When scheduling a group of events, there are constraint checks done to
make sure all events can go in a group. Example, one of the criteria is
that events in a group cannot use the same PMC. But platform specific
PMU supports alternative event for some of the event codes. During
perf_event_open(), if any event group doesn't match constraint check
criteria, further lookup is done to find alternative event.
By current design, the array of alternatives events in PMU code is
expected to be sorted by column 0. This is because in
find_alternative() the return criteria is based on event code
comparison. ie. "event < ev_alt[i][0])". This optimisation is there
since find_alternative() can be called multiple times. In power10 PMU
code, the alternative event array is not sorted properly and hence there
is breakage in finding alternative event.
To work with existing logic, fix the alternative event array to be
sorted by column 0 for power10-pmu.c
Results:
In case where an alternative event is not chosen when we could, events
will be multiplexed. ie, time sliced where it could actually run
concurrently.
Example, in power10 PM_INST_CMPL_ALT(0x00002) has alternative event,
PM_INST_CMPL(0x500fa). Without the fix, if a group of events with PMC1
to PMC4 is used along with PM_INST_CMPL_ALT, it will be time sliced
since all programmable PMC's are consumed already. But with the fix,
when it picks alternative event on PMC5, all events will run
concurrently.
Before:
# perf stat -e r00002,r100fc,r200fa,r300fc,r400fc
Performance counter stats for 'system wide':
328668935 r00002 (79.94%)
56501024 r100fc (79.95%)
49564238 r200fa (79.95%)
376 r300fc (80.19%)
660 r400fc (79.97%)
4.039150522 seconds time elapsed
With the fix, since alternative event is chosen to run on PMC6, events
will be run concurrently.
After:
# perf stat -e r00002,r100fc,r200fa,r300fc,r400fc
Performance counter stats for 'system wide':
23596607 r00002
4907738 r100fc
2283608 r200fa
135 r300fc
248 r400fc
1.664671390 seconds time elapsed
Fixes: a64e697cef ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Reviewed-by: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20220419114828.89843-2-atrajeev@linux.vnet.ibm.com
Some functions defined in 'arch/powerpc/perf' are deserving of an
`__init` macro attribute. These functions are only called by other
initialization functions and therefore should inherit the attribute.
Also, change function declarations in header files to include `__init`.
Signed-off-by: Nick Child <nick.child@ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211216220035.605465-5-nick.child@ibm.com
On power9 and earlier platforms, the default event used for cyles and
instructions is PM_CYC (0x0001e) and PM_INST_CMPL (0x00002)
respectively. These events use two programmable PMCs and by default will
count irrespective of the run latch state (idle state). But since they
use programmable PMCs, these events can lead to multiplexing with other
events, because there are only 4 programmable PMCs. Hence in power10,
performance monitoring unit (PMU) driver uses performance monitor
counter 5 (PMC5) and performance monitor counter6 (PMC6) for counting
instructions and cycles.
Currently on power10, the event used for cycles is PM_RUN_CYC (0x600F4)
and instructions uses PM_RUN_INST_CMPL (0x500fa). But counting of these
events in idle state is controlled by the CC56RUN bit setting in Monitor
Mode Control Register0 (MMCR0). If the CC56RUN bit is zero, PMC5/6 will
not count when CTRL[RUN] (run latch) is zero. This could lead to missing
some counts if a thread is in idle state during system wide profiling.
To fix it, set the CC56RUN bit in MMCR0 for power10, which makes PMC5
and PMC6 count instructions and cycles regardless of the run latch
state. Since this change make PMC5/6 count as PM_INST_CMPL/PM_CYC,
rename the event code 0x600f4 as PM_CYC instead of PM_RUN_CYC and event
code 0x500fa as PM_INST_CMPL instead of PM_RUN_INST_CMPL. The changes
are only for PMC5/6 event codes and will not affect the behaviour of
PM_RUN_CYC/PM_RUN_INST_CMPL if progammed in other PMC's.
Fixes: a64e697cef ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.cm>
Reviewed-by: Madhavan Srinivasan <maddy@linux.ibm.com>
[mpe: Tweak change log wording for style and consistency]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20211007075121.28497-1-atrajeev@linux.vnet.ibm.com
Add platform specific attr.config value checks. Patch
includes checks for both power9 and power10.
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210408074504.248211-2-maddy@linux.ibm.com
Thresholding, a performance monitoring unit feature, can be
used to identify marked instructions which take more than
expected cycles between start event and end event.
Threshold compare (thresh_cmp) bits are programmed in MMCRA
register. In Power9, thresh_cmp bits were part of the
event code. But in case of P10, thresh_cmp are not part of
event code due to inclusion of MMCR3 bits.
Patch here adds an option to use attr.config1 variable
to be used to pass thresh_cmp value to be programmed in
MMCRA register. A new ppmu flag called PPMU_HAS_ATTR_CONFIG1
has been added and this flag is used to notify the use of
attr.config1 variable.
Patch has extended the parameter list of 'compute_mmcr',
to include power_pmu's 'flags' element and parameter list of
get_constraint to include attr.config1 value. It also extend
parameter list of power_check_constraints inorder to pass
perf_event list.
As stated by commit ef0e3b650f ("powerpc/perf: Fix Threshold
Event Counter Multiplier width for P10"), constraint bits for
thresh_cmp is also needed to be increased to 11 bits, which is
handled as part of this patch. We added bit number 53 as part
of constraint bits of thresh_cmp for power10 to make it an
11 bit field.
