Librem5 issueshttps://source.puri.sm/groups/Librem5/-/issues2021-01-13T14:07:30Zhttps://source.puri.sm/Librem5/developer.puri.sm/-/issues/177Outdated documentation for plasma mobile on librem 52021-01-13T14:07:30ZGeorge AndreouOutdated documentation for plasma mobile on librem 5I was trying to install Plasma Mobile on Librem 5 (Dogwood) using documentation in https://developer.puri.sm/Librem5/Development_Environment/Boards/HowTo/Install_Plasma_Mobile.html#devkit-howto-install-plasma-mobile.
However when running...I was trying to install Plasma Mobile on Librem 5 (Dogwood) using documentation in https://developer.puri.sm/Librem5/Development_Environment/Boards/HowTo/Install_Plasma_Mobile.html#devkit-howto-install-plasma-mobile.
However when running `sudo apt update` I get a 404 error for https://repo.kaidan.im/debpm.
I asked in the forum (https://forums.puri.sm/t/problem-installing-plasma-mobile-on-librem-5-dogwood/11682) and it seems that the documentation is outdated.https://source.puri.sm/Librem5/debs/gnome-settings-daemon/-/issues/20Cancel SIM PIN entry dialog when the modem goes away2021-01-12T13:19:59ZSebastian KrzyszkowiakCancel SIM PIN entry dialog when the modem goes awayIf I turn the modem off with a killswitch while on the SIM PIN entry dialog, the dialog remains visible until either cancelled or any PIN is submitted (which, interestingly, doesn't error out at all and from the user perspective looks li...If I turn the modem off with a killswitch while on the SIM PIN entry dialog, the dialog remains visible until either cancelled or any PIN is submitted (which, interestingly, doesn't error out at all and from the user perspective looks like it succeeds :))https://source.puri.sm/Librem5/linux/-/issues/252kernel error messages during startup (with HKS off)2021-10-14T09:57:01ZMartin Kepplingerkernel error messages during startup (with HKS off)I'm using a v5.11 based kernel in "next" to work on this.
* HKS all "on": (for the upcoming rebase) we easily get down to the following error messages that are left:
```
[ 0.046059] debugfs: Directory 'dummy-syscon@30340000' with par...I'm using a v5.11 based kernel in "next" to work on this.
* HKS all "on": (for the upcoming rebase) we easily get down to the following error messages that are left:
```
[ 0.046059] debugfs: Directory 'dummy-syscon@30340000' with parent 'regmap' already present!
[ 0.927927] debugfs: Directory 'dummy-reset-controller@30390000' with parent 'regmap' already present!
[ 0.962538] spi_imx 30820000.spi: bitbang start failed with -517
[ 3.662065] clk: failed to reparent audio_pll1 to audio_pll1_out: -22
[ 3.937627] wm8962 2-001a: ASoC: error at soc_component_read_no_lock on wm8962.2-001a: -16
[ 3.946823] wm8962 2-001a: ASoC: error at soc_component_read_no_lock on wm8962.2-001a: -16
[ 5.305254] sd 0:0:0:0: [sda] No Caching mode page found
[ 5.314240] sd 0:0:0:0: [sda] Assuming drive cache: write through
```
the debugfs errors have to be fixed at the caller sites (syscon and reset-controller) upstream. the last 2 sd "errors" are long-standing scsi message that probably shouldn't be printed to error. For the remaining 4 I'm not sure yet, but let's keep track of them here.
