Skip to content

Latest commit

 

History

History
4210 lines (3838 loc) · 75.2 KB

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

File metadata and controls

4210 lines (3838 loc) · 75.2 KB

Pystats results

  • benchmark: unpickle_list
  • 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 Miss ratio
LOAD_FAST 8,640 33.3% 33.3%
PUSH_NULL 4,480 17.3% 50.6%
POP_TOP 4,160 16.0% 66.7%
CALL_BUILTIN_FAST_WITH_KEYWORDS 3,860 14.9% 81.6%
STORE_FAST 720 2.8% 84.3%
CALL 600 2.3% 86.7%
FOR_ITER_RANGE 380 1.5% 88.1%
JUMP_BACKWARD 340 1.3% 89.4%
LOAD_ATTR_MODULE 300 1.2% 90.6%
LOAD_ATTR 240 0.9% 91.5%
LOAD_DEREF 240 0.9% 92.4%
LOAD_GLOBAL_MODULE 240 0.9% 93.4%
LOAD_GLOBAL 200 0.8% 94.1%
CALL_NON_PY_GENERAL 180 0.7% 94.8%
RETURN_VALUE 160 0.6% 95.4%
CALL_FUNCTION_EX 160 0.6% 96.1%
RESUME_CHECK 120 0.5% 96.5%
GET_ITER 80 0.3% 96.8%
NOP 80 0.3% 97.1%
BUILD_LIST 80 0.3% 97.5%
CALL_INTRINSIC_1 80 0.3% 97.8%
COPY_FREE_VARS 80 0.3% 98.1%
LIST_EXTEND 80 0.3% 98.4%
ENTER_EXECUTOR 60 0.2% 98.6%
BINARY_OP_SUBTRACT_FLOAT 60 0.2% 98.8%
CALL_BUILTIN_CLASS 60 0.2% 99.1%
LOAD_ATTR_WITH_HINT 60 0.2% 99.3%
LOAD_GLOBAL_BUILTIN 60 0.2% 99.5%
BINARY_OP 40 0.2% 99.7%
FOR_ITER 40 0.2% 99.8%
RESUME 40 0.2% 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
PUSH_NULL LOAD_FAST 4,160 16.0% 16.0%
LOAD_FAST PUSH_NULL 4,000 15.4% 31.5%
CALL_BUILTIN_FAST_WITH_KEYWORDS POP_TOP 3,800 14.7% 46.1%
POP_TOP LOAD_FAST 3,600 13.9% 60.0%
LOAD_FAST CALL_BUILTIN_FAST_WITH_KEYWORDS 3,600 13.9% 73.9%
STORE_FAST LOAD_FAST 560 2.2% 76.1%
LOAD_FAST CALL 440 1.7% 77.8%
FOR_ITER_RANGE STORE_FAST 380 1.5% 79.2%
POP_TOP JUMP_BACKWARD 340 1.3% 80.6%
JUMP_BACKWARD FOR_ITER_RANGE 300 1.2% 81.7%
LOAD_ATTR_MODULE PUSH_NULL 240 0.9% 82.6%
CALL POP_TOP 220 0.8% 83.5%
CALL CALL_BUILTIN_FAST_WITH_KEYWORDS 220 0.8% 84.3%
LOAD_DEREF PUSH_NULL 160 0.6% 85.0%
PUSH_NULL CALL 120 0.5% 85.4%
PUSH_NULL CALL_NON_PY_GENERAL 120 0.5% 85.9%
LOAD_FAST LOAD_ATTR 120 0.5% 86.3%
LOAD_GLOBAL_MODULE LOAD_ATTR_MODULE 120 0.5% 86.8%
LOAD_ATTR LOAD_ATTR_MODULE 100 0.4% 87.2%
NOP LOAD_DEREF 80 0.3% 87.5%
POP_TOP NOP 80 0.3% 87.8%
RETURN_VALUE RETURN_VALUE 80 0.3% 88.1%
BUILD_LIST LOAD_DEREF 80 0.3% 88.4%
CALL_FUNCTION_EX COPY_FREE_VARS 80 0.3% 88.7%
CALL_INTRINSIC_1 CALL_FUNCTION_EX 80 0.3% 89.0%
LIST_EXTEND CALL_INTRINSIC_1 80 0.3% 89.4%
LOAD_ATTR PUSH_NULL 80 0.3% 89.7%
LOAD_DEREF LIST_EXTEND 80 0.3% 90.0%
LOAD_FAST GET_ITER 80 0.3% 90.3%
