Description of problem: Version-Release number of selected component (if applicable): 2.6.22.1-41 2.6.22.4-65 How reproducible: every time Steps to Reproduce: 1. boot 2. S06cpuspeed causes it 3. acpi=off solves (by disabling the module) Actual results: typed from screen capture, apologies for errors: kernel BUG at drivers/acpi/osl.c:428! invalid opcode: 0000 [#1] SMP last sysfs file: /devices/pnp0/00:00/id Modules linked in: acpi_cpufreq dm_mirror dm_multipath dm_mod video sbs button ock battery ac floppy snd_hda_intel snd_seq_dummy snd_seq_oss snd_seq_midi_eve snd_seq parport_pc snd_seq_device snd_pcm_oss snd_mixer_oss snd_pcm r8169 parport snd_timer iTCO_wdt snd iTCO_vendor_support i2c_i801 soundcore snd_page_alloc i2c_core rtc_cmos sg ata_generic ata_piix libata sd_mod scsi_mod ext3 jbd mbcase ehci_hcd ohci_hcd uhci_hcd CPU: 0 EIP: 0060:[<c050e018>] Not tainted VLI EFLAGS: 00210297 *2.6.22.4-65.fc7 #1) EIP is at acpi_os_write_port+0x2a/0x32 eax: 00000001 ebx: 00000832 ecx: 00000001 edx: 00000001 esi: de834cb4 edi: 00000000 ebp: 00000000 esp: de834c7c ds: 007b es: 007b fs: 00d8 gs: 0033 ss: 0068 Process modprobe (pid: 1295, ti=de834000 task=de1d38000 task.ti=de8340000) Stack: ... Call Trace: [<e028a4d6>] acpi_cpufreq_target+0x274/0x34e [acpi_cpufreq] ... [<e007e0c5>] acpi_cpufreq_init+0xc5/0xca [acpi_cpufreq] ... Expected results: Should not crash Additional info: I can provide the rest of the trace if desired (preferably as screenshots). I can also reproduce/test patches/fiddle with code if pointed in the right direction.
Created attachment 174781 [details] lspci -vv for the machine
Please post the full trace (screenshots are fine, just attach a jpeg.)
Created attachment 177521 [details] crash screenshot part 1
Created attachment 177541 [details] crash screenshot part 2
Created attachment 177561 [details] crash screenshot part 3
Hello Mark, I'm reviewing this bug as part of the kernel bug triage project, an attempt to isolate current bugs in the fedora kernel. http://fedoraproject.org/wiki/KernelBugTriage I am CC'ing myself to this bug and will try and assist you in resolving it if I can. There is a patch in the Fedora 2.6.23 kernel which outputs additional debugging information for this issue. If you are willing and able, can you test and post back - the line should say something like: writing 1234 bits to port 1234 or some such. If the problem no longer exists then please close this bug as CURRENTRELEASE indicating what resolved it for you or I'll do so in a few days if there is no additional information lodged.
Should be fixed by commit 49fbabf56dc715bbb51e59742e82ba762790aac0 [ACPI: Handle I/O access width requestst that are not a multiple of 8 bits]
Sorry for the long delay. I have tested: 2.6.23.12-52.fc7 and the result is the same. However I don't see any additional debug output, is there something required to enable it?
Hi Mark, The above commit was tagged for 2.6.24-rc3 so you won't see it yet. If you are able you could pull a development kernel from rawhide and test that with: # yum update kernel --enablerepo=development although 2.6.24 final is just around the corner... Cheers Chris
Works fine now, thanks.