* HKS all "off": adds these messages:
```
[ 3.874228] st_lsm6dsx_i2c 1-006a: failed to read whoami register
[ 3.894339] s5k3l6xx: i2c_read: error during transfer (-6)
[ 3.899928] s5k3l6xx 3-002d: model mismatch: 0x80 != 0x30
[ 3.908741] s5k3l6xx 3-002d: model mismatch: 0x0 != 0xC6
[ 3.914526] s5k3l6xx 3-002d: revision mismatch: 0x0 != 0xB0
[ 4.390661] hi846 2-0020: i2c read error: -6
[ 4.395076] hi846 2-0020: failed to find sensor: -5
```
because the IMU and cameras are not powered, but the drivers probe. Suggestions on how to fix this are welcome.https://source.puri.sm/Librem5/purple-mm-sms/-/issues/22Chatty stops receiving SMS after sending an SMS2021-01-11T20:17:59ZGergely HorváthChatty stops receiving SMS after sending an SMSThis issue was first published as https://gitlab.com/mobian1/issues/-/issues/215 and later on https://gitlab.freedesktop.org/mobile-broadband/ModemManager/-/issues/299. On the MM page it seemed that we triangled the problem to Chatty. He...This issue was first published as https://gitlab.com/mobian1/issues/-/issues/215 and later on https://gitlab.freedesktop.org/mobile-broadband/ModemManager/-/issues/299. On the MM page it seemed that we triangled the problem to Chatty. Here are my experiences:
I'm on Mobian+Phosh on PinePhone, latest version as of 2021-01-04 (just did apt update && apt upgrade). *Update*: system was updated on 2021-01-10. Receiving SMS is unreliable. I'm using a T-Mobile SIM in Hungary. Mobile setting currently is 2g/3g/4g (4g preferred). 2g only, 3g only, 4g only, 3g/4g (pref:4g) was also tested.
**Actual problem:**
The problem after some debugging (see in the ModemManager issue) seems to be that after sending an SMS from Chatty, it is unable to receive more. These issues does not surface when Chatty is killed. I could reproduce the issue anytime by sending an SMS (or more) from another phone, replying to it, and no more SMS arrives. I tried sending SMS from other numbers as well (bank 2FA) with the same result. When Chatty was shut down, sending has no effect on receiving the SMS messages.
No MMS were sent from the phone and to be fair I haven't received one in years.
Thanks in advance for looking into it!
Gergelyhttps://source.puri.sm/Librem5/librem5-devkit-tools/-/issues/33pytests: Separate out devkit and phone specific data2021-01-11T11:09:04ZGuido Guntherpytests: Separate out devkit and phone specific datawe have things like max brightness and several paths coded in the tests. Splitting these out into data structures make adding new devices / batches simpler.we have things like max brightness and several paths coded in the tests. Splitting these out into data structures make adding new devices / batches simpler.https://source.puri.sm/Librem5/debs/gtk/-/issues/28Wrong surrounding text in text input messages2021-01-10T11:24:08ZDorota CzaplejewiczWrong surrounding text in text input messagesGTK messes up the ordering of changes vs commit events. That messes up text correction/prediction rather badly.
```
[134915.127] -> zwp_text_input_v3@35.enable()
[134915.157] -> zwp_text_input_v3@35.set_surrounding_text("", 0, 0)
[134...GTK messes up the ordering of changes vs commit events. That messes up text correction/prediction rather badly.
