fix(profiling): fix crash due to Frame cache eviction by KowalskiThomas · Pull Request #17456 · DataDog/dd-trace-py

KowalskiThomas · 2026-04-10T11:29:20Z

Description

https://datadoghq.atlassian.net/browse/PROF-13112

Crash

This fixes a crash in the Profiler codebase.
The crash was caused by ThreadInfo::unwind_task trying to use Frame objects that it had a reference to (Frame::Ref) while those same Frames were being evicted from the cache, which can happen under high load.

Error UnixSignal: Process terminated with SI_KERNEL (SIGSEGV)
#0   0x00007f620376cfa2 std::make_unique<StackInfo, unsigned long&, bool&>
#1   0x00007f62037654bb is_uvloop_wrapper_frame
#2   0x00007f620376c9b3 ThreadInfo::unwind_tasks
#3   0x00007f620376cdb9 ThreadInfo::sample
#4   0x00007f620376ceee std::_Function_handler<void (_ts*, ThreadInfo&), Datadog::Sampler::sampling_thread(unsigned long)::{lambda(InterpreterInfo&)#1}::operator()(InterpreterInfo&) const::{lambda(_ts*, ThreadInfo&)#1}>::_M_invoke
#5   0x00007f62037678b5 for_each_thread
#6   0x00007f620376794d std::_Function_handler<void (InterpreterInfo&), Datadog::Sampler::sampling_thread(unsigned long)::{lambda(InterpreterInfo&)#1}>::_M_invoke
#7   0x00007f6203766257 for_each_interp
#8   0x00007f6203767cc0 Datadog::Sampler::sampling_thread
#9   0x00007f6203767f27 call_sampling_thread

Proposed fix

The proposed fix is to change FrameStack to extend std::deque<Frame> instead of std::deque<Frame::Ref>, which means pushing frames to FrameStack actually copies them and eliminates that race condition.
This should be OK performance-wise because Frame is tiny and only has trivial fields to copy (no memory allocations, only integral types).

Another possible fix would be to "lock" certain Frame objects in the cache (to prevent them from being evicted while any Task using them is being unwound) but this would be much more complicated to get right and although it would lead to less copies, the performance picture is blurry for more subtle reasons.
So I suggest we don't go ahead with the alternative unless we absolutely need to (which we don't seem to).

Performance

Running DoE with this change on the Enterprise archetype with asyncio/FastAPI (n=10) shows no significant difference (either in memory or CPU usage / latency), so I think this is safe to merge performance-wise.

Details

Raw runs

Commit	Run	p50 (ms)	p99 (ms)	CPU%	Mem (MB)
`90de7913` (after)	1	150.43	152.88	172.68	274.9
`90de7913` (after)	2	150.45	152.38	172.17	266.2
`90de7913` (after)	3	150.43	153.91	172.92	266.2
`90de7913` (after)	4	150.46	154.16	172.83	265.8
`90de7913` (after)	5	150.44	153.01	172.80	267.4
`90de7913` (after)	6	150.43	153.28	172.70	265.8
`90de7913` (after)	7	150.54	157.48	173.32	265.8
`90de7913` (after)	8	150.51	156.61	172.18	265.5
`90de7913` (after)	9	150.53	157.05	172.18	265.8
`90de7913` (after)	10	150.45	152.84	172.08	265.5
`c26e6181` (before)	1	150.51	152.80	173.63	265.5
`c26e6181` (before)	2	150.48	152.41	173.49	265.4
`c26e6181` (before)	3	150.47	156.31	172.53	265.9
`c26e6181` (before)	4	150.49	152.60	172.86	265.5
`c26e6181` (before)	5	150.40	152.86	172.99	266.0
`c26e6181` (before)	6	150.46	152.66	172.53	265.5
`c26e6181` (before)	7	150.49	154.91	172.74	265.3
`c26e6181` (before)	8	150.46	152.46	171.93	265.7
`c26e6181` (before)	9	150.52	156.70	172.96	265.7
`c26e6181` (before)	10	150.43	153.02	172.52	265.8

Aggregated (mean ± stdev, n=10)

