Solved Trackpoint works randomly.

This is my first post but I think this falls under peripheral hardware. I have been working my way though the handbook but am only up to chapter 4 and sporatically have read some of the other sections.

hardware: lenovo x130e netbook
os: FreeBSD version 12.1
DE: XFCE4 (installed with pkg)

Issue: Trackpoint doesn't work most the time, the trackpad always works though. I have the xf86-input-libinput package installed and before this I tried some things according to posts by others with thinkpads; I attempted to install and use xf86-input-synaptics although this did nothing that I could tell. I also have it enabled in the /boot/loader.conf file, and have set the sensitivity in the/etc/sysctl.conf file.

I have gathered as much info from my system as I know how at the moment and this is the output I am getting from dmesg | grep psm while booted with the verbose option selected.

From my best understanding it keeps that atkbdc0 is probably my trackpoint and the things that is causing so many issues.

Code:
psm0: <PS/2 Mouse> irq 12 on atkbdc0
psm0: [GIANT-LOCKED]
psm0: model Synaptics Touchpad, device ID 0
psm0: <PS/2 Mouse> irq 12 on atkbdc0
psm0: [GIANT-LOCKED]
psm0: model Synaptics Touchpad, device ID 0
psm0: unable to allocate IRQ
psmcpnp0: <PS/2 mouse port> irq 12 on acpi0
psm0: current command byte:0067
psm: status 00 02 64
psm: status 00 00 64
psm: status 00 03 64
psm: status 00 03 64
psm: data 08 00 00
psm: status 00 47 18
psm: status 01 e2 b1
psm: status d0 01 a3
psm: status 41 80 b4
psm: status 94 03 00
psm: status 12 3c 00
psm: status b4 63 a4
psm: status 24 8c 12
psm: status 1c 7d 40
psm: status 00 02 64
psm0: <PS/2 Mouse> flags 0x3000 irq 12 on atkbdc0
psm0: [GIANT-LOCKED]
psm0: model Synaptics Touchpad, device ID 0-00, 4 buttons
psm0: config:00007000, flags:00000008, packet size:6
psm0: syncmask:c0, syncbits:00

I have also attached my config files from my system and have a acpidump if that would be helpful. If anyone can point me in a direction on what to do next even that would be most helpful as I have reached the current limit of my understanding. Thanks.
 

Attachments

  • rc.conf
    621 bytes · Views: 208
  • loader.conf
    470 bytes · Views: 194
  • sysctl.conf
    700 bytes · Views: 188
Did more testing. After about six reboots in a row I managed to have a working trackpoint and pad until I rebooted the system again. Managed to save the dmesg output from it though.

I started to wonder if maybe it was a issue with the hardware or a connection issue, so I pulled out the keyboard and checked the ribbion cable. The contacts were clean but I cleaned them with alcohol anyway and then reconnected and assembled the laptop. I have also booted with a live usb and had the trackpoint working before... so I'm going forward with the assumption it's not a hardware issue.


For now it seems that every 10 boots of the system the trackpoint actually works when the probed. I have attached the dmesg output from two seprate working boots.
 

Attachments

  • v11_working_everything.txt
    70 KB · Views: 158
  • v10_working_everything.txt
    57.1 KB · Views: 158
Yes please post short *.conf files inline (CODE). In general, sporadic failures indicate a hardware issue (to the software, all what changes is the system clock). It's summer now, have a look at the surrounding air temperature where your system is.
 
Thank you for responding Elazar, I'll try to use pastebin or keep it shorter from now on if at all possible, if you think it's a good idea I can edit the original post to do such. I'm looking at the link you added and will try to make and install it to see if it is communicating over i2c. The thing that is confusing me is that it works sometimes but not others. I haven't been able to find to many details on this specific laptop as being as under powered as it was I think very few people actually liked them.

mjollnir, As for the temperature I had the same thought before too, I had checked the core temp and it was below 40C in a air conditioned room so I think that's okay. I will test again with a live usb and see if it works.

while I was writing this actually the dd with a live image finished and upon testing by rebooting from the drive a couple of times the trackpointer and pad just work every time. shows up as
Code:
IBM Trackpoint firmware: 0x0e
under the live usb.
 
With under-powered you mean the CPU is rather slow e.g. Intel Atom, or do you mean it's power supply is at the edge of it's specification? Obviously, this can be another source of random failures (e.g. varying voltage, short outages of electrical current (Ampère)). EDIT All these issues can also have their root cause in cheap electronic parts built into the laptop, who fail sporadically. Don't know about your model. I guess it's not a business-class ThinkPad, and Lenovo's reputation is not the best when it comes to consumer grade equipment?
 
