Pystats results

benchmark: unpickle
fork: python
ref: a19bb261a327e1008f219b62f6465941c981899b
commit hash: a19bb26
commit date: 2024-06-15T19:10:50+00:00

Execution counts

Execution counts for Tier 1 instructions.

The "miss ratio" column shows the percentage of times the instruction executed that it deoptimized. When this happens, the base unspecialized instruction is not counted.

Name	Count	Self	Cumulative
POP_TOP	53,420	34.4%	34.4%
ENTER_EXECUTOR	42,180	27.1%	61.5%
LOAD_FAST	26,240	16.9%	78.4%
PUSH_NULL	13,040	8.4%	86.8%
CALL_BUILTIN_FAST_WITH_KEYWORDS	12,180	7.8%	94.7%
STORE_FAST	1,580	1.0%	95.7%
CALL	1,080	0.7%	96.4%
FOR_ITER_TUPLE	700	0.5%	96.8%
JUMP_BACKWARD	680	0.4%	97.3%
GET_ITER	480	0.3%	97.6%
LOAD_ATTR_MODULE	420	0.3%	97.8%
LOAD_ATTR	400	0.3%	98.1%
FOR_ITER_RANGE	380	0.2%	98.3%
LOAD_GLOBAL_MODULE	360	0.2%	98.6%
LOAD_GLOBAL	280	0.2%	98.8%
LOAD_DEREF	240	0.2%	98.9%
CALL_NON_PY_GENERAL	180	0.1%	99.0%
LOAD_ATTR_WITH_HINT	180	0.1%	99.1%
RETURN_VALUE	160	0.1%	99.2%
CALL_FUNCTION_EX	160	0.1%	99.3%
RESUME_CHECK	120	0.1%	99.4%
NOP	80	0.1%	99.5%
BUILD_LIST	80	0.1%	99.5%
BUILD_TUPLE	80	0.1%	99.6%
CALL_INTRINSIC_1	80	0.1%	99.6%
COPY_FREE_VARS	80	0.1%	99.7%
FOR_ITER	80	0.1%	99.7%
LIST_EXTEND	80	0.1%	99.8%
LOAD_FAST_LOAD_FAST	80	0.1%	99.8%
BINARY_OP_SUBTRACT_FLOAT	60	0.0%	99.9%
CALL_BUILTIN_CLASS	60	0.0%	99.9%
LOAD_GLOBAL_BUILTIN	60	0.0%	99.9%
BINARY_OP	40	0.0%	100.0%
RESUME	40	0.0%	100.0%

Pair counts

Pair counts for top 100 opcode pairs

Pairs of specialized operations that deoptimize and are then followed by the corresponding unspecialized instruction are not counted as pairs.