Version	p50 (ms)	p99 (ms)	CPU%	Mem (MB)
`c26e6181` (before)	150.470 ± 0.034	153.67 ± 1.66	172.82 ± 0.50	265.6 ± 0.2
`90de7913` (after)	150.465 ± 0.043	154.36 ± 1.94	172.59 ± 0.41	266.9 ± 2.9
Δ (after − before)	−0.005ms (0.00%)	+0.69ms (+0.45%)	−0.23pp (−0.13%)	+1.2MB (+0.47%)

pr-commenter · 2026-04-10T12:00:56Z

Performance SLOs

Comparing candidate kowalski/fix-profiling-fix-crash-due-to-frame-cache-eviction (90de791) with baseline main (c26e618)

📈 Performance Regressions (3 suites)

📈 iastaspects - 118/118

✅ add_aspect

Time: ✅ 103.453µs (SLO: <130.000µs 📉 -20.4%) vs baseline: +0.9%

Memory: ✅ 43.906MB (SLO: <46.000MB -4.6%) vs baseline: +5.6%

✅ add_inplace_aspect

Time: ✅ 101.932µs (SLO: <130.000µs 📉 -21.6%) vs baseline: -2.0%

Memory: ✅ 43.890MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ add_inplace_noaspect

Time: ✅ 28.286µs (SLO: <40.000µs 📉 -29.3%) vs baseline: -0.4%

Memory: ✅ 43.971MB (SLO: <46.000MB -4.4%) vs baseline: +5.4%

✅ add_noaspect

Time: ✅ 49.437µs (SLO: <70.000µs 📉 -29.4%) vs baseline: ~same

Memory: ✅ 43.889MB (SLO: <46.000MB -4.6%) vs baseline: +5.4%

✅ bytearray_aspect

Time: ✅ 261.200µs (SLO: <400.000µs 📉 -34.7%) vs baseline: +0.9%

Memory: ✅ 43.935MB (SLO: <46.000MB -4.5%) vs baseline: +5.6%

✅ bytearray_extend_aspect

Time: ✅ 665.136µs (SLO: <800.000µs 📉 -16.9%) vs baseline: +2.7%

Memory: ✅ 43.846MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ bytearray_extend_noaspect

Time: ✅ 275.805µs (SLO: <400.000µs 📉 -31.0%) vs baseline: +5.5%

Memory: ✅ 43.861MB (SLO: <46.000MB -4.7%) vs baseline: +5.2%

✅ bytearray_noaspect

Time: ✅ 144.126µs (SLO: <300.000µs 📉 -52.0%) vs baseline: +2.1%

Memory: ✅ 43.878MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ bytes_aspect

Time: ✅ 220.359µs (SLO: <300.000µs 📉 -26.5%) vs baseline: -0.5%

Memory: ✅ 43.844MB (SLO: <46.000MB -4.7%) vs baseline: +5.4%

✅ bytes_noaspect

Time: ✅ 134.346µs (SLO: <200.000µs 📉 -32.8%) vs baseline: ~same

Memory: ✅ 43.870MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ bytesio_aspect

Time: ✅ 3.868ms (SLO: <5.000ms 📉 -22.6%) vs baseline: +0.9%

Memory: ✅ 43.769MB (SLO: <46.000MB -4.8%) vs baseline: +5.1%

✅ bytesio_noaspect

Time: ✅ 318.748µs (SLO: <420.000µs 📉 -24.1%) vs baseline: -0.8%

Memory: ✅ 43.887MB (SLO: <46.000MB -4.6%) vs baseline: +5.4%

✅ capitalize_aspect

Time: ✅ 88.446µs (SLO: <300.000µs 📉 -70.5%) vs baseline: -0.8%

Memory: ✅ 43.879MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ capitalize_noaspect

Time: ✅ 270.103µs (SLO: <300.000µs -10.0%) vs baseline: +5.8%

Memory: ✅ 43.894MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ casefold_aspect

Time: ✅ 87.908µs (SLO: <500.000µs 📉 -82.4%) vs baseline: -0.9%

Memory: ✅ 43.880MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ casefold_noaspect

Time: ✅ 317.849µs (SLO: <500.000µs 📉 -36.4%) vs baseline: +2.0%

Memory: ✅ 43.758MB (SLO: <46.000MB -4.9%) vs baseline: +5.3%

✅ decode_aspect

Time: ✅ 87.339µs (SLO: <100.000µs 📉 -12.7%) vs baseline: +0.8%

Memory: ✅ 44.014MB (SLO: <46.000MB -4.3%) vs baseline: +5.8%

✅ decode_noaspect

Time: ✅ 156.709µs (SLO: <210.000µs 📉 -25.4%) vs baseline: +1.3%

Memory: ✅ 43.871MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ encode_aspect

