On 05/08/2022 23:09, Rob Clark wrote:
On Fri, Aug 5, 2022 at 2:37 AM Dmitry Baryshkov
<dmitry.baryshkov@xxxxxxxxxx> wrote:
On 03/08/2022 23:37, Mark Pearson wrote:
The following changes since commit 150864a4d73e8c448eb1e2c68e65f07635fe1a66:
amdgpu partially revert "amdgpu: update beige goby to release 22.20"
(2022-07-25 14:16:04 -0400)
are available in the Git repository at:
https://github.com/mrhpearson/linux-firmware lenovo-thinkpad-x13s
for you to fetch changes up to 4ae4ae88918928e15006eb129ad981aa58216b59:
qcom: Add firmware for Lenovo ThinkPad X13s (2022-08-03 16:29:07 -0400)
Mark, it's so great to see firmware packages for Lenovo laptops being
submitted!
----------------------------------------------------------------
Mark Pearson (1):
qcom: Add firmware for Lenovo ThinkPad X13s
WHENCE | 8 ++++++++
qcom/LENOVO/21BX/adspr.jsn | 28 ++++++++++++++++++++++++++++
qcom/LENOVO/21BX/adspua.jsn | 29 +++++++++++++++++++++++++++++
qcom/LENOVO/21BX/battmgr.jsn | 22 ++++++++++++++++++++++
qcom/LENOVO/21BX/cdspr.jsn | 22 ++++++++++++++++++++++
qcom/LENOVO/21BX/qcadsp8280.mbn | Bin 0 -> 14367860 bytes
qcom/LENOVO/21BX/qccdsp8280.mbn | Bin 0 -> 3575808 bytes
qcom/LENOVO/21BX/qcdxkmsuc8280.mbn | Bin 0 -> 14392 bytes
qcom/LENOVO/21BX/qcslpi8280.mbn | Bin 0 -> 6213632 bytes
9 files changed, 109 insertions(+)
create mode 100644 qcom/LENOVO/21BX/adspr.jsn
create mode 100644 qcom/LENOVO/21BX/adspua.jsn
create mode 100644 qcom/LENOVO/21BX/battmgr.jsn
create mode 100644 qcom/LENOVO/21BX/cdspr.jsn
create mode 100755 qcom/LENOVO/21BX/qcadsp8280.mbn
create mode 100755 qcom/LENOVO/21BX/qccdsp8280.mbn
create mode 100755 qcom/LENOVO/21BX/qcdxkmsuc8280.mbn
create mode 100755 qcom/LENOVO/21BX/qcslpi8280.mbn
I think it's the time to settle onto firmware paths for the
Qualcomm-based devices. At some point it was proposed to use
qcom/<SoC>/device/*mbn.
So, in your case it could be qcom/sc8280xp/Lenovo_21BX/
I don't think it is a good idea.. we should stick to a path that can
be constructed from SMBIOS fields if we ever want a chance to get ACPI
boot with acceleration working
Yes, it's also true. I do not quite like the idea that we are going to
end up with two different path types. One using the DMI/SMBIOS ids,
another one using SoC and machine id. Maybe we can settle on using
SoC+id and then using DMI-based links.
There is another point to remember. I can imagine that once WoA/LoA
laptops are more common, there can be several different DMI models
sharing common firmware. So, there will be symlinks involved anyway.
--
With best wishes
Dmitry