Pair	Count	Self	Cumulative
ENTER_EXECUTOR POP_TOP	40,920	26.3%	26.3%
POP_TOP ENTER_EXECUTOR	40,900	26.3%	52.7%
PUSH_NULL LOAD_FAST	12,560	8.1%	60.7%
LOAD_FAST PUSH_NULL	12,400	8.0%	68.7%
CALL_BUILTIN_FAST_WITH_KEYWORDS POP_TOP	12,000	7.7%	76.5%
POP_TOP LOAD_FAST	11,780	7.6%	84.0%
LOAD_FAST CALL_BUILTIN_FAST_WITH_KEYWORDS	11,600	7.5%	91.5%
STORE_FAST LOAD_FAST	1,340	0.9%	92.4%
ENTER_EXECUTOR ENTER_EXECUTOR	1,260	0.8%	93.2%
LOAD_FAST CALL	840	0.5%	93.7%
FOR_ITER_TUPLE STORE_FAST	600	0.4%	94.1%
POP_TOP JUMP_BACKWARD	580	0.4%	94.5%
LOAD_FAST GET_ITER	480	0.3%	94.8%
CALL CALL_BUILTIN_FAST_WITH_KEYWORDS	460	0.3%	95.1%
CALL POP_TOP	420	0.3%	95.4%
GET_ITER FOR_ITER_TUPLE	380	0.2%	95.6%
FOR_ITER_RANGE STORE_FAST	380	0.2%	95.8%
LOAD_ATTR_MODULE PUSH_NULL	360	0.2%	96.1%
JUMP_BACKWARD FOR_ITER_RANGE	300	0.2%	96.3%
JUMP_BACKWARD FOR_ITER_TUPLE	300	0.2%	96.5%
LOAD_FAST LOAD_ATTR	280	0.2%	96.6%
CALL_BUILTIN_FAST_WITH_KEYWORDS STORE_FAST	180	0.1%	96.8%
LOAD_GLOBAL_MODULE LOAD_FAST	180	0.1%	96.9%
LOAD_DEREF PUSH_NULL	160	0.1%	97.0%
LOAD_FAST LOAD_ATTR_MODULE	160	0.1%	97.1%
LOAD_ATTR LOAD_ATTR_MODULE	140	0.1%	97.2%
PUSH_NULL CALL	120	0.1%	97.2%
PUSH_NULL LOAD_GLOBAL	120	0.1%	97.3%
PUSH_NULL CALL_NON_PY_GENERAL	120	0.1%	97.4%
PUSH_NULL LOAD_GLOBAL_MODULE	120	0.1%	97.5%
LOAD_ATTR PUSH_NULL	120	0.1%	97.6%
LOAD_FAST LOAD_ATTR_WITH_HINT	120	0.1%	97.6%
LOAD_GLOBAL LOAD_GLOBAL_MODULE	120	0.1%	97.7%
LOAD_ATTR_WITH_HINT CALL_BUILTIN_FAST_WITH_KEYWORDS	120	0.1%	97.8%
LOAD_GLOBAL_MODULE LOAD_ATTR_MODULE	120	0.1%	97.9%
CALL STORE_FAST	100	0.1%	97.9%
FOR_ITER_TUPLE JUMP_BACKWARD	100	0.1%	98.0%
NOP LOAD_DEREF	80	0.1%	98.0%
POP_TOP NOP	80	0.1%	98.1%
RETURN_VALUE RETURN_VALUE	80	0.1%	98.1%
BUILD_LIST LOAD_DEREF	80	0.1%	98.2%
BUILD_TUPLE STORE_FAST	80	0.1%	98.2%
CALL_FUNCTION_EX COPY_FREE_VARS	80	0.1%	98.3%
CALL_INTRINSIC_1 CALL_FUNCTION_EX	80	0.1%	98.4%
LIST_EXTEND CALL_INTRINSIC_1	80	0.1%	98.4%
LOAD_DEREF LIST_EXTEND	80	0.1%	98.5%
LOAD_FAST BUILD_LIST	80	0.1%	98.5%
LOAD_FAST BUILD_TUPLE	80	0.1%	98.6%
LOAD_FAST CALL_FUNCTION_EX	80	0.1%	98.6%
LOAD_FAST_LOAD_FAST LOAD_FAST	80	0.1%	98.7%
LOAD_GLOBAL LOAD_FAST	80	0.1%	98.7%
STORE_FAST LOAD_FAST_LOAD_FAST	80	0.1%	98.8%
STORE_FAST LOAD_GLOBAL	80	0.1%	98.8%
GET_ITER FOR_ITER_RANGE	60	0.0%	98.9%
CALL CALL_NON_PY_GENERAL	60	0.0%	98.9%
CALL_FUNCTION_EX RESUME_CHECK	60	0.0%	98.9%
COPY_FREE_VARS RESUME_CHECK	60	0.0%	99.0%
LOAD_ATTR CALL	60	0.0%	99.0%
LOAD_ATTR LOAD_ATTR_WITH_HINT	60	0.0%	99.0%
LOAD_GLOBAL LOAD_ATTR	60	0.0%	99.1%
BINARY_OP_SUBTRACT_FLOAT RETURN_VALUE	60	0.0%	99.1%
CALL_BUILTIN_CLASS STORE_FAST	60	0.0%	99.2%
CALL_NON_PY_GENERAL POP_TOP	60	0.0%	99.2%
CALL_NON_PY_GENERAL LOAD_FAST	60	0.0%	99.2%
CALL_NON_PY_GENERAL STORE_FAST	60	0.0%	99.3%
LOAD_ATTR_MODULE STORE_FAST	60	0.0%	99.3%
LOAD_ATTR_WITH_HINT CALL	60	0.0%	99.4%
LOAD_GLOBAL_BUILTIN LOAD_FAST	60	0.0%	99.4%
LOAD_GLOBAL_MODULE LOAD_ATTR	60	0.0%	99.4%
RESUME_CHECK LOAD_DEREF	60	0.0%	99.5%
RESUME_CHECK LOAD_FAST	60	0.0%	99.5%
GET_ITER FOR_ITER	40	0.0%	99.5%
POP_TOP LOAD_GLOBAL	40	0.0%	99.6%
POP_TOP LOAD_GLOBAL_MODULE	40	0.0%	99.6%
RETURN_VALUE LOAD_GLOBAL	40	0.0%	99.6%
RETURN_VALUE LOAD_GLOBAL_MODULE	40	0.0%	99.6%
FOR_ITER STORE_FAST	40	0.0%	99.7%
JUMP_BACKWARD FOR_ITER	40	0.0%	99.7%
LOAD_FAST BINARY_OP	40	0.0%	99.7%
LOAD_FAST BINARY_OP_SUBTRACT_FLOAT	40	0.0%	99.7%
LOAD_FAST CALL_BUILTIN_CLASS	40	0.0%	99.8%
STORE_FAST LOAD_GLOBAL_BUILTIN	40	0.0%	99.8%
STORE_FAST LOAD_GLOBAL_MODULE	40	0.0%	99.8%
BINARY_OP RETURN_VALUE	20	0.0%	99.8%
BINARY_OP BINARY_OP_SUBTRACT_FLOAT	20	0.0%	99.8%
CALL LOAD_FAST	20	0.0%	99.9%
CALL CALL_BUILTIN_CLASS	20	0.0%	99.9%
CALL_FUNCTION_EX RESUME	20	0.0%	99.9%
COPY_FREE_VARS RESUME	20	0.0%	99.9%
FOR_ITER FOR_ITER_RANGE	20	0.0%	99.9%
FOR_ITER FOR_ITER_TUPLE	20	0.0%	99.9%
JUMP_BACKWARD POP_TOP	20	0.0%	99.9%
JUMP_BACKWARD ENTER_EXECUTOR	20	0.0%	99.9%
LOAD_ATTR STORE_FAST	20	0.0%	100.0%
LOAD_GLOBAL LOAD_GLOBAL_BUILTIN	20	0.0%	100.0%
RESUME LOAD_DEREF	20	0.0%	100.0%
RESUME LOAD_FAST	20	0.0%	100.0%

Predecessor/Successor Pairs

Top 5 predecessors and successors of each Tier 1 opcode.

This does not include the unspecialized instructions that occur after a specialized instruction deoptimizes.

GET_ITER

Successors and predecessors for GET_ITER

Predecessors	Count	Percentage
LOAD_FAST	480	100.0%

Successors	Count	Percentage
FOR_ITER_TUPLE	380	79.2%
FOR_ITER_RANGE	60	12.5%
FOR_ITER	40	8.3%

NOP

Successors and predecessors for NOP

Predecessors	Count	Percentage
POP_TOP	80	100.0%

Successors	Count	Percentage
LOAD_DEREF	80	100.0%

POP_TOP

Successors and predecessors for POP_TOP

Predecessors	Count	Percentage
ENTER_EXECUTOR	40,920	76.6%
CALL_BUILTIN_FAST_WITH_KEYWORDS	12,000	22.5%
CALL	420	0.8%
CALL_NON_PY_GENERAL	60	0.1%
JUMP_BACKWARD	20	0.0%

Successors	Count	Percentage
ENTER_EXECUTOR	40,900	76.6%
LOAD_FAST	11,780	22.1%
JUMP_BACKWARD	580	1.1%
NOP	80	0.1%
LOAD_GLOBAL	40	0.1%

PUSH_NULL

Successors and predecessors for PUSH_NULL

Predecessors	Count	Percentage
LOAD_FAST	12,400	95.1%
LOAD_ATTR_MODULE	360	2.8%
LOAD_DEREF	160	1.2%
LOAD_ATTR	120	0.9%

Successors	Count	Percentage
LOAD_FAST	12,560	96.3%
CALL	120	0.9%
LOAD_GLOBAL	120	0.9%
CALL_NON_PY_GENERAL	120	0.9%
LOAD_GLOBAL_MODULE	120	0.9%

RETURN_VALUE

Successors and predecessors for RETURN_VALUE

Predecessors	Count	Percentage
RETURN_VALUE	80	50.0%
BINARY_OP_SUBTRACT_FLOAT	60	37.5%
BINARY_OP	20	12.5%

Successors	Count	Percentage
RETURN_VALUE	80	50.0%
LOAD_GLOBAL	40	25.0%
LOAD_GLOBAL_MODULE	40	25.0%

BINARY_OP

Successors and predecessors for BINARY_OP

Predecessors	Count	Percentage
LOAD_FAST	40	100.0%

Successors	Count	Percentage
RETURN_VALUE	20	50.0%
BINARY_OP_SUBTRACT_FLOAT	20	50.0%

BUILD_LIST

Successors and predecessors for BUILD_LIST

Predecessors	Count	Percentage
LOAD_FAST	80	100.0%

Successors	Count	Percentage
LOAD_DEREF	80	100.0%

BUILD_TUPLE

Successors and predecessors for BUILD_TUPLE

Predecessors	Count	Percentage
LOAD_FAST	80	100.0%

Successors	Count	Percentage
STORE_FAST	80	100.0%

CALL

Successors and predecessors for CALL

Predecessors	Count	Percentage
LOAD_FAST	840	77.8%
PUSH_NULL	120	11.1%
LOAD_ATTR	60	5.6%
LOAD_ATTR_WITH_HINT	60	5.6%

Successors	Count	Percentage
CALL_BUILTIN_FAST_WITH_KEYWORDS	460	42.6%
POP_TOP	420	38.9%
STORE_FAST	100	9.3%
CALL_NON_PY_GENERAL	60	5.6%
LOAD_FAST	20	1.9%

CALL_FUNCTION_EX

Successors and predecessors for CALL_FUNCTION_EX

Predecessors	Count	Percentage
CALL_INTRINSIC_1	80	50.0%
LOAD_FAST	80	50.0%

Successors	Count	Percentage
COPY_FREE_VARS	80	50.0%
RESUME_CHECK	60	37.5%
RESUME	20	12.5%

CALL_INTRINSIC_1

Successors and predecessors for CALL_INTRINSIC_1

Predecessors	Count	Percentage
LIST_EXTEND	80	100.0%

Successors	Count	Percentage
CALL_FUNCTION_EX	80	100.0%

COPY_FREE_VARS

Successors and predecessors for COPY_FREE_VARS

Predecessors	Count	Percentage
CALL_FUNCTION_EX	80	100.0%

Successors	Count	Percentage
RESUME_CHECK	60	75.0%
RESUME	20	25.0%

ENTER_EXECUTOR

Successors and predecessors for ENTER_EXECUTOR

Predecessors	Count	Percentage
POP_TOP	40,900	97.0%
ENTER_EXECUTOR	1,260	3.0%
JUMP_BACKWARD	20	0.0%

Successors	Count	Percentage
POP_TOP	40,920	97.0%
ENTER_EXECUTOR	1,260	3.0%

FOR_ITER

Successors and predecessors for FOR_ITER

Predecessors	Count	Percentage
GET_ITER	40	50.0%
JUMP_BACKWARD	40	50.0%

Successors	Count	Percentage
STORE_FAST	40	50.0%
FOR_ITER_RANGE	20	25.0%
FOR_ITER_TUPLE	20	25.0%

JUMP_BACKWARD

Successors and predecessors for JUMP_BACKWARD

Predecessors	Count	Percentage
POP_TOP	580	85.3%
FOR_ITER_TUPLE	100	14.7%

Successors	Count	Percentage
FOR_ITER_RANGE	300	44.1%
FOR_ITER_TUPLE	300	44.1%
FOR_ITER	40	5.9%
POP_TOP	20	2.9%
ENTER_EXECUTOR	20	2.9%

LIST_EXTEND

Successors and predecessors for LIST_EXTEND

Predecessors	Count	Percentage
LOAD_DEREF	80	100.0%

Successors	Count	Percentage
CALL_INTRINSIC_1	80	100.0%

LOAD_ATTR

Successors and predecessors for LOAD_ATTR

Predecessors	Count	Percentage
LOAD_FAST	280	70.0%
LOAD_GLOBAL	60	15.0%
LOAD_GLOBAL_MODULE	60	15.0%

Successors	Count	Percentage
LOAD_ATTR_MODULE	140	35.0%
PUSH_NULL	120	30.0%
CALL	60	15.0%
LOAD_ATTR_WITH_HINT	60	15.0%
STORE_FAST	20	5.0%

LOAD_DEREF

Successors and predecessors for LOAD_DEREF

Predecessors	Count	Percentage
NOP	80	33.3%
BUILD_LIST	80	33.3%
RESUME_CHECK	60	25.0%
RESUME	20	8.3%

Successors	Count	Percentage
PUSH_NULL	160	66.7%
LIST_EXTEND	80	33.3%

LOAD_FAST

Successors and predecessors for LOAD_FAST

Predecessors	Count	Percentage
PUSH_NULL	12,560	47.9%
POP_TOP	11,780	44.9%
STORE_FAST	1,340	5.1%
LOAD_GLOBAL_MODULE	180	0.7%
LOAD_FAST_LOAD_FAST	80	0.3%

Successors	Count	Percentage
PUSH_NULL	12,400	47.3%
CALL_BUILTIN_FAST_WITH_KEYWORDS	11,600	44.2%
CALL	840	3.2%
GET_ITER	480	1.8%
LOAD_ATTR	280	1.1%

LOAD_FAST_LOAD_FAST

Successors and predecessors for LOAD_FAST_LOAD_FAST

Predecessors	Count	Percentage
STORE_FAST	80	100.0%

Successors	Count	Percentage
LOAD_FAST	80	100.0%

LOAD_GLOBAL

Successors and predecessors for LOAD_GLOBAL

Predecessors	Count	Percentage
PUSH_NULL	120	42.9%
STORE_FAST	80	28.6%
POP_TOP	40	14.3%
RETURN_VALUE	40	14.3%

Successors	Count	Percentage
LOAD_GLOBAL_MODULE	120	42.9%
LOAD_FAST	80	28.6%
LOAD_ATTR	60	21.4%
LOAD_GLOBAL_BUILTIN	20	7.1%

STORE_FAST

Successors and predecessors for STORE_FAST

Predecessors	Count	Percentage
FOR_ITER_TUPLE	600	38.0%
FOR_ITER_RANGE	380	24.1%
CALL_BUILTIN_FAST_WITH_KEYWORDS	180	11.4%
CALL	100	6.3%
BUILD_TUPLE	80	5.1%

Successors	Count	Percentage
LOAD_FAST	1,340	84.8%
LOAD_FAST_LOAD_FAST	80	5.1%
LOAD_GLOBAL	80	5.1%
LOAD_GLOBAL_BUILTIN	40	2.5%
LOAD_GLOBAL_MODULE	40	2.5%

RESUME

Successors and predecessors for RESUME

Predecessors	Count	Percentage
CALL_FUNCTION_EX	20	50.0%
COPY_FREE_VARS	20	50.0%

Successors	Count	Percentage
LOAD_DEREF	20	50.0%
LOAD_FAST	20	50.0%

BINARY_OP_SUBTRACT_FLOAT

Successors and predecessors for BINARY_OP_SUBTRACT_FLOAT

Predecessors	Count	Percentage
LOAD_FAST	40	66.7%
BINARY_OP	20	33.3%

Successors	Count	Percentage
RETURN_VALUE	60	100.0%

CALL_BUILTIN_CLASS

Successors and predecessors for CALL_BUILTIN_CLASS

Predecessors	Count	Percentage
LOAD_FAST	40	66.7%
CALL	20	33.3%

Successors	Count	Percentage
STORE_FAST	60	100.0%

CALL_BUILTIN_FAST_WITH_KEYWORDS

Successors and predecessors for CALL_BUILTIN_FAST_WITH_KEYWORDS

Predecessors	Count	Percentage
LOAD_FAST	11,600	95.2%
CALL	460	3.8%
LOAD_ATTR_WITH_HINT	120	1.0%

Successors	Count	Percentage
POP_TOP	12,000	98.5%
STORE_FAST	180	1.5%

CALL_NON_PY_GENERAL

Successors and predecessors for CALL_NON_PY_GENERAL

Predecessors	Count	Percentage
PUSH_NULL	120	66.7%
CALL	60	33.3%

Successors	Count	Percentage
POP_TOP	60	33.3%
LOAD_FAST	60	33.3%
STORE_FAST	60	33.3%

FOR_ITER_RANGE

Successors and predecessors for FOR_ITER_RANGE

Predecessors	Count	Percentage
JUMP_BACKWARD	300	78.9%
GET_ITER	60	15.8%
FOR_ITER	20	5.3%

Successors	Count	Percentage
STORE_FAST	380	100.0%

FOR_ITER_TUPLE

Successors and predecessors for FOR_ITER_TUPLE

Predecessors	Count	Percentage
GET_ITER	380	54.3%
JUMP_BACKWARD	300	42.9%
FOR_ITER	20	2.9%

Successors	Count	Percentage
STORE_FAST	600	85.7%
JUMP_BACKWARD	100	14.3%

LOAD_ATTR_MODULE

Successors and predecessors for LOAD_ATTR_MODULE

Predecessors	Count	Percentage
LOAD_FAST	160	38.1%
LOAD_ATTR	140	33.3%
LOAD_GLOBAL_MODULE	120	28.6%

Successors	Count	Percentage
PUSH_NULL	360	85.7%
STORE_FAST	60	14.3%

LOAD_ATTR_WITH_HINT

Successors and predecessors for LOAD_ATTR_WITH_HINT

Predecessors	Count	Percentage
LOAD_FAST	120	66.7%
LOAD_ATTR	60	33.3%

Successors	Count	Percentage
CALL_BUILTIN_FAST_WITH_KEYWORDS	120	66.7%
CALL	60	33.3%

LOAD_GLOBAL_BUILTIN

Successors and predecessors for LOAD_GLOBAL_BUILTIN

Predecessors	Count	Percentage
STORE_FAST	40	66.7%
LOAD_GLOBAL	20	33.3%

Successors	Count	Percentage
LOAD_FAST	60	100.0%

LOAD_GLOBAL_MODULE

Successors and predecessors for LOAD_GLOBAL_MODULE

Predecessors	Count	Percentage
PUSH_NULL	120	33.3%
LOAD_GLOBAL	120	33.3%
POP_TOP	40	11.1%
RETURN_VALUE	40	11.1%
STORE_FAST	40	11.1%

Successors	Count	Percentage
LOAD_FAST	180	50.0%
LOAD_ATTR_MODULE	120	33.3%
LOAD_ATTR	60	16.7%

RESUME_CHECK

Successors and predecessors for RESUME_CHECK

Predecessors	Count	Percentage
CALL_FUNCTION_EX	60	50.0%
COPY_FREE_VARS	60	50.0%

Successors	Count	Percentage
LOAD_DEREF	60	50.0%
LOAD_FAST	60	50.0%

Specialization stats

Specialization stats by family

BINARY_OP

specialization stats for BINARY_OP family

Kind	Count	Ratio
deferred ⓘ Lists the number of "deferred" (i.e. not specialized) instructions executed.	20	20.0%
hit ⓘ Specialized instructions that complete.	60	60.0%

Success	Count	Ratio
Success	20	100.0%
Failure	0	0.0%

CALL

specialization stats for CALL family

Kind	Count	Ratio
deferred ⓘ Lists the number of "deferred" (i.e. not specialized) instructions executed.	540	4.1%
hit ⓘ Specialized instructions that complete.	12,240	91.9%

Success	Count	Ratio
Success	540	100.0%
Failure	0	0.0%

FOR_ITER

specialization stats for FOR_ITER family

Kind	Count	Ratio
deferred ⓘ Lists the number of "deferred" (i.e. not specialized) instructions executed.	40	3.4%
hit ⓘ Specialized instructions that complete.	1,080	93.1%

Success	Count	Ratio
Success	40	100.0%
Failure	0	0.0%

LOAD_ATTR

specialization stats for LOAD_ATTR family

Kind	Count	Ratio
deferred ⓘ Lists the number of "deferred" (i.e. not specialized) instructions executed.	200	20.0%
hit ⓘ Specialized instructions that complete.	600	60.0%

Success	Count	Ratio
Success	200	100.0%
Failure	0	0.0%

LOAD_GLOBAL

specialization stats for LOAD_GLOBAL family

Kind	Count	Ratio
deferred ⓘ Lists the number of "deferred" (i.e. not specialized) instructions executed.	140	20.0%
hit ⓘ Specialized instructions that complete.	420	60.0%

Success	Count	Ratio
Success	140	100.0%
Failure	0	0.0%

Specialization effectiveness

specialization effectiveness

All entries are execution counts. Should add up to the total number of Tier 1 instructions executed.

Instructions	Count	Ratio
Basic ⓘ Instructions that are not and cannot be specialized, e.g. `LOAD_FAST`.	138,780	89.3%
Not specialized ⓘ Instructions that could be specialized but aren't, e.g. `LOAD_ATTR`, `BINARY_SLICE`.	1,880	1.2%
Specialized hits ⓘ Specialized instructions, e.g. `LOAD_ATTR_MODULE` that complete.	14,700	9.5%
Specialized misses ⓘ Specialized instructions, e.g. `LOAD_ATTR_MODULE` that deopt.	0	0.0%

Deferred by instruction

Breakdown of deferred (not specialized) instruction counts by family

Name	Count	Ratio
CALL	540	57.4%
LOAD_ATTR	200	21.3%
LOAD_GLOBAL	140	14.9%
FOR_ITER	40	4.3%
BINARY_OP	20	2.1%
BINARY_SLICE	0	0.0%
STORE_SLICE	0	0.0%
BINARY_SUBSCR	0	0.0%
GET_ITER	0	0.0%
NOP	0	0.0%

Misses by instruction

Breakdown of misses (specialized deopts) instruction counts by family

Call stats

Inlined calls and frame stats

This shows what fraction of calls to Python functions are inlined (i.e. not having a call at the C level) and for those that are not, where the call comes from. The various categories overlap.

Also includes the count of frame objects created.

	Count	Ratio
Calls to PyEval_EvalDefault	0	0.0%
Calls to Python functions inlined	160	100.0%
Calls via PyEval_EvalFrame (total)	0	0.0%
Calls via PyEval_EvalFrame (vector)	0	0.0%
Calls via PyEval_EvalFrame (generator)	0	0.0%
Calls via PyEval_EvalFrame (legacy)	0	0.0%
Calls via PyEval_EvalFrame (function vectorcall)	0	0.0%
Calls via PyEval_EvalFrame (build class)	0	0.0%
Calls via PyEval_EvalFrame (slot)	0	0.0%
Calls via PyEval_EvalFrame (function ex)	160	100.0%
Calls via PyEval_EvalFrame (api)	0	0.0%
Calls via PyEval_EvalFrame (method)	0	0.0%
Frame objects created	0	0.0%
Frames pushed	160	100.0%

Object stats

Allocations, frees and dict materializatons

Below, "allocations" means "allocations that are not from a freelist". Total allocations = "Allocations from freelist" + "Allocations".

"Inline values" is the number of values arrays inlined into objects.

The cache hit/miss numbers are for the MRO cache, split into dunder and other names.

	Count	Ratio
Allocations from freelist	14,746,400	7.0%
Frees to freelist	14,746,340
Allocations	195,033,580	93.0%
Allocations to 512 bytes	191,756,100	91.4%
Allocations to 4 kbytes	3,277,140	1.6%
Allocations over 4 kbytes	340	0.0%
Frees	202,401,546
Inline values	0
Interpreter increfs	5,163,440	1.1%
Interpreter decrefs	7,683,220	1.2%
Increfs	460,590,798	98.9%
Decrefs	641,630,647	98.8%
Materialize dict (on request)	0
Materialize dict (new key)	0
Materialize dict (too big)	0
Materialize dict (str subclass)	0
Method cache hits	3,961,085
Method cache misses	955,015
Method cache collisions	962,494
Method cache dunder hits	1,630,853
Method cache dunder misses	8,267

GC stats

GC collections and effectiveness

Collected/visits gives some measure of efficiency.

Generation	Collections	Objects collected	Object visits
0	0	0	0
1	0	0	0
2	0	0	0

Optimization (Tier 2) stats

statistics about the Tier 2 optimizer

	Count	Ratio
Optimization attempts ⓘ The number of times a potential trace is identified. Specifically, this occurs in the JUMP BACKWARD instruction when the counter reaches a threshold.	140
Traces created ⓘ The number of traces that were successfully created.	40	28.6%
Trace stack overflow ⓘ A trace is truncated because it would require more than 5 stack frames.	0	0.0%
Trace stack underflow ⓘ A potential trace is abandoned because it pops more frames than it pushes.	0	0.0%
Trace too long ⓘ A trace is truncated because it is longer than the instruction buffer.	0	0.0%
Trace too short ⓘ A potential trace is abandoced because it it too short.	100	71.4%
Inner loop found ⓘ A trace is truncated because it has an inner loop	120	85.7%
Recursive call ⓘ A trace is truncated because it has a recursive call.	0	0.0%
Low confidence ⓘ A trace is abandoned because the likelihood of the jump to top being taken is too low.	0	0.0%
Executors invalidated ⓘ The number of executors that were invalidated due to watched dictionary changes.	0	0.0%
Traces executed ⓘ The number of traces that were executed	123,740
Uops executed ⓘ The total number of uops (micro-operations) that were executed	20,907,140	16,896.0%

	Count	Ratio
Optimizer attempts ⓘ The number of times the trace optimizer (_Py_uop_analyze_and_optimize) was run.	40
Optimizer successes ⓘ The number of traces that were successfully optimized.	40	100.0%
Optimizer no memory ⓘ The number of optimizations that failed due to no memory.	0	0.0%
Remove globals builtins changed ⓘ The builtins changed during optimization	0	0.0%
Remove globals incorrect keys ⓘ The keys in the globals dictionary aren't what was expected	0	0.0%

Trace length histogram

trace length histogram

Range	Count	Ratio
<= 1	0	0.0%
<= 2	0	0.0%
<= 4	0	0.0%
<= 8	0	0.0%
<= 16	0	0.0%
<= 32	0	0.0%
<= 64	0	0.0%
<= 128	0	0.0%
<= 256	40	100.0%

Optimized trace length histogram

optimized trace length histogram

Range	Count	Ratio
<= 1	0	0.0%
<= 2	0	0.0%
<= 4	0	0.0%
<= 8	0	0.0%
<= 16	0	0.0%
<= 32	0	0.0%
<= 64	0	0.0%
<= 128	0	0.0%
<= 256	40	100.0%

Trace run length histogram

trace run length histogram

Range	Count	Ratio
<= 1	0	0.0%
<= 2	0	0.0%
<= 4	0	0.0%
<= 8	80	0.1%
<= 16	0	0.0%
<= 32	0	0.0%
<= 64	0	0.0%
<= 128	0	0.0%
<= 256	40,580	32.8%
<= 512	40,840	33.0%

Uop execution stats

uop execution stats

Name	Count	Self	Cumulative	Miss ratio
_SET_IP	2,485,760	11.9%	11.9%
_CHECK_VALIDITY	2,485,760	11.9%	23.8%
_LOAD_FAST	2,485,760	11.9%	35.7%
_POP_TOP	2,445,200	11.7%	47.4%
_PUSH_NULL	2,445,200	11.7%	59.1%
_CALL_BUILTIN_FAST_WITH_KEYWORDS	2,445,200	11.7%	70.8%
_CHECK_PERIODIC	2,445,200	11.7%	82.5%
_LOAD_FAST_7	2,445,200	11.7%	94.1%
_TIER2_RESUME_CHECK	163,200	0.8%	94.9%
_GUARD_NOT_EXHAUSTED_TUPLE	163,120	0.8%	95.7%	25.0%
_ITER_CHECK_TUPLE	163,120	0.8%	96.5%
_STORE_FAST	162,820	0.8%	97.3%
_ITER_NEXT_TUPLE	122,260	0.6%	97.9%
_JUMP_TO_TOP	81,700	0.4%	98.2%
_EXIT_TRACE	81,500	0.4%	98.6%
_START_EXECUTOR	81,500	0.4%	99.0%
_COLD_EXIT	42,240	0.2%	99.2%
_GUARD_NOT_EXHAUSTED_RANGE	40,640	0.2%	99.4%	0.2%
_ITER_CHECK_RANGE	40,640	0.2%	99.6%
_GET_ITER	40,560	0.2%	99.8%
_ITER_NEXT_RANGE	40,560	0.2%	100.0%

Pair counts

Pair counts for top 100 Non-JIT uop pairs

Pairs of specialized operations that deoptimize and are then followed by the corresponding unspecialized instruction are not counted as pairs.

