Re: AP6275 / bcm43752 pcie on mainline brcmf

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/9/22 19:35, Andy Green wrote:
> 
> On 11/9/22 18:26, Arend Van Spriel wrote:

>> Can you share any kernel logs preferably with debug=0x1416 passed 
>> during insmod.
>>
> I passed that in modprobe.d/ but the only logs are its usual ones and 
> the ones I added

It seems the debug logging needs a few additional config options 
additionally.

[    3.469000] brcmfmac: brcmfmac_module_init No platform data available.
[    3.469140] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
[    3.469179] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
[    3.470077] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 
base 0x18000000 wrap 0x18100000
[    3.470079] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 
base 0x18001000 wrap 0x18101000
[    3.470081] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 
base 0x18002000 wrap 0x18102000
[    3.470082] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 
base 0x18003000 wrap 0x18103000
[    3.470084] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 
base 0x18004000 wrap 0x18104000
[    3.470085] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 
base 0x18005000 wrap 0x18105000
[    3.470087] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 
base 0x18010000 wrap 0x00000000
[    3.470088] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 
base 0x18012000 wrap 0x00000000
[    3.470090] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 
base 0x00000000 wrap 0x1810a000
[    3.470091] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 
base 0x00000000 wrap 0x18112000
[    3.470092] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 
base 0x00000000 wrap 0x18113000
[    3.470094] brcmfmac: brcmf_chip_set_passive Enter
[    3.579150] brcmfmac: brcmf_chip_set_passive Enter
[    3.579630] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.579631] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.579655] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, 
pmucaps=0x84565f23
[    3.579657] brcmfmac: brcmf_get_module_param Enter, bus=2, 
chip=43752, rev=2
[    3.579663] brcmfmac: brcmf_alloc Enter
[    3.579675] brcmfmac: brcmf_fw_alloc_request: using 
brcm/fw_bcm43752a2_pcie for chip BCM43752/2
[    3.579678] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
[    3.579680] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
[    3.583510] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.bin found
[    3.583516] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
[    3.583726] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.txt found
[    3.583727] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
[    3.583782] brcmfmac: brcmf_fw_request_nvram_done nvram 
000000001685ccfe len 6040
[    3.583784] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
[    3.584081] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.clm_blob found
[    3.584278] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.584279] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.677816] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
[    3.678423] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa1a05e5
[    3.678425] brcmfmac: brcmf_chip_set_active Enter
[    9.707127] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
failed to initialize
[    9.707144] brcmfmac: brcmf_detach Enter
[    9.731123] brcmfmac: brcmf_bus_change_state 0 -> 0
[    9.731126] brcmfmac: brcmf_proto_detach Enter

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 483 bytes --]

On November 10, 2022 8:06:38 AM Andy Green <andy@warmcat.com> wrote:

> On 11/9/22 19:35, Andy Green wrote:
>>
>> On 11/9/22 18:26, Arend Van Spriel wrote:
>
>>> Can you share any kernel logs preferably with debug=0x1416 passed
>>> during insmod.
>> I passed that in modprobe.d/ but the only logs are its usual ones and
>> the ones I added
>
> It seems the debug logging needs a few additional config options
> additionally.

Right. I wasn't sure you were aware ;-)

Thanks,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Hello, I'm facing the same problem on sdio version of AP6275 module in mainline kernel. But I found a very old version from here could work with mainline brcmfmac driver. 
Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
Besides who I should ask to make proper firmware for it into linux-firmware repo?

Shengyu

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/13/22 08:26, Shengyu Qu wrote:

> Hello, I'm facing the same problem on sdio version of AP6275 module in mainline kernel. But I found a very old version from here could work with mainline brcmfmac driver.
> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107

Thanks for the info.

The firmware on the firefly thread doesn't seem to have a CLM blob that 
the mainline driver wants, just two types of the ~900KB main firmware 
for AP mode support or not.

It also doesn't have _sdio or _pcie in the filenames... I tried it but 
although it's different in what it read after downloading NVRAM

[    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa2c05d3

(the rockchip SDK firmware says 0xfa1a05e5)

it still times it out a few seconds later as unresponsive after the fw 
load... maybe it is for SDIO only, or the lack of CLM, present in the 
Rockchip fw, means it won't get anywhere.

> Besides who I should ask to make proper firmware for it into linux-firmware repo?

It would also be great to hear if the PCIe firmware from Rockchip aimed 
at the OOT driver was definitively incompatible with the mainline driver 
and we have to find a solution for that from Broadcom, or it is expected 
to basically work.  I realize it might not be easy to get an answer 
internally but this chipset is likely to appear in more RK3588 SBCs soon 
since it's the one on the evb.

-Andy

> Shengyu

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Actually you can use any other clm info blob with this firmware as I tried. I also uploaded this to armbian's firmware repo. Hope we could get the proper firmware uploaded by Broadcom soon.

Shengyu

于 2022年11月13日 GMT+08:00 下午5:03:06, Andy Green <andy@warmcat.com> 写到:
>
>
>On 11/13/22 08:26, Shengyu Qu wrote:
>
>> Hello, I'm facing the same problem on sdio version of AP6275 module in mainline kernel. But I found a very old version from here could work with mainline brcmfmac driver.
>> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
>
>Thanks for the info.
>
>The firmware on the firefly thread doesn't seem to have a CLM blob that the mainline driver wants, just two types of the ~900KB main firmware for AP mode support or not.
>
>It also doesn't have _sdio or _pcie in the filenames... I tried it but although it's different in what it read after downloading NVRAM
>
>[    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 0xfa2c05d3
>
>(the rockchip SDK firmware says 0xfa1a05e5)
>
>it still times it out a few seconds later as unresponsive after the fw load... maybe it is for SDIO only, or the lack of CLM, present in the Rockchip fw, means it won't get anywhere.
>
>> Besides who I should ask to make proper firmware for it into linux-firmware repo?
>
>It would also be great to hear if the PCIe firmware from Rockchip aimed at the OOT driver was definitively incompatible with the mainline driver and we have to find a solution for that from Broadcom, or it is expected to basically work.  I realize it might not be easy to get an answer internally but this chipset is likely to appear in more RK3588 SBCs soon since it's the one on the evb.
>
>-Andy
>
>> Shengyu

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/13/22 09:10, Shengyu Qu wrote:
> Actually you can use any other clm info blob with this firmware as I tried. I also uploaded this to armbian's firmware repo. Hope we could get the proper firmware uploaded by Broadcom soon.

Thank, I tried this, it did accept to download the clm part from the 
Rockchip SDK firmware.

# ls -l /lib/firmware/brcm
-rw-r--r-- 1 root root  30993 Oct  3 08:40 clm_bcm43752a2_pcie_ag.blob
-rw-r--r-- 1 root root 743607 Nov 13 08:45 fw_bcm43752a2_ag.bin
lrwxrwxrwx 1 root root     46 Nov 13 09:19 fw_bcm43752a2_pcie.clm_blob 
-> /lib/firmware/brcm/clm_bcm43752a2_pcie_ag.blob
lrwxrwxrwx 1 root root     20 Nov 13 08:38 
fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin -> fw_bcm43752a2_ag.bin
lrwxrwxrwx 1 root root     36 Nov 13 08:47 fw_bcm43752a2_pcie.txt -> 
/lib/firmware/brcm/nvram_ap6275s.txt
-rw-r--r-- 1 root root   8709 Nov 13 08:44 nvram_ap6275s.txt

But it's no happier after the NVRAM download part.

# dmesg | grep brcm
[    3.530116] brcmfmac: brcmfmac_module_init No platform data available.
[    3.530303] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
[    3.530362] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
[    3.531325] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 
base 0x18000000 wrap 0x18100000
[    3.531329] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 
base 0x18001000 wrap 0x18101000
[    3.531334] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 
base 0x18002000 wrap 0x18102000
[    3.531338] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 
base 0x18003000 wrap 0x18103000
[    3.531342] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 
base 0x18004000 wrap 0x18104000
[    3.531346] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 
base 0x18005000 wrap 0x18105000
[    3.531350] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 
base 0x18010000 wrap 0x00000000
[    3.531354] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 
base 0x18012000 wrap 0x00000000
[    3.531358] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 
base 0x00000000 wrap 0x1810a000
[    3.531362] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 
base 0x00000000 wrap 0x18112000
[    3.531365] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 
base 0x00000000 wrap 0x18113000
[    3.531369] brcmfmac: brcmf_chip_set_passive Enter
[    3.641635] brcmfmac: brcmf_chip_set_passive Enter
[    3.642156] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.642159] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.642187] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, 
pmucaps=0x84565f23
[    3.642193] brcmfmac: brcmf_get_module_param Enter, bus=2, 
chip=43752, rev=2
[    3.642207] brcmfmac: brcmf_alloc Enter
[    3.642231] brcmfmac: brcmf_fw_alloc_request: using 
brcm/fw_bcm43752a2_pcie for chip BCM43752/2
[    3.642235] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
[    3.642241] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
[    3.646995] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.bin found
[    3.647006] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
[    3.647308] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.txt found
[    3.647312] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
[    3.647484] brcmfmac: brcmf_fw_request_nvram_done nvram 
00000000c4269a30 len 5968
[    3.647491] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
[    3.647911] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.clm_blob found
[    3.648197] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.648200] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.723260] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
[    3.723895] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa2c05d3
[    3.723899] brcmfmac: brcmf_chip_set_active Enter
[    9.721628] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
failed to initialize: 0xfa2c05d3
[    9.721657] brcmfmac: brcmf_detach Enter
[    9.745615] brcmfmac: brcmf_bus_change_state 0 -> 0
[    9.745623] brcmfmac: brcmf_proto_detach Enter

CONFIG_BRCMUTIL=m
# CONFIG_BRCMSMAC is not set
CONFIG_BRCMFMAC=m
# CONFIG_BRCMFMAC_PROTO_MSGBUF is not set
# CONFIG_BRCMFMAC_SDIO is not set
# CONFIG_BRCMFMAC_USB is not set
CONFIG_BRCMFMAC_PCIE=y
CONFIG_BRCM_TRACING=y
CONFIG_BRCMDBG=y

I have the AP6275 module EN signal with the gpio wired to it as an 
"always-on" gpio regulator in DT, and don't use the hacky rfkill thing 
the rockchip SDK came with.

It's on top of 6.1.0-rc4.

Does your setup differ from this?

-Andy

> Shengyu
> 
> 于 2022年11月13日 GMT+08:00 下午5:03:06, Andy Green <andy@warmcat.com> 写到:
>>
>>
>> On 11/13/22 08:26, Shengyu Qu wrote:
>>
>>> Hello, I'm facing the same problem on sdio version of AP6275 module in mainline kernel. But I found a very old version from here could work with mainline brcmfmac driver.
>>> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
>>
>> Thanks for the info.
>>
>> The firmware on the firefly thread doesn't seem to have a CLM blob that the mainline driver wants, just two types of the ~900KB main firmware for AP mode support or not.
>>
>> It also doesn't have _sdio or _pcie in the filenames... I tried it but although it's different in what it read after downloading NVRAM
>>
>> [    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 0xfa2c05d3
>>
>> (the rockchip SDK firmware says 0xfa1a05e5)
>>
>> it still times it out a few seconds later as unresponsive after the fw load... maybe it is for SDIO only, or the lack of CLM, present in the Rockchip fw, means it won't get anywhere.
>>
>>> Besides who I should ask to make proper firmware for it into linux-firmware repo?
>>
>> It would also be great to hear if the PCIe firmware from Rockchip aimed at the OOT driver was definitively incompatible with the mainline driver and we have to find a solution for that from Broadcom, or it is expected to basically work.  I realize it might not be easy to get an answer internally but this chipset is likely to appear in more RK3588 SBCs soon since it's the one on the evb.
>>
>> -Andy
>>
>>> Shengyu

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Almost same, except I'm using sdio driver (radxa rock 3a+their a6 module (AP6275S)). And I'm using mainline 5.19.8(self-compiled armbian edge) while I was testing.

Shengyu 

于 2022年11月13日 GMT+08:00 下午5:41:19, Andy Green <andy@warmcat.com> 写到:
>
>
>On 11/13/22 09:10, Shengyu Qu wrote:
>> Actually you can use any other clm info blob with this firmware as I tried. I also uploaded this to armbian's firmware repo. Hope we could get the proper firmware uploaded by Broadcom soon.
>
>Thank, I tried this, it did accept to download the clm part from the Rockchip SDK firmware.
>
># ls -l /lib/firmware/brcm
>-rw-r--r-- 1 root root  30993 Oct  3 08:40 clm_bcm43752a2_pcie_ag.blob
>-rw-r--r-- 1 root root 743607 Nov 13 08:45 fw_bcm43752a2_ag.bin
>lrwxrwxrwx 1 root root     46 Nov 13 09:19 fw_bcm43752a2_pcie.clm_blob -> /lib/firmware/brcm/clm_bcm43752a2_pcie_ag.blob
>lrwxrwxrwx 1 root root     20 Nov 13 08:38 fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin -> fw_bcm43752a2_ag.bin
>lrwxrwxrwx 1 root root     36 Nov 13 08:47 fw_bcm43752a2_pcie.txt -> /lib/firmware/brcm/nvram_ap6275s.txt
>-rw-r--r-- 1 root root   8709 Nov 13 08:44 nvram_ap6275s.txt
>
>But it's no happier after the NVRAM download part.
>
># dmesg | grep brcm
>[    3.530116] brcmfmac: brcmfmac_module_init No platform data available.
>[    3.530303] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
>[    3.530362] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
>[    3.531325] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 base 0x18000000 wrap 0x18100000
>[    3.531329] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 base 0x18001000 wrap 0x18101000
>[    3.531334] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 base 0x18002000 wrap 0x18102000
>[    3.531338] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 base 0x18003000 wrap 0x18103000
>[    3.531342] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 base 0x18004000 wrap 0x18104000
>[    3.531346] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 base 0x18005000 wrap 0x18105000
>[    3.531350] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 base 0x18010000 wrap 0x00000000
>[    3.531354] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 base 0x18012000 wrap 0x00000000
>[    3.531358] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 base 0x00000000 wrap 0x1810a000
>[    3.531362] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 base 0x00000000 wrap 0x18112000
>[    3.531365] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 base 0x00000000 wrap 0x18113000
>[    3.531369] brcmfmac: brcmf_chip_set_passive Enter
>[    3.641635] brcmfmac: brcmf_chip_set_passive Enter
>[    3.642156] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
>[    3.642159] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 size=1310720 (0x140000) sr=0 (0x0)
>[    3.642187] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, pmucaps=0x84565f23
>[    3.642193] brcmfmac: brcmf_get_module_param Enter, bus=2, chip=43752, rev=2
>[    3.642207] brcmfmac: brcmf_alloc Enter
>[    3.642231] brcmfmac: brcmf_fw_alloc_request: using brcm/fw_bcm43752a2_pcie for chip BCM43752/2
>[    3.642235] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
>[    3.642241] brcmfmac: brcm_alt_fw_path FW alt path: brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
>[    3.646995] brcmfmac: brcmf_fw_complete_request firmware brcm/fw_bcm43752a2_pcie.bin found
>[    3.647006] brcmfmac: brcm_alt_fw_path FW alt path: brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
>[    3.647308] brcmfmac: brcmf_fw_complete_request firmware brcm/fw_bcm43752a2_pcie.txt found
>[    3.647312] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
>[    3.647484] brcmfmac: brcmf_fw_request_nvram_done nvram 00000000c4269a30 len 5968
>[    3.647491] brcmfmac: brcm_alt_fw_path FW alt path: brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
>[    3.647911] brcmfmac: brcmf_fw_complete_request firmware brcm/fw_bcm43752a2_pcie.clm_blob found
>[    3.648197] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
>[    3.648200] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 size=1310720 (0x140000) sr=0 (0x0)
>[    3.723260] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
>[    3.723895] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 0xfa2c05d3
>[    3.723899] brcmfmac: brcmf_chip_set_active Enter
>[    9.721628] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW failed to initialize: 0xfa2c05d3
>[    9.721657] brcmfmac: brcmf_detach Enter
>[    9.745615] brcmfmac: brcmf_bus_change_state 0 -> 0
>[    9.745623] brcmfmac: brcmf_proto_detach Enter
>
>CONFIG_BRCMUTIL=m
># CONFIG_BRCMSMAC is not set
>CONFIG_BRCMFMAC=m
># CONFIG_BRCMFMAC_PROTO_MSGBUF is not set
># CONFIG_BRCMFMAC_SDIO is not set
># CONFIG_BRCMFMAC_USB is not set
>CONFIG_BRCMFMAC_PCIE=y
>CONFIG_BRCM_TRACING=y
>CONFIG_BRCMDBG=y
>
>I have the AP6275 module EN signal with the gpio wired to it as an "always-on" gpio regulator in DT, and don't use the hacky rfkill thing the rockchip SDK came with.
>
>It's on top of 6.1.0-rc4.
>
>Does your setup differ from this?
>
>-Andy
>
>> Shengyu
>> 
>> 于 2022年11月13日 GMT+08:00 下午5:03:06, Andy Green <andy@warmcat.com> 写到:
>>> 
>>> 
>>> On 11/13/22 08:26, Shengyu Qu wrote:
>>> 
>>>> Hello, I'm facing the same problem on sdio version of AP6275 module in mainline kernel. But I found a very old version from here could work with mainline brcmfmac driver.
>>>> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
>>> 
>>> Thanks for the info.
>>> 
>>> The firmware on the firefly thread doesn't seem to have a CLM blob that the mainline driver wants, just two types of the ~900KB main firmware for AP mode support or not.
>>> 
>>> It also doesn't have _sdio or _pcie in the filenames... I tried it but although it's different in what it read after downloading NVRAM
>>> 
>>> [    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 0xfa2c05d3
>>> 
>>> (the rockchip SDK firmware says 0xfa1a05e5)
>>> 
>>> it still times it out a few seconds later as unresponsive after the fw load... maybe it is for SDIO only, or the lack of CLM, present in the Rockchip fw, means it won't get anywhere.
>>> 
>>>> Besides who I should ask to make proper firmware for it into linux-firmware repo?
>>> 
>>> It would also be great to hear if the PCIe firmware from Rockchip aimed at the OOT driver was definitively incompatible with the mainline driver and we have to find a solution for that from Broadcom, or it is expected to basically work.  I realize it might not be easy to get an answer internally but this chipset is likely to appear in more RK3588 SBCs soon since it's the one on the evb.
>>> 
>>> -Andy
>>> 
>>>> Shengyu

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1954 bytes --]

On November 13, 2022 10:03:11 AM Andy Green <andy@warmcat.com> wrote:

> On 11/13/22 08:26, Shengyu Qu wrote:
>
>> Hello, I'm facing the same problem on sdio version of AP6275 module in 
>> mainline kernel. But I found a very old version from here could work with 
>> mainline brcmfmac driver.
>> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
>
> Thanks for the info.
>
> The firmware on the firefly thread doesn't seem to have a CLM blob that
> the mainline driver wants, just two types of the ~900KB main firmware
> for AP mode support or not.

