kernel/linux.git/tools/power/cpupower/utils, branch v4.4.69

cpupower: Fix turbo frequency reporting for pre-Sandy Bridge cores

2017-05-14T11:32:54+00:00

commit 4cca0457686e4ee1677d69469e4ddfd94d389a80 upstream. The switch that conditionally sets CPUPOWER_CAP_HAS_TURBO_RATIO and CPUPOWER_CAP_IS_SNB flags is missing a break, so all cores get both flags set and an assumed base clock of 100 MHz for turbo values. Reported-by: GSR Tested-by: GSR References: https://bugs.debian.org/859978 Fixes: 8fb2e440b223 (cpupower: Show Intel turbo ratio support via ...) Signed-off-by: Ben Hutchings Signed-off-by: Rafael J. Wysocki Signed-off-by: Greg Kroah-Hartman

Creating a common structure initialization pattern for struct option

2015-11-02T01:28:59+00:00

This patch tries to creates a common structure initialization within the cpupower tool. Previously the ``struct option`` was initialized using `designated initializer` technique which was not needed. There were conflicting initialization methods seen with bench/main.c & others. Signed-off-by: Sriram Raghunathan Signed-off-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: Enable disabled Cstates if they are below max latency

2015-11-02T01:28:59+00:00

cpupower idle-set -D currently only disables all C-states that have a higher latency than the specified . But if deep sleep states were already disabled and have a lower latency, they should get enabled again. For example: This call: cpupower idle-set -D 30 disables all C-states with a higher or equal latency than 30. If one then calls: cpupower idle-set -D 100 C-states with a latency between 30-99 will get enabled again with this patch now. It is ensured that only C-states with a latency of 100 and higher are disabled. Signed-off-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: Remove debug message when using cpupower idle-set -D switch

2015-11-02T01:28:59+00:00

Signed-off-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: cpupower monitor reports uninitialized values for offline cpus

2015-11-02T01:28:59+00:00

[root@hp-dl980g7-02 linux]# cpupower monitor ... 5472| 0| 1|******|******|******|******|| 0.00| 0.00| 0.00| 0.00| 0.00 *is offline 10567| 0| 159|******|******|******|******|| 0.00| 0.00| 0.00| 0.00| 0.00 *is offline 1661206560|859272560| 150|******|******|******|******|| 0.00| 0.00| 0.00| 0.00| 0.00 *is offline 1661206560|943093104| 140|******|******|******|******|| 0.00| 0.00| 0.00| 0.00| 0.00 *is offline because of this cpupower also holds the incorrect value for the number of physical packages in the machine Changed cpupower to initialize the values of an offline cpu's socket and core to -1, warn the user that one or more cpus is/are offline and not print statistics for offline cpus. This fix hides offlined cores where topology cannot be accessed. With a recent kernel patch suggested from Prarit Bhargava it may be possible that soft offlined cores' topology can still be parsed. This patch would then show which cores in which package/socket are offline, when sane toplogoy information is available. Signed-off-by: Jacob Tanenbaum Signed-off-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

tools: cpupower: Fix error when running cpupower monitor

2015-08-28T23:38:16+00:00

get_cpu_topology() tries to get topology info from all cpus by reading files in the topology sysfs dir. If a cpu is offlined, since it doesn't have topology dir, this function fails and returns -1. This causes functions relying on get_cpu_topology() to fail. For example- $ cpupower monitor Cannot read number of available processors Fix this by skipping fetching topology info for offline cpus. Signed-off-by: Shreyas B. Prabhu Reported-by: Pavaman Subramaniyam Acked-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: Do not change the frequency of offline cpu

2015-07-23T18:40:51+00:00

Check if the cpu is online before changing the frequency/governor of the cpu. Reported-by: Pavaman Subramaniyam Signed-off-by: Shilpasri G Bhat Reviewed-by: Gautham R. Shenoy Acked-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: mperf monitor: fix output in MAX_FREQ_SYSFS mode

2015-05-30T00:21:31+00:00

There is clearly wrong output when mperf monitor runs in MAX_FREQ_SYSFS mode: average frequency shows in kHz unit (despite the intended output to be in MHz), and percentages for C state information are all wrong (including high/negative values shown). The problem is that the max_frequency read on initialization isn't used where it should have been used on mperf_get_count_percent (to estimate the number of ticks in the given time period), and the value we read from sysfs is in kHz, so we must divide it to get the MHz value to use in current calculations. While at it, also I fixed another small issues in the debug output of max_frequency value in mperf_get_count_freq. Signed-off-by: Herton R. Krzesinski Acked-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

cpupower: fix breakage from libpci API change

2015-04-15T19:54:26+00:00

libpci 3.3.0 introduced an additional member in the pci_filter struct which needs to be initialized to -1 to get the same behavior as before the API change. The libpci internal helpers got updated accordingly, but as the cpupower pci helpers initialized the struct themselves the behavior changed. Use the libpci helper pci_filter_init() to fix this and guard against similar breakages in the future. This fixes probing of the AMD fam12h/14h cpuidle monitor on systems with libpci >= 3.3.0. Signed-off-by: Lucas Stach Acked-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki

tools / cpupower: Fix no idle state information return value

2014-12-19T22:01:12+00:00

sysfs_get_idlestate_count() returns an unsigned int. Returning -ENODEV is not the right thing to do here, and in any case is handled the same way as if there are no states found. Signed-off-by: Prarit Bhargava Acked-by: Thomas Renninger Signed-off-by: Rafael J. Wysocki