Comparing
cod-2714-investigate-pytest-flamegraph-regression
(72e5f60
) withmaster
(4a64bb2
)-1%
- Improvement
- 1
- Regression
- 1
- Untouched
- 54
- Skipped
- 4
Benchmarks
60 total
test_valgrind[valgrind.codspeed, stress-ng --cpu 1 --cpu-ops 10, no-inline]
bench/bench.py-13%
4.3 s5 s
test_valgrind[valgrind.codspeed, python3 testdata/test.py, no-inline]
bench/bench.py+13%
4.4 s3.9 s
test_valgrind[valgrind-3.25.1, python3 testdata/test.py, full-no-inline]
bench/bench.py+10%
6 s5.5 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 1 --cpu-ops 10, inline]
bench/bench.py+10%
5.3 s4.8 s
test_valgrind[valgrind-3.25.1, stress-ng --cpu 1 --cpu-ops 10, no-inline]
bench/bench.py+6%
4.4 s4.1 s
test_valgrind[valgrind.codspeed, python3 testdata/test.py, inline]
bench/bench.py-5%
3.9 s4.1 s
test_valgrind[valgrind-3.26.0, stress-ng --cpu 1 --cpu-ops 10, full-with-inline]
bench/bench.py-5%
8.5 s8.9 s
test_valgrind[valgrind.codspeed, python3 testdata/test.py, full-with-inline]
bench/bench.py-4%
5.7 s5.9 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 4 --cpu-ops 10, inline]
bench/bench.py-3%
2.2 s2.3 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 1 --cpu-ops 10, full-no-inline]
bench/bench.py+3%
8.4 s8.2 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 4 --cpu-ops 10, no-inline]
bench/bench.py-3%
2 s2 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 4 --cpu-ops 10, full-with-inline]
bench/bench.py-2%
3.5 s3.6 s
test_valgrind[valgrind-3.25.1, stress-ng --cpu 1 --cpu-ops 10, full-with-inline]
bench/bench.py-2%
8.2 s8.4 s
test_valgrind[valgrind-3.25.1, stress-ng --cpu 1 --cpu-ops 10, full-no-inline]
bench/bench.py+2%
8.2 s8.1 s
test_valgrind[valgrind.codspeed, python3 testdata/test.py, full-no-inline]
bench/bench.py-2%
5.6 s5.7 s
test_valgrind[valgrind.codspeed, echo Hello, World!, full-no-inline]
bench/bench.py-1%
578.6 ms585.7 ms
test_valgrind[valgrind-3.26.0, stress-ng --cpu 1 --cpu-ops 10, full-no-inline]
bench/bench.py-1%
8.2 s8.3 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 4 --cpu-ops 10, full-no-inline]
bench/bench.py-1%
3.3 s3.3 s
test_valgrind[valgrind.codspeed, stress-ng --cpu 1 --cpu-ops 10, full-with-inline]
bench/bench.py-1%
8.7 s8.8 s
test_valgrind[valgrind-3.26.0, stress-ng --cpu 1 --cpu-ops 10, no-inline]
bench/bench.py-1%
4.2 s4.2 s
test_valgrind[valgrind.codspeed, echo Hello, World!, no-inline]
bench/bench.py-1%
513 ms517.4 ms
test_valgrind[valgrind-3.26.0, python3 testdata/test.py, no-inline]
bench/bench.py+1%
3.7 s3.7 s
test_valgrind[valgrind-3.26.0, testdata/take_strings-aarch64 varbinview_non_null, full-no-inline]
bench/bench.py-1%
3.1 s3.1 s
test_valgrind[valgrind-3.26.0, stress-ng --cpu 4 --cpu-ops 10, no-inline]
bench/bench.py-1%
2 s2 s
test_valgrind[valgrind-3.26.0, testdata/take_strings-aarch64 varbinview_non_null, no-inline]
bench/bench.py-1%
2.5 s2.5 s
Commits
Click on a commit to change the comparison rangeBase
master
4a64bb2
-27.3%
test(callgrind): wire underflow + py3.14 reproducers, strengthen post-checks
dfe44cc
28 days ago
by not-matthias-20.25%
fix(callgrind/tests): link runtime_obj_skip libs via LDADD for --as-needed
9e62d5b
28 days ago
by not-matthias+45.91%
fix(callgrind/tests): normalize py314 prereq exit code for dash
939d007
28 days ago
by not-matthias+0.95%
fix(callgrind): skip frame 0 when seeding shadow call stack at START
caf4703
27 days ago
by not-matthias+12.11%
fix(callgrind): restore cxt on ret for seeded shadow-stack entries
e31aab9
27 days ago
by not-matthias-12.69%
fix(callgrind): don't seed cxt for anonymous (JIT) frames
72e5f60
27 days ago
by not-matthias