Dylan Van Assche (dylanvanassche) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
That's at least what I read/heard that NFC stuff relies on TrustZone, maybe for banking cards?
Would be nice to emulate NFC tags for public transport cards, building access card, parking access cards, etc.

biemster (biemster) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
What I kind of understand from the datasheets seems to indicate that the NFC part of the PN80 can obtain a key over the SPI bus from the SE part of it, or you can just pass it over the I2C bus from your app. I think I can also send RAW APDUs, which would pull everything including crypto stuff into userspace.

biemster (biemster) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
But that will require quite a bit more investigation, I'll start with the easier ones like the access cards and such 😄

biemster (biemster) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
I forgot, the libnfc-nci examples have a couple emulation examples as well, so this looks definitely doable without TrustZone.

Dylan Van Assche (dylanvanassche) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
Interesting :) I'm not well known with NFC bits:

- This libnfc-nci is a fork of libnfc right? Would NXP not send their patches upstream?
- Can it use neard and such as well as in the kernel? Might be important for application support to have well versed kernel interfaces

biemster (biemster) commented on issue #31 in Qualcomm Snapdragon 845 Mainline / Linux: OnePlus6 NFC
I did not get the impression that it's a fork of libnfc, at least it's nowhere mentioned (and I did not have a look at the libnfc sources yet). It seems to be written specifically for their pn5xx readers, and should be used with a different driver than the one in mainline linux, or directly over i2c (which is what we do here).
I don't think it plays ball with neard, at least I did not manage that.

To be honest, my experience with NFC is very limited as well so I might be off here and there.

[Qualcomm Snapdragon 845 Mainline / Linux] Issue #44 WIFI Firmware : opened by Martin Valba (Martinvlba)
* Hi, I'm trying to bring up my own aarch64 LFS distro and i noticed that ath10k_snoc is trying to laod firmware-N.bin where N is 2-5 as number.

* dir of "/lib/firmware/ath10k/WCN3990/hw1.0/" looks like this

board-2.bin  board-2.json  wlanmdsp.mbn

Feb 18 10:27:33 drunk kernel: ath10k_snoc 18800000.wifi: could not probe fw (-22)
Feb 18 10:27:33 drunk kernel: ath10k_snoc 18800000.wifi: could not fetch firmware files (-22)
Feb 18 10:27:33 drunk kernel: ath10k_snoc 18800000.wifi: Failed to find firmware-N.bin (N between 2 and 6) from ath10k/WCN3990/hw1.0: -22
Feb 18 10:27:33 drunk kernel: ath10k_snoc 18800000.wifi: invalid firmware magic
Feb 18 10:27:29 drunk kernel: ath10k_snoc 18800000.wifi: qmi fw_version 0x20090216 fw_build_timestamp 2019-12-18 08:15 fw_build_id QC_IMAGE_VERSION_STRING=WLAN.HL.2.0.c3-00534-QCAHLSWMTPLZ-1
Feb 18 10:27:29 drunk kernel: ath10k_snoc 18800000.wifi: qmi chip_id 0x30214 chip_family 0x4001 board_id 0xff soc_id 0x40030001
Feb 18 10:27:25 drunk kernel: ath10k_snoc 18800000.wifi: Adding to iommu group 24

Martin Valba (Martinvlba) commented on issue #44 in Qualcomm Snapdragon 845 Mainline / Linux: * WIFI Firmware : firmware-N.bin missing and ath10k_snoc wants it*
If any of you know where to find it or if its even needed then lemme know, searched pmos and elsewhere but only board-2.bin seems to be seen. Cant tell if board-2.bin even gets loaded or not by the dmesg and journalctl logs.

Arnaud Ferraris (a-wai) commented on issue #44 in Qualcomm Snapdragon 845 Mainline / Linux: * WIFI Firmware : firmware-N.bin missing and ath10k_snoc wants it*
Look at https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git

Alexey Min (minlexx) commented on issue #44 in Qualcomm Snapdragon 845 Mainline / Linux: * WIFI Firmware : firmware-N.bin missing and ath10k_snoc wants it*

* dir of "/lib/firmware/ath10k/WCN3990/hw1.0/" looks like this
board-2.bin board-2.json wlanmdsp.mbn

[Qualcomm Snapdragon 845 Mainline / Linux] Issue #45 ath10k monitor mode results in CrashDump mode opened by biemster (biemster)
Monitor mode for the ath10k seems to have issues, it does not work. Setting it up with the ip and iw tools to bring down/up the interface and enable monitor mode results in crashes in the kernel logging and no packets received with tcpdump. Using airmon-ng results in the Qualcomm CrashDump mode.
The info from dmesg is not very helpful:

