Summary: | [UXA IVB bisected] gnome-system-monitor perforamnce reduce by ~40% | ||
---|---|---|---|
Product: | xorg | Reporter: | ye.tian <yex.tian> |
Component: | Driver/intel | Assignee: | Chris Wilson <chris> |
Status: | CLOSED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
Severity: | major | ||
Priority: | low | ||
Version: | git | ||
Hardware: | x86 (IA32) | ||
OS: | Linux (All) | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
ye.tian
2013-01-05 07:24:12 UTC
Oh that is funny, performance testing using UXA. You might also add that UXA also renders it incorrectly to the list of its faults. Adjusting priority to reflect that apparently I am not allowed to fix performance bugs. (In reply to comment #1) > Oh that is funny, performance testing using UXA. You might also add that UXA > also renders it incorrectly to the list of its faults. Hi Chris, could you please tell me how to run cairo-trace without UXA, or give me a standard command? To test another X server, you have to run an alternative X server. To control which backend is being tested, use CAIRO_TEST_TARGET. (In reply to comment #2) > Adjusting priority to reflect that apparently I am not allowed to fix > performance bugs. I don't understand. So who should own this bug? And I don't feel this is a good reason to demote the priority. The bug is in the implementation of UXA. Given that the changes speeds up the accelerated drivers, cairo will optimise for them not a badly designed driver. so will you consider it won't ix? Quite the opposite. The fix has been floating around upstream for over a year. verified it. |
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.