LOAD_FAST BUILD_LIST 80 0.3% 90.6%
LOAD_FAST CALL_FUNCTION_EX 80 0.3% 90.9%
LOAD_FAST LOAD_ATTR_MODULE 80 0.3% 91.2%
LOAD_GLOBAL LOAD_GLOBAL_MODULE 80 0.3% 91.5%
STORE_FAST LOAD_GLOBAL 80 0.3% 91.8%
GET_ITER FOR_ITER_RANGE 60 0.2% 92.1%
POP_TOP ENTER_EXECUTOR 60 0.2% 92.3%
CALL STORE_FAST 60 0.2% 92.5%
CALL CALL_NON_PY_GENERAL 60 0.2% 92.7%
CALL_FUNCTION_EX RESUME_CHECK 60 0.2% 93.0%
COPY_FREE_VARS RESUME_CHECK 60 0.2% 93.2%
ENTER_EXECUTOR POP_TOP 60 0.2% 93.4%
LOAD_GLOBAL LOAD_ATTR 60 0.2% 93.7%
BINARY_OP_SUBTRACT_FLOAT RETURN_VALUE 60 0.2% 93.9%
CALL_BUILTIN_CLASS STORE_FAST 60 0.2% 94.1%
CALL_BUILTIN_FAST_WITH_KEYWORDS STORE_FAST 60 0.2% 94.4%
CALL_NON_PY_GENERAL POP_TOP 60 0.2% 94.6%
CALL_NON_PY_GENERAL LOAD_FAST 60 0.2% 94.8%
CALL_NON_PY_GENERAL STORE_FAST 60 0.2% 95.1%
LOAD_ATTR_MODULE STORE_FAST 60 0.2% 95.3%
LOAD_GLOBAL_BUILTIN LOAD_FAST 60 0.2% 95.5%
LOAD_GLOBAL_MODULE LOAD_ATTR 60 0.2% 95.8%
LOAD_GLOBAL_MODULE LOAD_FAST 60 0.2% 96.0%
RESUME_CHECK LOAD_DEREF 60 0.2% 96.2%
RESUME_CHECK LOAD_FAST 60 0.2% 96.5%
POP_TOP LOAD_GLOBAL 40 0.2% 96.6%
POP_TOP LOAD_GLOBAL_MODULE 40 0.2% 96.8%
PUSH_NULL LOAD_GLOBAL 40 0.2% 96.9%
PUSH_NULL LOAD_GLOBAL_MODULE 40 0.2% 97.1%
RETURN_VALUE LOAD_GLOBAL 40 0.2% 97.2%
RETURN_VALUE LOAD_GLOBAL_MODULE 40 0.2% 97.4%
LOAD_FAST BINARY_OP 40 0.2% 97.5%
LOAD_FAST BINARY_OP_SUBTRACT_FLOAT 40 0.2% 97.7%
LOAD_FAST CALL_BUILTIN_CLASS 40 0.2% 97.8%
LOAD_FAST LOAD_ATTR_WITH_HINT 40 0.2% 98.0%
LOAD_GLOBAL LOAD_FAST 40 0.2% 98.1%
STORE_FAST LOAD_GLOBAL_BUILTIN 40 0.2% 98.3%
STORE_FAST LOAD_GLOBAL_MODULE 40 0.2% 98.5%
LOAD_ATTR_WITH_HINT CALL_BUILTIN_FAST_WITH_KEYWORDS 40 0.2% 98.6%
GET_ITER FOR_ITER 20 0.1% 98.7%
BINARY_OP RETURN_VALUE 20 0.1% 98.8%
BINARY_OP BINARY_OP_SUBTRACT_FLOAT 20 0.1% 98.8%
CALL LOAD_FAST 20 0.1% 98.9%
CALL CALL_BUILTIN_CLASS 20 0.1% 99.0%
CALL_FUNCTION_EX RESUME 20 0.1% 99.1%
COPY_FREE_VARS RESUME 20 0.1% 99.2%
FOR_ITER STORE_FAST 20 0.1% 99.2%
FOR_ITER FOR_ITER_RANGE 20 0.1% 99.3%
JUMP_BACKWARD POP_TOP 20 0.1% 99.4%
JUMP_BACKWARD FOR_ITER 20 0.1% 99.5%
LOAD_ATTR CALL 20 0.1% 99.5%
LOAD_ATTR STORE_FAST 20 0.1% 99.6%
LOAD_ATTR LOAD_ATTR_WITH_HINT 20 0.1% 99.7%
LOAD_GLOBAL LOAD_GLOBAL_BUILTIN 20 0.1% 99.8%
RESUME LOAD_DEREF 20 0.1% 99.8%
RESUME LOAD_FAST 20 0.1% 99.9%
LOAD_ATTR_WITH_HINT CALL 20 0.1% 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 80 100.0%
Successors Count Percentage
FOR_ITER_RANGE 60 75.0%
FOR_ITER 20 25.0%

