Vanilla Dpup 11.0.x Development Builds

[dimkr]

Changes in 11.0.169:

• The ad blocker can be started without having to re-download the ad list (it's not cleared when stopped)

• save2flash offers more details when it encounters errors, instead of hiding errors and giving bad advice

• A syslog viewer menu entry in the System menu

• Faster init script - some things that happened at boot time now happen at build time

• More intuitive behavior when pmedia= is not specified: if needed, initrd will search for the main SFS and saves regardless of media (no need to specify pmedia=cd as in Puppy) and pmedia=usbflash no longer restricts the search to USB devices (so you can use pmedia=usbflash to activate PUPMODE 13 without other side effects) - to disable such search, specify pupsfs and psave

• Various init code quality improvements - redundant code is dropped, probepart is replaced with lsblk, ...

• The "pupsave upgrade" mechanism is mostly gone because it doesn't do anything useful and only slows down boot after upgrade

• yad patches are dropped because they're upstreamed (https://github.com/step-/yad/pulls?q=is ... s%3Aclosed)

[d-pupp]

I just installed dpup 11.0.169
The update of my save folder was much faster
The screen shot app works very nice and is a handy add-on

This old PC boots fast and CPU and memory use-age continue to be very low
Not bad for a dual-core Pentium from 2009

[retiredt00]

Tried to boot vanilladpup_11.0.169 in a Pentium4 machine along other puppies using LICK windows installer and it fails to find the partition where the vanilladpup files are

Similar to other puppies the menu entry in the windows partition where LICK resides is

Code: Select all

title Dpup 11
find --set-root --ignore-floppies /dpup11/vmlinuz
kernel /dpup11/vmlinuz pfix=fsckp psubdir=/dpup11 pupsfs=sdb2:/dpup11/puppy_vanilladpup_11.0.169.sfs zdrv=sdb2:/dpup11/zdrv_vanilladpup_11.0.169.sfs fdrv=sdb2:/dpup11/fdrv_vanilladpup_11.0.169.sfs psave=sdb2:/dpup11/vanilladpupsave
initrd /dpup11/ucode.cpio /dpup11/initrd.zst
boot

and vmlinuz ucode.cpio and initrd.zst are in the dpup11 folder at the root of the windows partition while all the vanilladpup files are on the sdb2 partition in the /dpup11 folder.

The message I get is the can not find the puppy sfs in the sdb2:/dpup11 folder
Interestingly after dpup initrd drops to console, I have no keyboard or mouse.

Is not clear to me if I do something wrong, if there is no intention to support LICK installer or if the new vmlinuz/initrd are missing the necessary components to understand the LICK menu entry.

[dimkr]

Is not clear to me if I do something wrong

This is the list of boot codes. Anything else is unsupported.

[retiredt00]

I see. So the drive specification accepts UUID and label but not device node
Thx

Since I'm away from this old clunker for few days, assuming sdb1 UUID=ad9e0ce4-3466-4ebd-9967-271aaf2cf789, should the entry below suffice?

Code: Select all

title Dpup 11
find --set-root --ignore-floppies /dpup11/vmlinuz
kernel /dpup11/vmlinuz pupsfs=ad9e0ce4-3466-4ebd-9967-271aaf2cf789 psave=ad9e0ce4-3466-4ebd-9967-271aaf2cf789 psubdir=/dpup11
initrd /dpup11/ucode.cpio /dpup11/initrd.zst
boot

Or you should/could specify both labeb/UUID and subfolder. ie psave=ad9e0ce4-3466-4ebd-9967-271aaf2cf789:/dpup11 that may allow different parts to be in different drives or folders?
I understand that you want to simplify the init, but in case this is not currently the case, the ability to separate the OS initrd/SFSs from user data in savefile/folder may be a good option to have.

[dimkr]

Since I'm away from this old clunker for few days, assuming sdb1 UUID=ad9e0ce4-3466-4ebd-9967-271aaf2cf789, should the entry below suffice?

