Pull perf updates from Ingo Molnar:
"On the kernel side the main changes in this cycle were:
- Add Intel Kaby Lake CPU support (Srinivas Pandruvada)
- AMD uncore driver updates for fam17 (Janakarajan Natarajan)
- Intel/PT updates and core events optimizations and cleanups
(Alexander Shishkin)
- cgroups events fixes (David Carrillo-Cisneros)
- kprobes improvements (Masami Hiramatsu)
- ... plus misc fixes and updates.
On the tooling side the main changes were:
- Support clang build in tools/{perf,lib/{bpf,traceevent,api}} with
CC=clang, to, for instance, take advantage of better warnings
(Arnaldo Carvalho de Melo):
- Introduce the 'delta-abs' 'perf diff' compute method, that orders
the histogram entries by the absolute value of the percentage delta
for a function in two perf.data files, i.e. the functions that
changed the most (increase or decrease in samples) comes first
(Namhyung Kim)
- Add support for parsing Intel uncore vendor event files and add
uncore vendor events for the Intel server processors (Haswell,
Broadwell, IvyBridge), Xeon Phi (Knights Landing) and Broadwell DE
(Andi Kleen)
- Introduce 'perf ftrace' a perf front end to the kernel's ftrace
function and function_graph tracer, defaulting to the
"function_graph" tracer, more work will be done in reviving this
effort, forward porting it from its initial patch submission
(Namhyung Kim)
- Add 'e' and 'c' hotkeys to expand/collapse call chains for a single
hist entry in the 'perf report' and 'perf top' TUI (Jiri Olsa)
- Account thread wait time (off CPU time) separately: sleep, iowait
and preempt, based on the prev_state of the last event, show the
breakdown when using "perf sched timehist --state" (Namhyumg Kim)
- Add more triggers to switch the output file (perf.data.TIMESTAMP).
Now, in addition to switching to a different output file when
receiving a SIGUSR2, one can also specify file size and time based
triggers:
perf record -a --switch-output=signal
is equivalent to what we had before:
perf record -a --switch-output
While we can also ask for the file to be "sliced" by size, taking
into account that that will happen only when we get woken up by the
kernel, i.e. one has to take into account the --mmap-pages (the
size of the perf mmap ring buffer):
perf record -a --switch-output=2G
will break the perf.data output into multiple files limited to 2GB
of samples, right when generating the output.
For time based samples, alert() will be used, so to have 1 minute
limited perf.data output files:
perf record -a --switch-output=1m
(Jiri Olsa)
- Improve 'perf trace' (Arnaldo Carvalho de Melo)
- 'perf kallsyms' toy tool to look for extended symbol information on
the running kernel and demonstrate the machine/thread/symbol APIs
for use in other tools, such as 'perf probe' (Arnaldo Carvalho de
Melo)
- ... plus tons of other changes, see the shortlog and Git log for
details"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (131 commits)
perf tools: Add missing parse_events_error() prototype
perf pmu: Fix check for unset alias->unit array
perf tools: Be consistent on the type of map->symbols[] interator
perf intel pt decoder: clang has no -Wno-override-init
perf evsel: Do not put a variable sized type not at the end of a struct
perf probe: Avoid accessing uninitialized 'map' variable
perf tools: Do not put a variable sized type not at the end of a struct
perf record: Do not put a variable sized type not at the end of a struct
perf tests: Synthesize struct instead of using field after variable sized type
perf bench numa: Make sure dprintf() is not defined
Revert "perf bench futex: Sanitize numeric parameters"
tools lib subcmd: Make it an error to pass a signed value to OPTION_UINTEGER
tools: Set the maximum optimization level according to the compiler being used
tools: Suppress request for warning options not existent in clang
samples/bpf: Reset global variables
samples/bpf: Ignore already processed ELF sections
samples/bpf: Add missing header
perf symbols: dso->name is an array, no need to check it against NULL
perf tests record: No need to test an array against NULL
perf symbols: No need to check if sym->name is NULL
...