<inode>
i have mainline linux 5.4 running on rk3229, which supossedly features a mali-400 gpu... i've built libdrm, mesa (from master), glmark2-es2, and xorg - yet if i make use of anything that involves libEGL like glmark2-es2-drm or Xorg's modesetting with glamor enable - the system locks up and i receive kernel messages about rcu_shed detecting stalls, and NMIs fired for seemingly unresponsive cores
<inode>
and before this, libEGL says "libEGL debug: EGL user error 0x3001 (EGL_NOT_INITIALIZED) in eglQueryString"
<inode>
does anyone have any ideas about what may be wrong?
yann has quit [Ping timeout: 268 seconds]
yuq825 has quit [Remote host closed the connection]
robertfoss has quit [Ping timeout: 240 seconds]
robertfoss has joined #lima
buzzmarshall has quit [Remote host closed the connection]
Wizzup has joined #lima
Wizzup has quit [Client Quit]
Wizzup has joined #lima
BenG83 has joined #lima
yann has joined #lima
<anarsoul>
inode: I'm not sure where rcu_sched stalls are coming from, but I'd suggest to check whether you have CPU and/or GPU voltage set correctly
<anarsoul>
lima doesn't configure it atm, so you have to make sure that initial values are good for GPU to work
<inode>
thanks, i'll try to confirm that they match what the vendor's kernel happens to use