Bug 1391212
| Summary: | Service fails if no switchable graphics present | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Adam Williamson <awilliam> |
| Component: | switcheroo-control | Assignee: | Kalev Lember <klember> |
| Status: | CLOSED ERRATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 25 | CC: | bnocera, gmarr, klember, kwizart, mp.x, rc040203, sgallagh, zbyszek |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | AcceptedFreezeException | ||
| Fixed In Version: | switcheroo-control-1.1-1.fc25 | Doc Type: | If docs needed, set a value |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2016-11-10 16:38:24 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
| Bug Depends On: | |||
| Bug Blocks: | 1277290 | ||
Whoever patches this should get rid of the useless 'err' variable... (In reply to Zbigniew Jędrzejewski-Szmek from comment #1) > Whoever patches this should get rid of the useless 'err' variable... It wouldn't be useless if errno wasn't thread-specific. Which it is. Bad habit. Please file upstream bugs in the future. Fixed both upstream. (In reply to Bastien Nocera from comment #2) > (In reply to Zbigniew Jędrzejewski-Szmek from comment #1) > > Whoever patches this should get rid of the useless 'err' variable... > > It wouldn't be useless if errno wasn't thread-specific. Which it is. Bad > habit. > > Please file upstream bugs in the future. > > Fixed both upstream. Can you please link to the fixes here and update this ticket with the Koji build that resolves it? We need to have a downstream ticket here because we're in Final Freeze and would need a Freeze Exception or Blocker status in order to pull this in. Oh, and +1 FE from me. It will need release notes changes too. Where would I do that? switcheroo-control-1.1-1.fc25 has been submitted as an update to Fedora 25. https://bodhi.fedoraproject.org/updates/FEDORA-2016-3d2ba1299d switcheroo-control-1.1-1.fc25 has been pushed to the Fedora 25 testing repository. If problems still persist, please make note of it in this bug report. See https://fedoraproject.org/wiki/QA:Updates_Testing for instructions on how to install test updates. You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2016-3d2ba1299d Discussed during the 2016-11-07 blocker review meeting: [1] The decision was made to classify this bug as an AcceptedFreezeException. [1] https://meetbot.fedoraproject.org/fedora-blocker-review/2016-11-07/f25-blocker-review.2016-11-07-17.01.txt switcheroo-control-1.1-1.fc25 has been pushed to the Fedora 25 stable repository. If problems still persist, please make note of it in this bug report. *** Bug 1393759 has been marked as a duplicate of this bug. *** (In reply to Fedora Update System from comment #7) > switcheroo-control-1.1-1.fc25 has been pushed to the Fedora 25 testing > repository. If problems still persist, please make note of it in this bug > report. > See https://fedoraproject.org/wiki/QA:Updates_Testing for > instructions on how to install test updates. > You can provide feedback for this update here: > https://bodhi.fedoraproject.org/updates/FEDORA-2016-3d2ba1299d I have the same problem ```$ systemctl status switcheroo-control.service ● switcheroo-control.service - Switcheroo Control Proxy service Loaded: loaded (/usr/lib/systemd/system/switcheroo-control.service; enabled; vendor preset: enabled) Active: failed (Result: exit-code) since Thu 2018-02-01 13:51:15 +03; 27min ago Process: 957 ExecStart=/usr/sbin/switcheroo-control (code=exited, status=1/FAILURE) Main PID: 957 (code=exited, status=1/FAILURE) Feb 01 13:51:12 localhost.localdomain systemd[1]: Starting Switcheroo Control Proxy service... Feb 01 13:51:12 localhost.localdomain switcheroo-cont[957]: switcheroo-control could not query vga_switcheroo status: Opera Feb 01 13:51:15 localhost.localdomain systemd[1]: switcheroo-control.service: Main process exited, code=exited, status=1/FA Feb 01 13:51:15 localhost.localdomain systemd[1]: Failed to start Switcheroo Control Proxy service. Feb 01 13:51:15 localhost.localdomain systemd[1]: switcheroo-control.service: Unit entered failed state. Feb 01 13:51:15 localhost.localdomain systemd[1]: switcheroo-control.service: Failed with result 'exit-code'.``` and dmesg error ``` [ 0.050375] ACPI Exception: Could not find/resolve named package element: AMD3 (20170728/dspkginit-381) [ 0.050406] ACPI Exception: Could not find/resolve named package element: AMD2 (20170728/dspkginit-381) [ 0.050431] ACPI Exception: Could not find/resolve named package element: AMD3 (20170728/dspkginit-381) [ 3.207193] kfd kfd: kgd2kfd_probe failed [ 3.210211] ACPI Error: [\_SB_.PCI0.GFX0.DD02._BCL] Namespace lookup failure, AE_NOT_FOUND (20170728/psargs-364) [ 3.210218] ACPI Error: Method parse/execution failed \_SB.PCI0.RP05.PEGP.DD02._BCL, AE_NOT_FOUND (20170728/psparse-550) [ 141.664156] sd 4:0:0:0: [sdb] No Caching mode page found [ 141.664160] sd 4:0:0:0: [sdb] Assuming drive cache: write through [Fedora@localhost ~]$ dmesg | grep -i switcheroo [ 2.937126] vga_switcheroo: detected switching method \_SB_.PCI0.GFX0.ATPX handle [ 2.971667] vga_switcheroo: enabled``` |
switcheroo-control is written to exit 1 if no switchable graphics hardware is present: /* Check for VGA switcheroo availability */ fd = open (SWITCHEROO_SYSFS_PATH, O_WRONLY); if (fd < 0) { int err = errno; switch (err) { case EACCES: g_warning ("switcheroo-control needs to run as root"); break; case ENOENT: g_debug ("No switcheroo support available"); break; default: g_warning ("switcheroo-control could not query vga_switcheroo status: %s", g_strerror (err)); } return 1; } This means the switcheroo-control service will very commonly fail on boot. This is bad, we don't want failed services, they cause users to worry (and we have an openQA test that checks for them which will fail forever if this is not fixed). This should be tweaked somehow so the service isn't considered 'failed' when no switchable graphics are present. Nominating as a freeze exception as I believe this is worth fixing for F25 Final and we're in freeze. There are several ways to fix this that shouldn't be too dangerous.