A separate CLM blob may not be necessary when it is built-in in firmware 
which can be overloaded by separate blob, but in that case it is optional.

> It also doesn't have _sdio or _pcie in the filenames... I tried it but
> although it's different in what it read after downloading NVRAM
>
> [    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr
> 0xfa2c05d3
>
> (the rockchip SDK firmware says 0xfa1a05e5)
>
> it still times it out a few seconds later as unresponsive after the fw
> load... maybe it is for SDIO only, or the lack of CLM, present in the
> Rockchip fw, means it won't get anywhere.

That sharedram_addr looks wrong. Probably the firmware is not booting for 
some reason. Can you provide a URL where I can download the rockchip 
firmware? My internet search skills are lacking.

>
>> Besides who I should ask to make proper firmware for it into linux-firmware 
>> repo?
>
> It would also be great to hear if the PCIe firmware from Rockchip aimed
> at the OOT driver was definitively incompatible with the mainline driver
> and we have to find a solution for that from Broadcom, or it is expected
> to basically work.  I realize it might not be easy to get an answer
> internally but this chipset is likely to appear in more RK3588 SBCs soon
> since it's the one on the evb.

Right now I don't have enough information to comment on this.

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

在 2022/11/13 18:56, Arend Van Spriel 写道:
> On November 13, 2022 10:03:11 AM Andy Green <andy@warmcat.com> wrote:
>
>> On 11/13/22 08:26, Shengyu Qu wrote:
>>
>>> Hello, I'm facing the same problem on sdio version of AP6275 module 
>>> in mainline kernel. But I found a very old version from here could 
>>> work with mainline brcmfmac driver.
>>> Link: https://bbs.t-firefly.com/forum.php?mod=viewthread&tid=3107
>>
>> Thanks for the info.
>>
>> The firmware on the firefly thread doesn't seem to have a CLM blob that
>> the mainline driver wants, just two types of the ~900KB main firmware
>> for AP mode support or not.
>
> A separate CLM blob may not be necessary when it is built-in in 
> firmware which can be overloaded by separate blob, but in that case it 
> is optional.
>
>> It also doesn't have _sdio or _pcie in the filenames... I tried it but
>> although it's different in what it read after downloading NVRAM
>>
>> [    3.775602] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr
>> 0xfa2c05d3
>>
>> (the rockchip SDK firmware says 0xfa1a05e5)
>>
>> it still times it out a few seconds later as unresponsive after the fw
>> load... maybe it is for SDIO only, or the lack of CLM, present in the
>> Rockchip fw, means it won't get anywhere.
>
> That sharedram_addr looks wrong. Probably the firmware is not booting 
> for some reason. Can you provide a URL where I can download the 
> rockchip firmware? My internet search skills are lacking.
>
That error report happens when using sdio version firmware for pcie 
card. If you want a new version of rockchip firmware, check this: 
https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10/firmware/broadcom

Currently seems only firmware from firefly bbs above works with SDIO 
version and mainline now. And no known version PCIE firmware can work 
with mainline driver.

>>
>>> Besides who I should ask to make proper firmware for it into 
>>> linux-firmware repo?
>>
>> It would also be great to hear if the PCIe firmware from Rockchip aimed
>> at the OOT driver was definitively incompatible with the mainline driver
>> and we have to find a solution for that from Broadcom, or it is expected
>> to basically work.  I realize it might not be easy to get an answer
>> internally but this chipset is likely to appear in more RK3588 SBCs soon
>> since it's the one on the evb.
>
> Right now I don't have enough information to comment on this.
Besides, could you upload the firmware that works with mainline driver 
to linux-firmware repo?
>
> Regards,
> Arend
>
>

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1098 bytes --]

On November 13, 2022 4:52:45 PM Shengyu Qu <wiagn233@outlook.com> wrote:

> 在 2022/11/13 18:56, Arend Van Spriel 写道:

[snip]

>
>
>>>
>>>> Besides who I should ask to make proper firmware for it into
>>>> linux-firmware repo?
>>>
>>> It would also be great to hear if the PCIe firmware from Rockchip aimed
>>> at the OOT driver was definitively incompatible with the mainline driver
>>> and we have to find a solution for that from Broadcom, or it is expected
>>> to basically work.  I realize it might not be easy to get an answer
>>> internally but this chipset is likely to appear in more RK3588 SBCs soon
>>> since it's the one on the evb.
>>
>> Right now I don't have enough information to comment on this.
> Besides, could you upload the firmware that works with mainline driver
> to linux-firmware repo?

I honestly don't know what entity provides the ampak wifi module. I guess 
it is coming from Synaptics, but I am not sure. I rather fix the mainline 
driver to make it work with the firmware (if possible) instead of releasing 
a specific firmware for mainline.

Regards,
Arend




[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/13/22 17:35, Arend Van Spriel wrote:
> On November 13, 2022 4:52:45 PM Shengyu Qu <wiagn233@outlook.com> wrote:
> 
>> 在 2022/11/13 18:56, Arend Van Spriel 写道:

>>>> to basically work.  I realize it might not be easy to get an answer
>>>> internally but this chipset is likely to appear in more RK3588 SBCs 
>>>> soon
>>>> since it's the one on the evb.
>>>
>>> Right now I don't have enough information to comment on this.
>> Besides, could you upload the firmware that works with mainline driver
>> to linux-firmware repo?
> 
> I honestly don't know what entity provides the ampak wifi module. I 
> guess it is coming from Synaptics, but I am not sure. I rather fix the 
> mainline driver to make it work with the firmware (if possible) instead 
> of releasing a specific firmware for mainline.

That'd be great... I have some amount of time to help with this if 
something useful I can do for pcie case.

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Got a new version of downstream driver and firmware here: https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4

于 2022年11月14日 GMT+08:00 上午1:35:49, Arend Van Spriel <arend.vanspriel@broadcom.com> 写到:
>On November 13, 2022 4:52:45 PM Shengyu Qu <wiagn233@outlook.com> wrote:
>
>> 在 2022/11/13 18:56, Arend Van Spriel 写道:
>
>[snip]
>
>> 
>> 
>>>> 
>>>>> Besides who I should ask to make proper firmware for it into
>>>>> linux-firmware repo?
>>>> 
>>>> It would also be great to hear if the PCIe firmware from Rockchip aimed
>>>> at the OOT driver was definitively incompatible with the mainline driver
>>>> and we have to find a solution for that from Broadcom, or it is expected
>>>> to basically work.  I realize it might not be easy to get an answer
>>>> internally but this chipset is likely to appear in more RK3588 SBCs soon
>>>> since it's the one on the evb.
>>> 
>>> Right now I don't have enough information to comment on this.
>> Besides, could you upload the firmware that works with mainline driver
>> to linux-firmware repo?
>
>I honestly don't know what entity provides the ampak wifi module. I guess it is coming from Synaptics, but I am not sure. I rather fix the mainline driver to make it work with the firmware (if possible) instead of releasing a specific firmware for mainline.
>
>Regards,
>Arend
>
>
>

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/16/22 12:25, Shengyu Qu wrote:

> Got a new version of downstream driver and firmware here: https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4

The AP6275 PCIE ones there are the firmware files from the rockchip Sept 
SDK drop that I started with... they give the behaviour that the 
sharedram address probed from the device is

[    3.744635] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa1a05e5

and shortly after sending the nvram data

[    9.876580] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
failed to initialize: 0xfa1a05e5

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 2164 bytes --]

On 11/16/2022 1:25 PM, Shengyu Qu wrote:
> Got a new version of downstream driver and firmware here: https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4

Thanks for that. The firmware indeed seems to originate from Synaptics 
and this module can be found on wifi-alliance product finder:

https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275

We have no known contacts there and they do not seem to have much 
interest in open-source. In that gitlab repo I see they are providing 
bcmdhd driver and I assume they consider that sufficient.

Will see what I can do. First have to think on how to approach this 
failure to get most out of a debug effort. I do not have the module over 
here so we would need to go back and forth. The fact that the issue is 
seen on both SDIO and PCIe is somewhat encouraging.

Regards,
Arend

> 于 2022年11月14日 GMT+08:00 上午1:35:49, Arend Van Spriel <arend.vanspriel@broadcom.com> 写到:
>> On November 13, 2022 4:52:45 PM Shengyu Qu <wiagn233@outlook.com> wrote:
>>
>>> 在 2022/11/13 18:56, Arend Van Spriel 写道:
>>
>> [snip]
>>
>>>
>>>
>>>>>
>>>>>> Besides who I should ask to make proper firmware for it into
>>>>>> linux-firmware repo?
>>>>>
>>>>> It would also be great to hear if the PCIe firmware from Rockchip aimed
>>>>> at the OOT driver was definitively incompatible with the mainline driver
>>>>> and we have to find a solution for that from Broadcom, or it is expected
>>>>> to basically work.  I realize it might not be easy to get an answer
>>>>> internally but this chipset is likely to appear in more RK3588 SBCs soon
>>>>> since it's the one on the evb.
>>>>
>>>> Right now I don't have enough information to comment on this.
>>> Besides, could you upload the firmware that works with mainline driver
>>> to linux-firmware repo?
>>
>> I honestly don't know what entity provides the ampak wifi module. I guess it is coming from Synaptics, but I am not sure. I rather fix the mainline driver to make it work with the firmware (if possible) instead of releasing a specific firmware for mainline.
>>
>> Regards,
>> Arend
>>
>>
>>

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/17/22 09:23, Arend van Spriel wrote:
> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>> Got a new version of downstream driver and firmware here: 
>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
> 
> Thanks for that. The firmware indeed seems to originate from Synaptics 
> and this module can be found on wifi-alliance product finder:
> 
> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
> 
> We have no known contacts there and they do not seem to have much 
> interest in open-source. In that gitlab repo I see they are providing 
> bcmdhd driver and I assume they consider that sufficient.
> 
> Will see what I can do. First have to think on how to approach this 
> failure to get most out of a debug effort. I do not have the module over 
> here so we would need to go back and forth. The fact that the issue is 
> seen on both SDIO and PCIe is somewhat encouraging.

Yes it seems that at some point with some firmware, SDIO has worked for, 
eg, Angus.  And, it should work well with bchdhd (but that does not seem 
to work for scan for me on v6.1-rc5 uplevel with the firmware load 
changed to use kernel apis; but the firmware has no timeout on bcmdhd). 
So it doesn't feel a million miles away.

I have the hw, and (paid) time to help with debugging this "pair of 
hands" style. I'm also a little familiar with the kernel code and can 
try jiggle things autonomously if you can prompt what is interesting.

-Andy

> Regards,
> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Angus Ainslie]

On 2022-11-17 04:01, Andy Green wrote:
> On 11/17/22 09:23, Arend van Spriel wrote:
>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>> Got a new version of downstream driver and firmware here: 
>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>> 
>> Thanks for that. The firmware indeed seems to originate from Synaptics 
>> and this module can be found on wifi-alliance product finder:
>> 
>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>> 
>> We have no known contacts there and they do not seem to have much 
>> interest in open-source. In that gitlab repo I see they are providing 
>> bcmdhd driver and I assume they consider that sufficient.
>> 
>> Will see what I can do. First have to think on how to approach this 
>> failure to get most out of a debug effort. I do not have the module 
>> over here so we would need to go back and forth. The fact that the 
>> issue is seen on both SDIO and PCIe is somewhat encouraging.
> 
> Yes it seems that at some point with some firmware, SDIO has worked
> for, eg, Angus.  And, it should work well with bchdhd (but that does
> not seem to work for scan for me on v6.1-rc5 uplevel with the firmware
> load changed to use kernel apis; but the firmware has no timeout on
> bcmdhd). So it doesn't feel a million miles away.
> 

This is the firmware we have been using for SDIO. It was sent to us 
directly by SparkLAN so not sure how it compares to other variants.

https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm

Angus

> I have the hw, and (paid) time to help with debugging this "pair of
> hands" style. I'm also a little familiar with the kernel code and can
> try jiggle things autonomously if you can prompt what is interesting.
> 
> -Andy
> 
>> Regards,
>> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 2179 bytes --]

On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> wrote:

> On 2022-11-17 04:01, Andy Green wrote:
>> On 11/17/22 09:23, Arend van Spriel wrote:
>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>> Got a new version of downstream driver and firmware here:
>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>
>>> Thanks for that. The firmware indeed seems to originate from Synaptics
>>> and this module can be found on wifi-alliance product finder:
>>>
>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>
>>> We have no known contacts there and they do not seem to have much
>>> interest in open-source. In that gitlab repo I see they are providing
>>> bcmdhd driver and I assume they consider that sufficient.
>>>
>>> Will see what I can do. First have to think on how to approach this
>>> failure to get most out of a debug effort. I do not have the module
>>> over here so we would need to go back and forth. The fact that the
>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>
>> Yes it seems that at some point with some firmware, SDIO has worked
>> for, eg, Angus.  And, it should work well with bchdhd (but that does
>> not seem to work for scan for me on v6.1-rc5 uplevel with the firmware
>> load changed to use kernel apis; but the firmware has no timeout on
>> bcmdhd). So it doesn't feel a million miles away.
>
> This is the firmware we have been using for SDIO. It was sent to us
> directly by SparkLAN so not sure how it compares to other variants.
>
> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm

It is a tad bit older:

$ strings brcmfmac43752-sdio.bin | grep FW-TAG
<FW-TAG>4362a2-roml/config_sdio_ampak_release_he Version=18.35.387.23.7 
(b328500@shgit) (wlan=r880297) Date=2020-07-13T10:34:00Z FWID=01-f50fdff0

Thanks,
Arend

>
> Angus
>
>> I have the hw, and (paid) time to help with debugging this "pair of
>> hands" style. I'm also a little familiar with the kernel code and can
>> try jiggle things autonomously if you can prompt what is interesting.
>>
>> -Andy
>>
>>> Regards,
>>> Arend




[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Angus Ainslie]

On 2022-11-17 10:12, Arend Van Spriel wrote:
> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> wrote:
> 
>> On 2022-11-17 04:01, Andy Green wrote:
>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>> Got a new version of downstream driver and firmware here:
>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>> 
>>>> Thanks for that. The firmware indeed seems to originate from 
>>>> Synaptics
>>>> and this module can be found on wifi-alliance product finder:
>>>> 
>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>> 
>>>> We have no known contacts there and they do not seem to have much
>>>> interest in open-source. In that gitlab repo I see they are 
>>>> providing
>>>> bcmdhd driver and I assume they consider that sufficient.
>>>> 
>>>> Will see what I can do. First have to think on how to approach this
>>>> failure to get most out of a debug effort. I do not have the module
>>>> over here so we would need to go back and forth. The fact that the
>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>> 
>>> Yes it seems that at some point with some firmware, SDIO has worked
>>> for, eg, Angus.  And, it should work well with bchdhd (but that does
>>> not seem to work for scan for me on v6.1-rc5 uplevel with the 
>>> firmware
>>> load changed to use kernel apis; but the firmware has no timeout on
>>> bcmdhd). So it doesn't feel a million miles away.
>> 
>> This is the firmware we have been using for SDIO. It was sent to us
>> directly by SparkLAN so not sure how it compares to other variants.
>> 
>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
> 
> It is a tad bit older:
> 
> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
> 

Currently it's broken for hotspot mode but I haven't had the chance to 
see if that is FW or something further up the stack.

Angus

> Thanks,
> Arend
> 
>> 
>> Angus
>> 
>>> I have the hw, and (paid) time to help with debugging this "pair of
>>> hands" style. I'm also a little familiar with the kernel code and can
>>> try jiggle things autonomously if you can prompt what is interesting.
>>> 
>>> -Andy
>>> 
>>>> Regards,
>>>> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 2713 bytes --]

On 11/17/2022 7:36 PM, Angus Ainslie wrote:
> On 2022-11-17 10:12, Arend Van Spriel wrote:
>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> wrote:
>>
>>> On 2022-11-17 04:01, Andy Green wrote:
>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>> Got a new version of downstream driver and firmware here:
>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>
>>>>> Thanks for that. The firmware indeed seems to originate from Synaptics
>>>>> and this module can be found on wifi-alliance product finder:
>>>>>
>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>
>>>>> We have no known contacts there and they do not seem to have much
>>>>> interest in open-source. In that gitlab repo I see they are providing
>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>
>>>>> Will see what I can do. First have to think on how to approach this
>>>>> failure to get most out of a debug effort. I do not have the module
>>>>> over here so we would need to go back and forth. The fact that the
>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>
>>>> Yes it seems that at some point with some firmware, SDIO has worked
>>>> for, eg, Angus.  And, it should work well with bchdhd (but that does
>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the firmware
>>>> load changed to use kernel apis; but the firmware has no timeout on
>>>> bcmdhd). So it doesn't feel a million miles away.
>>>
>>> This is the firmware we have been using for SDIO. It was sent to us
>>> directly by SparkLAN so not sure how it compares to other variants.
>>>
>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>
>> It is a tad bit older:
>>
>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>
> 
> Currently it's broken for hotspot mode but I haven't had the chance to 
> see if that is FW or something further up the stack.

Hi Angus,

When you added support for this device in brcmfmac you marked it as a 
Cypress device. Are you sure about that? AFAICT this device is coming 
from Synaptics.