Maybe, this depends on the way files are arranged but there could be bugs so I can't promise it's going to work. If it doesn't work, I'll need some logs to troubleshoot this.

Or you should/could specify both labeb/UUID and subfolder.

Right now you can specify a label or a UUID, and that's it. If you want to put files in a subdirectory, you can specify psubdir=.

You can put the OS in one partition by specifying pupsfs= (with all SFSs in this partition) and save to another that's specified by psave=.

[d-pupp]

@dimkr there is one thing I noticed with dpup 11.0.169
During cleanup you removed /usr/share/pixmaps/puppy. Some of those icons were is use in /usr/share/applications. Also it took out the icons I was using for my spacefm icons hack. I have redone it using Adwaita/scalable/mimetypes icons, however there is no icon for pdf. I added my own.

[dimkr]

@d-pupp True, I moved the icons to the gtkdialog package because they're mostly used in various old and unmaintained things that use gtkdialog. Debian testing just moved from GCC 13 to 14 but nobody seems to be interesting in fixing gtkdialog build failures against GCC 14 on 64 bit platforms, so I decided to remove it. It's there for backward compatibility and nothing preinstalled uses it (I ported everything to yad), so I can drop it and those who want it can build a "community addon". Smaller project scope makes development easier and gives me more time to focus on things that can be improved.

[d-pupp]

@dimkr I understand. Spacefm will use any icons even ones I add so this is not a problem.

I just noticed something else I want to report. When I first booted dpup 11.0.169 it did a update. I then saved my changes. When I shutdown for the day I said NO don't save the session. On the next restart it updated again. So I did a test... If I say NO to save the session at shutdown or reboot it doesn't save the updated changes. All other changes are saved as far as I can tell. If I say YES save the session at shutdown or reboot it does save the update changes.

[dimkr]

it doesn't save the updated changes.

What do you mean by "updated changes"? No matter if you save at shutdown or before shutdown, it saves all changes.

The update procedure that runs at boot each time you change the list of loaded SFSs only updates cache files. These changes to cache files are not saved automatically, only when you decide to save.

[d-pupp]

What do you mean by "updated changes"? No matter if you save at shutdown or before shutdown, it saves all changes.

Yes this was my understanding and the way it always worked before. However this time it didn't work that way.
Any changes I made were saved with the Save Changes button in the Puppy logout dialog box. However the update of my Save folder from version 11.0.164 to version 11.0.169 was not saved until I did a reboot and clicked Yes to save the session from the timed dialog box you get when you click shutdown or reboot.

So until I did that every time I started the pc I got the message about updating your gtk icon cache and in pink text the next boot will be faster

[retiredt00]

So

Code: Select all

title Dpup 11
find --set-root --ignore-floppies /dpup11/vmlinuz
kernel /dpup11/vmlinuz pupsfs=ad9e0ce4-3466-4ebd-9967-271aaf2cf789 psubdir=/dpup11
initrd /dpup11/ucode.cpio /dpup11/initrd.zst
boot

were everything is in the same folder (dpup11 in this case) in the specified UUID partition, works with LICK windows bootloader. Thanks

Interestingly if you take pupsave folder out of the dpup11 folder at the root of the partition and you specify psave=UUID it does not work. The savefolder is not loaded.
It would appear that the psubdir= take precedence and everything is expected to be there or that all files are expected in the same path.

A couple of other things that you might want to consider.
Currently does not offer to save on first shutdown
Interestingly if you make a pupsave at the first boot and manualy trigger a save from the menu, the folder is made but nothing is saved.
So basically you must manually make the savefolder and reboot to start saving/modifying the installation which is a bit cumbersome

[dimkr]

Interestingly if you take pupsave folder out of the dpup11 folder at the root of the partition and you specify psave=UUID it does not work. The savefolder is not loaded.
It would appear that the psubdir= take precedence and everything is expected to be there or that all files are expected in the same path.

