ath10k: "add target IRAM recovery feature support" breaks QCA9984 Firmware load capability

[sparks71]

[SRU Justification]

[Impact]

Commit 9af7c32ceca8 ("ath10k: add target IRAM recovery feature support")
in v5.11 fails firmware loading without IRAM recovery feature. However,
the latest firmware from
https://github.com/kvalo/ath10k-firmware/tree/master/QCA9984/hw1.0/3.9.0.2
doesn't support IRAM, it follows new official ath10k firmwares may no
longer be loadable for kernel newer than v5.11.

[Fix]

A fix is proposed to
http://lists.infradead.org/pipermail/ath10k/2021-July/012729.html that
turns the lack/failure of IRAM recovery support a warning instead.

[Test Case]

With this fix, kernel gives errors about the IRAM supportness, but
proceeds to remaining procedures as before.

  ath10k_pci 0000:04:00.0: No hardware memory
  ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
  ath10k_pci 0000:07:00.0: No hardware memory
  ath10k_pci 0000:07:00.0: failed to copy target iram contents: -12

[Where problems could occur]

So the IRAM recovery feature may or may not be supported in the
officially released firmware, and people meet problems when using
non-IRAM supported firmware may still have a fallback to use older, but
IRAM recovery supported ones.

========== original bug report ==========

Hello, I use Ubuntu 20.04.2 LTS.

The kernel was recently updated to 5.11

5.11.0-25-generic #27~20.04.1-Ubuntu SMP Tue Jul 13 17:41:23 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

Since then the ath10k_core driver for my QCA9984 card is broken and only loads ~3 firmware versions from the 3.9.0.2 branch

https://github.com/kvalo/ath10k-firmware/tree/master/QCA9984/hw1.0/3.9.0.2

The firmware versions 70 / 86 / 91 are functional (but these firmware versions are already 2 years old)

Responsible for this is the defective patch "ath10k: add target IRAM recovery feature support"

https://git.launchpad.net/~ubuntu-kernel/ubuntu/+source/linux/+git/focal/commit/?h=hwe-5.11&id=9af7c32ceca85da27867dd863697d2aafc80a3f8

If I understand the patch correctly then the driver tries to load firmware with the feature "iram recovery" into the memory of the host (for recvovery purposes).

If this fails, the firmware is discarded and the network controller remains unusable.

Of course, this is not a desirable behavior for optional firmware features.

dmesg:

[ 3.074952] ath: loading out-of-tree module taints kernel.
[ 3.090522] ath: module verification failed: signature and/or required key missing - tainting kernel
[ 3.110879] ath10k_pci 0000:03:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.178741] ath10k_pci 0000:04:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.229944] ath10k_pci 0000:03:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.229948] ath10k_pci 0000:03:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.230274] ath10k_pci 0000:03:00.0: firmware ver 10.4-3.9.0.2-00149 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 5f41acd7
[ 3.291329] ath10k_pci 0000:04:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.291333] ath10k_pci 0000:04:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.291678] ath10k_pci 0000:04:00.0: firmware ver 10.4-3.9.0.2-00149 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 5f41acd7
[ 4.463156] ath10k_pci 0000:03:00.0: board_file api 2 bmi_id 0:1 crc32 85498734
[ 4.520649] ath10k_pci 0000:04:00.0: board_file api 2 bmi_id 0:2 crc32 85498734
[ 7.047910] ath10k_pci 0000:03:00.0: failed to copy target iram contents: -12
[ 7.103260] ath10k_pci 0000:03:00.0: could not init core (-12)
[ 7.103267] ath10k_pci 0000:03:00.0: could not probe fw (-12)
[ 7.108570] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
[ 7.163247] ath10k_pci 0000:04:00.0: could not init core (-12)
[ 7.163268] ath10k_pci 0000:04:00.0: could not probe fw (-12)

In the ath10k mailing list I found a second patch which makes it possible to load the latest firmware on the card again.

https://<email address hidden>/msg13655.html

I have tested the patch and it works.

dmesg:

[18245.610112] ath10k_pci 0000:03:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[18245.674800] ath10k_pci 0000:04:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[18245.731606] ath10k_pci 0000:03:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[18245.731620] ath10k_pci 0000:03:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[18245.733172] ath10k_pci 0000:03:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[18245.790380] ath10k_pci 0000:04:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[18245.790383] ath10k_pci 0000:04:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[18245.790703] ath10k_pci 0000:04:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[18246.963760] ath10k_pci 0000:03:00.0: board_file api 2 bmi_id 0:1 crc32 85498734
[18247.019699] ath10k_pci 0000:04:00.0: board_file api 2 bmi_id 0:2 crc32 85498734
[18249.514054] ath10k_pci 0000:03:00.0: No hardware memory
[18249.514057] ath10k_pci 0000:03:00.0: failed to copy target iram contents: -12
[18249.570423] ath10k_pci 0000:04:00.0: No hardware memory
[18249.570426] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
[18249.620637] ath10k_pci 0000:03:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[18249.678035] ath10k_pci 0000:04:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1

