Latest working kernel version: n/a Earliest failing kernel version: at least 2.6.21 Distribution: opensuse 10.3 Hardware Environment: vmware Software Environment: opensuse 10.3 + kernel 2.6.26rc1 Problem Description: when doing some module regression testing (load/unload most available modules), cfag12864bfb driver always kills the local console on load and locks you out. visible area of vc shrinks to 16x4 and output freezing. this happens if you don`t have appropriate hardware attached, cannot test with hw attached (don`t own such). no vc switching afterwards. cannot unload module from remote (ssh) either. need reboot. Steps to reproduce: save all of your important data, switch to textconsole, modprobe cfag12864bfb dmesg shows: Console: switching to mono frame buffer device 16x8 fb0: cfag12864b frame buffer device contacted the author who confirmed the issue and tells that this didn`t happen initially. iirc, he thinks this is a regression due to change in framebuffer layer. didn`t get a response since then, maybe he`s lacking time. i`m putting it here so it`s being marked as a ToDo and won`t get lost.
Reply-To: akpm@linux-foundation.org (switched to email. Please respond via emailed reply-to-all, not via the bugzilla web interface). On Sun, 11 May 2008 05:10:41 -0700 (PDT) bugme-daemon@bugzilla.kernel.org wrote: > http://bugzilla.kernel.org/show_bug.cgi?id=10666 > > Summary: cfag12864bfb messing up vc on load > Product: Drivers > Version: 2.5 > KernelVersion: 2.6.21+ > Platform: All > OS/Version: Linux > Tree: Mainline > Status: NEW > Severity: low > Priority: P1 > Component: Console/Framebuffers > AssignedTo: jsimmons@infradead.org > ReportedBy: devzero@web.de > > > Latest working kernel version: n/a > > Earliest failing kernel version: at least 2.6.21 > > Distribution: opensuse 10.3 > > Hardware Environment: vmware > > Software Environment: opensuse 10.3 + kernel 2.6.26rc1 > > Problem Description: > when doing some module regression testing (load/unload most available > modules), > cfag12864bfb driver always kills the local console on load and locks you out. > visible area of vc shrinks to 16x4 and output freezing. this happens if you > don`t have appropriate hardware attached, cannot test with hw attached (don`t > own such). no vc switching afterwards. cannot unload module from remote (ssh) > either. need reboot. > > Steps to reproduce: > save all of your important data, switch to textconsole, modprobe cfag12864bfb > > dmesg shows: > Console: switching to mono frame buffer device 16x8 > fb0: cfag12864b frame buffer device > > contacted the author who confirmed the issue and tells that this didn`t > happen > initially. iirc, he thinks this is a regression due to change in framebuffer > layer. didn`t get a response since then, maybe he`s lacking time. > i`m putting it here so it`s being marked as a ToDo and won`t get lost. > > > -- > Configure bugmail: http://bugzilla.kernel.org/userprefs.cgi?tab=email > ------- You are receiving this mail because: ------- > You are on the CC list for the bug, or are watching someone who is.
apparently, i cannot reproduce this on 2.6.28-rc2-git3 as nothing appropriate changed in the driver code, there must be some change in the console/framebuffer-layer which has fixed that.
I tested long ago the driver at 2.6.25 (even sooner, but that is the first time I remember clearly) and it worked nicely again. I will perform a re-test on 2.6.27. Roland Kletzing is right, the code hasn't changed since years ago (as far as I know), so maybe someone changed some fb layer behaviour for a couple of months?
It appears to have been resolved