```
[134915.127] -> zwp_text_input_v3@35.enable()
[134915.157] -> zwp_text_input_v3@35.set_surrounding_text("", 0, 0)
[134915.171] -> zwp_text_input_v3@35.set_text_change_cause(0)
[134915.185] -> zwp_text_input_v3@35.set_content_type(0, 0)
[134915.204] -> zwp_text_input_v3@35.set_cursor_rectangle(107, 62, 0, 30)
[134915.218] -> zwp_text_input_v3@35.commit()
[153744.724] zwp_text_input_v3@35.done(1)
[153744.881] -> zwp_text_input_v3@35.set_surrounding_text("", 0, 0)
[153745.014] -> zwp_text_input_v3@35.set_text_change_cause(0)
[153745.233] zwp_text_input_v3@35.commit_string("aa")
[153745.320] zwp_text_input_v3@35.done(1)
[153745.647] -> zwp_text_input_v3@35.set_cursor_rectangle(125, 62, 0, 30)
[153745.808] -> zwp_text_input_v3@35.commit() // premature commit but ok
[153745.877] -> zwp_text_input_v3@35.set_surrounding_text("aa", 2, 2)
[153745.992] -> zwp_text_input_v3@35.set_text_change_cause(0)
----- here should be a commit
```
I have tested this only on Fedora's shipped GTK so far, so maybe it's not an issue at all. Still, I'm marking this in case of unexpected distraction.https://source.puri.sm/Librem5/purple-mm-sms/-/issues/21Must only remove SMS it created itself2021-01-08T13:41:33ZGuido GuntherMust only remove SMS it created itselfSeems chatty removes SMS created via mmcli from the modem as well, see: https://gitlab.freedesktop.org/mobile-broadband/ModemManager/-/issues/294#note_749979Seems chatty removes SMS created via mmcli from the modem as well, see: https://gitlab.freedesktop.org/mobile-broadband/ModemManager/-/issues/294#note_749979https://source.puri.sm/Librem5/librem5-base/-/issues/42Adjust ucm profiles to `Librem 5`2021-03-10T20:54:57ZGuido GuntherAdjust ucm profiles to `Librem 5`that's the upstream name and what we want to use from byzantium on. Needed to unbreak audio. Instead of doing this here we can drop it and patch `alsa-ucm-conf` (and forward things upstream).that's the upstream name and what we want to use from byzantium on. Needed to unbreak audio. Instead of doing this here we can drop it and patch `alsa-ucm-conf` (and forward things upstream).https://source.puri.sm/Librem5/OS-issues/-/issues/178Flipping wifi hardware killswitch disables cellular modem at software level2021-01-14T13:03:10ZJoao AzevedoFlipping wifi hardware killswitch disables cellular modem at software level---
# What problem did you encounter
Disabling the wifi via hardware killswitch also disables the `wwan` at software level. Re enabling the wifi via HKS does not re enable the `wwan`.
To have a cellular connection again you need to re...---
# What problem did you encounter
Disabling the wifi via hardware killswitch also disables the `wwan` at software level. Re enabling the wifi via HKS does not re enable the `wwan`.
To have a cellular connection again you need to re enable the modem at software level either in the power settings or in the mobile settings
## What is the actual behaviour?
The one described above.
## What is the expected behaviour?
Using the wifi killswitch should only affect the wifi module and not the `wwan` module.
## How to reproduce
Please provide steps to reproduce the issue. If it's a graphical issue please
attach screenshots.
# What hardware are you running on?
- [ ] amd64 qemu image
- [ ] Librem5 devkit
- [x] other (Librem 5 - Evergreen)
# Relevant OS information
## Which kernel are you using?
`Linux pureos 5.9.0-1-librem5 #1 SMP PREEMPT Tue Dec 22 10:22:25 PST 2020 aarch64 GNU/Linux`
## Which OS are you using?
```
purism@pureos:~$ cat /etc/os-release
ID=pureos
NAME=PureOS
PRETTY_NAME=PureOS
VERSION_ID=9.0
VERSION_CODENAME=amber
HOME_URL="https://pureos.net/"
SUPPORT_URL="https://puri.sm/faq/#faq-WherecanIfindoutmoreaboutPureOS"
BUG_REPORT_URL="https://tracker.pureos.net/"
```
## Any other information that may be helpful?
Several other team members have reported this issue, (so it is not an hardware issue) : @angus.ainslie @francois.techene perhaps they can also contribute.
@sebastian.krzyszkowiak L5 that is running a different kernel is not affected by this issue.
