As described in the original Debian bug report, a user fails to run a simple `fbset` command. Turnning drm.debug=0x3f does not show the root of the failure making it impossible to track the issue.
> rbthomas@bigal:~$ dmesg > /tmp/dmesg-before.out > > rbthomas@bigal:~$ fbset > > mode "1680x1050" > geometry 1680 1050 1680 1050 32 > timings 0 0 0 0 0 0 0 > accel true > rgba 8/16,8/8,8/0,0/0 > endmode > > rbthomas@bigal:~$ fbset 800x600-60 > ioctl FBIOPUT_VSCREENINFO: Invalid argument > rbthomas@bigal:~$ grep 800x600 /etc/fb.modes > # 800x600, 48 Hz, Interlaced (36.00 MHz dotclock) > mode "800x600-48-lace" > # 800x600, 56 Hz, Non-Interlaced (36.00 MHz dotclock) > mode "800x600-56" > # 800x600, 60 Hz, Non-Interlaced (40.00 MHz dotclock) > mode "800x600-60" > # 800x600, 70 Hz, Non-Interlaced (44.90 MHz dotclock) > mode "800x600-70" > # 800x600, 72 Hz, Non-Interlaced (50.00 MHz dotclock) > mode "800x600-72" > # 800x600, 75 Hz, Non-Interlaced (49.50 MHz dotclock) > mode "800x600-75" > # 800x600, 90 Hz, Non-Interlaced (56.64 MHz dotclock) > mode "800x600-90" > # 800x600, 100 Hz, Non-Interlaced (67.50 MHz dotclock) > mode “800x600-100" > > rbthomas@bigal:~$ dmesg > /tmp/dmesg-after.out > > rbthomas@bigal:~$ diff /tmp/dmesg-before.out /tmp/dmesg-after.out > > rbthomas@bigal:~$ cat /proc/cmdline > root=/dev/mapper/bigal-root ro drm.debug=0xe > > rbthomas@bigal:~$ lsmod | grep drm > drm_kms_helper 82001 1 nouveau > drm 325466 6 ttm,drm_kms_helper,nouveau,sil164 > agpgart 39217 2 drm,ttm > i2c_core 31058 10 > drm,i2c_powermac,windfarm_smu_sat,drm_kms_helper, > > i2c_algo_bit,windfarm_lm75_sensor,nouveau,sil164, > windfarm_max6690_sensor,snd_aoa_codec_onyx
Thanks! I’ll give them a couple of days to respond, then bring it to their attention and offer to help with testing if they are willing to take it on. Thanks very much for all your help so far! Enjoy! Rick On Jul 5, 2016, at 12:54 AM, bugzilla-daemon@bugzilla.kernel.org wrote: > https://bugzilla.kernel.org/show_bug.cgi?id=121411 > > Mathieu Malaterre <mathieu.malaterre@gmail.com> changed: > > What |Removed |Added > ---------------------------------------------------------------------------- > CC| |rbthomas@pobox.com > > -- > You are receiving this mail because: > You are on the CC list for the bug.
This is reproducible on v5.4.6 also. I did a little more debugging and looks like that since the bpp of the emulated framebuffer can not be changed. In my testing I had to add a compatible mode in fb.modes file with the same bpp and fbset worked again.