Please include this patch in Ubuntu so that we can load current firmware versions on our QCA9984 cards again.

additional reference:

https://lists.infradead.org/pipermail/ath10k/2021-May/012626.html

best regards

[You-Sheng Yang (vicamo)]

PPA: https://launchpad.net/~vicamo/+archive/ubuntu/ppa-1939937

Hi, could you help have tests with kernels being built in my ppa? You may do:

  $ sudo add-apt-repository ppa:vicamo/ppa-1939937
  $ sudo apt-get install \
        linux-image-unsigned-5.11.0-2033-generic \
        linux-modules-extra-5.11.0-2033-generic \
        linux-headers-5.11.0-2033-generic

It may take a few hours from now to complete compiling though.

[sparks71 (sparks71)]

$ sudo apt-get install \
> linux-image-unsigned-5.11.0-2033-generic \
> linux-modules-extra-5.11.0-2033-generic \
> linux-headers-5.11.0-2033-generic

Holen:1 http://ppa.launchpad.net/vicamo/ppa-1939937/ubuntu focal/main amd64 linux-hwe-5.11-headers-5.11.0-2033 all 5.11.0-2033.35~20.04.1+lp1939937.1.ath10k.iram [12,0 MB]
Holen:2 http://ppa.launchpad.net/vicamo/ppa-1939937/ubuntu focal/main amd64 linux-headers-5.11.0-2033-generic amd64 5.11.0-2033.35~20.04.1+lp1939937.1.ath10k.iram [1.863 kB]
Holen:3 http://ppa.launchpad.net/vicamo/ppa-1939937/ubuntu focal/main amd64 linux-modules-5.11.0-2033-generic amd64 5.11.0-2033.35~20.04.1+lp1939937.1.ath10k.iram [16,2 MB]
Holen:4 http://ppa.launchpad.net/vicamo/ppa-1939937/ubuntu focal/main amd64 linux-image-unsigned-5.11.0-2033-generic amd64 5.11.0-2033.35~20.04.1+lp1939937.1.ath10k.iram [10,3 MB]
Holen:5 http://ppa.launchpad.net/vicamo/ppa-1939937/ubuntu focal/main amd64 linux-modules-extra-5.11.0-2033-generic amd64 5.11.0-2033.35~20.04.1+lp1939937.1.ath10k.iram [42,4 MB]
Es wurden 82,8 MB in 8 s geholt (10,2 MB/s).

uname -a
... 5.11.0-2033-generic #35~20.04.1+lp1939937.1.ath10k.iram-Ubuntu SMP Thu Aug 19 16:37: x86_64 x86_64 x86_64 GNU/Linux

dmesg | grep ath
[ 3.279110] ath10k_pci 0000:03:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.355016] ath10k_pci 0000:04:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.393327] ath10k_pci 0000:03:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.393330] ath10k_pci 0000:03:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.393655] ath10k_pci 0000:03:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 3.475304] ath10k_pci 0000:04:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.475308] ath10k_pci 0000:04:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.475639] ath10k_pci 0000:04:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 4.624937] ath10k_pci 0000:03:00.0: board_file api 2 bmi_id 0:1 crc32 85498734
[ 4.704618] ath10k_pci 0000:04:00.0: board_file api 2 bmi_id 0:2 crc32 85498734
[ 7.202830] ath10k_pci 0000:03:00.0: No hardware memory
[ 7.202833] ath10k_pci 0000:03:00.0: failed to copy target iram contents: -12
[ 7.309047] ath10k_pci 0000:03:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[ 7.371208] ath: EEPROM regdomain sanitized
[ 7.371210] ath: EEPROM regdomain: 0x64
[ 7.371211] ath: EEPROM indicates we should expect a direct regpair map
[ 7.371212] ath: Country alpha2 being used: 00
[ 7.371213] ath: Regpair used: 0x64
[ 7.373346] ath10k_pci 0000:03:00.0 wlp3s0: renamed from wlan0
[ 7.382821] ath10k_pci 0000:04:00.0: No hardware memory
[ 7.382824] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
[ 7.501740] ath10k_pci 0000:04:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal ...

[You-Sheng Yang (vicamo)]

Hi, don't worry. linux/focal here means 5.4-generic, which is certainly not affected. linux/hirsute will then become the basis of hwe-5.11 in focal. So as long as linux/hirsute is being worked on, you'll have the fix in hwe-5.11/focal later.

Since the results are positive as expected, I'll send out SRU patches later.

[sparks71 (sparks71)]

thanks!

[You-Sheng Yang (vicamo)]

SRU: https://lists.ubuntu.com/archives/kernel-team/2021-August/123436.html (oem-5.13, hirsute, impish, unstable)

[sparks71 (sparks71)]

"Acked-by: Tim Gardner <tim.gardner at canonical.com>

I think for Hirsute this could wait until its at least in linux-next."

I have looked into the Hirsute Linux firmware package

linux-firmware_1.197_all.deb

The header of the QCA9984 firmware file says:

QCA-ATH10K 10.4-3.9.0.2-00131

This means that QCA9984 cards are not functional under Hirsute for the time being.

so users have only the possibility to downgrade the firmware to 70 to 91 to get a working card

With the packages from Hirsute alone, the cards are definitely not functional.

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-focal' to 'verification-done-focal'. If the problem still exists, change the tag 'verification-needed-focal' to 'verification-failed-focal'.

If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[sparks71 (sparks71)]

Which kernel should I test?
Please be more precise.

[sparks71 (sparks71)]

So "focal-proposed" offers me the update "linux-image-5.11.0-34-generic"

5.11.0-34-generic #36~20.04.1-Ubuntu SMP Fri Aug 27 08:06:32 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

[ 7.234159] ath10k_pci 0000:03:00.0: failed to copy target iram contents: -12
[ 7.242923] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
[ 7.291070] ath10k_pci 0000:03:00.0: could not init core (-12)
[ 7.291076] ath10k_pci 0000:03:00.0: could not probe fw (-12)
[ 7.299146] ath10k_pci 0000:04:00.0: could not init core (-12)
[ 7.299169] ath10k_pci 0000:04:00.0: could not probe fw (-12)

it was clear that the patch is not in the kernel and thanks for wasting my time.
Without concrete information I test nothing at all.

[sparks71 (sparks71)]

 5.13.0-1011-oem #15-Ubuntu SMP Fri Aug 27 09:45:09 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

[ 6.954194] ath10k_pci 0000:03:00.0: No hardware memory
[ 6.954197] ath10k_pci 0000:03:00.0: failed to copy target iram contents: -12
[ 6.985209] ath10k_pci 0000:04:00.0: No hardware memory
[ 6.985213] ath10k_pci 0000:04:00.0: failed to copy target iram contents: -12
[ 7.060527] ath10k_pci 0000:03:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[ 7.106011] ath10k_pci 0000:04:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1

works

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux-oem-5.13 - 5.13.0-1012.16

---------------
linux-oem-5.13 (5.13.0-1012.16) focal; urgency=medium

  * focal/linux-oem-5.13: 5.13.0-1012.16 -proposed tracker (LP: #1942861)

  * Fix sluggish r8169 by disaling ASPM L1.2 (LP: #1942830)
    - SAUCE: r8169: Disable ASPM L1.2

  * CVE-2021-3609
    - can: bcm: delay release of struct bcm_op after synchronize_rcu()
    - can: j1939: j1939_sk_init(): set SOCK_RCU_FREE to call sk_destruct() after
      RCU is done

 -- Timo Aaltonen <email address hidden> Tue, 07 Sep 2021 11:49:54 +0300

[sparks71 (sparks71)]

A small update:

a new patch has been released on the ath10k mailing list and has been merged into the ath git

the patch is currently in "master-pending"

https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=master-pending&id=973de582639a1e45276e4e3e2f3c2d82a04ad0a6

Edit: new URL:

https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=master-pending&id=139838bdd36e20b30dc725dfb24d7f505a436765

have applied the patch to my linux-hwe-5.11 kernel and it works
with the patch the error message is completely gone

5.11.0-34-generic #36~20.04.1-Ubuntu SMP Fri Aug 27 08:06:32 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux

[ 3.262867] ath10k_pci 0000:03:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.322339] ath10k_pci 0000:04:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.379884] ath10k_pci 0000:03:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.379888] ath10k_pci 0000:03:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.380218] ath10k_pci 0000:03:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 3.438075] ath10k_pci 0000:04:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.438079] ath10k_pci 0000:04:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.438418] ath10k_pci 0000:04:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 4.617452] ath10k_pci 0000:03:00.0: board_file api 2 bmi_id 0:1 crc32 85498734
[ 4.671543] ath10k_pci 0000:04:00.0: board_file api 2 bmi_id 0:2 crc32 85498734
[ 7.301463] ath10k_pci 0000:03:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[ 7.362170] ath: EEPROM regdomain sanitized
[ 7.362172] ath: EEPROM regdomain: 0x6c
[ 7.362173] ath: EEPROM indicates we should expect a direct regpair map
[ 7.362174] ath: Country alpha2 being used: 00
[ 7.362174] ath: Regpair used: 0x6c
[ 7.362258] ath: EEPROM regdomain: 0x8114
[ 7.362259] ath: EEPROM indicates we should expect a country code
[ 7.362259] ath: doing EEPROM country->regdmn map search
[ 7.362260] ath: country maps to regdmn code: 0x37
[ 7.362260] ath: Country alpha2 being used: DE
[ 7.362261] ath: Regpair used: 0x37
......