Bellow is a video of the issue. It corresponds to:
- Disabling the wifi via HSK
- Cellular connection gets disabled
- Enabling the wifi again
- Cellular is not enabled back, and needs to be enabled at software level
![vid](/uploads/7160646854940187a81baae0b0df86f0/vid.mp4)https://source.puri.sm/Librem5/debs/firefox-esr-mobile-config/-/issues/1Missing in byzantium2021-01-13T12:43:38ZGuido GuntherMissing in byzantiumThis is missing in byzantium which breaks librem5-base's `non-gnome-all`.This is missing in byzantium which breaks librem5-base's `non-gnome-all`.https://source.puri.sm/Librem5/geary/-/issues/17Always tries to unlock default keyring2021-01-11T10:45:13ZSam BullAlways tries to unlock default keyringEvery time Geary starts up it tries to unlock the default keyring, even though there are no credentials in there. Geary continues to run correctly if I cancel the unlock, but it's annoying being prompted for it every time I turn the phon...Every time Geary starts up it tries to unlock the default keyring, even though there are no credentials in there. Geary continues to run correctly if I cancel the unlock, but it's annoying being prompted for it every time I turn the phone on.
Expected behaviour is that it only tries to unlock the keyring which contains the credentials.
On my laptop, I have the same keyring setup, the same GNOME online accounts setup, but use Evolution for mail and everything works as expected there.
Steps I took (not sure how many are needed to reproduce):
1. Setup an email account in GNOME online accounts in the settings app.
2. Change the keyring password in seahorse to make it passwordless.
3. Add a new keyring in seahorse (with a password).
4. Set the new keyring as default in seahorse.https://source.puri.sm/Librem5/debs/webkit/-/issues/13Audio playback stops before end of audio2023-06-17T00:49:14ZSam BullAudio playback stops before end of audioWhen playing an <audio> element, the end of the audio never plays (e.g. the last second or half a second).
I can reproduce the issue with this MDN example: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/audio
This is causing...When playing an <audio> element, the end of the audio never plays (e.g. the last second or half a second).
I can reproduce the issue with this MDN example: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/audio
This is causing me major issues with AnkiWeb, as many cards have audio clips for a single word, so most of the time it only plays the first syllable of a word, or nothing at all, rendering it useless.
I'm not encountering the issue when testing on my laptop (running 3.36.4).https://source.puri.sm/Librem5/3D_designs/-/issues/3Evergreen case too tall2021-08-24T16:52:48ZAngus Ainslieangus.ainslie@puri.smEvergreen case too tallThere is about 3mm of space above the phone and the lip around the screen
https://source.puri.sm/Librem5/3D_designs/-/blob/master/Librem5-Case/librem5_case_Evergreen.stl
/cc @david.hamnerThere is about 3mm of space above the phone and the lip around the screen
https://source.puri.sm/Librem5/3D_designs/-/blob/master/Librem5-Case/librem5_case_Evergreen.stl
/cc @david.hamnerhttps://source.puri.sm/Librem5/OS-issues/-/issues/1772.4 GHz wifi and bluetooth headphones disconnect after a couple minutes2021-01-03T15:11:01ZAdrian2.4 GHz wifi and bluetooth headphones disconnect after a couple minutesWhen watching youtube videos over a 2.4 GHz wifi and using Sony WH-1000XM3 headphones over bluetooth, after a couple minutes both the wifi and the bluetooth disconnect. After that, neither wifi networks are shown in the list of available...When watching youtube videos over a 2.4 GHz wifi and using Sony WH-1000XM3 headphones over bluetooth, after a couple minutes both the wifi and the bluetooth disconnect. After that, neither wifi networks are shown in the list of available networks nor are bluetooth devices listed under the Bluetooth settings tab. After that, I have to restart the phone for them to work. I also checked using only wifi, or only bluetooth headphones, and then this problem does not reproduce.
Batch: Evergreenhttps://source.puri.sm/Librem5/gnome-control-center/-/issues/138Location Services does not set Weather automatic-location2021-02-28T15:49:54ZDavid HamnerLocation Services does not set Weather automatic-locationWith location services off (Settings -> Privacy -> Location Services), the Weather app still pulls an automatic location.
This is because "org.gnome.Weather automatic-location" is not kept in sync with the main toggle.
To turn off Weat...With location services off (Settings -> Privacy -> Location Services), the Weather app still pulls an automatic location.
