Eating my own dogfood – dynamic page table allocations

I don’t know if I’ve ever eaten my own dogfood that smells this risky. A few days ago, I published patches to support dynamic page table allocation and tear-down in the i915 driver This work will eventually help us support expanded page tables (similar to how things work for normal Linux page tables). The […]

i915 command submission via gem_exec_nop

EDIT1 (2014-07-12): Apologies to planets for update. Change b->B (bits to bytes) in the state walkthrough (thanks to Bernard Kilarski) Convert SVG images to PNG because they weren’t being rendered properly. Added TOC Use new style footnotes NOTE: With command parser merged, and execlists on the way – this post is already somewhat outdated. Disclaimer:¬†Everything […]

i915.ko authors, by the numbers

Because of some event I cannot remember, I was curious to see which developers had how many lines of code. It turned out to be more of a challenge than I thought (mostly because of Unicode). Here is the top 10 from the very latest drm-intel-nightly. >./ drivers/gpu/drm/i915/ total lines counted: 67417 (compare with git […]

Power regression? Moi?

I may have caused a “massive” 3.5 power regression: It’s not the first time I’ve been mentioned for breaking something, however it’s definitely the most sensational accusation thrown at me. On a serious note, it is unlikely that semaphores on their own would cause such a large power regression. In theory, semaphores should be […]