Regards,
Arend

> Angus
> 
>> Thanks,
>> Arend
>>
>>>
>>> Angus
>>>
>>>> I have the hw, and (paid) time to help with debugging this "pair of
>>>> hands" style. I'm also a little familiar with the kernel code and can
>>>> try jiggle things autonomously if you can prompt what is interesting.
>>>>
>>>> -Andy
>>>>
>>>>> Regards,
>>>>> Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Angus Ainslie]

On 2022-11-29 11:48, Arend van Spriel wrote:
> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> wrote:
>>> 
>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>> 
>>>>>> Thanks for that. The firmware indeed seems to originate from 
>>>>>> Synaptics
>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>> 
>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>> 
>>>>>> We have no known contacts there and they do not seem to have much
>>>>>> interest in open-source. In that gitlab repo I see they are 
>>>>>> providing
>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>> 
>>>>>> Will see what I can do. First have to think on how to approach 
>>>>>> this
>>>>>> failure to get most out of a debug effort. I do not have the 
>>>>>> module
>>>>>> over here so we would need to go back and forth. The fact that the
>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>> 
>>>>> Yes it seems that at some point with some firmware, SDIO has worked
>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that 
>>>>> does
>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the 
>>>>> firmware
>>>>> load changed to use kernel apis; but the firmware has no timeout on
>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>> 
>>>> This is the firmware we have been using for SDIO. It was sent to us
>>>> directly by SparkLAN so not sure how it compares to other variants.
>>>> 
>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>> 
>>> It is a tad bit older:
>>> 
>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>> 
>> 
>> Currently it's broken for hotspot mode but I haven't had the chance to 
>> see if that is FW or something further up the stack.
> 
> Hi Angus,
> 
> When you added support for this device in brcmfmac you marked it as a
> Cypress device. Are you sure about that? AFAICT this device is coming
> from Synaptics.
> 

Hi Arend,

I assumed Cypress based on the fact that it most closely resembled the 
4373 from the driver side.

How would I figure out whether it's Cypress or Synaptics ?

When I search the web for it I only find SparkLAN or "AMPAK technology" 
references.

Thanks
Angus


> Regards,
> Arend
> 
>> Angus
>> 
>>> Thanks,
>>> Arend
>>> 
>>>> 
>>>> Angus
>>>> 
>>>>> I have the hw, and (paid) time to help with debugging this "pair of
>>>>> hands" style. I'm also a little familiar with the kernel code and 
>>>>> can
>>>>> try jiggle things autonomously if you can prompt what is 
>>>>> interesting.
>>>>> 
>>>>> -Andy
>>>>> 
>>>>>> Regards,
>>>>>> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 3352 bytes --]

On December 3, 2022 2:56:39 PM Angus Ainslie <angus@akkea.ca> wrote:

> On 2022-11-29 11:48, Arend van Spriel wrote:
>> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> wrote:
>>>>
>>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>>>
>>>>>>> Thanks for that. The firmware indeed seems to originate from
>>>>>>> Synaptics
>>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>>>
>>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>>>
>>>>>>> We have no known contacts there and they do not seem to have much
>>>>>>> interest in open-source. In that gitlab repo I see they are
>>>>>>> providing
>>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>>>
>>>>>>> Will see what I can do. First have to think on how to approach
>>>>>>> this
>>>>>>> failure to get most out of a debug effort. I do not have the
>>>>>>> module
>>>>>>> over here so we would need to go back and forth. The fact that the
>>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>>>
>>>>>> Yes it seems that at some point with some firmware, SDIO has worked
>>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that
>>>>>> does
>>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the
>>>>>> firmware
>>>>>> load changed to use kernel apis; but the firmware has no timeout on
>>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>>>
>>>>> This is the firmware we have been using for SDIO. It was sent to us
>>>>> directly by SparkLAN so not sure how it compares to other variants.
>>>>>
>>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>>>
>>>> It is a tad bit older:
>>>>
>>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>>
>>> Currently it's broken for hotspot mode but I haven't had the chance to
>>> see if that is FW or something further up the stack.
>>
>> Hi Angus,
>>
>> When you added support for this device in brcmfmac you marked it as a
>> Cypress device. Are you sure about that? AFAICT this device is coming
>> from Synaptics.
>
> Hi Arend,
>
> I assumed Cypress based on the fact that it most closely resembled the
> 4373 from the driver side.

Right. Both are based on the same chip architecture that they brought from 
Broadcom so the similarities are not surprising.

> How would I figure out whether it's Cypress or Synaptics ?
>
> When I search the web for it I only find SparkLAN or "AMPAK technology"
> references.

Earlier my colleague Franky found this link to WFA certificate:

https://api.cert.wi-fi.org/api/certificate/download/public?variantId=103641

WFA and FCC are usually good sources of info. Also the product pages of the 
vendors are worth checking. I have them bookmarked out of professional 
interest ;-)

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Angus Ainslie]

On 2022-12-03 07:22, Arend Van Spriel wrote:
> On December 3, 2022 2:56:39 PM Angus Ainslie <angus@akkea.ca> wrote:
> 
>> On 2022-11-29 11:48, Arend van Spriel wrote:
>>> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>>>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca> 
>>>>> wrote:
>>>>> 
>>>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>>>> 
>>>>>>>> Thanks for that. The firmware indeed seems to originate from
>>>>>>>> Synaptics
>>>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>>>> 
>>>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>>>> 
>>>>>>>> We have no known contacts there and they do not seem to have 
>>>>>>>> much
>>>>>>>> interest in open-source. In that gitlab repo I see they are
>>>>>>>> providing
>>>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>>>> 
>>>>>>>> Will see what I can do. First have to think on how to approach
>>>>>>>> this
>>>>>>>> failure to get most out of a debug effort. I do not have the
>>>>>>>> module
>>>>>>>> over here so we would need to go back and forth. The fact that 
>>>>>>>> the
>>>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>>>> 
>>>>>>> Yes it seems that at some point with some firmware, SDIO has 
>>>>>>> worked
>>>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that
>>>>>>> does
>>>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the
>>>>>>> firmware
>>>>>>> load changed to use kernel apis; but the firmware has no timeout 
>>>>>>> on
>>>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>>>> 
>>>>>> This is the firmware we have been using for SDIO. It was sent to 
>>>>>> us
>>>>>> directly by SparkLAN so not sure how it compares to other 
>>>>>> variants.
>>>>>> 
>>>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>>>> 
>>>>> It is a tad bit older:
>>>>> 
>>>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>>> 
>>>> Currently it's broken for hotspot mode but I haven't had the chance 
>>>> to
>>>> see if that is FW or something further up the stack.
>>> 
>>> Hi Angus,
>>> 
>>> When you added support for this device in brcmfmac you marked it as a
>>> Cypress device. Are you sure about that? AFAICT this device is coming
>>> from Synaptics.
>> 
>> Hi Arend,
>> 
>> I assumed Cypress based on the fact that it most closely resembled the
>> 4373 from the driver side.
> 
> Right. Both are based on the same chip architecture that they brought
> from Broadcom so the similarities are not surprising.
> 
>> How would I figure out whether it's Cypress or Synaptics ?
>> 
>> When I search the web for it I only find SparkLAN or "AMPAK 
>> technology"
>> references.
> 
> Earlier my colleague Franky found this link to WFA certificate:
> 
> https://api.cert.wi-fi.org/api/certificate/download/public?variantId=103641
> 
> WFA and FCC are usually good sources of info. Also the product pages
> of the vendors are worth checking. I have them bookmarked out of
> professional interest ;-)
> 

I did find FCC references but those seemed ambiguous as well. Thanks for 
the wi-fi.org link.

Should I send a patch renaming SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752 -> 
SDIO_DEVICE_ID_BROADCOM_43752 ?

Thanks
Angus

> Regards,
> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 4002 bytes --]

On December 3, 2022 4:37:19 PM Angus Ainslie <angus@akkea.ca> wrote:

> On 2022-12-03 07:22, Arend Van Spriel wrote:
>> On December 3, 2022 2:56:39 PM Angus Ainslie <angus@akkea.ca> wrote:
>>
>>> On 2022-11-29 11:48, Arend van Spriel wrote:
>>>> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>>>>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>>>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca>
>>>>>> wrote:
>>>>>>
>>>>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>>>>>
>>>>>>>>> Thanks for that. The firmware indeed seems to originate from
>>>>>>>>> Synaptics
>>>>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>>>>>
>>>>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>>>>>
>>>>>>>>> We have no known contacts there and they do not seem to have
>>>>>>>>> much
>>>>>>>>> interest in open-source. In that gitlab repo I see they are
>>>>>>>>> providing
>>>>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>>>>>
>>>>>>>>> Will see what I can do. First have to think on how to approach
>>>>>>>>> this
>>>>>>>>> failure to get most out of a debug effort. I do not have the
>>>>>>>>> module
>>>>>>>>> over here so we would need to go back and forth. The fact that
>>>>>>>>> the
>>>>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>>>>>
>>>>>>>> Yes it seems that at some point with some firmware, SDIO has
>>>>>>>> worked
>>>>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that
>>>>>>>> does
>>>>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the
>>>>>>>> firmware
>>>>>>>> load changed to use kernel apis; but the firmware has no timeout
>>>>>>>> on
>>>>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>>>>>
>>>>>>> This is the firmware we have been using for SDIO. It was sent to
>>>>>>> us
>>>>>>> directly by SparkLAN so not sure how it compares to other
>>>>>>> variants.
>>>>>>>
>>>>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>>>>>
>>>>>> It is a tad bit older:
>>>>>>
>>>>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>>>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>>>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>>>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>>>>
>>>>> Currently it's broken for hotspot mode but I haven't had the chance
>>>>> to
>>>>> see if that is FW or something further up the stack.
>>>>
>>>> Hi Angus,
>>>>
>>>> When you added support for this device in brcmfmac you marked it as a
>>>> Cypress device. Are you sure about that? AFAICT this device is coming
>>>> from Synaptics.
>>>
>>> Hi Arend,
>>>
>>> I assumed Cypress based on the fact that it most closely resembled the
>>> 4373 from the driver side.
>>
>> Right. Both are based on the same chip architecture that they brought
>> from Broadcom so the similarities are not surprising.
>>
>>> How would I figure out whether it's Cypress or Synaptics ?
>>>
>>> When I search the web for it I only find SparkLAN or "AMPAK
>>> technology"
>>> references.
>>
>> Earlier my colleague Franky found this link to WFA certificate:
>>
>> https://api.cert.wi-fi.org/api/certificate/download/public?variantId=103641
>>
>> WFA and FCC are usually good sources of info. Also the product pages
>> of the vendors are worth checking. I have them bookmarked out of
>> professional interest ;-)
>
> I did find FCC references but those seemed ambiguous as well. Thanks for
> the wi-fi.org link.
>
> Should I send a patch renaming SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752 ->
> SDIO_DEVICE_ID_BROADCOM_43752 ?

Feel free to do so. I have some cleanup work queued up so I can also take 
care of it.

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Angus Ainslie]

On 2022-12-03 09:08, Arend Van Spriel wrote:
> On December 3, 2022 4:37:19 PM Angus Ainslie <angus@akkea.ca> wrote:
> 
>> On 2022-12-03 07:22, Arend Van Spriel wrote:
>>> On December 3, 2022 2:56:39 PM Angus Ainslie <angus@akkea.ca> wrote:
>>> 
>>>> On 2022-11-29 11:48, Arend van Spriel wrote:
>>>>> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>>>>>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>>>>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca>
>>>>>>> wrote:
>>>>>>> 
>>>>>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>>>>>> 
>>>>>>>>>> Thanks for that. The firmware indeed seems to originate from
>>>>>>>>>> Synaptics
>>>>>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>>>>>> 
>>>>>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>>>>>> 
>>>>>>>>>> We have no known contacts there and they do not seem to have
>>>>>>>>>> much
>>>>>>>>>> interest in open-source. In that gitlab repo I see they are
>>>>>>>>>> providing
>>>>>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>>>>>> 
>>>>>>>>>> Will see what I can do. First have to think on how to approach
>>>>>>>>>> this
>>>>>>>>>> failure to get most out of a debug effort. I do not have the
>>>>>>>>>> module
>>>>>>>>>> over here so we would need to go back and forth. The fact that
>>>>>>>>>> the
>>>>>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>>>>>> 
>>>>>>>>> Yes it seems that at some point with some firmware, SDIO has
>>>>>>>>> worked
>>>>>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that
>>>>>>>>> does
>>>>>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the
>>>>>>>>> firmware
>>>>>>>>> load changed to use kernel apis; but the firmware has no 
>>>>>>>>> timeout
>>>>>>>>> on
>>>>>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>>>>>> 
>>>>>>>> This is the firmware we have been using for SDIO. It was sent to
>>>>>>>> us
>>>>>>>> directly by SparkLAN so not sure how it compares to other
>>>>>>>> variants.
>>>>>>>> 
>>>>>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>>>>>> 
>>>>>>> It is a tad bit older:
>>>>>>> 
>>>>>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>>>>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>>>>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>>>>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>>>>> 
>>>>>> Currently it's broken for hotspot mode but I haven't had the 
>>>>>> chance
>>>>>> to
>>>>>> see if that is FW or something further up the stack.
>>>>> 
>>>>> Hi Angus,
>>>>> 
>>>>> When you added support for this device in brcmfmac you marked it as 
>>>>> a
>>>>> Cypress device. Are you sure about that? AFAICT this device is 
>>>>> coming
>>>>> from Synaptics.
>>>> 
>>>> Hi Arend,
>>>> 
>>>> I assumed Cypress based on the fact that it most closely resembled 
>>>> the
>>>> 4373 from the driver side.
>>> 
>>> Right. Both are based on the same chip architecture that they brought
>>> from Broadcom so the similarities are not surprising.
>>> 
>>>> How would I figure out whether it's Cypress or Synaptics ?
>>>> 
>>>> When I search the web for it I only find SparkLAN or "AMPAK
>>>> technology"
>>>> references.
>>> 
>>> Earlier my colleague Franky found this link to WFA certificate:
>>> 
>>> https://api.cert.wi-fi.org/api/certificate/download/public?variantId=103641
>>> 
>>> WFA and FCC are usually good sources of info. Also the product pages
>>> of the vendors are worth checking. I have them bookmarked out of
>>> professional interest ;-)
>> 
>> I did find FCC references but those seemed ambiguous as well. Thanks 
>> for
>> the wi-fi.org link.
>> 
>> Should I send a patch renaming SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752 
>> ->
>> SDIO_DEVICE_ID_BROADCOM_43752 ?
> 
> Feel free to do so. I have some cleanup work queued up so I can also
> take care of it.
> 

If you're already going to be in there it would be great if you could 
clean it up.

Thanks
Angus

> Regards,
> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 4478 bytes --]

On December 4, 2022 3:33:02 PM Angus Ainslie <angus@akkea.ca> wrote:

> On 2022-12-03 09:08, Arend Van Spriel wrote:
>> On December 3, 2022 4:37:19 PM Angus Ainslie <angus@akkea.ca> wrote:
>>
>>> On 2022-12-03 07:22, Arend Van Spriel wrote:
>>>> On December 3, 2022 2:56:39 PM Angus Ainslie <angus@akkea.ca> wrote:
>>>>
>>>>> On 2022-11-29 11:48, Arend van Spriel wrote:
>>>>>> On 11/17/2022 7:36 PM, Angus Ainslie wrote:
>>>>>>> On 2022-11-17 10:12, Arend Van Spriel wrote:
>>>>>>>> On November 17, 2022 5:06:25 PM Angus Ainslie <angus@akkea.ca>
>>>>>>>> wrote:
>>>>>>>>
>>>>>>>>> On 2022-11-17 04:01, Andy Green wrote:
>>>>>>>>>> On 11/17/22 09:23, Arend van Spriel wrote:
>>>>>>>>>>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>>>>>>>>>>> Got a new version of downstream driver and firmware here:
>>>>>>>>>>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>>>>>>>>>>
>>>>>>>>>>> Thanks for that. The firmware indeed seems to originate from
>>>>>>>>>>> Synaptics
>>>>>>>>>>> and this module can be found on wifi-alliance product finder:
>>>>>>>>>>>
>>>>>>>>>>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>>>>>>>>>>
>>>>>>>>>>> We have no known contacts there and they do not seem to have
>>>>>>>>>>> much
>>>>>>>>>>> interest in open-source. In that gitlab repo I see they are
>>>>>>>>>>> providing
>>>>>>>>>>> bcmdhd driver and I assume they consider that sufficient.
>>>>>>>>>>>
>>>>>>>>>>> Will see what I can do. First have to think on how to approach
>>>>>>>>>>> this
>>>>>>>>>>> failure to get most out of a debug effort. I do not have the
>>>>>>>>>>> module
>>>>>>>>>>> over here so we would need to go back and forth. The fact that
>>>>>>>>>>> the
>>>>>>>>>>> issue is seen on both SDIO and PCIe is somewhat encouraging.
>>>>>>>>>>
>>>>>>>>>> Yes it seems that at some point with some firmware, SDIO has
>>>>>>>>>> worked
>>>>>>>>>> for, eg, Angus.  And, it should work well with bchdhd (but that
>>>>>>>>>> does
>>>>>>>>>> not seem to work for scan for me on v6.1-rc5 uplevel with the
>>>>>>>>>> firmware
>>>>>>>>>> load changed to use kernel apis; but the firmware has no
>>>>>>>>>> timeout
>>>>>>>>>> on
>>>>>>>>>> bcmdhd). So it doesn't feel a million miles away.
>>>>>>>>>
>>>>>>>>> This is the firmware we have been using for SDIO. It was sent to
>>>>>>>>> us
>>>>>>>>> directly by SparkLAN so not sure how it compares to other
>>>>>>>>> variants.
>>>>>>>>>
>>>>>>>>> https://source.puri.sm/Librem5/firmware-brcm43752-nonfree/-/tree/master/firmware/brcm
>>>>>>>>
>>>>>>>> It is a tad bit older:
>>>>>>>>
>>>>>>>> $ strings brcmfmac43752-sdio.bin | grep FW-TAG
>>>>>>>> <FW-TAG>4362a2-roml/config_sdio_ampak_release_he
>>>>>>>> Version=18.35.387.23.7 (b328500@shgit) (wlan=r880297)
>>>>>>>> Date=2020-07-13T10:34:00Z FWID=01-f50fdff0
>>>>>>>
>>>>>>> Currently it's broken for hotspot mode but I haven't had the
>>>>>>> chance
>>>>>>> to
>>>>>>> see if that is FW or something further up the stack.
>>>>>>
>>>>>> Hi Angus,
>>>>>>
>>>>>> When you added support for this device in brcmfmac you marked it as
>>>>>> a
>>>>>> Cypress device. Are you sure about that? AFAICT this device is
>>>>>> coming
>>>>>> from Synaptics.
>>>>>
>>>>> Hi Arend,
>>>>>
>>>>> I assumed Cypress based on the fact that it most closely resembled
>>>>> the
>>>>> 4373 from the driver side.
>>>>
>>>> Right. Both are based on the same chip architecture that they brought
>>>> from Broadcom so the similarities are not surprising.
>>>>
>>>>> How would I figure out whether it's Cypress or Synaptics ?
>>>>>
>>>>> When I search the web for it I only find SparkLAN or "AMPAK
>>>>> technology"
>>>>> references.
>>>>
>>>> Earlier my colleague Franky found this link to WFA certificate:
>>>>
>>>> https://api.cert.wi-fi.org/api/certificate/download/public?variantId=103641
>>>>
>>>> WFA and FCC are usually good sources of info. Also the product pages
>>>> of the vendors are worth checking. I have them bookmarked out of
>>>> professional interest ;-)
>>>
>>> I did find FCC references but those seemed ambiguous as well. Thanks
>>> for
>>> the wi-fi.org link.
>>>
>>> Should I send a patch renaming SDIO_DEVICE_ID_BROADCOM_CYPRESS_43752
>>> ->
>>> SDIO_DEVICE_ID_BROADCOM_43752 ?
>>
>> Feel free to do so. I have some cleanup work queued up so I can also
>> take care of it.
>
> If you're already going to be in there it would be great if you could
> clean it up.

