]> CyberLeo.Net >> Repos - FreeBSD/stable/8.git/commit
projects / FreeBSD / stable / 8.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: e8f4641) | patch
MFC r209328,r209330:
authormav <mav@ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f>
Sun, 11 Jul 2010 11:58:46 +0000 (11:58 +0000)
committermav <mav@ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f>
Sun, 11 Jul 2010 11:58:46 +0000 (11:58 +0000)
commitfee16f56ae2bdda99e56e2a37c6b9a001827b30a
tree43511294eca58a1e8655cdcb05abb642e5de82adtree | snapshot
parente8f4641aa9d1e6ebf5e72572d2b2bde6b2d01e64commit | diff
MFC r209328,r209330:
While we indeed can't precisely measure time spent in C1, we can consider
measured interval as upper bound. It should be more precise then just
assuming 1/(2*hz). For idle CPU it should be quite precise, for busy - not
worse then before.

git-svn-id: svn://svn.freebsd.org/base/stable/8@209897 ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f
sys/dev/acpica/acpi_cpu.c diff | blob | history
8-STABLE