From: Rob Clark <robdclark(a)chromium.org>
Now that we can deal gracefully with bootloader (firmware) initialized
display on aarch64 laptops[1], the next step is to deal with the fact
that the same model of laptop can have one of multiple different panels.
(For the yoga c630 that I have, I know of at least two possible panels,
there might be a third.)
This is actually a scenario that comes up frequently in phones and
tablets as well, so it is useful to have an upstream solution for this.
The basic idea is to add a 'panel-id' property in dt chosen node, and
use that to pick the endpoint we look at when loading the panel driver,
e.g.
/ {
chosen {
panel-id = <0xc4>;
};
ivo_panel {
compatible = "ivo,m133nwf4-r0";
power-supply = <&vlcm_3v3>;
no-hpd;
ports {
port {
ivo_panel_in_edp: endpoint {
remote-endpoint = <&sn65dsi86_out_ivo>;
};
};
};
};
boe_panel {
compatible = "boe,nv133fhm-n61";
power-supply = <&vlcm_3v3>;
no-hpd;
ports {
port {
boe_panel_in_edp: endpoint {
remote-endpoint = <&sn65dsi86_out_boe>;
};
};
};
};
sn65dsi86: bridge@2c {
compatible = "ti,sn65dsi86";
...
ports {
#address-cells = <1>;
#size-cells = <0>;
...
port@1 {
#address-cells = <1>;
#size-cells = <0>;
reg = <1>;
endpoint@c4 {
reg = <0xc4>;
remote-endpoint = <&boe_panel_in_edp>;
};
endpoint@c5 {
reg = <0xc5>;
remote-endpoint = <&ivo_panel_in_edp>;
};
};
};
}
};
Note that the panel-id is potentially a sparse-int. The values I've
seen so far on aarch64 laptops are:
* 0xc2
* 0xc3
* 0xc4
* 0xc5
* 0x8011
* 0x8012
* 0x8055
* 0x8056
At least on snapdragon aarch64 laptops, they can be any u32 value.
However, on these laptops, the bootloader/firmware is not populating the
chosen node, but instead providing an "UEFIDisplayInfo" variable, which
contains the panel id. Unfortunately EFI variables are only available
before ExitBootServices, so the second patch checks for this variable
before EBS and populates the /chosen/panel-id variable.
[1] https://patchwork.freedesktop.org/series/63001/
Rob Clark (4):
dt-bindings: chosen: document panel-id binding
efi/libstub: detect panel-id
drm: add helper to lookup panel-id
drm/bridge: ti-sn65dsi86: use helper to lookup panel-id
Documentation/devicetree/bindings/chosen.txt | 69 ++++++++++++++++++++
drivers/firmware/efi/libstub/arm-stub.c | 49 ++++++++++++++
drivers/firmware/efi/libstub/efistub.h | 2 +
drivers/firmware/efi/libstub/fdt.c | 9 +++
drivers/gpu/drm/bridge/ti-sn65dsi86.c | 5 +-
drivers/gpu/drm/drm_of.c | 21 ++++++
include/drm/drm_of.h | 7 ++
7 files changed, 160 insertions(+), 2 deletions(-)
--
2.20.1
This patchset builds on top of Leif's DtbLoader, adding support to
pick a laptop specific dtb based on SMBIOS tables, and patching
for /chosen/panel-id based on UEFIDisplayInfo. I'll send a new
version of the kernel side patchset that uses /chosen/panel-id to
pick the appropriate panel driver in the near future (ie. I'll
probably get to it tomorrow)
Rob Clark (5):
DtbLoader: refactor out helper to try loading dtb
DtbLoader: Try to pick dtb based on SMBIOS tables
DtbLoader: move CRC calculation
DtbLoader: resize fdt
DtbLoader: add panel-id detection/fixup for qcom devices
.../Application/ConfigTableLoader/Common.h | 2 +
.../Application/ConfigTableLoader/DtbLoader.c | 437 +++++++++++++++---
.../ConfigTableLoader/DtbLoader.inf | 2 +
.../Application/ConfigTableLoader/Qcom.c | 94 ++++
.../Application/ConfigTableLoader/Qcom.h | 30 ++
5 files changed, 506 insertions(+), 59 deletions(-)
create mode 100644 EmbeddedPkg/Application/ConfigTableLoader/Qcom.c
create mode 100644 EmbeddedPkg/Application/ConfigTableLoader/Qcom.h
--
2.23.0
I'm curious, how have folks been adding the devicetree command to the
grub menu entry? So far I've been manually editing the grub.cfg since
that worked well with my development flow, but now things have changed
a bit for me, and I'm looking for a way that it would be appended
automatically when the config is autogenerated when installing a new
kernel.
-Jeff
hi All,
Is anyone running Debian user-space on a Yoga C630 or similar machine?
I am interested in understanding how you got it to work and if you have
written about your experience anywhere.
best regards,
Richard
--
Richard.Henwood(a)arm.com
Server Software Eco-System
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
I was able to get the Lenovo C630 to boot from the flash drive and have the
trackpad work. Awesome!
I'm now thinking the build looks good enough to install on the UFS drive.
I tried the "enabled installer".
I've updated WIndows. There are no pending updates.
I've shrunk the partiion.
I"m not running S mode
I've disabled secure boot.