Sure. No problem.

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend Van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1584 bytes --]

On November 17, 2022 1:01:29 PM Andy Green <andy@warmcat.com> wrote:

> On 11/17/22 09:23, Arend van Spriel wrote:
>> On 11/16/2022 1:25 PM, Shengyu Qu wrote:
>>> Got a new version of downstream driver and firmware here:
>>> https://gitlab.com/rk3588_linux/linux/external/rkwifibt/-/tree/linux-5.10-gen-rkr3.4
>>
>> Thanks for that. The firmware indeed seems to originate from Synaptics
>> and this module can be found on wifi-alliance product finder:
>>
>> https://www.wi-fi.org/product-finder-results?sort_by=certified&sort_order=desc&keywords=ap6275
>>
>> We have no known contacts there and they do not seem to have much
>> interest in open-source. In that gitlab repo I see they are providing
>> bcmdhd driver and I assume they consider that sufficient.
>>
>> Will see what I can do. First have to think on how to approach this
>> failure to get most out of a debug effort. I do not have the module over
>> here so we would need to go back and forth. The fact that the issue is
>> seen on both SDIO and PCIe is somewhat encouraging.
>
> Yes it seems that at some point with some firmware, SDIO has worked for,
> eg, Angus.  And, it should work well with bchdhd (but that does not seem
> to work for scan for me on v6.1-rc5 uplevel with the firmware load
> changed to use kernel apis; but the firmware has no timeout on bcmdhd).
> So it doesn't feel a million miles away.

I see. So could you load bcmdhd and provide log. Preferably with higher 
debug level. It also has a module parameter for it. I think it is called 
dhd_msg_level and please set it to 0x817.

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/17/22 21:52, Arend Van Spriel wrote:

> I see. So could you load bcmdhd and provide log. Preferably with higher 
> debug level. It also has a module parameter for it. I think it is called 
> dhd_msg_level and please set it to 0x817.

It's here:

https://warmcat.com/ap6275p-dhd-log1.txt

after boot he shows a wlan0

wlan0: flags=-28669<UP,BROADCAST,MULTICAST,DYNAMIC>  mtu 1500
         inet6 fe80::d69c:ddff:fef5:bf7e  prefixlen 64  scopeid 0x20<link>
         ether d4:9c:dd:f5:bf:7e  txqueuelen 1000  (Ethernet)
         RX packets 0  bytes 0 (0.0 B)
         RX errors 0  dropped 0  overruns 0  frame 0
         TX packets 11  bytes 866 (866.0 B)
         TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

The dhd OOT driver comes with a rockchip-hacked rfkill driver which has 
all sorts of things in it related to wakelocks I removed, I built that 
too and rfkill reports

# rfkill
ID TYPE DEVICE           SOFT      HARD
  0 wlan phy0          blocked unblocked
  1 wlan brcmfmac-wifi blocked unblocked

"brcmfmac-wifi" is a literal in the dhd driver sources, confusingly. 
Mainline brcmfmac is not built or inserted as a module for this test.

# rfkill unblock

shows everything unblocked, but connmanctl can start scans but never get 
any results.  But it's happy with the firmware after hacking it to use 
updated apis.

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 2417 bytes --]

On 11/18/2022 6:41 AM, Andy Green wrote:
> 
> 
> On 11/17/22 21:52, Arend Van Spriel wrote:
> 
>> I see. So could you load bcmdhd and provide log. Preferably with 
>> higher debug level. It also has a module parameter for it. I think it 
>> is called dhd_msg_level and please set it to 0x817.
> 
> It's here:
> 
> https://warmcat.com/ap6275p-dhd-log1.txt

Hi Andy,

Thanks for the log. Together with the dhd driver sources I could make 
some sense of it. Perhaps you can try the following hack and see if we 
make it over the first hurdle.

Regards,
Arend
---
diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c 
b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
index 97f0f13dfe50..ba436d71a7ec 100644
--- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
+++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/pcie.c
@@ -11,6 +11,7 @@
  #include <linux/delay.h>
  #include <linux/interrupt.h>
  #include <linux/bcma/bcma.h>
+#include <linux/random.h>
  #include <linux/sched.h>
  #include <linux/io.h>
  #include <asm/unaligned.h>
@@ -1527,6 +1528,33 @@ brcmf_pcie_init_share_ram_info(struct 
brcmf_pciedev_info *devinfo,
  	return 0;
  }

+#define BRCMF_SEED_SIGNATURE		0xFEEDC0DE
+#define BRCMF_ENTROPY_HOST_SEED_LEN	128
+
+struct brcmf_entropy_metadata {
+	__le32 count;
+	__le32 signature;
+};
+
+static void brcmf_pcie_write_entropy_bytes(struct brcmf_pciedev_info 
*devinfo,
+					   u32 nvram_len)
+{
+	struct brcmf_entropy_metadata entropy_info;
+	u8 seed[BRCMF_ENTROPY_HOST_SEED_LEN] = { 0 };
+	u32 count = sizeof(seed);
+	u32 addr = devinfo->ci->rambase + devinfo->ci->ramsize - nvram_len;
+
+	entropy_info.signture = cpu_to_le32(BRCMF_SEED_SIGNATURE);
+	entropy_info.count = cpu_to_le32(BRCMF_ENTROPY_HOST_SEED_LEN);
+
+	addr -= sizeof(entropy_info);
+	memcpy_toio(devinfo->tcm + addr, &entropy_info, sizeof(entropy_info));
+
+	get_random_bytes(seed, count);
+
+	addr -= sizeof(count);
+	memcpy_toio(devinfo->tcm + addr, seed, count);
+}

  static int brcmf_pcie_download_fw_nvram(struct brcmf_pciedev_info 
*devinfo,
  					const struct firmware *fw, void *nvram,
@@ -1568,6 +1596,8 @@ static int brcmf_pcie_download_fw_nvram(struct 
brcmf_pciedev_info *devinfo,
  			  devinfo->nvram_name);
  	}

+	brcmf_pcie_write_entropy_bytes(devinfo, nvram_len);
+
  	sharedram_addr_written = brcmf_pcie_read_ram32(devinfo,
  						       devinfo->ci->ramsize -
  						       4);

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 09:42, Arend van Spriel wrote:
> On 11/18/2022 6:41 AM, Andy Green wrote:
>>
>>
>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>
>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>> higher debug level. It also has a module parameter for it. I think it 
>>> is called dhd_msg_level and please set it to 0x817.
>>
>> It's here:
>>
>> https://warmcat.com/ap6275p-dhd-log1.txt
> 
> Hi Andy,
> 
> Thanks for the log. Together with the dhd driver sources I could make 
> some sense of it. Perhaps you can try the following hack and see if we 
> make it over the first hurdle.

Thanks, I tried it, but although it looks necessary, there's evidently 
something else since the result was the same.

> +
> +    entropy_info.signture = cpu_to_le32(BRCMF_SEED_SIGNATURE);

This should be ... .signature = ...

The debug logs

[    3.487772] brcmfmac: brcmfmac_module_init No platform data available.
[    3.487858] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
[    3.487896] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
[    3.488799] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 
base 0x18000000 wrap 0x18100000
[    3.488801] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 
base 0x18001000 wrap 0x18101000
[    3.488803] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 
base 0x18002000 wrap 0x18102000
[    3.488804] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 
base 0x18003000 wrap 0x18103000
[    3.488805] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 
base 0x18004000 wrap 0x18104000
[    3.488807] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 
base 0x18005000 wrap 0x18105000
[    3.488808] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 
base 0x18010000 wrap 0x00000000
[    3.488810] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 
base 0x18012000 wrap 0x00000000
[    3.488811] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 
base 0x00000000 wrap 0x1810a000
[    3.488813] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 
base 0x00000000 wrap 0x18112000
[    3.488814] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 
base 0x00000000 wrap 0x18113000
[    3.488815] brcmfmac: brcmf_chip_set_passive Enter
[    3.594733] brcmfmac: brcmf_chip_set_passive Enter
[    3.595214] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.595215] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.595238] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, 
pmucaps=0x84565f23
[    3.595241] brcmfmac: brcmf_get_module_param Enter, bus=2, 
chip=43752, rev=2
[    3.595246] brcmfmac: brcmf_alloc Enter
[    3.595254] brcmfmac: brcmf_fw_alloc_request: using 
brcm/fw_bcm43752a2_pcie for chip BCM43752/2
[    3.595257] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
[    3.595259] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
[    3.595555] brcmfmac 0002:21:00.0: Direct firmware load for 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin failed with error -2
[    3.600006] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.bin found
[    3.600010] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
[    3.600323] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.txt found
[    3.600324] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
[    3.600378] brcmfmac: brcmf_fw_request_nvram_done nvram 
00000000eea1e0fd len 6040
[    3.600381] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
[    3.600641] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.clm_blob found
[    3.600836] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.600837] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
[    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 6040 
entropy bytes
[    3.694910] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa1a05e5
[    3.694911] brcmfmac: brcmf_chip_set_active Enter
[    9.774716] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
failed to initialize: 0xfa1a05e5
[    9.774733] brcmfmac: brcmf_detach Enter
[    9.798714] brcmfmac: brcmf_bus_change_state 0 -> 0
[    9.798716] brcmfmac: brcmf_proto_detach Enter

I added a log showing it sending the entropy bytes.

Firmware situation

# ls -l /lib/firmware/brcm/
total 956
-rw-r--r-- 1 root root  30993 Nov 16 12:31 clm_bcm43752a2_pcie_ag.blob
lrwxrwxrwx 1 root root     44 Nov 24 10:00 fw_bcm43752a2_pcie.bin -> 
/lib/firmware/brcm/fw_bcm43752a2_pcie_ag.bin
lrwxrwxrwx 1 root root     46 Nov 24 10:02 fw_bcm43752a2_pcie.clm_blob 
-> /lib/firmware/brcm/clm_bcm43752a2_pcie_ag.blob
lrwxrwxrwx 1 root root     36 Nov 24 10:02 fw_bcm43752a2_pcie.txt -> 
/lib/firmware/brcm/nvram_AP6275P.txt
-rw-r--r-- 1 root root 936074 Nov 16 12:31 fw_bcm43752a2_pcie_ag.bin
-rw-r--r-- 1 root root   7458 Nov 16 12:31 nvram_AP6275P.txt

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 5047 bytes --]

On 11/24/2022 11:20 AM, Andy Green wrote:
> 
> 
> On 11/24/22 09:42, Arend van Spriel wrote:
>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>
>>>
>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>
>>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>>> higher debug level. It also has a module parameter for it. I think 
>>>> it is called dhd_msg_level and please set it to 0x817.
>>>
>>> It's here:
>>>
>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>
>> Hi Andy,
>>
>> Thanks for the log. Together with the dhd driver sources I could make 
>> some sense of it. Perhaps you can try the following hack and see if we 
>> make it over the first hurdle.
> 
> Thanks, I tried it, but although it looks necessary, there's evidently 
> something else since the result was the same.
> 
>> +
>> +    entropy_info.signture = cpu_to_le32(BRCMF_SEED_SIGNATURE);
> 
> This should be ... .signature = ...

That's what you get if you get too confident... :-(

> The debug logs
> 
> [    3.487772] brcmfmac: brcmfmac_module_init No platform data available.
> [    3.487858] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
> [    3.487896] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
> [    3.488799] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 base 
> 0x18000000 wrap 0x18100000
> [    3.488801] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 base 
> 0x18001000 wrap 0x18101000
> [    3.488803] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 base 
> 0x18002000 wrap 0x18102000
> [    3.488804] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 base 
> 0x18003000 wrap 0x18103000
> [    3.488805] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 base 
> 0x18004000 wrap 0x18104000
> [    3.488807] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 base 
> 0x18005000 wrap 0x18105000
> [    3.488808] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 base 
> 0x18010000 wrap 0x00000000
> [    3.488810] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 base 
> 0x18012000 wrap 0x00000000
> [    3.488811] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 base 
> 0x00000000 wrap 0x1810a000
> [    3.488813] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 base 
> 0x00000000 wrap 0x18112000
> [    3.488814] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 base 
> 0x00000000 wrap 0x18113000
> [    3.488815] brcmfmac: brcmf_chip_set_passive Enter
> [    3.594733] brcmfmac: brcmf_chip_set_passive Enter
> [    3.595214] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
> [    3.595215] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
> size=1310720 (0x140000) sr=0 (0x0)
> [    3.595238] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, 
> pmucaps=0x84565f23
> [    3.595241] brcmfmac: brcmf_get_module_param Enter, bus=2, 
> chip=43752, rev=2
> [    3.595246] brcmfmac: brcmf_alloc Enter
> [    3.595254] brcmfmac: brcmf_fw_alloc_request: using 
> brcm/fw_bcm43752a2_pcie for chip BCM43752/2
> [    3.595257] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
> [    3.595259] brcmfmac: brcm_alt_fw_path FW alt path: 
> brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
> [    3.595555] brcmfmac 0002:21:00.0: Direct firmware load for 
> brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin failed with error -2
> [    3.600006] brcmfmac: brcmf_fw_complete_request firmware 
> brcm/fw_bcm43752a2_pcie.bin found
> [    3.600010] brcmfmac: brcm_alt_fw_path FW alt path: 
> brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
> [    3.600323] brcmfmac: brcmf_fw_complete_request firmware 
> brcm/fw_bcm43752a2_pcie.txt found
> [    3.600324] brcmfmac: brcmf_fw_request_nvram_done enter: 
> dev=0002:21:00.0
> [    3.600378] brcmfmac: brcmf_fw_request_nvram_done nvram 
> 00000000eea1e0fd len 6040
> [    3.600381] brcmfmac: brcm_alt_fw_path FW alt path: 
> brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
> [    3.600641] brcmfmac: brcmf_fw_complete_request firmware 
> brcm/fw_bcm43752a2_pcie.clm_blob found
> [    3.600836] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
> [    3.600837] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
> size=1310720 (0x140000) sr=0 (0x0)
> [    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
> [    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 6040 
> entropy bytes
> [    3.694910] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
> 0xfa1a05e5
> [    3.694911] brcmfmac: brcmf_chip_set_active Enter
> [    9.774716] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
> failed to initialize: 0xfa1a05e5
> [    9.774733] brcmfmac: brcmf_detach Enter
> [    9.798714] brcmfmac: brcmf_bus_change_state 0 -> 0
> [    9.798716] brcmfmac: brcmf_proto_detach Enter
> 
> I added a log showing it sending the entropy bytes.

Ok. Something is off there. Maybe I tried to be to smart. How does that 
log statement look which you added?

Regards,
Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 10:49, Arend van Spriel wrote:
> On 11/24/2022 11:20 AM, Andy Green wrote:

>>> Thanks for the log. Together with the dhd driver sources I could make 
>>> some sense of it. Perhaps you can try the following hack and see if 
>>> we make it over the first hurdle.
>>
>> Thanks, I tried it, but although it looks necessary, there's evidently 
>> something else since the result was the same.
>>
>>> +
>>> +    entropy_info.signture = cpu_to_le32(BRCMF_SEED_SIGNATURE);
>>
>> This should be ... .signature = ...
> 
> That's what you get if you get too confident... :-(

Just mention it in case anyone following along at home with the patches.

>> I added a log showing it sending the entropy bytes.
> 
> Ok. Something is off there. Maybe I tried to be to smart. How does that 
> log statement look which you added?

It's inline with the other logs, it seems to be coming before the actual 
NVRAM download part

 >> [    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
 >> [    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 6040
 >> entropy bytes

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1327 bytes --]

On 11/24/2022 12:04 PM, Andy Green wrote:
> 
> 
> On 11/24/22 10:49, Arend van Spriel wrote:
>> On 11/24/2022 11:20 AM, Andy Green wrote:
> 
>>>> Thanks for the log. Together with the dhd driver sources I could 
>>>> make some sense of it. Perhaps you can try the following hack and 
>>>> see if we make it over the first hurdle.
>>>
>>> Thanks, I tried it, but although it looks necessary, there's 
>>> evidently something else since the result was the same.
>>>
>>>> +
>>>> +    entropy_info.signture = cpu_to_le32(BRCMF_SEED_SIGNATURE);
>>>
>>> This should be ... .signature = ...
>>
>> That's what you get if you get too confident... :-(
> 
> Just mention it in case anyone following along at home with the patches.
> 
>>> I added a log showing it sending the entropy bytes.
>>
>> Ok. Something is off there. Maybe I tried to be to smart. How does 
>> that log statement look which you added?
> 
> It's inline with the other logs, it seems to be coming before the actual 
> NVRAM download part
> 
>  >> [    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
>  >> [    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 6040
>  >> entropy bytes

Sorry. I meant how the source code look like. There should be 128 bytes 
written. 6040 is length of nvram so I am a bit confused.

Regards,
Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 11:07, Arend van Spriel wrote:
> On 11/24/2022 12:04 PM, Andy Green wrote:

>>> Ok. Something is off there. Maybe I tried to be to smart. How does 
>>> that log statement look which you added?
>>
>> It's inline with the other logs, it seems to be coming before the 
>> actual NVRAM download part
>>
>>  >> [    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
>>  >> [    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 6040
>>  >> entropy bytes
> 
> Sorry. I meant how the source code look like. There should be 128 bytes 
> written. 6040 is length of nvram so I am a bit confused.