[Fri Mar 15 16:37:47 2024] qcom-q6v5-mss 4080000.remoteproc: fatal error received: err_qdi.c:456:EF:wlan_process:1:cmnos_thread.c:3921:Asserted in wlan_mgmt_local_txrx.c:_wlan_mgmt_local_send_by_pmac:409
[Fri Mar 15 16:37:47 2024] remoteproc remoteproc3: crash detected in 4080000.remoteproc: type fatal error
[Fri Mar 15 16:37:47 2024] remoteproc remoteproc3: handling crash #1 in 4080000.remoteproc
[Fri Mar 15 16:37:47 2024] remoteproc remoteproc3: recovering 4080000.remoteproc
[Fri Mar 15 16:37:47 2024] ipa 1e40000.ipa: received modem crashed event
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: firmware crashed! (guid 1d6fd8a2-d4cd-47e5-970a-496d2675d90a)
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: wcn3990 hw1.0 target 0x00000008 chip_id 0x00000000 sub 0000:0000
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: kconfig debug 1 debugfs 1 tracing 0 dfs 0 testmode 0
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: firmware ver 1.0.0.570 api 5 features wowlan,mgmt-tx-by-reference,non-bmi crc32 b3d4b790
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: board_file api 2 bmi_id N/A crc32 00000000
[Fri Mar 15 16:37:47 2024] ath10k_snoc 18800000.wifi: htt-ver 3.56 wmi-op 4 htt-op 3 cal file max-sta 32 raw 0 hwcrypto 1
[Fri Mar 15 16:37:48 2024] qcom-q6v5-mss 4080000.remoteproc: port failed halt
[Fri Mar 15 16:37:48 2024] ipa 1e40000.ipa: received modem offline event
[Fri Mar 15 16:37:48 2024] remoteproc remoteproc3: stopped remote processor 4080000.remoteproc
[Fri Mar 15 16:37:48 2024] ipa 1e40000.ipa: received modem starting event
[Fri Mar 15 16:37:48 2024] qcom-q6v5-mss 4080000.remoteproc: MBA booted without debug policy, loading mpss

Unbalanced enable for IRQ 185
WARNING: CPU: 0 PID: 108 at kernel/irq/manage.c:789 __enable_irq+0x4c/0x7c

Martin Valba (Martinvlba) commented on issue #44 in Qualcomm Snapdragon 845 Mainline / Linux: * WIFI Firmware : firmware-N.bin missing and ath10k_snoc wants it*
Hehe, so I actually tried this before but device hard crashed.

Now when i tried to debug what happened is that:

without the firmware-5.bin = device boots and i get sddm to login into palsma-mobile

with firmware-5.bin = device crashes before jumping to sddm

[Qualcomm Snapdragon 845 Mainline / Linux] Issue #44 WIFI Firmware : firmware-N.bin missing and ath10k_snoc wants it closed by Martin Valba (Martinvlba)

heyarne (heyarne) commented on issue #43 in Qualcomm Snapdragon 845 Mainline / Linux: Flaky wifi on Poco F1
I have noticed another issue; I have rebuilt the image and flashed everything again. I can no longer use iw reg set DE to set the region code correctly. Ever sine then, wifi seems to have become even more unstable, and it's common for it to stop working after two days or so.

# iw reg get
global
country 00: DFS-UNSET
    (2402 - 2472 @ 40), (6, 20), (N/A)
    (2457 - 2482 @ 20), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
    (2474 - 2494 @ 20), (6, 20), (N/A), NO-OFDM, PASSIVE-SCAN
    (5170 - 5250 @ 80), (6, 20), (N/A), AUTO-BW, PASSIVE-SCAN
    (5250 - 5330 @ 80), (6, 20), (0 ms), DFS, AUTO-BW, PASSIVE-SCAN
    (5490 - 5730 @ 160), (6, 20), (0 ms), DFS, PASSIVE-SCAN
    (5735 - 5835 @ 80), (6, 20), (N/A), PASSIVE-SCAN
    (57240 - 63720 @ 2160), (N/A, 0), (N/A)

phy#0
country 99: DFS-UNSET
    (2402 - 2472 @ 40), (N/A, 20), (N/A)
    (5140 - 5360 @ 80), (N/A, 30), (N/A), PASSIVE-SCAN
    (5715 - 5860 @ 80), (N/A, 30), (N/A), PASSIVE-SCAN

# apk list --installed | grep iw
iw-6.7-r0 aarch64 {iw} (ISC) [installed]
iw-doc-6.7-r0 aarch64 {iw} (ISC) [installed]
# apk list --installed | grep reg
wireless-regdb-2024.01.23-r0 aarch64 {wireless-regdb} (ISC) [installed]
wireless-regdb-doc-2024.01.23-r0 aarch64 {wireless-regdb} (ISC) [installed]

[   18.384973] ath: EEPROM regdomain: 0x809c
[   18.384981] ath: EEPROM indicates we should expect a country code
[   18.384986] ath: doing EEPROM country->regdmn map search
[   18.384989] ath: country maps to regdmn code: 0x52
[   18.384994] ath: Country alpha2 being used: CN
[   18.384998] ath: Regpair used: 0x52

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
almost identical to an issue seen here: https://github.com/gregkh/commit_tree/blob/6fb1d30ba5c5851e46fa26bbd57684014c35a76c/changes/5.5/f4fe2e53349f1072d33c69f484dbf9d77bb8f45a#L8
But not really identical so probably unrelated, but it might help pin this down.