Are you sure this ever worked for you? As far as I see, the init script will only look for save files and folders under psubdir if specified: never under both psubdir and the partition root.

Currently does not offer to save on first shutdown

This is intentional: many users seem to run Puppy non-persistently (without any kind of saving) and use various hacks to disable this prompt.

So basically you must manually make the savefolder and reboot to start saving/modifying the installation which is a bit cumbersome

That's true, the pupsave tool only creates new save file/folders and you need to reboot to start using them. I'll see if there's any elegant way to trigger saving right after creation.

[retiredt00]

Interestingly if you take pupsave folder out of the dpup11 folder at the root of the partition and you specify psave=UUID it does not work. The savefolder is not loaded.
It would appear that the psubdir= take precedence and everything is expected to be there or that all files are expected in the same path.

Are you sure this ever worked for you? As far as I see, the init script will only look for save files and folders under psubdir if specified: never under both psubdir and the partition root.

So if you specify psubdir you can not have the savefolder anywhere else, as it can be done with Bookwoem, 10.x etc?

[dimkr]

So if you specify psubdir you can not have the savefolder anywhere else, as it can be done with Bookwoem, 10.x etc?

Right now - if you use psubdir, everything (SFSs and saves) must be under psubdir (not necessarily in the same partition).

[dimkr]

11.0.170 will:
1. Bump the kernel to 6.10.x
2. Add a save/no save prompt to pupsave, when creating a save under PUPMODE 5: it will save the current system state (you'll need to reboot to start using the created save; no save/no save prompt at shutdown)

[d-pupp]

@dimkr I want to revisit your waybar launcher as a way to quickly launch the calculator, screenshot tool, and L3afpad. However I'm having trouble finding code examples. Can you help me out?

[dimkr]

@d-pupp If you only want to add more items, you have enough examples in ~/.config/waybar/config.jsonc and can achieve this using copy-paste and some common sense. If you need help, man waybar-custom should work, and you can find examples in https://github.com/Alexays/Waybar/wiki/Configuration (specifically https://github.com/Alexays/Waybar/wiki/Module:-Custom and https://github.com/Alexays/Waybar/wiki/ ... om:-Simple).

[d-pupp]

@Dimkr Thanks. I found what I was looking for in the man pages.
I tried the menu first no luck some critical error. Next I tried groups, that worked but not what I wanted. Then I tried group drawer, That's close to what I wanted.
However waybar is not that stable, sometimes it just quits working. It's not crashed it just doesn't do anything.
Do you know off hand if labwc support multiple menus or of a real simple app launcher without dependencies?

[dimkr]

Do you know off hand if labwc support multiple menus

See man labwc-menu and search man labwc-actions for ShowMenu.

Regarding stability of waybar - I haven't experienced any issues. It only refuses to run if the configuration files are invalid, so maybe try to run it from a terminal - it will spit errors if you have a syntax error.

[d-pupp]

@dimkr I ran waybar from the terminal and recreated the problem by clicking the menu and then not selecting anything.
There was no error from waybar. However dmesg showed the following.
The segfault was when I copy and pasted in the code to use the menu