It's just my log is showing the wrong thing.  I was mainly wanting to 
make sure it went through there so I didn't look at what it was saying 
any closer.

         pr_err("%s: written %u entropy bytes\n", __func__, nvram_len);

Showing count instead

[    3.734102] brcmfmac: brcmf_pcie_write_entropy_bytes: written 128 
entropy bytes

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1266 bytes --]

On 11/24/2022 12:11 PM, Andy Green wrote:
> 
> 
> On 11/24/22 11:07, Arend van Spriel wrote:
>> On 11/24/2022 12:04 PM, Andy Green wrote:
> 
>>>> Ok. Something is off there. Maybe I tried to be to smart. How does 
>>>> that log statement look which you added?
>>>
>>> It's inline with the other logs, it seems to be coming before the 
>>> actual NVRAM download part
>>>
>>>  >> [    3.694272] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
>>>  >> [    3.694895] brcmfmac: brcmf_pcie_write_entropy_bytes: written 
>>> 6040
>>>  >> entropy bytes
>>
>> Sorry. I meant how the source code look like. There should be 128 
>> bytes written. 6040 is length of nvram so I am a bit confused.
> 
> It's just my log is showing the wrong thing.  I was mainly wanting to 
> make sure it went through there so I didn't look at what it was saying 
> any closer.
> 
>          pr_err("%s: written %u entropy bytes\n", __func__, nvram_len);
> 
> Showing count instead
> 
> [    3.734102] brcmfmac: brcmf_pcie_write_entropy_bytes: written 128 
> entropy bytes

I did not have any hopes that would give any improvement. Just a glitch 
in my matrix.

Could you try and modify the nvram file, ie. change value for 
boardflags3 to 0x44002180.

Regards,
Arend



[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 11:18, Arend van Spriel wrote:

> Could you try and modify the nvram file, ie. change value for 
> boardflags3 to 0x44002180.

Yes, it was  0x40002180 before, here's the modified file

# AP6275P_NVRAM_V1.2_20210918A
# AP6271P_V00 board, iPA version.
# nvram copied and edited from AP6271P_EVB_V02 EVB board //
# SSID generated using Alberto's boardssid.py script:
# ********************SUMMARY********************
# Board Name: AP6271P_V00
#SSID: 0x086d
#macmid: 0x02df
# Successfully made SSID entry in sromdefs.tcl.
# Successfully made macmid entry in sromdefs.tcl.
# Successfully made SSID entry in tblssid.py.
# *************************************************
# $ Copyright Broadcom $
#
#
# <<Broadcom-WL-IPTag/Proprietary:>>
NVRAMRev=$Rev: 874188 $
sromrev=11
boardrev=0x1213
boardtype=0x08ed
boardflags=0x00400201
boardflags2=0xc0800000
boardflags3=0x44002180
#boardnum=57410
macaddr=00:90:4c:12:d0:01
jtag_irw=38

#Regulatory specific
ccode=0
regrev=0

# Board specific
vendid=0x14e4
devid=0x449d
manfid=0x2d0
antswitch=0
pdgain5g=0
pdgain2g=0
aa2g=3
aa5g=3
agbg0=2
agbg1=2
aga0=2
aga1=2
extpagain2g=2
extpagain5g=2
rxgains2gelnagaina0=0
rxgains2gtrisoa0=0
rxgains2gtrelnabypa0=0
rxgains5gelnagaina0=0
rxgains5gtrisoa0=0
rxgains5gtrelnabypa0=0
rxgains5gmelnagaina0=0
rxgains5gmtrisoa0=0
rxgains5gmtrelnabypa0=0
rxgains5ghelnagaina0=0
rxgains5ghtrisoa0=0
rxgains5ghtrelnabypa0=0
rxgains2gelnagaina1=0
rxgains2gtrisoa1=0
rxgains2gtrelnabypa1=0
rxgains5gelnagaina1=0
rxgains5gtrisoa1=0
rxgains5gtrelnabypa1=0
rxgains5gmelnagaina1=0
rxgains5gmtrisoa1=0
rxgains5gmtrelnabypa1=0
rxgains5ghelnagaina1=0
rxgains5ghtrisoa1=0
rxgains5ghtrelnabypa1=0

#RSSI related
# 2G
rssicorrnorm_c0=4,4
rssicorrnorm_c1=4,4
# 5G
rssicorrnorm5g_c0=5,5,5,5,5,5,5,5,5,5,5,5
rssicorrnorm5g_c1=4,4,4,4,4,4,4,4,4,4,4,4


#Two range TSSI
tworangetssi2g=0
tworangetssi5g=0
lowpowerrange2g=0
lowpowerrange5g=0
low_adc_rate_en=1

nocrc=1
otpimagesize=502

xtalfreq=37400

txchain=3
rxchain=3

cckdigfilttype=2

#bit mask for slice capability bit 0:2G bit 1:5G
bandcap=3

#TXBF Related
rpcal2g=0
rpcal5gb0=0
rpcal5gb1=0
rpcal5gb2=0
rpcal5gb3=0


#FDSS Related
fdss_level_2g=4,4
fdss_interp_en=1
fdss_level_5g=3,3
fdss_level_11ax_2g=3,3
fdss_level_11ax_2g_ch1=3,3
fdss_level_11ax_2g_ch11=3,3
fdss_level_11ax_5g=3,3

#Tempsense Related
tempthresh=255
tempoffset=40
rawtempsense=0x1ff
phycal_tempdelta=15
temps_period=15
temps_hysteresis=15

#------------- TSSI Related -------------
tssipos2g=1
tssipos5g=1
AvVmid_c0=2,127,4,92,4,91,4,91,4,94
AvVmid_c1=2,127,4,93,4,93,4,95,3,110

# CCK in case of multi mode 2
pa2gccka0=-51,9141,-1039
pa2gccka1=-106,8172,-949
# OFDM in case of multi_mode 2

pa2ga0=-31,8047,-863
pa2ga1=-3,8006,-803

pa5ga0=-184,5375,-671,-174,5517,-677,-193,5122,-649,-177,5303,-666
pa5ga1=-207,5022,-635,-168,5741,-695,-169,5544,-687,-182,5435,-687


# Max power and offsets
maxp2ga0=88
maxp2ga1=88
maxp5ga0=68,68,68,68
maxp5ga1=69,69,66,66
subband5gver=0x4
paparambwver=3
cckpwroffset0=0
cckpwroffset1=0
pdoffset40ma0=0x4444
pdoffset80ma0=0x4444
pdoffset40ma1=0x4444
pdoffset80ma1=0x4444
cckbw202gpo=0x4444
cckbw20ul2gpo=0
mcsbw202gpo=0xBB977665
mcsbw402gpo=0xBB977665
dot11agofdmhrbw202gpo=0x7766
ofdmlrbw202gpo=0x0055
mcsbw205glpo=0x88331100
mcsbw405glpo=0xC8332200
mcsbw805glpo=0xCC443320
mcsbw1605glpo=0
mcsbw205gmpo=0x88331100
mcsbw405gmpo=0xC8332200
mcsbw805gmpo=0xCC443320
mcsbw1605gmpo=0
mcsbw205ghpo=0x88331100
mcsbw405ghpo=0xC8333300
mcsbw805ghpo=0xCC443320
powoffs2gtna0=0,0,0,0,0,0,0,0,0,0,0,0,0,0
powoffs2gtna1=0,0,0,0,0,0,0,0,0,0,0,0,0,0
mcs1024qam2gpo=0xEEEE
mcs1024qam5glpo=0xFFFFDD
mcs1024qam5gmpo=0xFFFFDD
mcs1024qam5ghpo=0xFFFFDD
mcs1024qam5gx1po=0xFFFFDD
mcs1024qam5gx2po=0xFFFFDD
mcs8poexp=0
mcs9poexp=0
mcs10poexp=0

# 5G power offset per channel for band edge channel
powoffs5g20mtna0=0,0,0,0,0,0,0
powoffs5g20mtna1=0,0,0,0,0,0,0
powoffs5g40mtna0=0,0,0,0,0
powoffs5g40mtna1=0,0,0,0,0
powoffs5g80mtna0=0,0,0,0,0
powoffs5g80mtna1=0,0,0,0,0
mcs11poexp=0

#LTE Coex Related
ltecxmux=0
ltecxpadnum=0x0504
ltecxfnsel=0x44
ltecxgcigpio=0x04
#OOB params
#device_wake_opt=1
host_wake_opt=0

# SWCTRL Related

swctrlmap_2g=0x10101010,0x06030401,0x04011010,0x000000,0x3FF
swctrlmapext_2g=0x00000000,0x00000000,0x00000000,0x000000,0x000
swctrlmap_5g=0x80408040,0x21240120,0x01208040,0x000000,0x3FF
swctrlmapext_5g=0x00000000,0x00000000,0x00000000,0x000000,0x000
clb2gslice0core0=0x01b
clb2gslice1core0=0x000
clb5gslice0core0=0x064
clb5gslice1core0=0x000
clb2gslice0core1=0x056
clb2gslice1core1=0x000
clb5gslice0core1=0x0a1
clb5gslice1core1=0x000
btc_prisel_ant_mask=0x2
clb_swctrl_smask_ant0=0x27f
clb_swctrl_smask_ant1=0x2f7
muxenab=1

#BT Coex 1:TDM
btc_mode=1

# --- PAPD Cal related params ----
txwbpapden=0 # 0:NBPAPD 1:WBPAPD
# NB PAPD Cal params
nb_eps_offset=470,470
nb_bbmult=66,66
nb_papdcalidx=6,6
nb_txattn=2,2
nb_rxattn=1,1
nb_eps_stopidx=63
epsilonoff_5g20_c0=0,0,0,0
epsilonoff_5g40_c0=0,0,0,0
epsilonoff_5g80_c0=0,0,0,0
epsilonoff_5g20_c1=0,0,-2,-3
epsilonoff_5g40_c1=0,0,-2,-3
epsilonoff_5g80_c1=0,0,-2,-3
epsilonoff_2g20_c0=0
epsilonoff_2g20_c1=0

# energy detect threshold
ed_thresh2g=-67
ed_thresh5g=-67
# energy detect threshold for EU
eu_edthresh2g=-67
eu_edthresh5g=-67

#rpcal coef for imptxbf
rpcal5gb0=238
rpcal5gb1=228
rpcal5gb2=222
rpcal5gb3=229
rpcal2g=15
ocl=1
bt_coex_chdep_div=1

# OLPC Related
disable_olpc=0
olpc_thresh5g=32
olpc_anchor5g=40
olpc_thresh2g=32
olpc_anchor2g=40

#PAPR related
paprdis=0
paprrmcsgamma2g=500,550,550,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain2g=128,128,128,0,0,0,0,0,0,0,0,0
paprrmcsgamma2g_ch13=500,550,550,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain2g_ch13=128,128,128,0,0,0,0,0,0,0,0,0
paprrmcsgamma2g_ch1=500,550,550,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain2g_ch1=128,128,128,0,0,0,0,0,0,0,0,0
paprrmcsgamma5g20=500,500,500,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain5g20=128,128,128,0,0,0,0,0,0,0,0,0
paprrmcsgamma5g40=600,600,600,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain5g40=128,128,128,0,0,0,0,0,0,0,0,0
paprrmcsgamma5g80=-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1
paprrmcsgain5g80=0,0,0,0,0,0,0,0,0,0,0,0

# Enable papd for cck when target pwr ge 16dBm
cckpapd_pwrthresh=64

## ULOFDMA Board limit PPRs for 2G20 ##
ruppr2g20bpska0=0x00021084
ruppr2g20bpska1=0x00021084
ruppr2g20qpska0=0x00021084
ruppr2g20qpska1=0x00021084
ruppr2g20qam16a0=0x000294A5
ruppr2g20qam16a1=0x000294A5
ruppr2g20qam64a0=0x00039CE7
ruppr2g20qam64a1=0x00039CE7
ruppr2g20qam256a0=0x0005AD6B
ruppr2g20qam256a1=0x0005AD6B
ruppr2g20qam1024a0=0x0005AD6B
ruppr2g20qam1024a1=0x0005AD6B
## ULOFDMA Board limit PPRs for 5G20 ##
ruppr5g20bpska0=0x00008421
ruppr5g20bpska1=0x00008421
ruppr5g20qpska0=0x00008421
ruppr5g20qpska1=0x00008421
ruppr5g20qam16a0=0x00010842
ruppr5g20qam16a1=0x00010842
ruppr5g20qam64a0=0x00021084
ruppr5g20qam64a1=0x00021084
ruppr5g20qam256a0=0x00042108
ruppr5g20qam256a1=0x00042108
ruppr5g20qam1024a0=0x0006318C
ruppr5g20qam1024a1=0x0006318C
## ULOFDMA Board limit PPRs for 5G40 ##
ruppr5g40bpska0=0x00108421
ruppr5g40bpska1=0x00108421
ruppr5g40qpska0=0x00108421
ruppr5g40qpska1=0x00108421
ruppr5g40qam16a0=0x00210842
ruppr5g40qam16a1=0x00210842
ruppr5g40qam64a0=0x00421084
ruppr5g40qam64a1=0x00421084
ruppr5g40qam256a0=0x00C6318C
ruppr5g40qam256a1=0x00C6318C
ruppr5g40qam1024a0=0x01084210
ruppr5g40qam1024a1=0x01084210
## ULOFDMA Board limit PPRs for 5G80 ##
ruppr5g80bpska0=0x04108421
ruppr5g80bpska1=0x04108421
ruppr5g80qpska0=0x04108421
ruppr5g80qpska1=0x04108421
ruppr5g80qam16a0=0x0C318C63
ruppr5g80qam16a1=0x0C318C63
ruppr5g80qam64a0=0x10421084
ruppr5g80qam64a1=0x10421084
ruppr5g80qam256a0=0x30C6318C
ruppr5g80qam256a1=0x30C6318C
ruppr5g80qam1024a0=0x41084210
ruppr5g80qam1024a1=0x41084210


It is the same result.  Logs from where it loads the NVRAM file

[    3.534708] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.txt found
[    3.534709] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
[    3.534771] brcmfmac: brcmf_fw_request_nvram_done nvram 
0000000007e422eb len 6040
[    3.534774] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
[    3.535350] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.clm_blob found
[    3.535558] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.535559] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.628168] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
[    3.628767] brcmfmac: brcmf_pcie_write_entropy_bytes: written 128 
entropy bytes
[    3.628787] brcmfmac: brcmf_pcie_download_fw_nvram: sharedram_addr 
0xfa1a05e5
[    3.628789] brcmfmac: brcmf_chip_set_active Enter
[    9.689983] brcmfmac 0002:21:00.0: brcmf_pcie_download_fw_nvram: FW 
failed to initialize: 0xfa1a05e5

Since it mentions the vid:pid, here's the verbose lspci for it

0002:21:00.0 0280: 14e4:449d (rev 02)
	Subsystem: 14e4:aae8
	Control: I/O- Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- 
Stepping- SERR- FastB2B- DisINTx-
	Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- 
<MAbort- >SERR- <PERR- INTx-
	Interrupt: pin A routed to IRQ 47
	Region 0: Memory at f2200000 (64-bit, non-prefetchable) [size=64K]
	Region 2: Memory at f2400000 (64-bit, non-prefetchable) [size=4M]
	Capabilities: [48] Power Management version 3
		Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=0mA PME(D0+,D1+,D2+,D3hot+,D3cold+)
		Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=2 PME-
	Capabilities: [58] MSI: Enable- Count=1/32 Maskable- 64bit+
		Address: 0000000000000000  Data: 0000
	Capabilities: [68] Vendor Specific Information: Len=38 <?>
	Capabilities: [ac] Express (v2) Endpoint, MSI 00
		DevCap:	MaxPayload 512 bytes, PhantFunc 0, Latency L0s <4us, L1 unlimited
			ExtTag- AttnBtn- AttnInd- PwrInd- RBE+ FLReset+ SlotPowerLimit 0.000W
		DevCtl:	CorrErr+ NonFatalErr+ FatalErr+ UnsupReq+
			RlxdOrd+ ExtTag- PhantFunc- AuxPwr+ NoSnoop+ FLReset-
			MaxPayload 128 bytes, MaxReadReq 512 bytes
		DevSta:	CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr+ TransPend-
		LnkCap:	Port #0, Speed 2.5GT/s, Width x1, ASPM L1, Exit Latency L1 <32us
			ClockPM+ Surprise- LLActRep- BwNot- ASPMOptComp+
		LnkCtl:	ASPM Disabled; RCB 64 bytes, Disabled- CommClk-
			ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
		LnkSta:	Speed 2.5GT/s (ok), Width x1 (ok)
			TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
		DevCap2: Completion Timeout: Range ABCD, TimeoutDis+ NROPrPrP- LTR+
			 10BitTagComp- 10BitTagReq- OBFF Via WAKE#, ExtFmt- EETLPPrefix-
			 EmergencyPowerReduction Not Supported, EmergencyPowerReductionInit-
			 FRS- TPHComp- ExtTPHComp-
			 AtomicOpsCap: 32bit- 64bit- 128bitCAS-
		DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- LTR+ OBFF Disabled,
			 AtomicOpsCtl: ReqEn-
		LnkCap2: Supported Link Speeds: 2.5GT/s, Crosslink- Retimer- 
2Retimers- DRS-
		LnkCtl2: Target Link Speed: 2.5GT/s, EnterCompliance- SpeedDis-
			 Transmit Margin: Normal Operating Range, EnterModifiedCompliance- 
ComplianceSOS-
			 Compliance De-emphasis: -6dB
		LnkSta2: Current De-emphasis Level: -3.5dB, EqualizationComplete- 
EqualizationPhase1-
			 EqualizationPhase2- EqualizationPhase3- LinkEqualizationRequest-
			 Retimer- 2Retimers- CrosslinkRes: unsupported
	Capabilities: [100 v1] Advanced Error Reporting
		UESta:	DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- 
MalfTLP- ECRC- UnsupReq- ACSViol-
		UEMsk:	DLP- SDES- TLP- FCP- CmpltTO- CmpltAbrt- UnxCmplt- RxOF- 
MalfTLP- ECRC- UnsupReq- ACSViol-
		UESvrt:	DLP+ SDES+ TLP- FCP+ CmpltTO- CmpltAbrt- UnxCmplt- RxOF+ 
MalfTLP+ ECRC- UnsupReq- ACSViol-
		CESta:	RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr-
		CEMsk:	RxErr- BadTLP- BadDLLP- Rollover- Timeout- AdvNonFatalErr+
		AERCap:	First Error Pointer: 00, ECRCGenCap+ ECRCGenEn- ECRCChkCap+ 
ECRCChkEn-
			MultHdrRecCap- MultHdrRecEn- TLPPfxPres- HdrLogCap-
		HeaderLog: 00000000 00000000 00000000 00000000
	Capabilities: [13c v1] Device Serial Number 00-00-00-ff-ff-00-00-00
	Capabilities: [150 v1] Power Budgeting <?>
	Capabilities: [160 v1] Virtual Channel
		Caps:	LPEVC=0 RefClk=100ns PATEntryBits=1
		Arb:	Fixed- WRR32- WRR64- WRR128-
		Ctrl:	ArbSelect=Fixed
		Status:	InProgress-
		VC0:	Caps:	PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
			Arb:	Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
			Ctrl:	Enable+ ID=0 ArbSelect=Fixed TC/VC=ff
			Status:	NegoPending- InProgress-
	Capabilities: [1b0 v1] Latency Tolerance Reporting
		Max snoop latency: 0ns
		Max no snoop latency: 0ns
	Capabilities: [220 v1] Physical Resizable BAR
		BAR 2: current size: 8MB, supported: 1MB 2MB 4MB
	Capabilities: [240 v1] L1 PM Substates
		L1SubCap: PCI-PM_L1.2+ PCI-PM_L1.1+ ASPM_L1.2+ ASPM_L1.1+ L1_PM_Substates+
			  PortCommonModeRestoreTime=8us PortTPowerOnTime=50us
		L1SubCtl1: PCI-PM_L1.2- PCI-PM_L1.1- ASPM_L1.2- ASPM_L1.1-
			   T_CommonMode=0us LTR1.2_Threshold=0ns
		L1SubCtl2: T_PwrOn=10us
	Kernel modules: brcmfmac


-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 09:42, Arend van Spriel wrote:
> On 11/18/2022 6:41 AM, Andy Green wrote:
>>
>>
>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>
>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>> higher debug level. It also has a module parameter for it. I think it 
>>> is called dhd_msg_level and please set it to 0x817.
>>
>> It's here:
>>
>> https://warmcat.com/ap6275p-dhd-log1.txt
> 
> Hi Andy,
> 
> Thanks for the log. Together with the dhd driver sources I could make 
> some sense of it. Perhaps you can try the following hack and see if we 
> make it over the first hurdle.