I used Dimitri's Bionic installer. md5sum bde45aa67f19333955d00349aa6b2077
I flashed the installer onto a USB stick using gnome-disks
WIth the flash drive in the left USB-C port, it boots straight to Windows.
This is the same machine that was booting successfully from the flash drive
into Linux.
Any ideas?
Is there a way to copy the flash drive image that does work onto the UFS
drive?
Thanks
On Wed, 04 Sep 2019, David wrote:
> I am using the same USB flash drive, putting different images on it.
>
> I found and ran the "WIndows 10 update assistant" for May 2019. It starts
> up, briefly says "Checking for updates", then immediately says, "Thank you
> for updating to the latest version of WIndows". I assume that means I'm
> already at that level since it didn't seem to do anything. The filename is
> "Windows10Upgrade9252.exe"
Not sure what's going on then. The installer has worked on all of the
5-6 laptops (from 2-3 different batches) I've worked on.
Maybe there is something wrong with the way it was flashed. Have you
tried mounting it on your development machine. What does it look like
in `fdisk` and `gparted`?
> On Wed, Sep 4, 2019 at 5:05 AM Lee Jones <lee.jones(a)linaro.org> wrote:
>
> > On Tue, 03 Sep 2019, David wrote:
> >
> > > I was able to get the Lenovo C630 to boot from the flash drive and have
> > the
> > > trackpad work. Awesome!
> > >
> > > I'm now thinking the build looks good enough to install on the UFS drive.
> > >
> > > I tried the "enabled installer".
> > > I've updated WIndows. There are no pending updates.
> > > I've shrunk the partiion.
> > > I"m not running S mode
> > > I've disabled secure boot.
> > > I used Dimitri's Bionic installer. md5sum
> > bde45aa67f19333955d00349aa6b2077
> > > I flashed the installer onto a USB stick using gnome-disks
> > >
> > > WIth the flash drive in the left USB-C port, it boots straight to
> > Windows.
> > >
> > > This is the same machine that was booting successfully from the flash
> > drive
> > > into Linux.
> >
> > Are you using the same USB drive?
> >
> > > Is there a way to copy the flash drive image that does work onto the UFS
> > > drive?
> >
> > I wouldn't do that.
> >
> > The installer has been tested many times using the instructions you
> > quote above. Not entirely sure why it's not working for you though.
> >
> > It might be worth your time properly updating Windows. Boot back into
> > Windows and search for "Windows 10 Update Assistant". The version I'm
> > using is [May 2019 (1903)].
> >
> >
--
Lee Jones [李琼斯]
Linaro Services Technical Lead
Linaro.org │ Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog
These are cherrypicks from the aarch64-laptops [1] effort. I have
pinged those maintainers pointing out that these patches have not been
submitted upstream. In the future it should be... Otherwise, I'll keep
rebasing this.
This is useful on the AArch64 laptops [2] booting in ACPI mode, as they
are Win10 laptops they use/publish WMI, hence it should stop being x86
only and be enabled on arm64 too.
[1] https://github.com/aarch64-laptops/linux/tree/laptops-ubuntu
[2] ASUS NovaGo TP370QL, HP Envy x2, Lenovo Mixx 630 & Yoga C630, and
Samsung Book2
Ard Biesheuvel (2):
UBUNTU: SAUCE: acpi: move WMI subsystem to generic code
UBUNTU: SAUCE: acpi/wmi: move BMOF driver to generic code
Dimitri John Ledkov (1):
UBUNTU: [Config] Update WMI_BMOF annotations.
.../abi/5.3.0-0.1/arm64/generic.modules | 2 ++
debian.master/config/annotations | 4 +--
drivers/acpi/Kconfig | 33 +++++++++++++++++++
drivers/acpi/Makefile | 2 ++
drivers/{platform/x86 => acpi}/wmi-bmof.c | 0
drivers/{platform/x86 => acpi}/wmi.c | 0
drivers/platform/x86/Kconfig | 33 -------------------
drivers/platform/x86/Makefile | 2 --
8 files changed, 39 insertions(+), 37 deletions(-)
rename drivers/{platform/x86 => acpi}/wmi-bmof.c (100%)
rename drivers/{platform/x86 => acpi}/wmi.c (100%)
--
2.20.1
Hello, this appears to be a development-only mailing list, so pardon me if I’m disturbing anyone. I’m running a Samsung Galaxy Book 2 (using SD850 + QUALCOMM x20 modem + AMOLED touch screen, detachable KB, and stylus) and am wondering if the work being done mostly on Lenovo SD850 laptops apparently and older devices will apply to this hardware, or should I not hold my breath? Don’t worry, this isn’t an ETA or a complaint, I appreciate everything you’re doing and wish I knew more about programming so I could help out.
Sent from Mail for Windows 10
So, we still have sensor-hub and acpi-button disabled using kernel config.
I could patch the quirks into the kernel to skip loading those based
on DMI data.
However, that's a bit counter-productive as that's compiled in code
and enumerates SKUs.
Can I instead use kernel command line to skip initializing those
built-in and compiled modules?
Ie. would something like:
module_blacklist=hid-sensor-hub initcall_blacklist=acpi_button_driver_init
work, even when the kernel is configured with CONFIG_HID_SENSOR_HUB=m
and CONFIG_ACPI_BUTTON=y ?
Also not sure if I got the initcall function name right.
I am asking because, it may be easier for me to control cmdline
reliably, instead of distro kernel config.
--
Regards,
Dimitri.