Bug 77941
Summary: | Bug with intel_pm fills my logs | ||
---|---|---|---|
Product: | Drivers | Reporter: | contact |
Component: | Video(DRI - Intel) | Assignee: | Chris Wilson (chris) |
Status: | RESOLVED INVALID | ||
Severity: | normal | CC: | intel-gfx-bugs |
Priority: | P3 | ||
Hardware: | x86-64 | ||
OS: | Linux | ||
Kernel Version: | 3.14.6-1-ARCH | Subsystem: | |
Regression: | No | Bisected commit-id: |
Description
contact
2014-06-15 12:22:06 UTC
Actually, the powerdown service, with (likely) this line was responsible: for i in /sys/kernel/debug/dri/*/i915_max_freq; do opt "$i" 350; done Bug solved. (In reply to contact from comment #1) > Actually, the powerdown service, with (likely) this line was responsible: > > for i in /sys/kernel/debug/dri/*/i915_max_freq; do opt "$i" 350; done Uhm, what? Where is this from exactly? Oh sorry, it's from the powerdown script. It should be located at /usr/bin/powerdown, on Archlinux. We've added better bounds checks for debugfs in v3.15, which means you'll no longer see the warns in dmesg. *However* the magic 350 in the powerdown service will still be as much bogus as it ever was. Maybe it was something that worked for the author's machine at some point in time, but really it's debugfs and all bets are off. Similarly for various power tuning i915 module parameters (fbc, rc6, etc). They have safe per platform defaults for a reason. |