RCA Cambio W101 v2 firmware.fw Install

[cjohnson80]

I'm struggling to get this driver working on this tablet. I feel like I'm missing something here.
I can modify the touchscreen_dmi.c but struggle with recompiling the kernel. Is there an up to date walkthrough or guide for this?

[onitake]

Which distribution are you using?
Are you getting compilation errors?

[cjohnson80]

I finished compiling but something still isn't right. Haven't had much time to troubleshoot yet but hopefully will get it working.
I am, however, intrigued by your kernel. I'll give that a go and see what results I can get out of it too

[cjohnson80]

I started from scratch and tried out your kernel by installing as per your generic instructions, but I had no success.
The output from dmesg is this
[ 79.394131] silead_ts i2c-MSSL1680:00: supply vddio not found, using dummy regulator
[ 79.394940] silead_ts i2c-MSSL1680:00: supply avdd not found, using dummy regulator
[ 79.395264] i2c_designware 80860F41:03: i2c_dw_handle_tx_abort: lost arbitration
[ 80.457836] i2c_designware 80860F41:03: controller timed out
[ 80.457857] silead_ts i2c-MSSL1680:00: Chip ID read error -110
[ 80.458683] silead_ts: probe of i2c-MSSL1680:00 failed with error -110

[cjohnson80]

I thihnk I copied the wrong firmware into the /lib/firmware/silead folder. After making that change I have some touch happening.
I continued along the list with running evtest
Found the variables for min/max x & y
sudo rmmod silead
sudo modprobe silead settings=touchscreen-min-x=4;touchscreen-min-y=20;touchscreen-size-x=1644;touchscreen-size-y=874;touchscreen-swapped-x-y

This is the output as follows:

bash: touchscreen-min-y=20: command not found...
bash: touchscreen-size-x=1644: command not found...
bash: touchscreen-size-y=874: command not found...
bash: touchscreen-swapped-x-y: command not found...

[jwrdegoede]

Ah, my bad. (ba)sh thinks the commadline ends at the first ; so you need to put single or double quotes around the entire settings= argument. Please give that a try.

[cjohnson80]

Ok cool. That did the trick! (I gave it a quick go before heading out) Everything seems to be working as expected touch wise but I’ll have to give it some further testing when I get back home.
Then I’ll see if I can duplicate it again.

[cjohnson80]

Here are the settings I used for my device: modprobe silead settings="touchscreen-min-x=4;touchscreen-min-y=20;touchscreen-size-x=1644;touchscreen-size-y=874;touchscreen-swapped-x-y"
What information do you need from dmidecode? or just everything?

[cjohnson80]

Here is the full output
dmidecode RCA Cambio.txt

[jwrdegoede]

Hmm, that DMI info is surprisingly clean/useful (good). Attached here is the patch which I have cooked up. Note when running a kernel with this patch, the kernel will expect the firmware file to be named /lib/firmware/silead/gsl1680-rca-cambio-w101-v2.fw so please make a copy of your firmware there.

0001-platform-x86-touchscreen_dmi-Add-info-for-the-RCA-Ca.patch.txt

A Fedora kernel with this patch added is now building here:

https://koji.fedoraproject.org/koji/taskinfo?taskID=92563362

(finished in about 2 hours from now I guess)

Please give this a try. If this work I will push the patch to the official mainline Linux kernels. If this works can you please submit a pull-request here, adding your firmware as firmware/linux/silead/gsl1680-rca-cambio-w101-v2.fw ?

[jwrdegoede]

p.s. Note the new kernel no longer has the settings= patch ..., so if it does not work, don't expect the settings= thing to work with this new kernel build.

[cjohnson80]

Not sure if I'm doing something wrong but this doesn't seem to work.

[jwrdegoede]

What do uname -a and dmesg | grep silead give as output (when run from a terminal) ?

[cjohnson80]

uname -a && dmesg | grep silead
Linux fedora 5.19.12-200.gsl_fw_dmi.fc36.x86_64 #1 SMP PREEMPT_DYNAMIC Mon Oct 3 12:24:20 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
[ 15.236761] silead_ts i2c-MSSL1680:00: supply vddio not found, using dummy regulator
[ 15.237102] silead_ts i2c-MSSL1680:00: supply avdd not found, using dummy regulator
[ 15.239780] silead_ts i2c-MSSL1680:00: Silead chip ID: 0x50910000
[ 16.831876] input: silead_ts as /devices/platform/80860F41:03/i2c-3/i2c-MSSL1680:00/input/input10

[jwrdegoede]

That evtest output looks incomplete ? Also did the re-install help ?

[cjohnson80]

It didn't help. still the same results. running evtest and selecting the event* for silead_ts produces no touch feedback either.

[jwrdegoede]

Did you try the rmmod + modprobe thing? That will execute the ACPI _PS3 (off) + _PS0 (on) methods, which may be what did the trick last time ...