OH!  I found that the line below should be count, not sizeof(count).

> +    addr -= sizeof(count);
> +    memcpy_toio(devinfo->tcm + addr, seed, count);
> +}

With that, you were right on the money, the firmware comes up

[    3.796062] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43752/2 
wl0: Jul 12 2022 18:54:54 version 18.35.387.23.146 (g412cc5ec) FWID 
01-93c53be6

along with

wlP2p33s0: flags=4098<BROADCAST,MULTICAST>  mtu 1500
         ether d4:9c:dd:f5:bf:7e  txqueuelen 1000  (Ethernet)
         RX packets 0  bytes 0 (0.0 B)
         RX errors 0  dropped 0  overruns 0  frame 0
         TX packets 0  bytes 0 (0.0 B)
         TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

I will try get it to scan...

... any way, good job!

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1783 bytes --]

On 11/24/2022 1:14 PM, Andy Green wrote:
> 
> 
> On 11/24/22 09:42, Arend van Spriel wrote:
>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>
>>>
>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>
>>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>>> higher debug level. It also has a module parameter for it. I think 
>>>> it is called dhd_msg_level and please set it to 0x817.
>>>
>>> It's here:
>>>
>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>
>> Hi Andy,
>>
>> Thanks for the log. Together with the dhd driver sources I could make 
>> some sense of it. Perhaps you can try the following hack and see if we 
>> make it over the first hurdle.
> 
> OH!  I found that the line below should be count, not sizeof(count).

Good job! I was testing you obviously. You passed. Nah, I just feel stupid.

>> +    addr -= sizeof(count);
>> +    memcpy_toio(devinfo->tcm + addr, seed, count);
>> +}
> 
> With that, you were right on the money, the firmware comes up
> 
> [    3.796062] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43752/2 
> wl0: Jul 12 2022 18:54:54 version 18.35.387.23.146 (g412cc5ec) FWID 
> 01-93c53be6
> 
> along with
> 
> wlP2p33s0: flags=4098<BROADCAST,MULTICAST>  mtu 1500
>          ether d4:9c:dd:f5:bf:7e  txqueuelen 1000  (Ethernet)
>          RX packets 0  bytes 0 (0.0 B)
>          RX errors 0  dropped 0  overruns 0  frame 0
>          TX packets 0  bytes 0 (0.0 B)
>          TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
> 
> I will try get it to scan...
> 
> ... any way, good job!

Thanks. Will clean it up and formally submit it. I also had a quick look 
at the SDIO variant, but not seeing anything obvious there in terms of 
firmware loading.

Regards,
Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Great work! If you need to test SDIO variant, I could help with it.
Shengyu 

于 2022年11月24日 GMT+08:00 下午8:22:09, Arend van Spriel <arend.vanspriel@broadcom.com> 写到:
>On 11/24/2022 1:14 PM, Andy Green wrote:
>> 
>> 
>> On 11/24/22 09:42, Arend van Spriel wrote:
>>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>> 
>>>> 
>>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>> 
>>>>> I see. So could you load bcmdhd and provide log. Preferably with higher debug level. It also has a module parameter for it. I think it is called dhd_msg_level and please set it to 0x817.
>>>> 
>>>> It's here:
>>>> 
>>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>> 
>>> Hi Andy,
>>> 
>>> Thanks for the log. Together with the dhd driver sources I could make some sense of it. Perhaps you can try the following hack and see if we make it over the first hurdle.
>> 
>> OH!  I found that the line below should be count, not sizeof(count).
>
>Good job! I was testing you obviously. You passed. Nah, I just feel stupid.
>
>>> +    addr -= sizeof(count);
>>> +    memcpy_toio(devinfo->tcm + addr, seed, count);
>>> +}
>> 
>> With that, you were right on the money, the firmware comes up
>> 
>> [    3.796062] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43752/2 wl0: Jul 12 2022 18:54:54 version 18.35.387.23.146 (g412cc5ec) FWID 01-93c53be6
>> 
>> along with
>> 
>> wlP2p33s0: flags=4098<BROADCAST,MULTICAST>  mtu 1500
>>          ether d4:9c:dd:f5:bf:7e  txqueuelen 1000  (Ethernet)
>>          RX packets 0  bytes 0 (0.0 B)
>>          RX errors 0  dropped 0  overruns 0  frame 0
>>          TX packets 0  bytes 0 (0.0 B)
>>          TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0
>> 
>> I will try get it to scan...
>> 
>> ... any way, good job!
>
>Thanks. Will clean it up and formally submit it. I also had a quick look at the SDIO variant, but not seeing anything obvious there in terms of firmware loading.
>
>Regards,
>Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 12:22, Arend van Spriel wrote:
> On 11/24/2022 1:14 PM, Andy Green wrote:
>>
>>
>> On 11/24/22 09:42, Arend van Spriel wrote:
>>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>>
>>>>
>>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>>
>>>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>>>> higher debug level. It also has a module parameter for it. I think 
>>>>> it is called dhd_msg_level and please set it to 0x817.
>>>>
>>>> It's here:
>>>>
>>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>>
>>> Hi Andy,
>>>
>>> Thanks for the log. Together with the dhd driver sources I could make 
>>> some sense of it. Perhaps you can try the following hack and see if 
>>> we make it over the first hurdle.
>>
>> OH!  I found that the line below should be count, not sizeof(count).
> 
> Good job! I was testing you obviously. You passed. Nah, I just feel stupid.

No it's great, the point is it directly identified the root cause 
without access to the platform, it's not so easy.

>> ... any way, good job!
> 
> Thanks. Will clean it up and formally submit it. I also had a quick look 
> at the SDIO variant, but not seeing anything obvious there in terms of 
> firmware loading.

I only have the pcie module not the SDIO one.

For the scan situation, he understands what is being asked via 
connmanctl "enable wifi", but he feels something wrong

connmanctl> enable wifi

connmanctl>
Enabled wifi
connmanctl> [  272.174946] ieee80211 phy0: brcmf_run_escan: error (-52)
[  272.174956] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)


connmanctl> [  275.133157] ieee80211 phy0: brcmf_run_escan: error (-52)
[  275.133166] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)


connmanctl> [  284.138852] ieee80211 phy0: brcmf_run_escan: error (-52)
[  284.138856] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)

52 is EBADE, I recall you mentioned that the scan api had been changed 
recently... is this possibly related to that?  Grepping around for EBADE 
near the function it mentions, it tries to take evasive action if it 
sees one in some places already.

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Shengyu Qu]

Seems SDIO variant load firmware correctly without any modification, but also reports this -52 error while scanning.
Shengyu 

于 2022年11月24日 GMT+08:00 下午8:26:45, Andy Green <andy@warmcat.com> 写到:
>
>
>On 11/24/22 12:22, Arend van Spriel wrote:
>> On 11/24/2022 1:14 PM, Andy Green wrote:
>>> 
>>> 
>>> On 11/24/22 09:42, Arend van Spriel wrote:
>>>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>>> 
>>>>> 
>>>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>>> 
>>>>>> I see. So could you load bcmdhd and provide log. Preferably with higher debug level. It also has a module parameter for it. I think it is called dhd_msg_level and please set it to 0x817.
>>>>> 
>>>>> It's here:
>>>>> 
>>>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>>> 
>>>> Hi Andy,
>>>> 
>>>> Thanks for the log. Together with the dhd driver sources I could make some sense of it. Perhaps you can try the following hack and see if we make it over the first hurdle.
>>> 
>>> OH!  I found that the line below should be count, not sizeof(count).
>> 
>> Good job! I was testing you obviously. You passed. Nah, I just feel stupid.
>
>No it's great, the point is it directly identified the root cause without access to the platform, it's not so easy.
>
>>> ... any way, good job!
>> 
>> Thanks. Will clean it up and formally submit it. I also had a quick look at the SDIO variant, but not seeing anything obvious there in terms of firmware loading.
>
>I only have the pcie module not the SDIO one.
>
>For the scan situation, he understands what is being asked via connmanctl "enable wifi", but he feels something wrong
>
>connmanctl> enable wifi
>
>connmanctl>
>Enabled wifi
>connmanctl> [  272.174946] ieee80211 phy0: brcmf_run_escan: error (-52)
>[  272.174956] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
>
>
>connmanctl> [  275.133157] ieee80211 phy0: brcmf_run_escan: error (-52)
>[  275.133166] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
>
>
>connmanctl> [  284.138852] ieee80211 phy0: brcmf_run_escan: error (-52)
>[  284.138856] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
>
>52 is EBADE, I recall you mentioned that the scan api had been changed recently... is this possibly related to that?  Grepping around for EBADE near the function it mentions, it tries to take evasive action if it sees one in some places already.
>
>-Andy
>
>

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 279 bytes --]

On 11/24/2022 1:35 PM, Shengyu Qu wrote:
> Seems SDIO variant load firmware correctly without any modification, but also reports this -52 error while scanning.
> Shengyu

Thanks for letting me know. So we can focus on scanning. I will reply to 
Andy about that.

Regards,
Arend


[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 2518 bytes --]

On 11/24/2022 1:26 PM, Andy Green wrote:
> 
> 
> On 11/24/22 12:22, Arend van Spriel wrote:
>> On 11/24/2022 1:14 PM, Andy Green wrote:
>>>
>>>
>>> On 11/24/22 09:42, Arend van Spriel wrote:
>>>> On 11/18/2022 6:41 AM, Andy Green wrote:
>>>>>
>>>>>
>>>>> On 11/17/22 21:52, Arend Van Spriel wrote:
>>>>>
>>>>>> I see. So could you load bcmdhd and provide log. Preferably with 
>>>>>> higher debug level. It also has a module parameter for it. I think 
>>>>>> it is called dhd_msg_level and please set it to 0x817.
>>>>>
>>>>> It's here:
>>>>>
>>>>> https://warmcat.com/ap6275p-dhd-log1.txt
>>>>
>>>> Hi Andy,
>>>>
>>>> Thanks for the log. Together with the dhd driver sources I could 
>>>> make some sense of it. Perhaps you can try the following hack and 
>>>> see if we make it over the first hurdle.
>>>
>>> OH!  I found that the line below should be count, not sizeof(count).
>>
>> Good job! I was testing you obviously. You passed. Nah, I just feel 
>> stupid.
> 
> No it's great, the point is it directly identified the root cause 
> without access to the platform, it's not so easy.
> 
>>> ... any way, good job!
>>
>> Thanks. Will clean it up and formally submit it. I also had a quick 
>> look at the SDIO variant, but not seeing anything obvious there in 
>> terms of firmware loading.
> 
> I only have the pcie module not the SDIO one.
> 
> For the scan situation, he understands what is being asked via 
> connmanctl "enable wifi", but he feels something wrong
> 
> connmanctl> enable wifi
> 
> connmanctl>
> Enabled wifi
> connmanctl> [  272.174946] ieee80211 phy0: brcmf_run_escan: error (-52)
> [  272.174956] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
> 
> 
> connmanctl> [  275.133157] ieee80211 phy0: brcmf_run_escan: error (-52)
> [  275.133166] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
> 
> 
> connmanctl> [  284.138852] ieee80211 phy0: brcmf_run_escan: error (-52)
> [  284.138856] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
> 
> 52 is EBADE, I recall you mentioned that the scan api had been changed 
> recently... is this possibly related to that?  Grepping around for EBADE 
> near the function it mentions, it tries to take evasive action if it 
> sees one in some places already.

EBADE is a generic error message indicating something went wrong 
communicating with firmware. Not very informative. Were you using any 
debug level? Anyway, please load the driver with module parameter 
debug=0x5416 and try scanning using iw utility.

Regards,
Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 12:44, Arend van Spriel wrote:

>> 52 is EBADE, I recall you mentioned that the scan api had been changed 
>> recently... is this possibly related to that?  Grepping around for 
>> EBADE near the function it mentions, it tries to take evasive action 
>> if it sees one in some places already.
> 
> EBADE is a generic error message indicating something went wrong 
> communicating with firmware. Not very informative. Were you using any 
> debug level? Anyway, please load the driver with module parameter 
> debug=0x5416 and try scanning using iw utility.

My kernel crashed on boot while I was trying things earlier 
necessitating a rootfs reflash, I lost the /etc/modprobe.d entry for the 
debug.

