You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Basically, SDL 2.0.4 doesn't seem to capture any input on Raspberry Pi 2. I built SDL from source and then wrote the program shown in the first link above. (Lemme know if you want the code migrated here.) Did I code some part it wrong? It's a fairly minimal program.
As noted in the second link, if the program is terminated without properly exiting, the keyboard stops working entirely (despite never acknowledging any events). It doesn't respond in the Raspian terminal at all. I was able to reproduce that.
On 2016-01-11 04:19:51 +0000, TheBuzzSaw wrote:
To be clear, if the program exits normally, the keyboard resumes normal operation. So, I added the int n to get the program to exit even if there is no input detected. (I was sick of having to reboot the Pi to fix the problem.)
On 2016-01-16 22:02:53 +0000, Joshua Bodine wrote:
Make sure you have libudev development files installed. Try:
sudo apt-get install libudev-dev
Then, rebuild and reinstall SDL 2. Mouse/keyboard input started working for me after I did that.
There appears to be an issue with the makefile or configure.ac where it searches for libudev files, doesn't find them, but still says at the end that it will use libudev. ./configure --disable-libudev does appear to make it correctly show that libudev will not be used. Regardless, it appears that you do need libudev to get mouse/keyboard input to work on the Raspberry Pi.
On 2016-01-17 18:56:30 +0000, TheBuzzSaw wrote:
Thank you so much. That was indeed my problem. I thought I ensured evdev was there at least, but I forgot. Installing libudev fixed it. It does seem strange that one can compile SDL without input support, but whatever.
On 2016-01-17 21:11:45 +0000, Joshua Bodine wrote:
Created attachment 2367
patch to fix the issue
I'm going to reopen this because configure should still accurately report whether libudev will be used. Right now it just tests whether it's enabled as an argument, not whether configure was successful in finding it. Here's a patch that fixes the issue.
On 2016-01-17 21:12:57 +0000, Joshua Bodine wrote:
This bug report was migrated from our old Bugzilla tracker.
These attachments are available in the static archive:
Reported in version: 2.0.4
Reported for operating system, platform: Linux, ARM
Comments on the original bug report:
On 2016-01-11 04:15:57 +0000, TheBuzzSaw wrote:
On 2016-01-11 04:19:51 +0000, TheBuzzSaw wrote:
On 2016-01-16 22:02:53 +0000, Joshua Bodine wrote:
On 2016-01-17 18:56:30 +0000, TheBuzzSaw wrote:
On 2016-01-17 21:11:45 +0000, Joshua Bodine wrote:
On 2016-01-17 21:12:57 +0000, Joshua Bodine wrote:
On 2016-11-01 17:43:00 +0000, Sam Lantinga wrote:
The text was updated successfully, but these errors were encountered: