[PATCH v4 0/5] powerpc/perf: IMC trace-mode support
Anju T Sudhakar
anju at linux.vnet.ibm.com
Tue Apr 16 20:07:59 AEST 2019
On 4/16/19 3:14 PM, Anju T Sudhakar wrote:
> Hi,
>
> Kindly ignore this series, since patch 5/5 in this series doesn't
> incorporate the event-format change
>
> that I've done in v4 of this series.
>
>
> Apologies for the inconvenience. I will post the updated v5 soon.
>
>
s/v5/v4
> Thanks,
>
> Anju
>
> On 4/15/19 3:41 PM, Anju T Sudhakar wrote:
>> IMC (In-Memory collection counters) is a hardware monitoring facility
>> that collects large number of hardware performance events.
>> POWER9 support two modes for IMC which are the Accumulation mode and
>> Trace mode. In Accumulation mode, event counts are accumulated in system
>> Memory. Hypervisor then reads the posted counts periodically or when
>> requested. In IMC Trace mode, the 64 bit trace scom value is initialized
>> with the event information. The CPMC*SEL and CPMC_LOAD in the trace
>> scom, specifies
>> the event to be monitored and the sampling duration. On each overflow
>> in the
>> CPMC*SEL, hardware snapshots the program counter along with event counts
>> and writes into memory pointed by LDBAR. LDBAR has bits to indicate
>> whether
>> hardware is configured for accumulation or trace mode.
>> Currently the event monitored for trace-mode is fixed as cycle.
>>
>> Trace-IMC Implementation:
>> --------------------------
>> To enable trace-imc, we need to
>>
>> * Add trace node in the DTS file for power9, so that the new trace
>> node can
>> be discovered by the kernel.
>>
>> Information included in the DTS file are as follows, (a snippet from
>> the ima-catalog)
>>
>> TRACE_IMC: trace-events {
>> #address-cells = <0x1>;
>> #size-cells = <0x1>;
>> event at 10200000 {
>> event-name = "cycles" ;
>> reg = <0x10200000 0x8>;
>> desc = "Reference cycles" ;
>> };
>> };
>> trace at 0 {
>> compatible = "ibm,imc-counters";
>> events-prefix = "trace_";
>> reg = <0x0 0x8>;
>> events = < &TRACE_IMC >;
>> type = <0x2>;
>> size = <0x40000>;
>> };
>>
>> OP-BUILD changes needed to include the "trace node" is already pulled in
>> to the ima-catalog repo.
>>
>> ps://github.com/open-power/op-build/commit/d3e75dc26d1283d7d5eb444bff1ec9e40d5dfc07
>>
>>
>> * Enchance the opal_imc_counters_* calls to support this new trace mode
>> in imc. Add support to initialize the trace-mode scom.
>>
>> TRACE_IMC_SCOM bit representation:
>>
>> 0:1 : SAMPSEL
>> 2:33 : CPMC_LOAD
>> 34:40 : CPMC1SEL
>> 41:47 : CPMC2SEL
>> 48:50 : BUFFERSIZE
>> 51:63 : RESERVED
>>
>> CPMC_LOAD contains the sampling duration. SAMPSEL and CPMC*SEL
>> determines
>> the event to count. BUFFRSIZE indicates the memory range. On each
>> overflow,
>> hardware snapshots program counter along with event counts and update
>> the
>> memory and reloads the CMPC_LOAD value for the next sampling duration.
>> IMC hardware does not support exceptions, so it quietly wraps around if
>> memory buffer reaches the end.
>>
>> OPAL support for IMC trace mode is already upstream.
>>
>> * Set LDBAR spr to enable imc-trace mode.
>> LDBAR Layout:
>> 0 : Enable/Disable
>> 1 : 0 -> Accumulation Mode
>> 1 -> Trace Mode
>> 2:3 : Reserved
>> 4-6 : PB scope
>> 7 : Reserved
>> 8:50 : Counter Address
>> 51:63 : Reserved
>>
>> ----------------------
>>
>> PMI interrupt handling is avoided, since IMC trace mode snapshots the
>> program counter and update to the memory. And this also provide a way
>> for
>> the operating system to do instruction sampling in real time without
>> PMI(Performance Monitoring Interrupts) processing overhead.
>> Performance data using 'perf top' with and without trace-imc event:
>>
>> PMI interrupts count when `perf top` command is executed without
>> trace-imc event.
>>
>> # cat /proc/interrupts (a snippet from the output)
>> 9944 1072 804 804 1644 804 1306
>> 804 804 804 804 804 804 804
>> 804 804 1961 1602 804 804 1258
>> [-----------------------------------------------------------------]
>> 803 803 803 803 803 803 803
>> 803 803 803 803 804 804 804
>> 804 804 804 804 804 804 803
>> 803 803 803 803 803 1306 803
>> 803 Performance monitoring interrupts
>>
>>
>> `perf top` with trace-imc (executed right after 'perf top' without
>> trace-imc event):
>>
>> # perf top -e trace_imc/trace_cycles/
>> 12.50% [kernel] [k] arch_cpu_idle
>> 11.81% [kernel] [k] __next_timer_interrupt
>> 11.22% [kernel] [k] rcu_idle_enter
>> 10.25% [kernel] [k] find_next_bit
>> 7.91% [kernel] [k] do_idle
>> 7.69% [kernel] [k] rcu_dynticks_eqs_exit
>> 5.20% [kernel] [k] tick_nohz_idle_stop_tick
>> [-----------------------]
>>
>> # cat /proc/interrupts (a snippet from the output)
>>
>> 9944 1072 804 804 1644 804 1306
>> 804 804 804 804 804 804 804
>> 804 804 1961 1602 804 804 1258
>> [-----------------------------------------------------------------]
>> 803 803 803 803 803 803 803
>> 803 803 803 804 804 804 804
>> 804 804 804 804 804 804 803
>> 803 803 803 803 803 1306 803
>> 803 Performance monitoring interrupts
>>
>> The PMI interrupts count remains the same.
>>
>>
>> Changelog:
>> ----------
>> From v3 -> v4:
>>
>> * trace_imc_refc is introduced. So that even if, core-imc
>> is disabled, trace-imc can be used.
>>
>> * trace_imc_pmu_sched_task is removed and opal start/stop
>> is invoked in trace_imc_event_add/del function.
>>
>>
>> Suggestions/comments are welcome.
>>
>> Anju T Sudhakar (4):
>> powerpc/include: Add data structures and macros for IMC trace mode
>> powerpc/perf: Rearrange setting of ldbar for thread-imc
>> powerpc/perf: Trace imc events detection and cpuhotplug
>> powerpc/perf: Trace imc PMU functions
>>
>> Madhavan Srinivasan (1):
>> powerpc/perf: Add privileged access check for thread_imc
>>
>> arch/powerpc/include/asm/imc-pmu.h | 39 +++
>> arch/powerpc/include/asm/opal-api.h | 1 +
>> arch/powerpc/perf/imc-pmu.c | 318 +++++++++++++++++++++-
>> arch/powerpc/platforms/powernv/opal-imc.c | 3 +
>> include/linux/cpuhotplug.h | 1 +
>> 5 files changed, 351 insertions(+), 11 deletions(-)
>>
More information about the Linuxppc-dev
mailing list