Code: Select all

[ 1165.441478] waybar[1438]: segfault at 2 ip 00007f5a634b9d0d sp 00007fff6998ced0 error 4 in libgtk-3.so.0.2411.32[7f5a631ed000+38c000] likely on CPU 0 (core 0, socket 0)
[ 1165.441510] Code: 0b 00 e8 d6 3b d3 ff 31 c0 5b c3 66 90 f3 0f 1e fa 53 48 89 fb e8 63 82 ff ff 48 85 db 74 3e 48 89 c6 48 8b 03 48 85 c0 74 05 <48> 39 30 74 0c 48 89 df e8 b6 8f d3 ff 85 c0 74 22 48 89 df e8 6a
[ 1607.256917] perf: interrupt took too long (2526 > 2500), lowering kernel.perf_event_max_sample_rate to 79000
[ 1899.584678] input: altf1 as /devices/virtual/input/input12
[ 1911.630129] perf: interrupt took too long (3162 > 3157), lowering kernel.perf_event_max_sample_rate to 63250
[ 1915.164630] waybar[1886]: segfault at 55c900000016 ip 00007fcb23664d28 sp 00007fffbd259628 error 4 in libgobject-2.0.so.0.8101.0[7fcb23638000+36000] likely on CPU 0 (core 0, socket 0)
[ 1915.164662] Code: 8d 15 7c 66 02 00 48 c1 e9 02 4c 8b 0c ca 4d 85 c9 74 07 41 f6 41 16 20 75 36 48 8b 38 48 81 ff fc 03 00 00 76 3c 48 83 e7 fc <f6> 47 16 10 74 05 4d 85 c9 75 4d 31 c0 c3 66 2e 0f 1f 84 00 00 00
[ 3909.703876] perf: interrupt took too long (3967 > 3952), lowering kernel.perf_event_max_sample_rate to 50250
[ 6290.903162] input: altf1 as /devices/virtual/input/input13
[ 6341.167715] input: altf1 as /devices/virtual/input/input14
[ 6343.279605] input: altf1 as /devices/virtual/input/input15
[ 6345.623026] input: altf1 as /devices/virtual/input/input16
[ 6404.310279] input: altf1 as /devices/virtual/input/input17
[ 6406.885385] input: altf1 as /devices/virtual/input/input18
[ 6423.057784] input: altf1 as /devices/virtual/input/input19
[ 6424.998693] input: altf1 as /devices/virtual/input/input20
[ 6427.477618] input: altf1 as /devices/virtual/input/input21
[ 6446.515419] input: altf1 as /devices/virtual/input/input22
[ 6454.990141] input: altf1 as /devices/virtual/input/input23
[ 6511.219493] input: altf1 as /devices/virtual/input/input24
[ 7330.626709] perf: interrupt took too long (4970 > 4958), lowering kernel.perf_event_max_sample_rate to 40000
[ 7722.075514] input: altf1 as /devices/virtual/input/input25
[ 7729.668096] input: altf1 as /devices/virtual/input/input26
[ 7748.487322] input: altf1 as /devices/virtual/input/input27
[ 7756.677105] input: altf1 as /devices/virtual/input/input28
[ 7758.902653] input: altf1 as /devices/virtual/input/input29
[ 7763.958264] input: altf1 as /devices/virtual/input/input30
[ 8039.850971] input: altf1 as /devices/virtual/input/input31
[19793.629266] input: altf1 as /devices/virtual/input/input32
[19862.933825] input: altf1 as /devices/virtual/input/input33
[19864.280901] input: altf1 as /devices/virtual/input/input34
[19865.235760] input: altf1 as /devices/virtual/input/input35
[20047.166136] input: altf1 as /devices/virtual/input/input36
[20080.072214] input: altf1 as /devices/virtual/input/input37
[20094.020948] input: altf1 as /devices/virtual/input/input38
[20212.725199] input: altf1 as /devices/virtual/input/input39
[20214.168808] input: altf1 as /devices/virtual/input/input40

[d-pupp]

To clarify the above post The segfault has nothing to do with the problem with waybar. That was me working on the launcher menu I wanted

[dimkr]

@d-pupp I'm sure this waybar crash will get fixed easily and quickly if you submit a detailed bug report with instructions how to reproduce it.

