Release [WIP] Master Control - Testers needed


Thanks for your feedbacks. I found a critical bug in the new plugin system, where the HID descriptors are assembled incorrectly. This error affects both, USB and Bluetooth. I will upload a hotfixed version tomorrow!

Download will be disabled until then.
Oh hey thanks! This is a really cool program and I wish you luck with it! :)
 
Just happened that even with "Grab input" deselected it was still sending keys to the PC, while writing on the Pandora itself... searching for the cause...

ok, found:

if you connect (bluetooth) with "Grab input" deselected, it will start to grab without blocking the pandora, and even if you select/deselect it nothing changes.

If you start the connection with "Grab input" selected, all works fine if you deselect/select it
 
Just happened that even with "Grab input" deselected it was still sending keys to the PC, while writing on the Pandora itself... searching for the cause...
"Grab input" does not stop sending events to the host. It just blocks input to to the Pandora's X system.

if you connect (bluetooth) with "Grab input" deselected, it will start to grab without blocking the pandora, and even if you select/deselect it nothing changes.

If you start the connection with "Grab input" selected, all works fine if you deselect/select it
Are you sure, that nothing changes? It should allow/disallow input to the Pandora. Maybe you misunderstood the option. Or maybe I should call it somehow else :)
 
Last edited by a moderator:
Simply want to note that I played YGOPro on my Arch Linux over Wine with Master Control.

Also used it to navgate throgh folders.

So Mouse and Keyboard over USB work fine here.

:)  
 
Works perfectly on my fresh NAND install (which I only use for testing). Segfaults immediately on my customized SD-card install. Any chance you'd release the sources, so I can try to get it working on my normal install? Probably a simple recompile would be enough to fix the problem.
 
customized SD-card install
Do you mean SZ running off SD-card? What did you change in the distribution ? 
Yes. Mostly installing extra stuff and manually compiling some things to get more recent versions. Usually that doesn't break PNDs, but in this case something does break - it's just a bit hard to tell what goes wrong, if all you've got is a stripped binary.
 
Just happened that even with "Grab input" deselected it was still sending keys to the PC, while writing on the Pandora itself... searching for the cause...
"Grab input" does not stop sending events to the host. It just blocks input to to the Pandora's X system.

if you connect (bluetooth) with "Grab input" deselected, it will start to grab without blocking the pandora, and even if you select/deselect it nothing changes.

If you start the connection with "Grab input" selected, all works fine if you deselect/select it
Are you sure, that nothing changes? It should allow/disallow input to the Pandora. Maybe you misunderstood the option. Or maybe I should call it somehow else :)
I mean I was actually writing at the same time on Mousepad on the Pandora AND Notepad on Windows indipendently of the state of "Grab input"
 
customized SD-card install
 
Do you mean SZ running off SD-card? What did you change in the distribution ?
 
Yes. Mostly installing extra stuff and manually compiling some things to get more recent versions. Usually that doesn't break PNDs, but in this case something does break - it's just a bit hard to tell what goes wrong, if all you've got is a stripped binary.
What does the pnd out says?
 
customized SD-card install
 
Do you mean SZ running off SD-card? What did you change in the distribution ?
 
Yes. Mostly installing extra stuff and manually compiling some things to get more recent versions. Usually that doesn't break PNDs, but in this case something does break - it's just a bit hard to tell what goes wrong, if all you've got is a stripped binary.
What does the pnd out says?
Only that it segfaulted. If I make it dump core and backtrace with gdb, it looks like the problem occurs somewhere in a call to Qt, so it probably has something to do with the fact that my Qt is not exactly the same as the one on a default firmware (it's the same version though, but I did manually compile it, probably with different options). A simple recompile would probably fix it for me.

Anyway, I don't mind rebooting to a stock SZ to use this, but it's of course a bit inconvenient. Also, this is something that runs with root privileges, so it wouldn't hurt to check the source for malware, security issues and bugs. Especially if we want this included in the "official" firmware.
 
Only that it segfaulted. If I make it dump core and backtrace with gdb, it looks like the problem occurs somewhere in a call to Qt, so it probably has something to do with the fact that my Qt is not exactly the same as the one on a default firmware (it's the same version though, but I did manually compile it, probably with different options). A simple recompile would probably fix it for me.
I know what you're talking about. This is a weird thing. When using tooltips or popup menus, Qt gonna crash when showing these. The backtrace always goes down to 'strlen' in libc. I guess strlen is called with a null pointer somehow. BUT: This only

happens, when I compile MC with the '-O2' flag. Using '-O1' or '-O0' seems to prevent these crashes.