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
CALL_BUILTIN_FAST_WITH_KEYWORDS 3,800 91.3%
CALL 220 5.3%
ENTER_EXECUTOR 60 1.4%
CALL_NON_PY_GENERAL 60 1.4%
JUMP_BACKWARD 20 0.5%
Successors Count Percentage
LOAD_FAST 3,600 86.5%
JUMP_BACKWARD 340 8.2%
NOP 80 1.9%
ENTER_EXECUTOR 60 1.4%
LOAD_GLOBAL 40 1.0%

PUSH_NULL

Successors and predecessors for PUSH_NULL
Predecessors Count Percentage
LOAD_FAST 4,000 89.3%
LOAD_ATTR_MODULE 240 5.4%
LOAD_DEREF 160 3.6%
LOAD_ATTR 80 1.8%
Successors Count Percentage
LOAD_FAST 4,160 92.9%
CALL 120 2.7%
CALL_NON_PY_GENERAL 120 2.7%
LOAD_GLOBAL 40 0.9%
LOAD_GLOBAL_MODULE 40 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%

CALL

Successors and predecessors for CALL
Predecessors Count Percentage
LOAD_FAST 440 73.3%
PUSH_NULL 120 20.0%
LOAD_ATTR 20 3.3%
LOAD_ATTR_WITH_HINT 20 3.3%
Successors Count Percentage
POP_TOP 220 36.7%
CALL_BUILTIN_FAST_WITH_KEYWORDS 220 36.7%
STORE_FAST 60 10.0%
CALL_NON_PY_GENERAL 60 10.0%
LOAD_FAST 20 3.3%

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 60 100.0%
Successors Count Percentage
POP_TOP 60 100.0%

FOR_ITER

Successors and predecessors for FOR_ITER
Predecessors Count Percentage
GET_ITER 20 50.0%
JUMP_BACKWARD 20 50.0%
Successors Count Percentage
STORE_FAST 20 50.0%
FOR_ITER_RANGE 20 50.0%

JUMP_BACKWARD

Successors and predecessors for JUMP_BACKWARD
Predecessors Count Percentage
POP_TOP 340 100.0%
Successors Count Percentage
FOR_ITER_RANGE 300 88.2%
POP_TOP 20 5.9%
FOR_ITER 20 5.9%

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 120 50.0%
LOAD_GLOBAL 60 25.0%
LOAD_GLOBAL_MODULE 60 25.0%
Successors Count Percentage
LOAD_ATTR_MODULE 100 41.7%
PUSH_NULL 80 33.3%
CALL 20 8.3%
STORE_FAST 20 8.3%
LOAD_ATTR_WITH_HINT 20 8.3%

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 4,160 48.1%
POP_TOP 3,600 41.7%
STORE_FAST 560 6.5%
CALL_NON_PY_GENERAL 60 0.7%
LOAD_GLOBAL_BUILTIN 60 0.7%
Successors Count Percentage
PUSH_NULL 4,000 46.3%
CALL_BUILTIN_FAST_WITH_KEYWORDS 3,600 41.7%
CALL 440 5.1%
LOAD_ATTR 120 1.4%
GET_ITER 80 0.9%