[cjohnson80]

Yeah here are the results:
[chris@fedora ~]$ sudo rmmod silead
[sudo] password for chris:
[chris@fedora ~]$ sudo modprobe silead
[chris@fedora ~]$ dmesg | tail
[11105.928212] perf: interrupt took too long (6422 > 6405), lowering kernel.perf_event_max_sample_rate to 31000
[16862.062175] perf: interrupt took too long (8037 > 8027), lowering kernel.perf_event_max_sample_rate to 24000
[24319.161535] hid-generic 0005:3938:1001.0006: unknown main item tag 0x0
[24319.162042] input: Bluetooth Mouse as /devices/platform/80860F0A:00/serial0/serial0-0/bluetooth/hci0/hci0:2/0005:3938:1001.0006/input/input20
[24319.163261] input: Bluetooth Mouse as /devices/platform/80860F0A:00/serial0/serial0-0/bluetooth/hci0/hci0:2/0005:3938:1001.0006/input/input22
[24319.163576] hid-generic 0005:3938:1001.0006: input,hidraw2: BLUETOOTH HID v0.1e Mouse [Bluetooth Mouse] on 38:a2:8c:65:a9:63
[26690.811502] silead_ts i2c-MSSL1680:00: supply vddio not found, using dummy regulator
[26690.812428] silead_ts i2c-MSSL1680:00: supply avdd not found, using dummy regulator
[26690.813234] silead_ts i2c-MSSL1680:00: Silead chip ID: 0x50910000
[26693.963134] input: silead_ts as /devices/platform/80860F41:03/i2c-3/i2c-MSSL1680:00/input/input23
[chris@fedora ~]$ sudo evtest
No device specified, trying to scan all of /dev/input/event*
Available devices:
/dev/input/event0: Windows USB Keyboard
/dev/input/event1: Windows USB Keyboard System Control
/dev/input/event2: Windows USB Keyboard Consumer Control
/dev/input/event3: Windows USB Keyboard Mouse
/dev/input/event4: Video Bus
/dev/input/event5: Intel HDMI/DP LPE Audio HDMI/DP,pcm=0
/dev/input/event6: Intel HDMI/DP LPE Audio HDMI/DP,pcm=1
/dev/input/event7: axp20x-pek
/dev/input/event8: silead_ts
/dev/input/event9: gpio-keys
/dev/input/event10: gpio-keys
/dev/input/event11: Bluetooth Mouse
/dev/input/event12: Bluetooth Mouse
Select the device event number [0-12]: 8
Input driver version is 1.0.1
Input device ID: bus 0x18 vendor 0x0 product 0x0 version 0x0
Input device name: "silead_ts"
Supported events:
Event type 0 (EV_SYN)
Event type 1 (EV_KEY)
Event code 330 (BTN_TOUCH)
Event type 3 (EV_ABS)
Event code 0 (ABS_X)
Value 0
Min 20
Max 873
Event code 1 (ABS_Y)
Value 0
Min 4
Max 1643
Event code 47 (ABS_MT_SLOT)
Value 0
Min 0
Max 9
Event code 53 (ABS_MT_POSITION_X)
Value 0
Min 20
Max 873
Event code 54 (ABS_MT_POSITION_Y)
Value 0
Min 4
Max 1643
Event code 57 (ABS_MT_TRACKING_ID)
Value 0
Min 0
Max 65535
Properties:
Property type 1 (INPUT_PROP_DIRECT)
Testing ... (interrupt to exit)

[jwrdegoede]

Did you try to see if touches now get registered or not ?

If touches still do not get registered, what happens if you install my previous kernel from: https://koji.fedoraproject.org/koji/taskinfo?taskID=92396301 and then do rmmod + "modprobe silead settings=..." again. IOW go back to what was a known working combination ?

[cjohnson80]

There are no touches after that. I'll try going back to that kernel

[jwrdegoede]

Did you rebuild the vmlinuz make bzImage and then copy arch/x86/boot/vmlinuz to /boot/vmlinuz-<version> ? The touchscreen_dmi.c file gets build into the main kernel image, it is not a module.

[cjohnson80]

Yes I did that but still don’t have results. I did a few fresh kernel rebuilds too just to be sure

[jwrdegoede]

Weird, I'm afraid I'm all out of ideas then.

[jwrdegoede]

I'll go and submit the touchscreen_dmi.c patch upstream as is then, since the addition of the "silead,stuck-controller-bug" property does not help.

[cjohnson80]

Yeah I don’t think it would be a bad idea. I’m still playing around with different variations and recompiling kernels. I’ll post here if I happen to figure something out.

[chrisma82]

Unrelated sorry but can anyone provide some driver files from this device? Rca support is not helpful. Thanks!