perf/x86/intel: Support adaptive PEBS v4
authorKan Liang <kan.liang@linux.intel.com>
Tue, 2 Apr 2019 19:45:02 +0000 (12:45 -0700)
committerIngo Molnar <mingo@kernel.org>
Tue, 16 Apr 2019 10:25:47 +0000 (12:25 +0200)
commitc22497f5838c237e3094a4dfb99d1c5de6353239
tree146038d425f37c7b0f32233aaf1e423f424a7e0b
parent477f00f9617009a9a3a9271885231573b728ca4f
perf/x86/intel: Support adaptive PEBS v4

Adaptive PEBS is a new way to report PEBS sampling information. Instead
of a fixed size record for all PEBS events it allows to configure the
PEBS record to only include the information needed. Events can then opt
in to use such an extended record, or stay with a basic record which
only contains the IP.

The major new feature is to support LBRs in PEBS record.
Besides normal LBR, this allows (much faster) large PEBS, while still
supporting callstacks through callstack LBR. So essentially a lot of
profiling can now be done without frequent interrupts, dropping the
overhead significantly.

The main requirement still is to use a period, and not use frequency
mode, because frequency mode requires reevaluating the frequency on each
overflow.

The floating point state (XMM) is also supported, which allows efficient
profiling of FP function arguments.

Introduce specific drain function to handle variable length records.
Use a new callback to parse the new record format, and also handle the
STATUS field now being at a different offset.

Add code to set up the configuration register. Since there is only a
single register, all events either get the full super set of all events,
or only the basic record.

Originally-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: acme@kernel.org
Cc: jolsa@kernel.org
Link: https://lkml.kernel.org/r/20190402194509.2832-6-kan.liang@linux.intel.com
[ Renamed GPRS => GP. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
arch/x86/events/intel/core.c
arch/x86/events/intel/ds.c
arch/x86/events/intel/lbr.c
arch/x86/events/perf_event.h
arch/x86/include/asm/msr-index.h
arch/x86/include/asm/perf_event.h