Pair	Count	Self	Cumulative
_LOAD_FAST _SET_IP	2,485,760	11.9%	11.9%
_PUSH_NULL _LOAD_FAST	2,445,200	11.7%	23.6%
_SET_IP _CALL_BUILTIN_FAST_WITH_KEYWORDS	2,445,200	11.7%	35.3%
_CALL_BUILTIN_FAST_WITH_KEYWORDS _CHECK_PERIODIC	2,445,200	11.7%	47.0%
_CHECK_PERIODIC _CHECK_VALIDITY	2,445,200	11.7%	58.7%
_CHECK_VALIDITY _POP_TOP	2,445,200	11.7%	70.4%
_LOAD_FAST_7 _PUSH_NULL	2,445,200	11.7%	82.1%
_POP_TOP _LOAD_FAST_7	2,322,940	11.1%	93.2%
_ITER_CHECK_TUPLE _GUARD_NOT_EXHAUSTED_TUPLE	163,120	0.8%	94.0%
_TIER2_RESUME_CHECK _ITER_CHECK_TUPLE	122,560	0.6%	94.5%
_GUARD_NOT_EXHAUSTED_TUPLE _ITER_NEXT_TUPLE	122,260	0.6%	95.1%
_ITER_NEXT_TUPLE _STORE_FAST	122,260	0.6%	95.7%
_STORE_FAST _LOAD_FAST_7	122,260	0.6%	96.3%
_POP_TOP _JUMP_TO_TOP	81,700	0.4%	96.7%
_JUMP_TO_TOP _TIER2_RESUME_CHECK	81,700	0.4%	97.1%
_START_EXECUTOR _TIER2_RESUME_CHECK	81,500	0.4%	97.5%
_GUARD_NOT_EXHAUSTED_TUPLE _EXIT_TRACE	40,860	0.2%	97.7%
_ITER_CHECK_RANGE _GUARD_NOT_EXHAUSTED_RANGE	40,640	0.2%	97.9%
_TIER2_RESUME_CHECK _ITER_CHECK_RANGE	40,640	0.2%	98.0%
_GET_ITER _CHECK_VALIDITY	40,560	0.2%	98.2%
_POP_TOP _EXIT_TRACE	40,560	0.2%	98.4%
_SET_IP _GET_ITER	40,560	0.2%	98.6%
_CHECK_VALIDITY _ITER_CHECK_TUPLE	40,560	0.2%	98.8%
_GUARD_NOT_EXHAUSTED_RANGE _ITER_NEXT_RANGE	40,560	0.2%	99.0%
_ITER_NEXT_RANGE _STORE_FAST	40,560	0.2%	99.2%
_STORE_FAST _LOAD_FAST	40,560	0.2%	99.4%
_GUARD_NOT_EXHAUSTED_RANGE _EXIT_TRACE	80	0.0%	99.4%

Unsupported opcodes

unsupported opcodes

Optimizer errored out with opcode

Optimization stopped after encountering this opcode

Rare events

Counts of rare/unlikely events

Event	Count
set class ⓘ Setting an object's class, `obj.__class__ = ...`	0
set bases ⓘ Setting the bases of a class, `cls.__bases__ = ...`	0
set eval frame func ⓘ Setting the PEP 523 frame eval function `_PyInterpreterState_SetFrameEvalFunc()`	0
builtin dict ⓘ Modifying the builtins, `__builtins__.__dict__[var] = ...`	0
func modification ⓘ Modifying a function, e.g. `func.__defaults__ = ...`, etc.	0
watched dict modification ⓘ A watched dict has been modified	0
watched globals modification ⓘ A watched `globals()` dict has been modified	0

Meta stats

Meta statistics

	Count
Number of data files	20

Stats gathered on: 2024-06-16

Files

bm-20240615-azure-x86_64-python-a19bb261a327e1008f21-3.13.0b2+-a19bb26-pystats-unpickle.md

Latest commit

History

bm-20240615-azure-x86_64-python-a19bb261a327e1008f21-3.13.0b2+-a19bb26-pystats-unpickle.md

File metadata and controls

Pystats results

Execution counts

Pair counts

Predecessor/Successor Pairs

GET_ITER

NOP

POP_TOP

PUSH_NULL

RETURN_VALUE

BINARY_OP

BUILD_LIST

BUILD_TUPLE

CALL

CALL_FUNCTION_EX

CALL_INTRINSIC_1

COPY_FREE_VARS

ENTER_EXECUTOR

FOR_ITER

JUMP_BACKWARD

LIST_EXTEND

LOAD_ATTR

LOAD_DEREF

LOAD_FAST

LOAD_FAST_LOAD_FAST

LOAD_GLOBAL

STORE_FAST

RESUME

BINARY_OP_SUBTRACT_FLOAT

CALL_BUILTIN_CLASS

CALL_BUILTIN_FAST_WITH_KEYWORDS

CALL_NON_PY_GENERAL

FOR_ITER_RANGE

FOR_ITER_TUPLE

LOAD_ATTR_MODULE

LOAD_ATTR_WITH_HINT

LOAD_GLOBAL_BUILTIN

LOAD_GLOBAL_MODULE

RESUME_CHECK

Specialization stats

BINARY_OP

CALL

FOR_ITER

LOAD_ATTR

LOAD_GLOBAL

Specialization effectiveness

Deferred by instruction

Misses by instruction

Call stats

Object stats

GC stats

Optimization (Tier 2) stats

Trace length histogram

Optimized trace length histogram

Trace run length histogram

Uop execution stats

Pair counts

Unsupported opcodes

Optimizer errored out with opcode

Rare events

Meta stats