Adding it back in, it's a lot more informative

[    3.569646] brcmfmac: brcmfmac_module_init No platform data available.
[    3.569832] brcmfmac 0002:21:00.0: enabling device (0000 -> 0002)
[    3.569888] brcmfmac: brcmf_chip_recognition found AXI chip: BCM43752/2
[    3.570843] brcmfmac: brcmf_chip_cores_check  [1 ] core 0x800:65 
base 0x18000000 wrap 0x18100000
[    3.570848] brcmfmac: brcmf_chip_cores_check  [2 ] core 0x812:82 
base 0x18001000 wrap 0x18101000
[    3.570853] brcmfmac: brcmf_chip_cores_check  [3 ] core 0x83e:16 
base 0x18002000 wrap 0x18102000
[    3.570857] brcmfmac: brcmf_chip_cores_check  [4 ] core 0x83c:65 
base 0x18003000 wrap 0x18103000
[    3.570860] brcmfmac: brcmf_chip_cores_check  [5 ] core 0x829:26 
base 0x18004000 wrap 0x18104000
[    3.570864] brcmfmac: brcmf_chip_cores_check  [6 ] core 0x844:2 
base 0x18005000 wrap 0x18105000
[    3.570868] brcmfmac: brcmf_chip_cores_check  [7 ] core 0x840:15 
base 0x18010000 wrap 0x00000000
[    3.570872] brcmfmac: brcmf_chip_cores_check  [8 ] core 0x827:35 
base 0x18012000 wrap 0x00000000
[    3.570876] brcmfmac: brcmf_chip_cores_check  [9 ] core 0x240:0 
base 0x00000000 wrap 0x1810a000
[    3.570880] brcmfmac: brcmf_chip_cores_check  [10] core 0x135:0 
base 0x00000000 wrap 0x18112000
[    3.570884] brcmfmac: brcmf_chip_cores_check  [11] core 0x135:0 
base 0x00000000 wrap 0x18113000
[    3.570888] brcmfmac: brcmf_chip_set_passive Enter
[    3.678925] brcmfmac: brcmf_chip_set_passive Enter
[    3.679441] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.679443] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.679471] brcmfmac: brcmf_chip_setup ccrev=65, pmurev=35, 
pmucaps=0x84565f23
[    3.679477] brcmfmac: brcmf_get_module_param Enter, bus=2, 
chip=43752, rev=2
[    3.679495] brcmfmac: brcmf_alloc Enter
[    3.679512] brcmfmac: brcmf_fw_alloc_request: using 
brcm/fw_bcm43752a2_pcie for chip BCM43752/2
[    3.679517] brcmfmac: brcmf_fw_get_firmwares enter: dev=0002:21:00.0
[    3.679522] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin
[    3.679826] brcmfmac 0002:21:00.0: Direct firmware load for 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.bin failed with error -2
[    3.684423] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.bin found
[    3.684433] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.txt
[    3.684768] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.txt found
[    3.684772] brcmfmac: brcmf_fw_request_nvram_done enter: dev=0002:21:00.0
[    3.684947] brcmfmac: brcmf_fw_request_nvram_done nvram 
000000007fae8cbc len 6040
[    3.684955] brcmfmac: brcm_alt_fw_path FW alt path: 
brcm/fw_bcm43752a2_pcie.rockchip,rk3588-evb1-v10.clm_blob
[    3.685347] brcmfmac: brcmf_fw_complete_request firmware 
brcm/fw_bcm43752a2_pcie.clm_blob found
[    3.685625] brcmfmac: brcmf_chip_tcm_rambase: 43752 OK
[    3.685627] brcmfmac: brcmf_chip_get_raminfo RAM: base=0x170000 
size=1310720 (0x140000) sr=0 (0x0)
[    3.779392] brcmfmac: Download NVRAM brcm/fw_bcm43752a2_pcie.txt
[    3.780001] brcmfmac: brcmf_pcie_write_entropy_bytes: written 128 
entropy bytes
[    3.780016] brcmfmac: Bring ARM in running state
[    3.780018] brcmfmac: brcmf_chip_set_active Enter
[    3.780229] brcmfmac: Wait for FW init (0xfa1a05e5)
[    3.995343] brcmfmac: brcmf_attach Enter
[    3.995348] brcmfmac: brcmf_proto_attach Enter
[    3.997068] brcmfmac: brcmf_fweh_register event handler registered 
for PSM_WATCHDOG
[    3.997073] brcmfmac: brcmf_bus_started
[    3.997076] brcmfmac: brcmf_add_if Enter, bsscfgidx=0, ifidx=0
[    3.997080] brcmfmac: brcmf_add_if allocate netdev interface
[    3.997096] brcmfmac: brcmf_add_if  ==== pid:bf, if:wlan%d 
(00:00:00:00:00:00) created ===
[    3.997100] brcmfmac: brcmf_bus_change_state 0 -> 1
[    3.999755] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=cur_etheraddr, len=6, err=0
[    4.000059] brcmfmac: brcmf_fil_cmd_data_get ifidx=0, cmd=98, len=68, 
err=0
[    4.000077] brcmfmac: brcmf_c_process_clm_blob Enter
[    4.000083] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.000583] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.001002] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.001432] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.001852] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.002282] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.002713] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.003137] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.003569] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.003989] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.004420] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.004883] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.005163] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.005583] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.006008] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.006479] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.006864] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.007286] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.007718] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.008138] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.008568] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.008999] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=1415
[    4.009417] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=clmload, 
len=208
[    4.012350] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=ver, 
len=256, err=0
[    4.012368] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43752/2 
wl0: Jul 12 2022 18:54:54 version 18.35.387.23.146 (g412cc5ec) FWID 
01-93c53be6
[    4.012791] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=clmver, 
len=256, err=0
[    4.012807] brcmfmac: brcmf_c_preinit_dcmds CLM version = API: 20.0 
Data: 9.9.12_SS Compiler: 1.53.0 ClmImport: 1.52.0 Customization: 
V15_190709 Creation: 2021-10-12 17:03:28
[    4.012810] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=mpc, len=4
[    4.013199] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=join_pref, len=8
[    4.013759] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=event_msgs, len=18, err=0
[    4.013770] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=event_msgs, len=18
[    4.014108] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=185, value=40
[    4.014506] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=187, value=40
[    4.014930] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=txbf, len=4
[    4.015732] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=cap, 
len=768, err=0
[    4.015749] brcmfmac: brcmf_feat_firmware_capabilities [ ap sta wme 
802.11d 802.11h cqa cac dualband ampdu ampdu_tx ampdu_rx amsdurx amsdutx 
tdls radio_pwrsave p2p proptxstatus mchan vht-prop-rates 
single-user-beamformer multi-user-beamformee single-user-beamformee 
dfrts txpwrcache stbc-tx stbc-rx-1ss mfp ndoe wnm bsstrans txcap epno 
scanmac cptlv-4 ve fbtoverds fbt_adpt ecounters event_log cdef txcap ocl 
hchk logtrace d11status scancache apf ifver idauth ifst fie tvpm 
lpr_scan bkoff_evt etd_infoclm_restrict evt_ext btc_wifi_prot monitor 
mbo estm d3cbuf sae extsae wrxf0ovfl ]
[    4.015765] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: MCHAN
[    4.015768] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: P2P
[    4.015775] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: MONITOR
[    4.015794] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: DOT11H
[    4.015802] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: SAE
[    4.015808] brcmfmac: brcmf_feat_firmware_capabilities enabling 
feature: FWAUTH
[    4.015810] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=pfn_gscan_cfg, len=18
[    4.016360] brcmfmac: brcmf_fil_cmd_data Firmware error: BCME_VERSION 
(-37)
[    4.016364] brcmfmac: brcmf_feat_iovar_data_set enabling feature: GSCAN
[    4.016731] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=pfn, 
len=4, err=0
[    4.016739] brcmfmac: brcmf_feat_iovar_int_get enabling feature: PNO
[    4.017167] brcmfmac: brcmf_fil_cmd_data Firmware error: 
BCME_UNSUPPORTED (-23)
[    4.017170] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=wowl, 
len=4, err=-23
[    4.017177] brcmfmac: brcmf_feat_iovar_int_get WOWL feature check 
failed: -23
[    4.017456] brcmfmac: brcmf_fil_cmd_data Firmware error: 
BCME_UNSUPPORTED (-23)
[    4.017459] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=rsdb_mode, len=4, err=-23
[    4.017467] brcmfmac: brcmf_feat_iovar_int_get RSDB feature check 
failed: -23
[    4.017823] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=tdls_enable, len=4, err=0
[    4.017831] brcmfmac: brcmf_feat_iovar_int_get enabling feature: TDLS
[    4.018220] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=mfp, 
len=4, err=0
[    4.018228] brcmfmac: brcmf_feat_iovar_int_get enabling feature: MFP
[    4.018611] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=pfn_macaddr, len=8, err=0
[    4.018933] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=sup_wpa, 
len=4, err=0
[    4.018942] brcmfmac: brcmf_feat_iovar_int_get enabling feature: FWSUP
[    4.018947] brcmfmac: brcmf_alloc_vif allocating virtual interface 
(size=4864)
[    4.018961] brcmfmac: brcmf_fweh_register event handler registered 
for LINK
[    4.018964] brcmfmac: brcmf_fweh_register event handler registered 
for DEAUTH_IND
[    4.018966] brcmfmac: brcmf_fweh_register event handler registered 
for DEAUTH
[    4.018968] brcmfmac: brcmf_fweh_register event handler registered 
for DISASSOC_IND
[    4.018970] brcmfmac: brcmf_fweh_register event handler registered 
for ASSOC_IND
[    4.018972] brcmfmac: brcmf_fweh_register event handler registered 
for REASSOC_IND
[    4.018974] brcmfmac: brcmf_fweh_register event handler registered 
for ROAM
[    4.018976] brcmfmac: brcmf_fweh_register event handler registered 
for MIC_ERROR
[    4.018979] brcmfmac: brcmf_fweh_register event handler registered 
for SET_SSID
[    4.018981] brcmfmac: brcmf_fweh_register event handler registered 
for PFN_NET_FOUND
[    4.018983] brcmfmac: brcmf_fweh_register event handler registered for IF
[    4.018986] brcmfmac: brcmf_fweh_register event handler registered 
for P2P_PROBEREQ_MSG
[    4.018988] brcmfmac: brcmf_fweh_register event handler registered 
for P2P_DISC_LISTEN_COMPLETE
[    4.018990] brcmfmac: brcmf_fweh_register event handler registered 
for ACTION_FRAME_RX
[    4.018993] brcmfmac: brcmf_fweh_register event handler registered 
for ACTION_FRAME_COMPLETE
[    4.018995] brcmfmac: brcmf_fweh_register event handler registered 
for ACTION_FRAME_OFF_CHAN_COMPLETE
[    4.018997] brcmfmac: brcmf_fweh_register event handler registered 
for PSK_SUP
[    4.019000] brcmfmac: brcmf_fweh_register event handler registered 
for RSSI
[    4.019002] brcmfmac: brcmf_fweh_register event handler registered 
for ESCAN_RESULT
[    4.019441] brcmfmac: brcmf_fil_cmd_int_get ifidx=0, cmd=1, value=2
[    4.019748] brcmfmac: brcmf_fil_cmd_data_get ifidx=0, cmd=140, 
len=12, err=0
[    4.019763] brcmfmac: brcmf_cfg80211_attach Registering custom regulatory
[    4.021256] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=vhtmode, 
len=4, err=0
[    4.021585] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=nmode, 
len=4, err=0
[    4.021915] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=bw_cap, 
len=4, err=0
[    4.022278] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=bw_cap, 
len=4, err=0
[    4.022286] brcmfmac: brcmf_setup_wiphybands nmode=1, vhtmode=1, 
bw_cap=(1, 7)
[    4.022530] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=rxchain, 
len=4, err=0
[    4.022538] brcmfmac: brcmf_setup_wiphybands nchain=2
[    4.024274] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspecs, len=1536, err=0
[    4.024793] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.025312] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.025832] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.026188] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.026718] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.027264] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.027782] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.028302] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.028822] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.029341] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.029698] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.030232] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.030589] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.031116] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.031660] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.032185] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.032575] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.032937] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.033462] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.033823] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.034358] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.034930] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.035455] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.036001] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.036368] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.036885] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.037242] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.037799] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.038161] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.038521] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.039049] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.039405] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.039931] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.040291] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.040816] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.041175] brcmfmac: brcmf_fil_bsscfg_data_get ifidx=0, bsscfgidx=0, 
name=per_chan_info, len=4, err=0
[    4.041578] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=txstreams, len=4, err=0
[    4.041956] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=txbf_bfe_cap, len=4, err=0
[    4.042320] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=txbf_bfr_cap, len=4, err=0
[    4.042670] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=bw_cap, 
len=4, err=0
[    4.042678] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=bw_cap, 
len=8
[    4.043445] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspecs, len=1536, err=0
[    4.043464] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=obss_coex, len=4
[    4.043822] brcmfmac: brcmf_fweh_activate_events enable event SET_SSID
[    4.043825] brcmfmac: brcmf_fweh_activate_events enable event DEAUTH
[    4.043827] brcmfmac: brcmf_fweh_activate_events enable event DEAUTH_IND
[    4.043829] brcmfmac: brcmf_fweh_activate_events enable event ASSOC_IND
[    4.043831] brcmfmac: brcmf_fweh_activate_events enable event REASSOC_IND
[    4.043833] brcmfmac: brcmf_fweh_activate_events enable event 
DISASSOC_IND
[    4.043836] brcmfmac: brcmf_fweh_activate_events enable event LINK
[    4.043838] brcmfmac: brcmf_fweh_activate_events enable event MIC_ERROR
[    4.043840] brcmfmac: brcmf_fweh_activate_events enable event ROAM
[    4.043842] brcmfmac: brcmf_fweh_activate_events enable event 
PFN_NET_FOUND
[    4.043845] brcmfmac: brcmf_fweh_activate_events enable event 
PSM_WATCHDOG
[    4.043847] brcmfmac: brcmf_fweh_activate_events enable event PSK_SUP
[    4.043849] brcmfmac: brcmf_fweh_activate_events enable event IF
[    4.043851] brcmfmac: brcmf_fweh_activate_events enable event 
P2P_DISC_LISTEN_COMPLETE
[    4.043853] brcmfmac: brcmf_fweh_activate_events enable event RSSI
[    4.043856] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_COMPLETE
[    4.043858] brcmfmac: brcmf_fweh_activate_events enable event 
ESCAN_RESULT
[    4.043860] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_OFF_CHAN_COMPLETE
[    4.043862] brcmfmac: brcmf_fweh_activate_events enable event 
P2P_PROBEREQ_MSG
[    4.043864] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_RX
[    4.043867] brcmfmac: brcmf_fweh_activate_events enable event IF
[    4.043869] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=event_msgs, len=18
[    4.044164] brcmfmac: brcmf_btcoex_attach enter
[    4.044167] brcmfmac: brcmf_pno_attach enter
[    4.044171] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=tdls_enable, len=4
[    4.044520] brcmfmac: brcmf_fweh_register event handler registered 
for TDLS_PEER_EVENT
[    4.044523] brcmfmac: brcmf_fweh_activate_events enable event SET_SSID
[    4.044525] brcmfmac: brcmf_fweh_activate_events enable event DEAUTH
[    4.044527] brcmfmac: brcmf_fweh_activate_events enable event DEAUTH_IND
[    4.044529] brcmfmac: brcmf_fweh_activate_events enable event ASSOC_IND
[    4.044531] brcmfmac: brcmf_fweh_activate_events enable event REASSOC_IND
[    4.044533] brcmfmac: brcmf_fweh_activate_events enable event 
DISASSOC_IND
[    4.044536] brcmfmac: brcmf_fweh_activate_events enable event LINK
[    4.044537] brcmfmac: brcmf_fweh_activate_events enable event MIC_ERROR
[    4.044540] brcmfmac: brcmf_fweh_activate_events enable event ROAM
[    4.044542] brcmfmac: brcmf_fweh_activate_events enable event 
PFN_NET_FOUND
[    4.044544] brcmfmac: brcmf_fweh_activate_events enable event 
PSM_WATCHDOG
[    4.044546] brcmfmac: brcmf_fweh_activate_events enable event PSK_SUP
[    4.044548] brcmfmac: brcmf_fweh_activate_events enable event IF
[    4.044550] brcmfmac: brcmf_fweh_activate_events enable event 
P2P_DISC_LISTEN_COMPLETE
[    4.044552] brcmfmac: brcmf_fweh_activate_events enable event RSSI
[    4.044554] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_COMPLETE
[    4.044557] brcmfmac: brcmf_fweh_activate_events enable event 
ESCAN_RESULT
[    4.044559] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_OFF_CHAN_COMPLETE
[    4.044561] brcmfmac: brcmf_fweh_activate_events enable event 
P2P_PROBEREQ_MSG
[    4.044564] brcmfmac: brcmf_fweh_activate_events enable event 
ACTION_FRAME_RX
[    4.044566] brcmfmac: brcmf_fweh_activate_events enable event 
TDLS_PEER_EVENT
[    4.044568] brcmfmac: brcmf_fweh_activate_events enable event IF
[    4.044570] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=event_msgs, len=18
[    4.044864] brcmfmac: brcmf_net_attach Enter, bsscfgidx=0 
mac=d4:9c:dd:f5:bf:7e
[    4.046336] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    4.046350] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    4.046352] brcmfmac: check_vif_up device is not ready : status (0)
[    4.046386] brcmfmac: brcmf_net_attach wlan0: Broadcom Dongle Host Driver
[    4.052221] brcmfmac 0002:21:00.0 wlP2p33s0: renamed from wlan0
[    4.629677] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    4.629688] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    4.629689] brcmfmac: check_vif_up device is not ready : status (0)
[    4.629733] brcmfmac: brcmf_cfg80211_dump_station Enter, idx 0
[    4.630174] brcmfmac: brcmf_fil_cmd_data_get ifidx=0, cmd=159, 
len=772, err=0
[    4.970573] brcmfmac: brcmf_netdev_open Enter, bsscfgidx=0
[    4.972550] brcmfmac: brcmf_fil_cmd_data Firmware error: 
BCME_UNSUPPORTED (-23)
[    4.972552] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, name=toe_ol, 
len=4, err=-52
[    4.972558] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=2, value=0
[    4.999772] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=185, value=40
[    4.999812] brcmfmac: brcmf_fweh_event_worker event IF (54) ifidx 0 
bsscfg 0 addr d4:9c:dd:f5:bf:7e
[    4.999819] brcmfmac: brcmf_fweh_event_worker   version 2 flags 0 
status 0 reason 0
[    4.999829] brcmfmac: brcmf_fweh_handle_if_event action: 1 ifidx: 0 
bsscfgidx: 0 flags: 0 role: 0
[    4.999832] brcmfmac: brcmf_fweh_handle_if_event adding wl0 
(d4:9c:dd:f5:bf:7e)
[    4.999835] brcmfmac: brcmf_add_if Enter, bsscfgidx=0, ifidx=0
[    4.999838] brcmfmac: brcmf_add_if netdev:wlP2p33s0 ignore IF event
[    5.000315] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=187, value=40
[    5.000563] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=258, value=120
[    5.000922] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=86, value=2
[    5.001170] brcmfmac: brcmf_config_dongle power save set to enabled
[    5.001172] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=bcn_timeout, len=4
[    5.001399] brcmfmac: brcmf_dongle_roam Internal Roaming = On
[    5.001401] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=roam_off, len=4
[    5.001737] brcmfmac: brcmf_fil_cmd_data_set ifidx=0, cmd=55, len=8
[    5.001866] brcmfmac: brcmf_fil_cmd_data_set ifidx=0, cmd=57, len=8
[    5.002110] brcmfmac: brcmf_cfg80211_change_iface Enter, bsscfgidx=0, 
type=2
[    5.002112] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=20, value=1
[    5.002512] brcmfmac: brcmf_cfg80211_change_iface IF Type = Infra
[    5.002514] brcmfmac: brcmf_cfg80211_change_iface Exit
[    5.002515] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arp_ol, 
len=4
[    5.002722] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arpoe, len=4
[    5.003040] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ARP offload to 0x9
[    5.003042] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=ndoe, len=4
[    5.003358] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ND offload to 0x9
[    5.003360] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=219, value=1
[    5.003606] brcmfmac: _brcmf_set_multicast_list Enter, bsscfgidx=0
[    5.003608] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=mcast_list, len=10
[    5.003659] brcmfmac: brcmf_cfg80211_set_power_mgmt Enter
[    5.003660] brcmfmac: brcmf_cfg80211_set_power_mgmt power save enabled
[    5.003992] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=86, value=2
[    5.004224] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=allmulti, len=4
[    5.004614] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=pm2_sleep_ret, len=4
[    5.004824] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=10, value=0
[    5.004827] brcmfmac: brcmf_cfg80211_set_power_mgmt Exit
[    5.004954] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arp_ol, 
len=4
[    5.005270] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arpoe, len=4
[    5.005474] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ARP offload to 0x9
[    5.005475] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=ndoe, len=4
[    5.005788] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ND offload to 0x9
[    5.005790] brcmfmac: _brcmf_set_multicast_list Enter, bsscfgidx=0
[    5.005792] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=mcast_list, len=16
[    5.006050] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=allmulti, len=4
[    5.006293] brcmfmac: brcmf_fil_cmd_int_set ifidx=0, cmd=10, value=0
[    5.006420] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arp_ol, 
len=4
[    5.006626] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=arpoe, len=4
[    5.006835] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ARP offload to 0x9
[    5.006837] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=ndoe, len=4
[    5.007148] brcmfmac: brcmf_configure_arp_nd_offload successfully 
configured (1) ND offload to 0x9
[    5.008354] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    5.008362] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    5.008607] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=qtxpower, len=4, err=0
[    5.008610] brcmfmac: brcmf_cfg80211_get_tx_power Exit (0x7f 31)
[    5.008859] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    5.008862] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    5.009112] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=qtxpower, len=4, err=0
[    5.009114] brcmfmac: brcmf_cfg80211_get_tx_power Exit (0x7f 31)
[    5.009404] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    5.009408] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    5.009693] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=qtxpower, len=4, err=0
[    5.009695] brcmfmac: brcmf_cfg80211_get_tx_power Exit (0x7f 31)
[    5.009851] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    5.009855] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    5.010102] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=qtxpower, len=4, err=0
[    5.010105] brcmfmac: brcmf_cfg80211_get_tx_power Exit (0x7f 31)
[    5.010732] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=chanspec, len=4, err=0
[    5.010737] brcmfmac: brcmf_cfg80211_get_tx_power Enter
[    5.010984] brcmfmac: brcmf_fil_iovar_data_get ifidx=0, 
name=qtxpower, len=4, err=0
[    5.010987] brcmfmac: brcmf_cfg80211_get_tx_power Exit (0x7f 31)
[    5.035480] brcmfmac: brcmf_cfg80211_flush_pmksa Enter
[    5.035489] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, 
name=pmkid_info, len=356
[    5.035999] brcmfmac: brcmf_cfg80211_flush_pmksa Exit
[    5.083266] brcmfmac: brcmf_cfg80211_scan Enter
[    5.083271] brcmfmac: brcmf_cfg80211_scan START ESCAN
[    5.083272] brcmfmac: brcmf_vif_set_mgmt_ie bsscfgidx 0, pktflag : 0x10
[    5.083275] brcmfmac: brcmf_do_escan Enter
[    5.083277] brcmfmac: brcmf_run_escan E-SCAN START
[    5.083278] brcmfmac: brcmf_escan_prep ### List of channelspecs to 
scan ### 36
[    5.083280] brcmfmac: brcmf_escan_prep Chan : 1, Channel spec: 1001
[    5.083281] brcmfmac: brcmf_escan_prep Chan : 2, Channel spec: 1002
[    5.083282] brcmfmac: brcmf_escan_prep Chan : 3, Channel spec: 1003
[    5.083283] brcmfmac: brcmf_escan_prep Chan : 4, Channel spec: 1004
[    5.083284] brcmfmac: brcmf_escan_prep Chan : 5, Channel spec: 1005
[    5.083285] brcmfmac: brcmf_escan_prep Chan : 6, Channel spec: 1006
[    5.083286] brcmfmac: brcmf_escan_prep Chan : 7, Channel spec: 1007
[    5.083287] brcmfmac: brcmf_escan_prep Chan : 8, Channel spec: 1008
[    5.083287] brcmfmac: brcmf_escan_prep Chan : 9, Channel spec: 1009
[    5.083288] brcmfmac: brcmf_escan_prep Chan : 10, Channel spec: 100a
[    5.083289] brcmfmac: brcmf_escan_prep Chan : 11, Channel spec: 100b
[    5.083290] brcmfmac: brcmf_escan_prep Chan : 36, Channel spec: d024
[    5.083291] brcmfmac: brcmf_escan_prep Chan : 40, Channel spec: d028
[    5.083292] brcmfmac: brcmf_escan_prep Chan : 44, Channel spec: d02c
[    5.083293] brcmfmac: brcmf_escan_prep Chan : 48, Channel spec: d030
[    5.083294] brcmfmac: brcmf_escan_prep Chan : 52, Channel spec: d034
[    5.083294] brcmfmac: brcmf_escan_prep Chan : 56, Channel spec: d038
[    5.083295] brcmfmac: brcmf_escan_prep Chan : 60, Channel spec: d03c
[    5.083296] brcmfmac: brcmf_escan_prep Chan : 64, Channel spec: d040
[    5.083297] brcmfmac: brcmf_escan_prep Chan : 100, Channel spec: d064
[    5.083298] brcmfmac: brcmf_escan_prep Chan : 104, Channel spec: d068
[    5.083299] brcmfmac: brcmf_escan_prep Chan : 108, Channel spec: d06c
[    5.083300] brcmfmac: brcmf_escan_prep Chan : 112, Channel spec: d070
[    5.083301] brcmfmac: brcmf_escan_prep Chan : 116, Channel spec: d074
[    5.083301] brcmfmac: brcmf_escan_prep Chan : 120, Channel spec: d078
[    5.083303] brcmfmac: brcmf_escan_prep Chan : 124, Channel spec: d07c
[    5.083303] brcmfmac: brcmf_escan_prep Chan : 128, Channel spec: d080
[    5.083304] brcmfmac: brcmf_escan_prep Chan : 132, Channel spec: d084
[    5.083305] brcmfmac: brcmf_escan_prep Chan : 136, Channel spec: d088
[    5.083306] brcmfmac: brcmf_escan_prep Chan : 140, Channel spec: d08c
[    5.083307] brcmfmac: brcmf_escan_prep Chan : 144, Channel spec: d090
[    5.083308] brcmfmac: brcmf_escan_prep Chan : 149, Channel spec: d095
[    5.083309] brcmfmac: brcmf_escan_prep Chan : 153, Channel spec: d099
[    5.083310] brcmfmac: brcmf_escan_prep Chan : 157, Channel spec: d09d
[    5.083310] brcmfmac: brcmf_escan_prep Chan : 161, Channel spec: d0a1
[    5.083311] brcmfmac: brcmf_escan_prep Chan : 165, Channel spec: d0a5
[    5.083312] brcmfmac: brcmf_escan_prep ### List of SSIDs to scan ### 1
[    5.083313] brcmfmac: brcmf_escan_prep 0: Broadcast scan
[    5.083314] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=escan, 
len=180
[    5.083722] brcmfmac: brcmf_fil_cmd_data Firmware error: BCME_VERSION 
(-37)
[    8.316080] brcmfmac: brcmf_cfg80211_scan Enter
[    8.316085] brcmfmac: brcmf_cfg80211_scan START ESCAN
[    8.316087] brcmfmac: brcmf_vif_set_mgmt_ie bsscfgidx 0, pktflag : 0x10
[    8.316091] brcmfmac: brcmf_do_escan Enter
[    8.316092] brcmfmac: brcmf_run_escan E-SCAN START
[    8.316093] brcmfmac: brcmf_escan_prep ### List of channelspecs to 
scan ### 36
[    8.316095] brcmfmac: brcmf_escan_prep Chan : 1, Channel spec: 1001
[    8.316096] brcmfmac: brcmf_escan_prep Chan : 2, Channel spec: 1002
[    8.316097] brcmfmac: brcmf_escan_prep Chan : 3, Channel spec: 1003
[    8.316098] brcmfmac: brcmf_escan_prep Chan : 4, Channel spec: 1004
[    8.316099] brcmfmac: brcmf_escan_prep Chan : 5, Channel spec: 1005
[    8.316100] brcmfmac: brcmf_escan_prep Chan : 6, Channel spec: 1006
[    8.316101] brcmfmac: brcmf_escan_prep Chan : 7, Channel spec: 1007
[    8.316102] brcmfmac: brcmf_escan_prep Chan : 8, Channel spec: 1008
[    8.316103] brcmfmac: brcmf_escan_prep Chan : 9, Channel spec: 1009
[    8.316104] brcmfmac: brcmf_escan_prep Chan : 10, Channel spec: 100a
[    8.316104] brcmfmac: brcmf_escan_prep Chan : 11, Channel spec: 100b
[    8.316105] brcmfmac: brcmf_escan_prep Chan : 36, Channel spec: d024
[    8.316106] brcmfmac: brcmf_escan_prep Chan : 40, Channel spec: d028
[    8.316107] brcmfmac: brcmf_escan_prep Chan : 44, Channel spec: d02c
[    8.316108] brcmfmac: brcmf_escan_prep Chan : 48, Channel spec: d030
[    8.316109] brcmfmac: brcmf_escan_prep Chan : 52, Channel spec: d034
[    8.316110] brcmfmac: brcmf_escan_prep Chan : 56, Channel spec: d038
[    8.316111] brcmfmac: brcmf_escan_prep Chan : 60, Channel spec: d03c
[    8.316112] brcmfmac: brcmf_escan_prep Chan : 64, Channel spec: d040
[    8.316113] brcmfmac: brcmf_escan_prep Chan : 100, Channel spec: d064
[    8.316114] brcmfmac: brcmf_escan_prep Chan : 104, Channel spec: d068
[    8.316115] brcmfmac: brcmf_escan_prep Chan : 108, Channel spec: d06c
[    8.316115] brcmfmac: brcmf_escan_prep Chan : 112, Channel spec: d070
[    8.316116] brcmfmac: brcmf_escan_prep Chan : 116, Channel spec: d074
[    8.316117] brcmfmac: brcmf_escan_prep Chan : 120, Channel spec: d078
[    8.316118] brcmfmac: brcmf_escan_prep Chan : 124, Channel spec: d07c
[    8.316119] brcmfmac: brcmf_escan_prep Chan : 128, Channel spec: d080
[    8.316120] brcmfmac: brcmf_escan_prep Chan : 132, Channel spec: d084
[    8.316121] brcmfmac: brcmf_escan_prep Chan : 136, Channel spec: d088
[    8.316122] brcmfmac: brcmf_escan_prep Chan : 140, Channel spec: d08c
[    8.316123] brcmfmac: brcmf_escan_prep Chan : 144, Channel spec: d090
[    8.316124] brcmfmac: brcmf_escan_prep Chan : 149, Channel spec: d095
[    8.316125] brcmfmac: brcmf_escan_prep Chan : 153, Channel spec: d099
[    8.316125] brcmfmac: brcmf_escan_prep Chan : 157, Channel spec: d09d
[    8.316126] brcmfmac: brcmf_escan_prep Chan : 161, Channel spec: d0a1
[    8.316127] brcmfmac: brcmf_escan_prep Chan : 165, Channel spec: d0a5
[    8.316128] brcmfmac: brcmf_escan_prep ### List of SSIDs to scan ### 1
[    8.316129] brcmfmac: brcmf_escan_prep 0: Broadcast scan
[    8.316130] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=escan, 
len=180
[    8.317949] brcmfmac: brcmf_fil_cmd_data Firmware error: BCME_VERSION 
(-37)
[   17.202228] brcmfmac: brcmf_cfg80211_scan Enter
[   17.202237] brcmfmac: brcmf_cfg80211_scan START ESCAN
[   17.202240] brcmfmac: brcmf_vif_set_mgmt_ie bsscfgidx 0, pktflag : 0x10
[   17.202247] brcmfmac: brcmf_do_escan Enter
[   17.202250] brcmfmac: brcmf_run_escan E-SCAN START
[   17.202253] brcmfmac: brcmf_escan_prep ### List of channelspecs to 
scan ### 36
[   17.202257] brcmfmac: brcmf_escan_prep Chan : 1, Channel spec: 1001
[   17.202260] brcmfmac: brcmf_escan_prep Chan : 2, Channel spec: 1002
[   17.202263] brcmfmac: brcmf_escan_prep Chan : 3, Channel spec: 1003
[   17.202265] brcmfmac: brcmf_escan_prep Chan : 4, Channel spec: 1004
[   17.202267] brcmfmac: brcmf_escan_prep Chan : 5, Channel spec: 1005
[   17.202270] brcmfmac: brcmf_escan_prep Chan : 6, Channel spec: 1006
[   17.202272] brcmfmac: brcmf_escan_prep Chan : 7, Channel spec: 1007
[   17.202275] brcmfmac: brcmf_escan_prep Chan : 8, Channel spec: 1008
[   17.202277] brcmfmac: brcmf_escan_prep Chan : 9, Channel spec: 1009
[   17.202280] brcmfmac: brcmf_escan_prep Chan : 10, Channel spec: 100a
[   17.202283] brcmfmac: brcmf_escan_prep Chan : 11, Channel spec: 100b
[   17.202285] brcmfmac: brcmf_escan_prep Chan : 36, Channel spec: d024
[   17.202288] brcmfmac: brcmf_escan_prep Chan : 40, Channel spec: d028
[   17.202290] brcmfmac: brcmf_escan_prep Chan : 44, Channel spec: d02c
[   17.202293] brcmfmac: brcmf_escan_prep Chan : 48, Channel spec: d030
[   17.202295] brcmfmac: brcmf_escan_prep Chan : 52, Channel spec: d034
[   17.202298] brcmfmac: brcmf_escan_prep Chan : 56, Channel spec: d038
[   17.202300] brcmfmac: brcmf_escan_prep Chan : 60, Channel spec: d03c
[   17.202303] brcmfmac: brcmf_escan_prep Chan : 64, Channel spec: d040
[   17.202305] brcmfmac: brcmf_escan_prep Chan : 100, Channel spec: d064
[   17.202308] brcmfmac: brcmf_escan_prep Chan : 104, Channel spec: d068
[   17.202311] brcmfmac: brcmf_escan_prep Chan : 108, Channel spec: d06c
[   17.202313] brcmfmac: brcmf_escan_prep Chan : 112, Channel spec: d070
[   17.202316] brcmfmac: brcmf_escan_prep Chan : 116, Channel spec: d074
[   17.202318] brcmfmac: brcmf_escan_prep Chan : 120, Channel spec: d078
[   17.202321] brcmfmac: brcmf_escan_prep Chan : 124, Channel spec: d07c
[   17.202323] brcmfmac: brcmf_escan_prep Chan : 128, Channel spec: d080
[   17.202326] brcmfmac: brcmf_escan_prep Chan : 132, Channel spec: d084
[   17.202328] brcmfmac: brcmf_escan_prep Chan : 136, Channel spec: d088
[   17.202330] brcmfmac: brcmf_escan_prep Chan : 140, Channel spec: d08c
[   17.202333] brcmfmac: brcmf_escan_prep Chan : 144, Channel spec: d090
[   17.202335] brcmfmac: brcmf_escan_prep Chan : 149, Channel spec: d095
[   17.202338] brcmfmac: brcmf_escan_prep Chan : 153, Channel spec: d099
[   17.202340] brcmfmac: brcmf_escan_prep Chan : 157, Channel spec: d09d
[   17.202343] brcmfmac: brcmf_escan_prep Chan : 161, Channel spec: d0a1
[   17.202345] brcmfmac: brcmf_escan_prep Chan : 165, Channel spec: d0a5
[   17.202348] brcmfmac: brcmf_escan_prep ### List of SSIDs to scan ### 1
[   17.202350] brcmfmac: brcmf_escan_prep 0: Broadcast scan
[   17.202352] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=escan, 
len=180
[   17.204223] brcmfmac: brcmf_fil_cmd_data Firmware error: BCME_VERSION 
(-37)
root@inmusic-az04-mpc:~# [   44.202645] ieee80211 phy0: brcmf_run_escan: 
error (-52)
[   44.202649] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)

-Andy


> Regards,
> Arend

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Arend van Spriel]

[-- Attachment #1: Type: text/plain, Size: 1065 bytes --]

On 11/24/2022 1:54 PM, Andy Green wrote:
> [   17.202348] brcmfmac: brcmf_escan_prep ### List of SSIDs to scan ### 1
> [   17.202350] brcmfmac: brcmf_escan_prep 0: Broadcast scan
> [   17.202352] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=escan, 
> len=180
> [   17.204223] brcmfmac: brcmf_fil_cmd_data Firmware error: BCME_VERSION 
> (-37)
> root@inmusic-az04-mpc:~# [   44.202645] ieee80211 phy0: brcmf_run_escan: 
> error (-52)
> [   44.202649] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)

The root cause is easy. Basically the brcmfmac driver is not on par with 
the firmware. brcmfmac uses escan api version 1 and firmware expects 
version 2. Fixing it is going to be more difficult. Things start to 
diverge on firmware API level. I already had a patch on the shelf 
supporting version 2, but it does not match the API your firmware expects.

I was working on supporting multiple vendors, but I need some rework. 
The scanning issue confirms the need for per-vendor support. Hopefully I 
can spend some time on it.

Regards,
Arend

[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4219 bytes --]

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 13:47, Arend van Spriel wrote:
> On 11/24/2022 1:54 PM, Andy Green wrote:
>> [   17.202348] brcmfmac: brcmf_escan_prep ### List of SSIDs to scan ### 1
>> [   17.202350] brcmfmac: brcmf_escan_prep 0: Broadcast scan
>> [   17.202352] brcmfmac: brcmf_fil_iovar_data_set ifidx=0, name=escan, 
>> len=180
>> [   17.204223] brcmfmac: brcmf_fil_cmd_data Firmware error: 
>> BCME_VERSION (-37)
>> root@inmusic-az04-mpc:~# [   44.202645] ieee80211 phy0: 
>> brcmf_run_escan: error (-52)
>> [   44.202649] ieee80211 phy0: brcmf_cfg80211_scan: scan error (-52)
> 
> The root cause is easy. Basically the brcmfmac driver is not on par with 
> the firmware. brcmfmac uses escan api version 1 and firmware expects 
> version 2. Fixing it is going to be more difficult. Things start to 

It seems so.

> diverge on firmware API level. I already had a patch on the shelf 
> supporting version 2, but it does not match the API your firmware expects.
> 
> I was working on supporting multiple vendors, but I need some rework. 
> The scanning issue confirms the need for per-vendor support. Hopefully I 
> can spend some time on it.

If it makes sense for you, I'm willing take a look at your alpha patch 
with the idea to align with what the OOT driver does for scan, but I'm 
not sure how intricate wading in there is going to get.

I'd need at least some guidance about how to detect which api stack it 
wants dynamically if there's more than one "version 2" and likely some 
hand-holding about firmware or soc-specific things in places.

-Andy

Re: AP6275 / bcm43752 pcie on mainline brcmf

[Andy Green]

On 11/24/22 14:16, Andy Green wrote:

>> diverge on firmware API level. I already had a patch on the shelf 
>> supporting version 2, but it does not match the API your firmware 
>> expects.
>>
>> I was working on supporting multiple vendors, but I need some rework. 
>> The scanning issue confirms the need for per-vendor support. Hopefully 
>> I can spend some time on it.
> 
> If it makes sense for you, I'm willing take a look at your alpha patch 
> with the idea to align with what the OOT driver does for scan, but I'm 
> not sure how intricate wading in there is going to get.
> 
> I'd need at least some guidance about how to detect which api stack it 
> wants dynamically if there's more than one "version 2" and likely some 
> hand-holding about firmware or soc-specific things in places.

... there's the additional quirk that my rebase on mainline of the OOT 
driver starts up like mainline now, but doesn't return any scan results 
as it is.

So aligning to that may only take us so far without some possibly 
non-open knowledge about what the firmware really wants from its 
perspective directly without referencing dhd.

-Andy