Time: ✅ 85.123µs (SLO: <200.000µs 📉 -57.4%) vs baseline: +0.5%

Memory: ✅ 43.927MB (SLO: <46.000MB -4.5%) vs baseline: +5.3%

✅ encode_noaspect

Time: ✅ 144.834µs (SLO: <200.000µs 📉 -27.6%) vs baseline: +0.5%

Memory: ✅ 43.931MB (SLO: <46.000MB -4.5%) vs baseline: +5.5%

✅ format_aspect

Time: ✅ 14.605ms (SLO: <19.200ms 📉 -23.9%) vs baseline: ~same

Memory: ✅ 43.881MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ format_map_aspect

Time: ✅ 16.354ms (SLO: <21.500ms 📉 -23.9%) vs baseline: -0.9%

Memory: ✅ 43.967MB (SLO: <46.000MB -4.4%) vs baseline: +5.5%

✅ format_map_noaspect

Time: ✅ 364.961µs (SLO: <500.000µs 📉 -27.0%) vs baseline: -3.0%

Memory: ✅ 43.959MB (SLO: <46.000MB -4.4%) vs baseline: +5.7%

✅ format_noaspect

Time: ✅ 310.029µs (SLO: <500.000µs 📉 -38.0%) vs baseline: -1.8%

Memory: ✅ 43.931MB (SLO: <46.000MB -4.5%) vs baseline: +5.5%

✅ index_aspect

Time: ✅ 139.159µs (SLO: <300.000µs 📉 -53.6%) vs baseline: 📈 +10.0%

Memory: ✅ 43.824MB (SLO: <46.000MB -4.7%) vs baseline: +5.4%

✅ index_noaspect

Time: ✅ 40.469µs (SLO: <300.000µs 📉 -86.5%) vs baseline: +0.1%

Memory: ✅ 43.830MB (SLO: <46.000MB -4.7%) vs baseline: +5.2%

✅ join_aspect

Time: ✅ 206.696µs (SLO: <300.000µs 📉 -31.1%) vs baseline: -3.4%

Memory: ✅ 43.889MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ join_noaspect

Time: ✅ 141.754µs (SLO: <300.000µs 📉 -52.7%) vs baseline: -1.6%

Memory: ✅ 43.887MB (SLO: <46.000MB -4.6%) vs baseline: +5.2%

✅ ljust_aspect

Time: ✅ 502.977µs (SLO: <700.000µs 📉 -28.1%) vs baseline: -3.0%

Memory: ✅ 43.795MB (SLO: <46.000MB -4.8%) vs baseline: +5.2%

✅ ljust_noaspect

Time: ✅ 273.824µs (SLO: <300.000µs -8.7%) vs baseline: +2.7%

Memory: ✅ 43.831MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ lower_aspect

Time: ✅ 307.620µs (SLO: <500.000µs 📉 -38.5%) vs baseline: -1.1%

Memory: ✅ 43.794MB (SLO: <46.000MB -4.8%) vs baseline: +5.2%

✅ lower_noaspect

Time: ✅ 244.007µs (SLO: <300.000µs 📉 -18.7%) vs baseline: +1.5%

Memory: ✅ 43.889MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ lstrip_aspect

