search for: tfb_varying_locs

...9:06:12 hpprol2 kernel: nouveau 0000:0a:00.0: fifo: PBDMA0: 80000000 [] ch 30 [007e6ab000 kscreenlocker_g[4257]] subc 6 mthd 2878 data a0c02020 A few things wrong with this picture... first off, nouveau never uses subchannel 6. Method 2878 is: lookup -a c8 -d SUBCHAN -- -v obj-class GF100_3D 2878 TFB_VARYING_LOCS[0][0x1e] => 0 But those are all emitted in one fell swoop, starting from 2800. No way something could get into the middle of that and screw up the channel on just that one. (Since they don't have independent begin clauses). So this has to have come from elsewhere.... but where? Is this the...