Yes the cpu or apu I should say is the E-300 dual core 1.3Ghz version. I'm using the original battery from when it came out and it still gets around 3 hours of run time at full speed. The power supply is fine. It has more than enough amprage to keep the laptop fed. Also It's a x130e lenovo as I said in the first post.

Good news is that everytime I boot the devices showsup after running libinput list-devices or xinput as /dev/input/event4 everytime.

I have also set the device specifically according to ArchLinux-Trackpoint documentation for the xorg.conf file I generated from Xorg -configure. Although it still doesn't seem to have solved my issue. The device just doesn't seem to output anything under the libinput debugging or xinput. on a "bad" reboot


I guess I'll just keep trying things. I have already put a little over a week into getting the trackpoint to work, might as well keep going.
 
That sounds like a very dissatisfying experience. Usually running Xorg --configure is considered obsolete, since the automagic handles devices well. Sorry I can not help more than what I wrote above: random failure = bad hardware, very very likely.
 
Yeah I would be inclined to agree with you on hardware failure except it works just fine under well every other OS I have tried. I'm thinking I might just need to nuke it from orbit and start over. But I'm still having nightmares about the 7.5 hour compile time just for the wpa_gui I'll have to do all over again as well.

Could be worse I guess. I might pull out the old Oscope and start probing the trackpoint if I get desperate or just make a converter with a microcontroller to go from the trackpoint to a usb so I can just make it show up as a usb mouse.....I might almost prefer that over digging through the source at this point. There must be some reason why it doesn't establish communication with the sysmouse but after reading the man pages for moused, libinput, xinput, psm and another dozen old form posts I'm stumped.

I just don't think anyone ever used these with freeBSD much less anyother BSD version. At least the best of my google/duckduckgo Fu hasn't turned up anybody using the same laptop with freeBSD.
 
Usually the trackpoint is PS/2, it should work out of the box in most cases. For the touchpad to work properly (native, not using some mouse protocol), you need to give moused(8) the -l 2 flag, e.g.: sysrc moused_flags='-a 1.5 -A 1.5,2 -VH -L 3 -U 3 -l 2'. In sysctl.conf(5), you need kern.evdev.rcpt_mask=6. Then install x11-drivers/xf86-input-synaptics.

Standard disclaimer: install the docs: pkg install {de,en}-freebsd-doc, replace de with your native tongue, and point your favorite browser to /usr/local/share/doc/freebsd.
Add to the ALIAS section of /usr/local/etc/pkg.conf message: "query '[%C/%n] %M'", and read through all pkg message|less.
 
I already have tried the synaptics driver as well as alps and elentech. Still had the same issues. I also had already attempted kern.evdev.rcpt_mask=6 before and then set it to 12 after that didn't work according to another users attempts.

I already have messed with the moused_flags before and they had no affect on the consistency of being able to have a working mouse. Last night I reinstalled from scratch again only to have the trackpad fail after a rooboot and have the same random working/nonworking issues with both the trackpad and trackpoint. They both showup just fine under libinput but seem not to be output anything(usally the trackpoint).

Thank you mjollnir for sticking with me on this but for now I'm just going to have to leave FreeBSD on my servers and run something else on this laptop. For my usecase I need something that won't randomly change it's behavior in the field, I understand this is probably due to my odd hardware but none the less I don't see having to drag a usb mouse with me everywhere worth the trade off of being able to get away from linux for now.
 
Last resort would be some odd BIOS knob. But as you're fine with Linux, go for it on that laptop. 20-100-2fe uses Void Linux, he's an experienced UNIX guy, so I guess that's a reasonable choice.
 
Yup, I can always just SSH into my FreeBSD boxes when I have the urge so that should hold me over for now. I'm going to mark this solved for now, if I find another solution to make it work I will edit this post to reflect that.
 
You wrote about dd(1)'ing the image??? What's that? You should use the installer!
EDIT Sorry. Very likely you meant you dd'ed the downloaded installer image to a USB thumb drive.
 
This is my first post but I think this falls under peripheral hardware. I have been working my way though the handbook but am only up to chapter 4 and sporatically have read some of the other sections.

hardware: lenovo x130e netbook
os: FreeBSD version 12.1
DE: XFCE4 (installed with pkg)

Issue: Trackpoint doesn't work most the time, the trackpad always works though. I have the xf86-input-libinput package installed and before this I tried some things according to posts by others with thinkpads; I attempted to install and use xf86-input-synaptics although this did nothing that I could tell. I also have it enabled in the /boot/loader.conf file, and have set the sensitivity in the/etc/sysctl.conf file.

See https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=242542 . It looks very similar.
 
Back
Top