This is because "org.gnome.Weather automatic-location" is not kept in sync with the main toggle.
To turn off Weather location you can run:
```
gsettings set org.gnome.Weather automatic-location false
```https://source.puri.sm/Librem5/debs/gnome-settings-daemon/-/issues/19Support SIM PIN in GNOME Keyring2021-01-12T16:36:42ZKyle RankinSupport SIM PIN in GNOME KeyringSome Librem 5 users use a SIM PIN to lock and protect their SIM. The user must enter their PIN and unlock the SIM at each boot to use phone features. This adds additional friction to the boot process.
If the SIM PIN were managed by the ...Some Librem 5 users use a SIM PIN to lock and protect their SIM. The user must enter their PIN and unlock the SIM at each boot to use phone features. This adds additional friction to the boot process.
If the SIM PIN were managed by the GNOME keyring, then unlocking that keyring would be sufficient to unlock the SIM and users could enable the security feature on their SIM card without losing convenience. We could then even consider adding a workflow to set a SIM PIN as an option at initial phone set-up.https://source.puri.sm/Librem5/OS-issues/-/issues/176Support SIM PIN in GNOME Keyring2020-12-30T11:08:38ZKyle RankinSupport SIM PIN in GNOME KeyringSome Librem 5 users use a SIM PIN to lock and protect their SIM. The user must enter their PIN and unlock the SIM at each boot to use phone features. This adds additional friction to the boot process.
If the SIM PIN were managed by the ...Some Librem 5 users use a SIM PIN to lock and protect their SIM. The user must enter their PIN and unlock the SIM at each boot to use phone features. This adds additional friction to the boot process.
If the SIM PIN were managed by the GNOME keyring, then unlocking that keyring would be sufficient to unlock the SIM and users could enable the security feature on their SIM card without losing convenience. We could then even consider adding a workflow to set a SIM PIN as an option at initial phone set-up.https://source.puri.sm/Librem5/gedit/-/issues/5Save As dialog in Librem 5 Text Editor app does not fit2021-09-27T08:46:29ZElias RudbergSave As dialog in Librem 5 Text Editor app does not fitHow to reproduce:
- Open the "Text Editor" app (gedit) on the Librem 5
- Type some text
- Tap the "Save" button
- Now the "Save As" dialog is shown, but it does not fit
Version info:
gedit/amber-phone,now 3.34.0+12189+gitba0149a95-1pure...How to reproduce:
- Open the "Text Editor" app (gedit) on the Librem 5
- Type some text
- Tap the "Save" button
- Now the "Save As" dialog is shown, but it does not fit
Version info:
gedit/amber-phone,now 3.34.0+12189+gitba0149a95-1pureos0 arm64https://source.puri.sm/Librem5/debs/epiphany/-/issues/29Audio playback stops before end of audio2021-01-04T08:57:06ZSam BullAudio playback stops before end of audioWhen playing an <audio> element, the end of the audio never plays (e.g. the last second or half a second).
I can reproduce the issue with this MDN example: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/audio
This is causing...When playing an <audio> element, the end of the audio never plays (e.g. the last second or half a second).
I can reproduce the issue with this MDN example: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/audio
This is causing me major issues with AnkiWeb, as many cards have audio clips for a single word, so most of the time it only plays the first syllable of a word, or nothing at all, rendering it useless.
I'm not encountering the issue when testing on my laptop (running 3.36.4).https://source.puri.sm/Librem5/geary/-/issues/16"Accounts" view doesn't fit the screen when there's long account name2021-08-24T05:15:37ZSebastian Krzyszkowiak"Accounts" view doesn't fit the screen when there's long account nameAs seen on the screenshot:
![geary-fit](/uploads/8e8ad1908ccd5c1252c8134204d70361/geary-fit.png)
Editing the account name to be shorter makes it fit again.As seen on the screenshot:
![geary-fit](/uploads/8e8ad1908ccd5c1252c8134204d70361/geary-fit.png)
Editing the account name to be shorter makes it fit again.