best regards

[You-Sheng Yang (vicamo)]

It seems the updated url is https://git.kernel.org/pub/scm/linux/kernel/git/kvalo/ath.git/commit/?h=master-pending&id=37a86881331e1451ec357b1fc6738c34fe5230c1

Per review comments in https://lists.ubuntu.com/archives/kernel-team/2021-August/123442.html, switch to pending and wait until this being merged in linux-next.

[You-Sheng Yang (vicamo)]

Updated PPA with linux-hwe-5.11/focal version 5.11.0-2037.41~20.04.2+lp1939937.2.ath10k.iram, linux/hirsute version 5.11.0-2038.42+lp1939937.2.ath10k.iram and linux/impish version 5.13.0-2014.14+lp1939937.2.ath10k.iram for testing.

[sparks71 (sparks71)]

@vicamo

yes this is the right patch and thank you!

I have updated the link in comment #12

and tested your PPA

5.11.0-2037-generic #41~20.04.2+lp1939937.2.ath10k.iram-Ubuntu SMP Mon Sep 27 06:29: x86_64 x86_64 x86_64 GNU/Linux

[ 3.521734] ath10k_pci 0000:03:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.579723] ath10k_pci 0000:04:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0
[ 3.639858] ath10k_pci 0000:03:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.639862] ath10k_pci 0000:03:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.640195] ath10k_pci 0000:03:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 3.688745] ath10k_pci 0000:04:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe
[ 3.688749] ath10k_pci 0000:04:00.0: kconfig debug 0 debugfs 1 tracing 1 dfs 0 testmode 0
[ 3.689089] ath10k_pci 0000:04:00.0: firmware ver 10.4-3.9.0.2-00152 api 5 features no-p2p,mfp,peer-flow-ctrl,btcoex-param,allows-mesh-bcast,no-ps,peer-fixed-rate,iram-recovery crc32 723f9771
[ 4.875609] ath10k_pci 0000:03:00.0: board_file api 2 bmi_id 0:1 crc32 85498734
[ 4.918482] ath10k_pci 0000:04:00.0: board_file api 2 bmi_id 0:2 crc32 85498734
[ 7.578121] ath10k_pci 0000:03:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[ 7.598799] ath10k_pci 0000:04:00.0: htt-ver 2.2 wmi-op 6 htt-op 4 cal otp max-sta 512 raw 0 hwcrypto 1
[ 7.640829] ath: EEPROM regdomain sanitized
[ 7.640831] ath: EEPROM regdomain: 0x64
[ 7.640832] ath: EEPROM indicates we should expect a direct regpair map
[ 7.640833] ath: Country alpha2 being used: 00
[ 7.640834] ath: Regpair used: 0x64
[ 7.644614] ath10k_pci 0000:03:00.0 wlp3s0: renamed from wlan0
[ 7.660776] ath: EEPROM regdomain sanitized
[ 7.660778] ath: EEPROM regdomain: 0x64
[ 7.660779] ath: EEPROM indicates we should expect a direct regpair map
[ 7.660780] ath: Country alpha2 being used: 00
[ 7.660781] ath: Regpair used: 0x64
[ 7.675279] ath10k_pci 0000:04:00.0 wlp4s0: renamed from wlan0

works!

i will watch this and let you know as soon as the patch is in linux-next

i think the old (first patch) can be removed then

[sparks71 (sparks71)]

@vicamo

https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=6f8c8bf4c7c9be1c42088689fd4370e06b46608a

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=6f8c8bf4c7c9

"ath10k: fix module load regression with iram-recovery feature"

new patch....

tested locally and works

[Timo Aaltonen (tjaalton)]

seems to be fixed upstream by

6f8c8bf4c7c9 ath10k: fix module load regression with iram-recovery feature

which is at least in impish and up

[Brian Murray (brian-murray)]

The Hirsute Hippo has reached End of Life, so this bug will not be fixed for that release.

[You-Sheng Yang (vicamo)]

Re: comment #17, yes, in hirsute or newer actually.