diff options
author | Hubert Chrzaniuk <hubert.chrzaniuk@intel.com> | 2015-09-14 13:31:00 +0200 |
---|---|---|
committer | Len Brown <len.brown@intel.com> | 2015-09-26 00:50:54 -0400 |
commit | b2b34dfe4d9aa4c468fc363b3b666974783ed1f9 (patch) | |
tree | 1613d67821a25330cee5cdb348c41675f0cb177d /crypto/cts.c | |
parent | 756357b8e4b072fd5ee86421f794e071a348802b (diff) | |
download | linux-b2b34dfe4d9aa4c468fc363b3b666974783ed1f9.tar.gz linux-b2b34dfe4d9aa4c468fc363b3b666974783ed1f9.tar.bz2 linux-b2b34dfe4d9aa4c468fc363b3b666974783ed1f9.zip |
tools/power turbostat: KNL workaround for %Busy and Avg_MHz
KNL increments APERF and MPERF every 1024 clocks.
This is compliant with the architecture specification,
which requires that only the ratio of APERF/MPERF need be valid.
However, turbostat takes advantage of the fact that these
two MSRs increment every un-halted clock
at the actual and base frequency:
AVG_MHz = APERF_delta/measurement_interval
%Busy = MPERF_delta/TSC_delta
This quirk is needed for these calculations to also work on KNL,
which would otherwise show a value 1024x smaller than expected.
Signed-off-by: Hubert Chrzaniuk <hubert.chrzaniuk@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Diffstat (limited to 'crypto/cts.c')
0 files changed, 0 insertions, 0 deletions