given against this param will be
|
|
||||||
loaded into the latency timer register in PCI Config
|
|
||||||
space, else the register is left with its reset value.
|
|
||||||
|
|
||||||
II. Performance tuning.
|
|
||||||
By changing a few sysctl parameters.
|
1. Introduction:
|
||||||
Copy the following lines into a file and run the following command,
|
This Linux driver supports Neterion's Xframe I PCI-X 1.0 and
|
||||||
"sysctl -p <file_name>"
|
Xframe II PCI-X 2.0 adapters. It supports several features
|
||||||
### IPV4 specific settings
|
such as jumbo frames, MSI/MSI-X, checksum offloads, TSO, UFO and so on.
|
||||||
net.ipv4.tcp_timestamps = 0 # turns TCP timestamp support off, default 1, reduces CPU use
|
See below for complete list of features.
|
||||||
net.ipv4.tcp_sack = 0 # turn SACK support off, default on
|
All features are supported for both IPv4 and IPv6.
|
||||||
# on systems with a VERY fast bus -> memory interface this is the big gainer
|
|
||||||
net.ipv4.tcp_rmem = 10000000 10000000 10000000 # sets min/default/max TCP read buffer, default 4096 87380 174760
|
2. Identifying the adapter/interface:
|
||||||
net.ipv4.tcp_wmem = 10000000 10000000 10000000 # sets min/pressure/max TCP write buffer, default 4096 16384 131072
|
a. Insert the adapter(s) in your system.
|
||||||
net.ipv4.tcp_mem = 10000000 10000000 10000000 # sets min/pressure/max TCP buffer space, default 31744 32256 32768
|
b. Build and load driver
|
||||||
|
# insmod s2io.ko
|
||||||
### CORE settings (mostly for socket and UDP effect)
|
c. View log messages
|
||||||
net.core.rmem_max = 524287 # maximum receive socket buffer size, default 131071
|
# dmesg | tail -40
|
||||||
net.core.wmem_max = 524287 # maximum send socket buffer size, default 131071
|
You will see messages similar to:
|
||||||
net.core.rmem_default = 524287 # default receive socket buffer size, default 65535
|
eth3: Neterion Xframe I 10GbE adapter (rev 3), Version 2.0.9.1, Intr type INTA
|
||||||
net.core.wmem_default = 524287 # default send socket buffer size, default 65535
|
eth4: Neterion Xframe II 10GbE adapter (rev 2), Version 2.0.9.1, Intr type INTA
|
||||||
net.core.optmem_max = 524287 # maximum amount of option memory buffers, default 10240
|
eth4: Device is on 64 bit 133MHz PCIX(M1) bus
|
||||||
net.core.netdev_max_backlog = 300000 # number of unprocessed input packets before kernel starts dropping them, default 300
|
|
||||||
---End of performance tuning file---
|
The above messages identify the adapter type(Xframe I/II), adapter revision,
|
||||||
|
driver version, interface name(eth3, eth4), Interrupt type(INTA, MSI, MSI-X).
|
||||||
|
In case of Xframe II, the PCI/PCI-X bus width and frequency are displayed
|
||||||
|
as well.
|
||||||
|
|
||||||
|
To associate an interface with a physical adapter use "ethtool -p <ethX>".
|
||||||
|
The corresponding adapter's LED will blink multiple times.
|
||||||
|
|
||||||
|
3. Features supported:
|
||||||
|
a. Jumbo frames. Xframe I/II supports MTU upto 9600 bytes,
|
||||||
|
modifiable using ifconfig command.
|
||||||
|
|
||||||
|
b. Offloads. Supports checksum offload(TCP/UDP/IP) on transmit
|
||||||
|
and receive, TSO.
|
||||||
|
|
||||||
|
c. Multi-buffer receive mode. Scattering of packet across multiple
|
||||||
|
buffers. Currently driver supports 2-buffer mode which yields
|
||||||
|
significant performance improvement on certain platforms(SGI Altix,
|
||||||
|
IBM xSeries).
|
||||||
|
|
||||||
|
d. MSI/MSI-X. Can be enabled on platforms which support this feature
|
||||||
|
(IA64, Xeon) resulting in noticeable performance improvement(upto 7%
|
||||||
|
on certain platforms).
|
||||||
|
|
||||||
|
e. NAPI. Compile-time option(CONFIG_S2IO_NAPI) for better Rx interrupt
|
||||||
|
moderation.
|
||||||
|
|
||||||
|
f. Statistics. Comprehensive MAC-level and software statistics displayed
|
||||||
|
using "ethtool -S" option.
|
||||||
|
|
||||||
|
g. Multi-FIFO/Ring. Supports up to 8 transmit queues and receive rings,
|
||||||
|
with multiple steering options.
