Finally... The touchpad works great!

[moonwalkers]

Didn't work for me. I took a screenshot of the terminal output figuring it'd help but there's no way to attach images directly to the forum.

Code:

>>> Verify mismatch: type=f2, pass=e2, received=d2
EEE Touchpad verify data failed type:242 pass:226
EEE Touchpad erase failed
>>> USB device closed

The verify mismatch line gets spammed a hundred times before finally giving up and outputting the last three lines. My keyboard is ANSI, but my PBP is a first-run ISO model. I wouldn't think that there'd be ISO trackpads and ANSI trackpads that would cause the script to fail, but I honestly don't know. Right this second the trackpad doesn't work at all and the keyboard doesn't work but certain keys output letters not assigned to them, so it's probably not completely dead, hopefully, fingers crossed. 
Is it fubar or is there still hope yet?

I ran into something similar to this. Unfortunately, I didn't keep notes, but among the things I've done were:

• reboot machine
  
• clean and reset git tree to clean out modified keyboard layouts (I swap LCtrl and Fn to match ThinkPad layout), re-build updater tool
  

After one (or both - like I said, didn't keep notes) actions I retried step-2 and update succeeded. Afterwards I was able to modify the layout .hex again and re-apply it.

[poVoq]

Iso version with Manjaro repository provided updater worked without problems. Thanks!

[Pleonasmic]

I ran into something similar to this. Unfortunately, I didn't keep notes, but among the things I've done were:

• reboot machine
  
• clean and reset git tree to clean out modified keyboard layouts (I swap LCtrl and Fn to match ThinkPad layout), re-build updater tool
  

After one (or both - like I said, didn't keep notes) actions I retried step-2 and updated succeeded. Afterwards I was able to modify the layout .hex again and re-apply it.

In order, I tried:
1) Nuking the git tree, downloading the 0.0.3 tag tarball, building that, and running stage-2 again. Didn't work;
2) Playing with fwupd. The first command in the OP link that lists the devices available for firmware updates still showed the keyboard and trackpad, so I knew they and/or the IC controlling them couldn't be dead (yet);
3) Downloading the version in the manjaro arm repository, which was 0.0.3-2 for whatever that's worth, and running its stage-2. THAT ended up working.
I don't know if the -2 version bump was significant, or if it was something on my end, or if the computer spirits were just pranking me for an afternoon, but now it works. (And now that it works, I gotta admit, the trackpad is about as good as I could've hoped for. Especially with the small motions that were jittery before.) Thanks for the indirect help, I really didn't want to have to do another keyboard & trackpad swap if this thing was kill.

What's the deal with the LCtrl and Fn swap? I'm mildly interested in that.

[KC9UDX]

One success and one failure.

Success: Early 2020 PBP, default Manjaro, did have to update Manjaro, but otherwise everything went without a hitch.

Failure: Not blaming the developer. This has been a major debacle, partly due to a (relatively) long-term power outage here, and partly, I thought because of my personal clashing with Linux, but perhaps not.

First, I tried to update from NetBSD, and got the following output from step-1:

Code:

Please note that updating the firmware carries a small risk
of permanently disabling the trackpad and keyboard, effectively
bricking them.  Proceed at your own risk.

Press Ctrl+C within the next five seconds to cancel... 5 4 3 2 1

[x] Running STEP-1...
[*] Flashing keyboard updater firmware...
>>> Hex file data fixed
[*] Opening USB device in user mode...
>>> Trying to open VID:258a PID:001e...
>>> USB device not found
>>> Trying to open VID:258a PID:001f...
>>> USB device not found
>>> Trying to open VID:258a PID:000d...
>>> USB device not found
>>> Trying to open VID:258a PID:001e...
>>> USB device not found
>>> Trying to open VID:258a PID:001f...
>>> USB device not found
>>> Trying to open VID:258a PID:000d...
>>> USB device not found
>>> Trying to open VID:258a PID:001e...
>>> USB device not found
>>> Trying to open VID:258a PID:001f...
>>> USB device not found
>>> Trying to open VID:258a PID:000d...
>>> USB device not found
EEE Failed to open in user mode
[*] Opening USB device in boot mode...
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
>>> Trying to open VID:0603 PID:1020...
>>> USB device not found
EEE Failed to open in boot mode

I assumed that this was some deficiency of lib-usb-1 on my system. I don't exactly have a standard NetBSD installation, if there is such a thing.
Next, i was able to run step-1 from "Linux Debian-Desktop 4.4.207 #1 SMP Thu Dec 26 21:14:57 EST 2019 aarch64 GNU/Linux". Then the power went out, and Debian quit booting. Now, fortunately, step-1 had completely finished. I had done the poweroff after step-1 just before the power went out. I was able to reinstall Debian. However, I ran into a time-skew problem with 'make'. I resolved this by `find . | xargs touch`
Next, 'make' completed successfully. But,

Edit: nevermind; it just works in Debian (when the file dates are correct). So probably lib-usb-1 is just broken in my NetBSD.

[GingerDave]

I checked, and my Pinebook has the SH61F83 keyboard controller

I've already updated it before (maybe a few times as I remember having some trouble), so I'm not going to risk it again until we have confirmation that the SH61F83 and SH68F83 controllers have the same support for write cycles. Has anyone actually encountered an issue with writing to SH61F83 too many times? I'm itching to give this a go as it sounds like it makes a really nice change.

Many thanks to all involved in this!

[jsfrederick]

I updated my ANSI PBP today. No issues with the process, it was as expected.

Touchpad is better, but not great. Still not anywhere near mainstream touchpads.

At least it’s pretty usable now.


Sent from my iPad using Tapatalk

[KC9UDX]

Keep in mind that your OS/DE settings for the touchpad affect its performance greatly.

[jsfrederick]

Keep in mind that your OS/DE settings for the touchpad affect its performance greatly.

Good Point. I’ve been playing around with the touchpad setting in Arabian and it’s gotten a little better. Still going to tweak them a bit and see if I can get them more to my liking.


Sent from my iPad using Tapatalk

[Tharadash]

The firmware upgrade worked flawlessly, thanks!
(PBP ordered Aug 2020)

So far, the touchpad feels great, though I was not experiencing any issues before

[linuxad]

I wasn't able to compile the updater using the instructions on Github for Arch Linux. Compilation failed due to a missing file, sys/cdefs.h. I booted Debian via the sdcard and had no issues compiling.

I did, however, have issues using an external USB keyboard after stage 1. I use Arch Linux via https://github.com/SvenKiljan/archlinuxarm-pbp, which uses Tow-Boot. Tow-Boot requires either a CTRL-C or ESC to choose a different boot method. Since I have an ANSI PBP, I wasn't able to use the PBP keyboard or an external keyboard to issue CTRL-C/ESC. Fortunately the external keyboard started working only after Tow-Boot booted Arch, and then I was able to mount the sdcard and run stage 2 and the flash-kb command.

The touchpad is very much improved. That said, middle click via three finger tap is still difficult. The workaround is to tap very quickly. Like another poster, I'm also finding that if I gently brush the touchpad while typing, the touchpad unfortunately will register a click.

This is a major improvement. Thanks!