Co-authored-by: Zhi Yuan (Carl) Zhao <zyzhao@arista.com>
Why I did it
Arista 7060 platform has a rare and unreproduceable PCIe timeout that could possibly be solved with increasing the switch PCIe timeout value. To do this we'll call a script for this platform to increase the PCIe timeout on boot-up.
No issues would be expected from the setpci command. From the PCIe spec:
"Software is permitted to change the value in this field at any
time. For Requests already pending when the Completion
Timeout Value is changed, hardware is permitted to use either
the new or the old value for the outstanding Requests, and is
permitted to base the start time for each Request either on when
this value was changed or on when each request was issued. "
How I did it
Add "platform-init" support in swss docker similar to how "hwsku-init" is called, only this would be for any device belonging to a platform. Then the script would reside in device data folder.
Additionally, add pciutils dependency to docker-orchagent so it can run the setpci commands.
How to verify it
On bootup of an Arista 7060, can execute:
lspci -vv -s 01:00.0 | grep -i "devctl2"
In order to check that the timeout has changed.
Why I did it
Improve throughput and latency for 7260 deployments
How I did it
Update the dynamic threshold to 0 and ECN settings as 2mb/10mb/5%
How to verify it
With the new dir structure on 7260, updated the new alpha values and ecn settings in the appropriate files, loaded minigraph and verified that the new settings are applied
Added unit tests for rendering the qos template for 7260. Built sonic config engine wheel successfully
neethajohn added 6 commits 10 days ago
Signed-off-by: Neetha John <nejo@microsoft.com>
Why I did it
This was an ask by Microsoft to provide:
7260 config.bcm file for hardware sku Arista-7260CX3-D92C16 (Named Arista-7260CX3-D96C16).
There are 16 100G uplinks:
Ethernet13-20/1
Ethernet45-52/1
All other ports are breakout to 2 50G ports.
How I did it
Copied existing Arista-7260CX3-D108C8 HWSKU and altered the bcm.config and port_config.ini files.
How to verify it
The new 100G ports do come up with a 201811 image using this HWSKU.
Co-authored-by: Zhi Yuan (Carl) Zhao <zyzhao@arista.com>
Why I did it
7050 S4Q31 mmu configuration is missing ALPM configurations, causing not enough memory reserved for routes. Orchagent crashes on a nightly testbed with 6400 route entries.
How I did it
Add the missing ALPM configurations.
How to verify it
Load the configuration on testbed and verified new configuration exists and no more crash.
Signed-off-by: Ying Xie ying.xie@microsoft.com
This PR contains the following changes
Original Arista-7050-QX-32S sku (32x40G ports) has been renamed to Arista-7050QX32S-Q32
Arista-7050-QX-32S is symlinked to Arista-7050QX-32S-S4Q31 (4x10G, 31x40G ports)
Signed-off-by: Neetha John <nejo@microsoft.com>
Signed-off-by: Neetha John <nejo@microsoft.com>
Why I did it
Need proper MMU and Qos settings for Arista-7050QX-32S-S4Q31
How I did it
Updated the settings based on Arista-7050-QX-32S
Why I did it
PG profile settings need to be aligned with Arista-7050-QX-32S
How I did it
Copy over the current settings from Arista-7050-QX-32S and define params for 10G and 1G speeds as well
Signed-off-by: Neetha John <nejo@microsoft.com>
#### Why I did it
- xcvrd crash was seen in latest 201811 images.
- For Dell S6100,API 2.0 uses poll mode while 1.0 was still using interrupt mode.
#### How I did it
- Modified get_transceiver_change_event in 1.0 to poll mode in all the related branches.
Why I did it
To enable autonegotiation/link training on some Broadcom-based platforms (Arista 7060 and 7260, Celestica DX010)
How I did it
Add appropriate SOC property for enabling the feature to the Broadcom config files of appropriate platforms
Also convert line endings to UNIX format for one Celestica file
Why I did it
The S6000 devices, the cold reboot is abrupt and it is likely to cause issues which will cause the device to land into EFI shell. Hence the platform reboot will happen after graceful unmount of all the filesystems as in S6100.
How I did it
Moved the platform_reboot to platform_reboot_override and hooked it to the systemd shutdown services as in S6100.
Fixed the "/host unmount failed" issue as well in 201811.
How to verify it
Issue "reboot" command to verify if the reboot is happening gracefully.
Update fancontrol service for Seastone-DX010/E1031 device to support hysteresis temperature threshold and difference config for each unit fan direction type (B2F/F2B); follow master branch
Dynamic threshold setting changed to 0 and WRED profile green min threshold set to 250000 for Tomahawk devices
Changed the dynamic threshold settings in pg_profile_lookup.ini
Added a macro for WRED profiles in qos.json.j2 for Tomahawk devices
Necessary changes made in qos.config.j2 to use the macro if present
Signed-off-by: Neetha John <nejo@microsoft.com>
* Add new CIG device CS6436-54P and CS5435-54P, also update code for CS6436-56P
* Update port_config.ini
update port config file started from "Ethernet1"
* Update port_config.ini
Update port config file started from "Ethernet1"
* Update port_config.ini
Update the index column in the port config file
* Update sfputil.py
Update sfputil python file as the port index changed
* Update port_config.ini
Update the index column in the port config file
* Update sfputil.py
Update sfputil python file as the port index changed
* Remove the divide by 4 operation to the under the hood SAI
This is to avoid the need and thus the confusion for application program to know
the mmu internal architecture
This change must have support from SAI change to reach the correct
config
Signed-off-by: Wenda <wenni@microsoft.com>
* Relegate the divide by 4 operation to the under the hood SAI for egress
lossless pool
Extend to 7060 and 6100
Signed-off-by: Wenda <wenni@microsoft.com>
* Add more TH/TH2 hwskus
Signed-off-by: Wenda Ni <wenni@microsoft.com>
* Update config test
Signed-off-by: Wenda Ni <wenni@microsoft.com>
* Add TH2 ingress lossy profile
Signed-off-by: Wenda Ni <wenni@microsoft.com>
* Move the divide by 4 operation to SAI internal
Signed-off-by: Wenda Ni <wenni@microsoft.com>
* [bcm SAI] Upgrade Broadcom SAI to version 3.5.3.1-15
- Broadcom SAI 3.5 GA release 20190924.
Signed-off-by: Ying Xie <ying.xie@microsoft.com>
* add config.bcm for hlx
* modify config.bcm path for hlx
* Delete hx4-cel-hbtn-48x1G+4x10G.config.bcm
* add config.bcm and path
* update led for cxp
* Add new device data for dx010
There is a very low possibility that qsfp_status exists when opening it but doesn't when reading it a bit later. If this occurs an exception is raised.
However, this should be treated as "not presence" rather than causing an exception, which can be achieved by putting it into try block.
