| Summary: | Missing NULL check; radeon_drm_winsys.c | ||
|---|---|---|---|
| Product: | Mesa | Reporter: | Dave Gilbert <freedesktop> |
| Component: | Drivers/Gallium/r600 | Assignee: | Default DRI bug account <dri-devel> |
| Status: | RESOLVED MOVED | QA Contact: | |
| Severity: | normal | ||
| Priority: | medium | CC: | fredericg_99 |
| Version: | unspecified | ||
| Hardware: | Other | ||
| OS: | All | ||
| Whiteboard: | |||
| i915 platform: | i915 features: | ||
OK, I believe the case that's causing it to go down this route is a perms screw up somewhere; I was running this test in a separate X session (just logging as a normal user on the console and running X :2 ) But with a bit of debug I'm seeing: libGL: OpenDriver: trying /usr/lib64/dri/tls/r600_dri.so libGL: OpenDriver: trying /usr/lib64/dri/r600_dri.so libGL error: failed to open drm device: Permission denied libGL error: failed to load driver: r600 [dg@major ~]$ ls -lZ /dev/dri/card0 crw-rw----+ root video system_u:object_r:dri_device_t:s0 /dev/dri/card0 [dg@major ~]$ id -Z unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 So I'm not quite sure what the perm screw up is, but that's separate - this will seg in glxgears or anything if the drm device won't open. Dave and now understand the perm problem; I was running X :2 from vt 2, it was starting X visible on vt 3, when I changed to vt3 it removed the acl on /dev/dri/card0 so that it lost the perms when clients ran (although the server still had it open). So still say it's missing that NULL check. -- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/mesa/mesa/issues/476. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
I can sometime trigger a seg in do_winsys_init at radeon_drm_winsys.c on my HD4350 There seems to be a simple missing NULL check: 214 version = drmGetVersion(ws->fd); 215 if (version->version_major != 2 || 216 version->version_minor < 3) { All the other users of drmGetVersion do a NULL check, so I suggest adding: if (!version) { fprintf(stderr,"%s: drmGetVersion NULL - bad fd (%d)?\n", __FUNCTION__, ws->fd); return FALSE; } I believe this to be part of the reason for: https://bugzilla.redhat.com/show_bug.cgi?id=993463 why drmGetVersion is failing is a different matter (probably a bad fd at that point - but why?) , but at least the NULL check would stop the seg. Dave