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
This one is even more similar: https://www.spinics.net/lists/linux-wireless/msg182274.html

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
I'll keep a list here of random stuff I tried to work around this (since I've seen reports that monitor mode worked on the 6T on android):
1. The error message Asserted in wlan_mgmt_local_txrx.c:_wlan_mgmt_local_send_by_pmac:409 vaguely seemed to indicate a connection with the mgmt-tx-by-ref feature in firmware-5.bin, but removing that feature with qca-swiss-army-knife did not have any effect.

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
2. I noticed there is a /lib/firmware/qcom/sdm845/wlanmdsp.mbn and a /lib/firmware/qcom/sdm845/oneplus6/wlanmdsp.mbn, with /lib/firmware/ath10k/WCN3990/hw1.0/wlanmdsp.mbn being a symlink to the first. Changing the symlink to point to the /oneplus6/ one did not solve/change anything.

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
3. When using airmon-ng to put the interface in monitor mode no crashes happen at first. The device is monitoring channel 10 at that point, tcpdump is not receiving any packets. When changing the channel to 1 with iw dev wlan0mon set channel 1 crashes show up in dmesg immediately. So either changing the channel triggers this, or the reception of a packet (and channel 10 is silent around my bench)?

Arseniy Velikanov (adomerle) opened merge request !90 Draft: Add support for Xiaomi Mi 8 in Qualcomm Snapdragon 845 Mainline / Linux

biemster (biemster) commented on issue #45 in Qualcomm Snapdragon 845 Mainline / Linux: ath10k monitor mode results in CrashDump
4. Regarding point 2.: actually oneplus6/wlanmdsp.mbn is loaded by default (I don't really understand the fw loading mechanisms yet). This gives a firmware ver 1.0.0.570 and htt-ver 3.56 in dmesg. When replacing this file with ../wlanmdsp.mbn firmware ver 1.0.0.559 with htt-ver 3.53 gets loaded, but unfortunately the bug is still there. I've seen dmesg logs floating around sporting a ver 1.0.0.992, so I'll try to find one of those (although that might be for a totally different chip)

Marco Mattiolo (tiol11) commented on issue #21 in Qualcomm Snapdragon 845 Mainline / Linux: can you add camera sensor support IMX519 please ?
Here op6-enchilada on branch 6.8-camera, still stuck on VIDIOC_STREAMON returned -1 (Broken pipe) but dmesg gives something different

[    0.079370] platform acb3000.camss: Fixed dependency cycle(s) with /soc@0/cci@ac4a000/i2c-bus@0/camera@1a
[    1.219217] platform acb3000.camss: Fixed dependency cycle(s) with /soc@0/cci@ac4a000/i2c-bus@0/camera@1a
[    1.219966] i2c 16-001a: Fixed dependency cycle(s) with /soc@0/camss@acb3000
[    3.521281] qcom-camss acb3000.camss: Adding to iommu group 13
[    3.597876] qcom-camss acb3000.camss: supply vdda-phy not found, using dummy regulator
[    3.598032] qcom-camss acb3000.camss: supply vdda-pll not found, using dummy regulator
[    3.599422] qcom-camss acb3000.camss: supply vdda-phy not found, using dummy regulator
[    3.599472] qcom-camss acb3000.camss: supply vdda-pll not found, using dummy regulator
[    3.642686] qcom-camss acb3000.camss: supply vdda-phy not found, using dummy regulator
[    3.664079] qcom-camss acb3000.camss: supply vdda-pll not found, using dummy regulator
[    3.695631] qcom-camss acb3000.camss: CSIPHY0: D-PHY, 3 data lanes, with clock lane
[    3.902884] cam_cc_ife_0_clk_src: rcg didn't update its configuration.
[    3.906905] cam_cc_ife_0_clk_src: rcg didn't update its configuration.
[    3.909751] cam_cc_ife_0_clk_src: rcg didn't update its configuration.
[    3.911862] cam_cc_ife_0_clk_src: rcg didn't update its configuration.
[    3.914025] cam_cc_ife_lite_clk_src: rcg didn't update its configuration.
[    3.915727] cam_cc_ife_0_clk_src: rcg didn't update its configuration.
[    3.917932] cam_cc_ife_lite_clk_src: rcg didn't update its configuration.
[    3.920001] cam_cc_ife_0_clk_src: rcg didn't update its configuration.

jenneron (jenneron) opened merge request !91 Draft: Add support for AYN Odin in Qualcomm Snapdragon 845 Mainline / Linux

Caleb Connolly (calebccff) commented on merge request !91 in Qualcomm Snapdragon 845 Mainline / Linux: Draft: Add support for AYN Odin
I'm a bit wary of some of the hacks in here, I can pick stuff up which is in an upstreamable state and isn't likely to break on a rebase. But I'd much rather this was all sent upstream, I expect a few iterations are going to be needed there.

I'm wary of the lontium bridge hacks and the gamepad driver in particular.

jenneron (jenneron) commented on merge request !91 in Qualcomm Snapdragon 845 Mainline / Linux: Draft: Add support for AYN Odin
can you review gamepad driver and let me know what I can improve?

speaking of bridge, is there a better way to make it work without DDC?