Updated layout for p10:
/*
* Layout of constraint bits:
*
* 60 56 52 48 44 40 36 32
* | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - |
* [ fab_match ] [ thresh_cmp ] [ thresh_ctl ] [ ]
* | |
* [ thresh_cmp bits for p10] thresh_sel -*
*
* 28 24 20 16 12 8 4 0
* | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - | - - - - |
* [ ] | [ ] | [ sample ] [ ] [6] [5] [4] [3] [2] [1]
* | | | | |
* BHRB IFM -* | | |*radix_scope | Count of events for each PMC.
* EBB -* | | p1, p2, p3, p4, p5, p6.
* L1 I/D qualifier -* |
* nc - number of counters -*
*
* The PMC fields P1..P6, and NC, are adder fields. As we accumulate constraints
* we want the low bit of each field to be added to any existing value.
*
* Everything else is a value field.
*/
Result:
command#: cat /sys/devices/cpu/format/thresh_cmp
config1:0-17
ex. usage:
command#: perf record -I --weight -d -e
cpu/event=0x67340101EC,thresh_cmp=500/ ./ebizzy -S 2 -t 1 -s 4096
1826636 records/s
real 2.00 s
user 2.00 s
sys 0.00 s
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.038 MB perf.data (61 samples) ]
Signed-off-by: Kajol Jain <kjain@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20210209095234.837356-1-kjain@linux.ibm.com
Export l2l3 events (PM_L2_ST_MISS and PM_L2_ST) and LLC-prefetches
(PM_L3_PF_MISS_L3) via sysfs, and also add these to list of
cache_events.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-7-git-send-email-atrajeev@linux.vnet.ibm.com
Fix the event code for events: branch-instructions (to PM_BR_FIN),
branch-misses (to PM_MPRED_BR_FIN) and cache-misses (to
PM_LD_DEMAND_MISS_L1_FIN) for power10 PMU. Update the
list of generic events with this modified event code.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-6-git-send-email-atrajeev@linux.vnet.ibm.com
There are event code updates for some of the generic events
and cache events for power10. Inorder to maintain the current
event codes work with DD1 also, create a new array of generic_events,
cache_events and pmu_attr_groups with suffix _dd1, example,
power10_events_attr_dd1. So that further updates to event codes
can be made in the original list, ie, power10_events_attr. Update the
power10 pmu init code to pick the dd1 list while registering
the power PMU, based on the pvr (Processor Version Register) value.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-5-git-send-email-atrajeev@linux.vnet.ibm.com
power10 uses bit 9 of the raw event code as RADIX_SCOPE_QUAL.
This bit is used for enabling the radix process events.
Patch fixes the PMU counter support functions to program bit
18 of MMCR1 ( Monitor Mode Control Register1 ) with the
RADIX_SCOPE_QUAL bit value. Since this field is not per-pmc,
add this to PMU group constraints to make sure events in a
group will have same bit value for this field. Use bit 21 as
constraint bit field for radix_scope_qual. Patch also updates
the power10 raw event encoding layout information, format field
and constraints bit layout to include the radix_scope_qual bit.
Fixes: a64e697cef ("powerpc/perf: power10 Performance Monitoring support")
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1606409684-1589-2-git-send-email-atrajeev@linux.vnet.ibm.com
Add a new power PMU flag "PPMU_P10_DD1" which can be used to
conditionally add any code path for power10 DD1 processor version.
Also modify power10 PMU driver code to set this flag only for DD1,
based on the Processor Version Register (PVR) value.
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20201021085329.384535-1-maddy@linux.ibm.com
Sparse warns about all the init functions:
symbol init_ppc970_pmu was not declared. Should it be static?
symbol init_power5p_pmu was not declared. Should it be static?
symbol init_power5_pmu was not declared. Should it be static?
symbol init_power6_pmu was not declared. Should it be static?
symbol init_power7_pmu was not declared. Should it be static?
symbol init_power9_pmu was not declared. Should it be static?
symbol init_power8_pmu was not declared. Should it be static?
symbol init_generic_compat_pmu was not declared. Should it be static?
They're already declared in internal.h, so just make sure all the C
files include that directly or indirectly.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://lore.kernel.org/r/20200916115637.3100484-2-mpe@ellerman.id.au
Include capability flag PERF_PMU_CAP_EXTENDED_REGS for power10 and
expose MMCR3, SIER2, SIER3 registers as part of extended regs. Also
introduce PERF_REG_PMU_MASK_31 to define extended mask value at
runtime for power10.
Suggested-by: Ryan Grimm <grimm@linux.ibm.com>
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Tested-by: Nageswara R Sastry <nasastry@in.ibm.com>
Reviewed-by: Kajol Jain <kjain@linux.ibm.com>
Reviewed-and-tested-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1596794701-23530-3-git-send-email-atrajeev@linux.vnet.ibm.com
PowerISA v3.1 introduce filtering support for
PERF_SAMPLE_BRANCH_IND_CALL/COND. The patch adds BHRB filter
support for "ind_call" and "cond" in power10_bhrb_filter_map().
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1594996707-3727-11-git-send-email-atrajeev@linux.vnet.ibm.com
Base enablement patch to register performance monitoring hardware
support for power10. Patch introduce the raw event encoding format,
defines the supported list of events, config fields for the event
attributes and their corresponding bit values which are exported via
sysfs.
Patch also enhances the support function in isa207_common.c to include
power10 pmu hardware.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Signed-off-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/1594996707-3727-9-git-send-email-atrajeev@linux.vnet.ibm.com
Athira Rajeev