LOAD_GLOBAL

Successors and predecessors for LOAD_GLOBAL
Predecessors Count Percentage
STORE_FAST 80 40.0%
POP_TOP 40 20.0%
PUSH_NULL 40 20.0%
RETURN_VALUE 40 20.0%
Successors Count Percentage
LOAD_GLOBAL_MODULE 80 40.0%
LOAD_ATTR 60 30.0%
LOAD_FAST 40 20.0%
LOAD_GLOBAL_BUILTIN 20 10.0%

STORE_FAST

Successors and predecessors for STORE_FAST
Predecessors Count Percentage
FOR_ITER_RANGE 380 52.8%
CALL 60 8.3%
CALL_BUILTIN_CLASS 60 8.3%
CALL_BUILTIN_FAST_WITH_KEYWORDS 60 8.3%
CALL_NON_PY_GENERAL 60 8.3%
Successors Count Percentage
LOAD_FAST 560 77.8%
LOAD_GLOBAL 80 11.1%
LOAD_GLOBAL_BUILTIN 40 5.6%
LOAD_GLOBAL_MODULE 40 5.6%

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 3,600 93.3%
CALL 220 5.7%
LOAD_ATTR_WITH_HINT 40 1.0%
Successors Count Percentage
POP_TOP 3,800 98.4%
STORE_FAST 60 1.6%

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%

LOAD_ATTR_MODULE

Successors and predecessors for LOAD_ATTR_MODULE
Predecessors Count Percentage
LOAD_GLOBAL_MODULE 120 40.0%
LOAD_ATTR 100 33.3%
LOAD_FAST 80 26.7%
Successors Count Percentage
PUSH_NULL 240 80.0%
STORE_FAST 60 20.0%

LOAD_ATTR_WITH_HINT

Successors and predecessors for LOAD_ATTR_WITH_HINT
Predecessors Count Percentage
LOAD_FAST 40 66.7%
LOAD_ATTR 20 33.3%
Successors Count Percentage
CALL_BUILTIN_FAST_WITH_KEYWORDS 40 66.7%
CALL 20 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
LOAD_GLOBAL 80 33.3%
POP_TOP 40 16.7%
PUSH_NULL 40 16.7%
RETURN_VALUE 40 16.7%
STORE_FAST 40 16.7%
Successors Count Percentage
LOAD_ATTR_MODULE 120 50.0%
LOAD_ATTR 60 25.0%
LOAD_FAST 60 25.0%

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.

 300 6.6%
hit
ⓘ

Specialized instructions that complete.

 3,920 86.7%
Success Count Ratio
Success 300 100.0%
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.

 20 4.8%
hit
ⓘ

Specialized instructions that complete.

 380 90.5%
Success Count Ratio
Success 20 100.0%
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.

 120 20.0%
hit
ⓘ

Specialized instructions that complete.

 360 60.0%
Success Count Ratio
Success 120 100.0%
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.

 100 20.0%
hit
ⓘ

Specialized instructions that complete.

 300 60.0%
Success Count Ratio
Success 100 100.0%
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.

 19,480 75.2%
Not specialized
ⓘ

Instructions that could be specialized but aren't, e.g. LOAD_ATTR, BINARY_SLICE.

 1,120 4.3%
Specialized hits
ⓘ

Specialized instructions, e.g. LOAD_ATTR_MODULE that complete.

 5,320 20.5%
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 300 53.6%
LOAD_ATTR 120 21.4%
LOAD_GLOBAL 100 17.9%
BINARY_OP 20 3.6%
FOR_ITER 20 3.6%
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 101,580,740 61.9%
Frees to freelist 101,581,060
Allocations 62,403,780 38.1%
Allocations to 512 bytes 62,403,520 38.1%
Allocations to 4 kbytes 180 0.0%
Allocations over 4 kbytes 80 0.0%
Frees 116,470,565
Inline values 0
Interpreter increfs 3,279,000 2.0%
Interpreter decrefs 5,061,300 1.9%
Increfs 160,569,300 98.0%
Decrefs 265,427,025 98.1%
Materialize dict (on request) 0
Materialize dict (new key) 0
Materialize dict (too big) 0
Materialize dict (str subclass) 0
Method cache hits 1,638,583
Method cache misses 117
Method cache collisions 82
Method cache dunder hits 0
Method cache dunder misses 0

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.

 20
Traces created
ⓘ

The number of traces that were successfully created.

 20 100.0%
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.

