Nvidia will verify that the -proposed version works with the new
functionality as expected internally.
I can at least confirm that I don't see any regressions in the the
values in the power table on a similar SKU, and that are aligned with
those provided in MediaTek's example (for EHT26, EHT52, EHT106, EHT242)
with the proposed version of linux-firmware installed and the adapter
set to channel 149. I also observed no regressions when connecting to a
wifi network and pinging via the onboard adapter:
root@ubuntu:/sys/kernel/debug/ieee80211/phy0/mt76# iw wlP9s9 set channel
149^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^H^[[10Pcat txpower_sku^M
^[[?2004l^M 1m 2m 5m 11m^M
CCK (tmac) : 37 37 37 37^M
6m 9m 12m 18m 24m 36m 48m
54m^M
OFDM (tmac) : 44 44 43 43 40 38 38
36^M
mcs0 mcs1 mcs2 mcs3 mcs4 mcs5 mcs6
mcs7^M
HT20 (tmac) : 44 44 43 39 39 39 38
37^M
mcs0 mcs1 mcs2 mcs3 mcs4 mcs5 mcs6 mcs7
mcs32^M
HT40 (tmac) : 0 0 0 0 0 0 0 0
0^M
mcs0 mcs1 mcs2 mcs3 mcs4 mcs5 mcs6 mcs7
mcs8 mcs9 mcs10 mcs11^M
VHT20 (tmac) : 44 44 43 39 39 39 38 37
35 35 42 42^M
VHT40 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
VHT80 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
VHT160 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
HE26 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33^M
HE52 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33^M
HE106 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33^M
HE242 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33^M
HE484 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
HE996 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
HE996x2 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0^M
mcs0 mcs1 mcs2 mcs3 mcs4 mcs5 mcs6 mcs7
mcs8 mcs9 mcs10 mcs11 mcs12 mcs13 mcs14 mcs15^M
EHT26 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT52 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT106 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT242 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT484 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
EHT996 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
EHT996x2 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
EHT996x4 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
EHT26_52 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT26_106 (tmac) : 44 44 43 39 39 39 38 37
35 35 33 33 30 30 44 44^M
EHT484_242 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
EHT996_484 (tmac) : 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0^M
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-firmware in Ubuntu.
https://bugs.launchpad.net/bugs/2109796
Title:
[SRU][O/N] Support Wi-fi and Bluetooth for Nvidia DGX Spark Platform
Status in linux-firmware package in Ubuntu:
In Progress
Status in linux-firmware source package in Noble:
Fix Committed
Status in linux-firmware source package in Oracular:
Fix Committed
Bug description:
[SRU Justfication]
BugLink: https://bugs.launchpad.net/bugs/2109796
These binary firmware changes are required for full Wifi and Bluetooth
functionality on DGX Spark
[Impact]
Lack of some WiFi and Bluetooth functionality on Nvidia DGX Spark
platform. Specifically, this update is required to update the power
table as required for regulatory compliance in several countries (per
MediaTek).
[Fix]
Apply the following vendor-proposed patches from upstream:
WIFI
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/commit/?id=8b3109069b3b381788576f137896ab5067741d3f
BT
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/commit/?id=ae25a3cb0fd890ad83ab058cdb13e6056908be29
https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/commit/?id=7a074fed8691869665ae5b164f55d04252c605f4
[Test Case]
Verify that Wifi and BT connections can still be established with the
updated blobs, and verify that the status of the power table is
correct [0]
[Where problems could occur]
As this is an opaque binary update, this could impact all MT7925 and
MT7920 WiFi and Bluetooth devices
[Other Info]
This update originates from a support request by Nvidia
[0] MediaTek provided the following information on how to test this update:
The method of verification is to observe the status of the power table. By
connecting to 5G channel 149 under different firmware versions, it can be found
that the power is abnormal in the old version of the firmware, while it is
normal in the new version.
See the attached log for an example.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-firmware/+bug/2109796/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