[d-pupp]

@dimkr I have done more testing and traced the problem to the alt f1 used by the menu icon in waybar.
I changed alt f1 to alt f2 the menu icon won't open the run dialog as expected
I then changed it to alt z on both configs and tested that alt z opens the menu in labwc. The menu still doesn' t work in waybar. During both of these tests with the menu not working I had no problems with waybar.
When I put everything back to original the menu works but the problem with waybar not recognizing clicks returns .
Waybar is acting like it's waiting for something when the menu had been opened. In all other cases when I click on a icon the app opens and I get the waybar back.
Did you have to do something to make the alt f1 menu activation work?

dmesg is still showing

1035.903518] input: altf1 as /devices/virtual/input/input20

even with alt f1 changed in both configs

[dimkr]

@d-pupp I don't fully understand what you're trying to do and how exactly, but altf1 is a very simple tool that injects an alt+f1 key press event. Nothing more.

https://github.com/vanilla-dpup/woof-CE ... py/altf1.c

[d-pupp]

@dimkr Thanks for the information. I thought altf1 was a part of waybar I now see it is a executable that send alt f1 keystrokes.

What I am trying to do is figure out why I have been having trouble with waybar not recognizing that I am clicking on it.

The issue is if I open the menu from waybar and say open the calculator or decide not to open anything and click someplace to close it. I don't get the bar back right away. Sometimes I have to click several times before it starts working again. For me it has been this way for quite some time.

It only seems to happen sometimes and only when using the altf1 executable. Maybe it's something with my setup or hardware. I take that you can't reproduce the issue?

I'll run another test on the stock version and let you know what I find.

[d-pupp]

@dimkr I have testing waybar menu altf1 issue in a stock dpup 11.0.169 and can confirm it has the same issue on my system at least.

Here is how I tested.
Open apps from the waybar menu. They will stack on top or one another.
Each time you open an app use the task bar to bring the one behind forward.
On my system after two or three times it stops working and takes several clicks before it starts working again.
You will know it's not working if you get the system menu. It acts like it's not there and you just clicked on the desktop.

However I may have a bigger issue. When trying to test on a stock frugal install I could not boot at all without a save folder!
The main sfs would fail to load and dropped me to a minimal prompt. The only way out was a hard shutdown or reboot -f

[d-pupp]

@dimkr I'm not sure if you need this but I recreated the boot failure this morning.
I couldn't save the logs but I do have a shot of the screen on my phone.
It doesn't tell you much.
It said waiting for slow device 1 of 5 ...
Then Loading puppy main sfs file failed
swap mount failed It been doing that for a couple of versions
mounted and unmounted sda1 and sda2
*** is not OK
*** error is too critical, dropping out to console
Then it drops to Busybox prompt
No drives mounted. However no problem mounting sda1 and mkdir vanilladpupsave
It then booted normally

Hope this helps
d-pupp

[dimkr]

@d-pupp Without any details like blkid output, the kernel command line and logs I can't help much.

[d-pupp]

@dimkr Here is the information requested. Let me know if you need anything else and the commands to run.
blkid

Code: Select all

/dev/sda1: UUID="b226b3de-6248-4c96-9404-ae8b77953d40" TYPE="ext4"
/dev/sda2: LABEL="data" UUID="f4d27ff0-e0bc-45e5-bda9-91fe9808adfb" TYPE="ext4"
/dev/sda3: LABEL="swap" UUID="d1316763-2ecb-4b29-8920-11b260b1054a" TYPE="swap"

lsblk

Code: Select all

NAME   MAJ:MIN RM   SIZE RO TYPE MOUNTPOINTS
sda      8:0    0   149G  0 disk 
|-sda1   8:1    0   9.8G  0 part /mnt/dev_save
|-sda2   8:2    0 138.7G  0 part 
`-sda3   8:3    0     8M  0 part 
sr0     11:0    1  1024M  0 rom  

bootinit.log

Code: Select all

HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.
HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.
HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.
HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.
HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.
HAVE_PARTS=''
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
mount: mounting /dev/sda3 on /mnt/pdrv failed: No such device
sda3 on /mnt/pdrv as swap mount failed.

dmegs (fake.gz)
Edit Note: I mounted sda1 so that I could save these files