|
||||||
|
|
||||||
|
4. Command line parameters
|
||||||
|
a. tx_fifo_num
|
||||||
|
Number of transmit queues
|
||||||
|
Valid range: 1-8
|
||||||
|
Default: 1
|
||||||
|
|
||||||
|
b. rx_ring_num
|
||||||
|
Number of receive rings
|
||||||
|
Valid range: 1-8
|
||||||
|
Default: 1
|
||||||
|
|
||||||
|
c. tx_fifo_len
|
||||||
|
Size of each transmit queue
|
||||||
|
Valid range: Total length of all queues should not exceed 8192
|
||||||
|
Default: 4096
|
||||||
|
|
||||||
|
d. rx_ring_sz
|
||||||
|
Size of each receive ring(in 4K blocks)
|
||||||
|
Valid range: Limited by memory on system
|
||||||
|
Default: 30
|
||||||
|
|
||||||
|
e. intr_type
|
||||||
|
Specifies interrupt type. Possible values 1(INTA), 2(MSI), 3(MSI-X)
|
||||||
|
Valid range: 1-3
|
||||||
|
Default: 1
|
||||||
|
|
||||||
|
5. Performance suggestions
|
||||||
|
General:
|
||||||
|
a. Set MTU to maximum(9000 for switch setup, 9600 in back-to-back configuration)
|
||||||
|
b. Set TCP windows size to optimal value.
|
||||||
|
For instance, for MTU=1500 a value of 210K has been observed to result in
|
||||||
|
good performance.
|
||||||
|
# sysctl -w net.ipv4.tcp_rmem="210000 210000 210000"
|
||||||
|
# sysctl -w net.ipv4.tcp_wmem="210000 210000 210000"
|
||||||
|
For MTU=9000, TCP window size of 10 MB is recommended.
|
||||||
|
# sysctl -w net.ipv4.tcp_rmem="10000000 10000000 10000000"
|
||||||
|
# sysctl -w net.ipv4.tcp_wmem="10000000 10000000 10000000"
|
||||||
|
|
||||||
|
Transmit performance:
|
||||||
|
a. By default, the driver respects BIOS settings for PCI bus parameters.
|
||||||
|
However, you may want to experiment with PCI bus parameters
|
||||||
|
max-split-transactions(MOST) and MMRBC (use setpci command).
|
||||||
|
A MOST value of 2 has been found optimal for Opterons and 3 for Itanium.
|
||||||
|
It could be different for your hardware.
|
||||||
|
Set MMRBC to 4K**.
|
||||||
|
|
||||||
|
For example you can set
|
||||||
|
For opteron
|
||||||
|
#setpci -d 17d5:* 62=1d
|
||||||
|
For Itanium
|
||||||
|
#setpci -d 17d5:* 62=3d
|
||||||
|
|
||||||
|
For detailed description of the PCI registers, please see Xframe User Guide.
|
||||||
|
|
||||||
|
b. Ensure Transmit Checksum offload is enabled. Use ethtool to set/verify this
|
||||||
|
parameter.
|
||||||
|
c. Turn on TSO(using "ethtool -K")
|
||||||
|
# ethtool -K <ethX> tso on
|
||||||
|
|
||||||
|
Receive performance:
|
||||||
|
a. By default, the driver respects BIOS settings for PCI bus parameters.
|
||||||
|
However, you may want to set PCI latency timer to 248.
|
||||||
|
#setpci -d 17d5:* LATENCY_TIMER=f8
|
||||||
|
For detailed description of the PCI registers, please see Xframe User Guide.
|
||||||
|
b. Use 2-buffer mode. This results in large performance boost on
|
||||||
|
on certain platforms(eg. SGI Altix, IBM xSeries).
|
||||||
|
c. Ensure Receive Checksum offload is enabled. Use "ethtool -K ethX" command to
|
||||||
|
set/verify this option.
|
||||||
|
d. Enable NAPI feature(in kernel configuration Device Drivers ---> Network
|
||||||
|
device support ---> Ethernet (10000 Mbit) ---> S2IO 10Gbe Xframe NIC) to
|
||||||
|
bring down CPU utilization.
|
||||||
|
|
||||||
|
** For AMD opteron platforms with 8131 chipset, MMRBC=1 and MOST=1 are
|
||||||
|
recommended as safe parameters.
|
||||||
|
For more information, please review the AMD8131 errata at
|
||||||
|
http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/26310.pdf
|
||||||
|
|
||||||
|
6. Available Downloads
|
||||||
|
Neterion "s2io" driver in Red Hat and Suse 2.6-based distributions is kept up
|
||||||
|
to date, also the latest "s2io" code (including support for 2.4 kernels) is
|
||||||
|
available via "Support" link on the Neterion site: http://www.neterion.com.
|
||||||
|
|
||||||
|
For Xframe User Guide (Programming manual), visit ftp site ns1.s2io.com,
|
||||||
|
user: linuxdocs password: HALdocs
|
||||||
|
|
||||||
|
7. Support
|
||||||
|
For further support please contact either your 10GbE Xframe NIC vendor (IBM,
|
||||||
|
HP, SGI etc.) or click on the "Support" link on the Neterion site:
|
||||||
|
http://www.neterion.com.
|
||||||
|
|
||||||
|
|
|
||||||
Loading…
Reference in New Issue