| commit | d992c3771baa6d7578bf112b8dbea1326e461f01 | |
| author | Matthew Dillon <dillon@apollo.backplane.com> | |
| Mon, 24 Sep 2012 20:32:11 +0000 (24 13:32 -0700) | ||
| committer | Matthew Dillon <dillon@apollo.backplane.com> | |
| Mon, 24 Sep 2012 20:32:11 +0000 (24 13:32 -0700) | ||
| tree | 3fc568f4b968fd19dca18b6ac42454b74c2d0aa1 | treesnapshot (tar.gz zip) |
| parent | d9ae8717dd8b42d5e415cff4a15751d9d7608551 | commitdiff |
kernel - usched_dfly revamp (4), improve tail
* Improve tail performance (many more cpu-bound processes than available
cpus).
* Experiment with removing the LWKT priority adjustments for kernel vs user.
Instead give LWKT a hint about the user scheduler when scheduling a thread.
LWKT's round-robin is left unhinted to hopefully round-robin starved LWKTs
running in kernel mode.
* Implement a better calculation for the per-thread uload than the priority.
Instead, use estcpu.
* Adjust default weigntings for new uload calculation scale.
* Improve tail performance (many more cpu-bound processes than available
cpus).
* Experiment with removing the LWKT priority adjustments for kernel vs user.
Instead give LWKT a hint about the user scheduler when scheduling a thread.
LWKT's round-robin is left unhinted to hopefully round-robin starved LWKTs
running in kernel mode.
* Implement a better calculation for the per-thread uload than the priority.
Instead, use estcpu.
* Adjust default weigntings for new uload calculation scale.