 0 0.0%
Inner loop found
ⓘ

A trace is truncated because it has an inner loop

 0 0.0%
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

 160
Uops executed
ⓘ

The total number of uops (micro-operations) that were executed

 14,056,320 8,785,200.0%
Count Ratio
Optimizer attempts
ⓘ

The number of times the trace optimizer (_Py_uop_analyze_and_optimize) was run.

 20
Optimizer successes
ⓘ

The number of traces that were successfully optimized.

 20 100.0%
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 20 100.0%

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 20 100.0%

Trace run length histogram

trace run length histogram
Range Count Ratio
<= 1 0 0.0%
<= 2 0 0.0%
<= 4 0 0.0%
<= 8 0 0.0%
<= 16 0 0.0%
<= 32 0 0.0%
<= 64 0 0.0%
<= 128 0 0.0%
<= 256 0 0.0%
<= 512 0 0.0%
<= 1,024 0 0.0%
<= 2,048 0 0.0%
<= 4,096 0 0.0%
<= 8,192 0 0.0%
<= 16,384 0 0.0%
<= 32,768 0 0.0%
<= 65,536 0 0.0%
<= 131,072 0 0.0%
<= 262,144 80 50.0%

Uop execution stats

uop execution stats
Name Count Self Cumulative Miss ratio
_POP_TOP 1,634,400 11.6% 11.6%
_PUSH_NULL 1,634,400 11.6% 23.3%
_SET_IP 1,634,400 11.6% 34.9%
_CALL_BUILTIN_FAST_WITH_KEYWORDS 1,634,400 11.6% 46.5%
_CHECK_PERIODIC 1,634,400 11.6% 58.1%
_CHECK_VALIDITY 1,634,400 11.6% 69.8%
_LOAD_FAST_3 1,634,400 11.6% 81.4%
_LOAD_FAST_5 1,634,400 11.6% 93.0%
_GUARD_NOT_EXHAUSTED_RANGE 163,520 1.2% 94.2% 0.0%
_ITER_CHECK_RANGE 163,520 1.2% 95.3%
_TIER2_RESUME_CHECK 163,520 1.2% 96.5%
_ITER_NEXT_RANGE 163,440 1.2% 97.7%
_JUMP_TO_TOP 163,440 1.2% 98.8%
_STORE_FAST_7 163,440 1.2% 100.0%
_EXIT_TRACE 80 0.0% 100.0%
_COLD_EXIT 80 0.0% 100.0%
_START_EXECUTOR 80 0.0% 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
_PUSH_NULL _LOAD_FAST_3 1,634,400 11.6% 11.6%
_SET_IP _CALL_BUILTIN_FAST_WITH_KEYWORDS 1,634,400 11.6% 23.3%
_CALL_BUILTIN_FAST_WITH_KEYWORDS _CHECK_PERIODIC 1,634,400 11.6% 34.9%
_CHECK_PERIODIC _CHECK_VALIDITY 1,634,400 11.6% 46.5%
_CHECK_VALIDITY _POP_TOP 1,634,400 11.6% 58.1%
_LOAD_FAST_3 _SET_IP 1,634,400 11.6% 69.8%
_LOAD_FAST_5 _PUSH_NULL 1,634,400 11.6% 81.4%
_POP_TOP _LOAD_FAST_5 1,470,960 10.5% 91.9%
_ITER_CHECK_RANGE _GUARD_NOT_EXHAUSTED_RANGE 163,520 1.2% 93.0%
_TIER2_RESUME_CHECK _ITER_CHECK_RANGE 163,520 1.2% 94.2%
_POP_TOP _JUMP_TO_TOP 163,440 1.2% 95.3%
_GUARD_NOT_EXHAUSTED_RANGE _ITER_NEXT_RANGE 163,440 1.2% 96.5%
_ITER_NEXT_RANGE _STORE_FAST_7 163,440 1.2% 97.7%
_JUMP_TO_TOP _TIER2_RESUME_CHECK 163,440 1.2% 98.8%
_STORE_FAST_7 _LOAD_FAST_5 163,440 1.2% 100.0%
_GUARD_NOT_EXHAUSTED_RANGE _EXIT_TRACE 80 0.0% 100.0%
_START_EXECUTOR _TIER2_RESUME_CHECK 80 0.0% 100.0%

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