Looks familiar. Have you tried clearing the shadercache?
In ~/.cache folder, there's a cache called: 'qtshadercache'.
- 2
Looks familiar. Have you tried clearing the shadercache?
In ~/.cache folder, there's a cache called: 'qtshadercache'.
3 minutes ago, ahoeben said:Looks familiar. Have you tried clearing the shadercache?
In ~/.cache folder, there's a cache called: 'qtshadercache'.
Thank you, thank you, thank you!!
Sanity is restored and 'rm -rf ~/.cache/qtshadercache/*' is recorded in my Cura notes for next time.
Thanks for the logs. Looking at cura.log, the reported OpenGL version changed on the 14th of August (19.0.2 to 19.0.8) so it's not obvious what happened today and I can see that Aldo's suggestion worked for you. Happy days!
3 minutes ago, burtoogle said:Thanks for the logs. Looking at cura.log, the reported OpenGL version changed on the 14th of August (19.0.2 to 19.0.8) so it's not obvious what happened today and I can see that Aldo's suggestion worked for you. Happy days!
The update that caused the problem doesn't change the MESA version number. In the packaging, the old version (currently installed) is 19.0.8-0ubuntu0~18.04.1 and the update is 19.0.8-0ubuntu0~18.04.2.
The change log lists nothing which should cause worry:
* i965-revert-enabling-softpin.diff: Dropped, unnecessary by now and
broke Ice Lake which needs softpin. (LP: #1815172)
* rules: Force -DNDEBUG, meson < 0.47 is buggy and doesn't set it even
if we have b_ndebug=true. (LP: #1841412)
But then again, it wouldn't be the first time that a change log doesn't tell the whole story...
Had the exact same problem just now with the Mesa update. I've taken to generating a system snapshot before every 'upgrade' in Mint 19.2. Someone isn't paying attention lately. A simple restore of the snapshot cures the problem.
On 9/24/2019 at 4:13 PM, lali said:Had the exact same problem just now with the Mesa update. I've taken to generating a system snapshot before every 'upgrade' in Mint 19.2. Someone isn't paying attention lately. A simple restore of the snapshot cures the problem.
That's odd. Restoring the snapshot didn't fix it for me this time around, which is logical if the update causes QT's cached shader data (which is in my home directory and excluded from the snapshot) to become invalid.
I guess there's more than one thing that gets broken when MESA is updated.
That's very good to know. I was feeling so smug.
I've added ''rm -rf ~/.cache/qtshadercache/*" to my notes, also.
Recommended Posts
grizewald 3
Sure. Here's cura.log.gz
and stderr.log.gz
Both compressed with gzip to save my bandwidth. I couldn't upload them to the forum due to the 511MB file size limit.
Link to post
Share on other sites