Re: [gentoo-user] Loading a Firmware Module By hand?

[Hunter Jozwiak]

On 12/18/17, Floyd Anderson <f.a@××××.net> wrote:
> On Mon, 18 Dec 2017 19:15:04 +0000
> Hunter Jozwiak <hunter.t.joz@×××××.com> wrote:
>>On 12/18/17, Mick <michaelkintzios@×××××.com> wrote:
>>> On Monday, 18 December 2017 05:11:20 GMT Hunter Jozwiak wrote:
>>>> Hmm. I have kernel 4.14.7 and linux-firmware 20171206. I tried version
>>>> 999999999 as well, but that didn't help matters, either. Nor did
>>>> compiling the firmware into the kernel; either 4.14 is too old, or it
>>>> is too new.
>>>
>>> I'd think they are both too new?
>>>
>>>> I tried copying the firmware my live iso was using, but
>>>> that didn't help either.
>>>
>>> If the live iso works, start with using the same kernel release and
>>> linux-
>>> firmware version, to see if this works as expected on your installation.
>>> Then
>>> update kernel sources and firmware to the latest stable and see if this
>>> works
>>> too.
>>>
>>> From there on you can move into ~arch to find the version at which
>>> things
>>> break.
>>> --
>>> Regards,
>>> Mick
>>Okay, here are the dmesg messages:
>>https://paste.pound-python.org/show/nrNfBAEPfh9W7ZIeItJC/
>>The present kernel configuration, as of yesterday evening:
>>https://paste.pound-python.org/show/NmNB8nzLuEjmsz74kRVE/
>>Make and model of the offending card: Qualcomm 6174 revision 20.
>>Not exactly what the -2 error means, but I will try as Mick suggested
>>and work my way forwards to see what I can get working.
>>
>
> Hi,
>
> I was a little bit confused from your earlier mentioned
>
>     /lib/firmware/ath10k/QCABLEFAGD/HW3.0
>
> I wonder where this firmware come from. I cannot find it, neither with
> the identifier ‘QCABLEFAGD’ nor an upper cased ‘HW3.0’ directory in
> [1][2][3][4].
>
> Your log tells an other story now. As far as I can tell:
>
>     “Direct firmware load for […] failed with error -2”
>
> comes from _request_firmware and fw_get_filesystem_firmware functions in
> file [5]. The symbolic error name ENOENT, see `man 3 errno`, means
> something like ‘No such element’ or ‘No such file or directory’.
>
> Even CONFIG_EXTRA_FIRMWARE_DIR defaults to ‘/lib/firmware’, there should
> be whether an entry ‘# CONFIG_EXTRA_FIRMWARE_DIR is not set’ or
> something like ‘CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware"’ in your
> config. Its unclear to me why it is not there.
>
> If you want to build into the kernel again, check
> ‘/lib/firmware/ath10k/QCA6174/hw2.1/firmware-6.bin’ exists and set:
>
>     CONFIG_FIRMWARE_IN_KERNEL=y
>     CONFIG_EXTRA_FIRMWARE="ath10k/QCA6174/hw2.1/firmware-6.bin"
>     CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware"
>
> Hope that helps.
>
>
> [1]
> <https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/log/ath10k>
> [2]
> <https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/tree/drivers/net/wireless/ath/ath10k/hw.h?h=v4.14.7>
> [3] <https://github.com/kvalo/ath10k-firmware>
> [4] <https://wireless.wiki.kernel.org/en/users/Drivers/ath10k/firmware>
> [5]
> <https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/tree/drivers/base/firmware_class.c?h=v4.14.7#n386>
>
>
>
> --
> Regards,
> floyd
>
>
>
That particular firmware does not exist. I am trying to figure how far
back I need to rollback the kernel. Sabayon's live ISO is using
4.13.0, but that is no longer in the tree. Not sure if I need to go
back to the 4.12 ebuilds, or if 4.13.5 will suit my needs; perhaps
there is a bit of source code that I can look at to see what is being
used for the firmware?