Anyway, I don't mind rebooting to a stock SZ to use this, but it's of course a bit inconvenient. Also, this is something that runs with root privileges, so it wouldn't hurt to check the source for malware, security issues and bugs. Especially if we want this included in the "official" firmware.
I gonna upload the sources to sourceforge within the next days, so you might take a look.
 
I mean I was actually writing at the same time on Mousepad on the Pandora AND Notepad on Windows indipendently of the state of "Grab input"
Ah ok, that shouldn't happen. I'll check that.
Can't reproduce that. Works as intended for me.

Can you check the log file, please? Clicking "Grab input" should produce "Grabbing Mouse/Keyboard: Ok! / Failed!"
 
Still unable to get this to run. Updated to the latest version on the repo, and used the "Upgrade pandora OS" option to upgrade kernel and stuff,

It asks for root password then nothing appears.

It created mastercontrol folder in appdata, nothing in it though. No logs or anything.
 
I was gonna grab the log but it seemed to have been replaced with a new one, no errors in it. after restarting my pandy it worked fine and my pc found the usd hid devices. only tested the keyb and mouse but it worked.

will grab log next time it errors on me.
 
Still unable to get this to run. Updated to the latest version on the repo, and used the "Upgrade pandora OS" option to upgrade kernel and stuff,

It asks for root password then nothing appears.

It created mastercontrol folder in a

ppdata, nothing in it though. No logs or anything.
Looks like the startup script bails out somewhere. Please check /tmp/pndrun_mastercontrol.out for errors.
 
Still unable to get this to run. Updated to the latest version on the repo, and used the "Upgrade pandora OS" option to upgrade kernel and stuff,

It asks for root password then nothing appears.

It created mastercontrol folder in a

ppdata, nothing in it though. No logs or anything.
Looks like the startup script bails out somewhere. Please check /tmp/pndrun_mastercontrol.out for errors.
mount: unknown filesystem type 'gadgetfs' appears to be why it won't launch.
 
Ah, this damn script. Don't know, why it doesn't show an error message. Your error occurs, when there is a module loaded, that conflicts with gadgetfs. These modules should be unloaded by the script automatically.

Please type "lsmod | grep g_" in a console and then "sudo rmmod <module>" for all listed modules.
 
Last edited by a moderator:
Still unable to get this to run. Updated to the latest version on the repo, and used the "Upgrade pandora OS" option to upgrade kernel and stuff,

It asks for root password then nothing appears.

It created mastercontrol folder in a

ppdata, nothing in it though. No logs or anything.
Looks like the startup script bails out somewhere. Please check /tmp/pndrun_mastercontrol.out for errors.
mount: unknown filesystem type 'gadgetfs' appears to be why it won't launch.
check if the mounted pnd in the log is MasterControl, because this error happens sometimes when another open PND messes things up
 
It still doesn't work on Bluetooth for me.

Here's the new log: 

Code:
Config: Loaded main.conf!
Config: Loaded events.conf!
Found input device on '/dev/input/event0' (keypad)
Found input device on '/dev/input/event1' (power-button)
Found input device on '/dev/input/event2' (nub0)
Found input device on '/dev/input/event3' (nub1)
Found input device on '/dev/input/event4' (gpio-keys)
Found input device on '/dev/input/event5' (touchscreen)
Config: Loaded Keyboard.conf!
Config: Loaded Keyboard_us.conf!
Plugin: Loaded Keyboard!
Config: Loaded Mouse.conf!
Plugin: Loaded Mouse!
Config: Loaded Gamepad.conf!
Plugin: Loaded Gamepad!
SDP: Services registered!
Bluetooth: Server started!
Bluetooth: Listening for connections...
USB: Server started. Waiting for connections...
Bluetooth: Trying to connect to D4:20:6D:6D:0A:49...
Bluetooth: Failed to connect to control psm 17! (Connection refused)
Bluetooth: Connection failed! Connection refused
Bluetooth: Trying to connect to D4:20:6D:6D:0A:49...
Bluetooth: Failed to connect to control psm 17! (Connection refused)
Bluetooth: Connection failed! Connection refused
Bluetooth: Trying to connect to D4:20:6D:6D:0A:49...
Bluetooth: Failed to connect to control psm 17! (Connection refused)
Bluetooth: Connection failed! Connection refused
Bluetooth: Trying to connect to D4:20:6D:6D:0A:49...
Bluetooth: Failed to connect to control psm 17! (Connection refused)
Bluetooth: Connection failed! Connection refused
Bluetooth: Not longer listening...
SDP: Services unregistered!
Bluetooth: Server stopped!
Plugin: Unloaded Gamepad
Plugin: Unloaded Keyboard
Plugin: Unloaded Mouse
 
Back
Top