Time: ✅ 0.278ms (SLO: <3.000ms 📉 -90.7%) vs baseline: +0.2%

Memory: ✅ 43.888MB (SLO: <46.000MB -4.6%) vs baseline: +5.4%

✅ lstrip_noaspect

Time: ✅ 0.179ms (SLO: <3.000ms 📉 -94.0%) vs baseline: +0.5%

Memory: ✅ 43.814MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

✅ modulo_aspect

Time: ✅ 14.242ms (SLO: <18.750ms 📉 -24.0%) vs baseline: -0.6%

Memory: ✅ 43.886MB (SLO: <46.000MB -4.6%) vs baseline: +5.1%

✅ modulo_aspect_for_bytearray_bytearray

Time: ✅ 14.826ms (SLO: <19.350ms 📉 -23.4%) vs baseline: ~same

Memory: ✅ 43.935MB (SLO: <46.000MB -4.5%) vs baseline: +5.3%

✅ modulo_aspect_for_bytes

Time: ✅ 14.400ms (SLO: <18.900ms 📉 -23.8%) vs baseline: +0.1%

Memory: ✅ 43.835MB (SLO: <46.000MB -4.7%) vs baseline: +4.9%

✅ modulo_aspect_for_bytes_bytearray

Time: ✅ 14.658ms (SLO: <19.150ms 📉 -23.5%) vs baseline: +0.7%

Memory: ✅ 43.952MB (SLO: <46.000MB -4.5%) vs baseline: +5.1%

✅ modulo_noaspect

Time: ✅ 0.364ms (SLO: <3.000ms 📉 -87.9%) vs baseline: -1.8%

Memory: ✅ 43.832MB (SLO: <46.000MB -4.7%) vs baseline: +5.3%

✅ replace_aspect

Time: ✅ 18.368ms (SLO: <24.000ms 📉 -23.5%) vs baseline: -0.2%

Memory: ✅ 43.874MB (SLO: <46.000MB -4.6%) vs baseline: +5.1%

✅ replace_noaspect

Time: ✅ 285.876µs (SLO: <400.000µs 📉 -28.5%) vs baseline: +1.2%

Memory: ✅ 43.935MB (SLO: <46.000MB -4.5%) vs baseline: +5.6%

✅ repr_aspect

Time: ✅ 316.054µs (SLO: <420.000µs 📉 -24.7%) vs baseline: -2.7%

Memory: ✅ 43.850MB (SLO: <46.000MB -4.7%) vs baseline: +5.3%

✅ repr_noaspect

Time: ✅ 46.883µs (SLO: <90.000µs 📉 -47.9%) vs baseline: -0.7%

Memory: ✅ 43.857MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ rstrip_aspect

Time: ✅ 382.985µs (SLO: <500.000µs 📉 -23.4%) vs baseline: -2.0%

Memory: ✅ 43.980MB (SLO: <46.000MB -4.4%) vs baseline: +5.6%

✅ rstrip_noaspect

Time: ✅ 186.735µs (SLO: <300.000µs 📉 -37.8%) vs baseline: +2.4%

Memory: ✅ 43.911MB (SLO: <46.000MB -4.5%) vs baseline: +5.5%

✅ slice_aspect

Time: ✅ 186.011µs (SLO: <300.000µs 📉 -38.0%) vs baseline: +0.8%

Memory: ✅ 43.824MB (SLO: <46.000MB -4.7%) vs baseline: +5.1%

✅ slice_noaspect

Time: ✅ 54.352µs (SLO: <90.000µs 📉 -39.6%) vs baseline: +0.6%

Memory: ✅ 43.871MB (SLO: <46.000MB -4.6%) vs baseline: +5.2%

✅ stringio_aspect

Time: ✅ 4.459ms (SLO: <5.000ms 📉 -10.8%) vs baseline: 📈 +14.9%

Memory: ✅ 43.809MB (SLO: <46.000MB -4.8%) vs baseline: +4.8%

✅ stringio_noaspect

Time: ✅ 351.934µs (SLO: <500.000µs 📉 -29.6%) vs baseline: -0.7%

Memory: ✅ 43.855MB (SLO: <46.000MB -4.7%) vs baseline: +5.6%

✅ strip_aspect

Time: ✅ 275.634µs (SLO: <350.000µs 📉 -21.2%) vs baseline: +1.0%

Memory: ✅ 43.810MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

✅ strip_noaspect

Time: ✅ 179.645µs (SLO: <240.000µs 📉 -25.1%) vs baseline: +0.5%

Memory: ✅ 43.824MB (SLO: <46.000MB -4.7%) vs baseline: +5.3%

✅ swapcase_aspect

Time: ✅ 346.177µs (SLO: <500.000µs 📉 -30.8%) vs baseline: -0.8%

Memory: ✅ 43.955MB (SLO: <46.000MB -4.4%) vs baseline: +5.6%

✅ swapcase_noaspect

Time: ✅ 275.442µs (SLO: <400.000µs 📉 -31.1%) vs baseline: -0.3%

Memory: ✅ 43.852MB (SLO: <46.000MB -4.7%) vs baseline: +5.3%

✅ title_aspect

Time: ✅ 330.083µs (SLO: <500.000µs 📉 -34.0%) vs baseline: +0.7%

Memory: ✅ 43.913MB (SLO: <46.000MB -4.5%) vs baseline: +5.4%

✅ title_noaspect

Time: ✅ 267.079µs (SLO: <400.000µs 📉 -33.2%) vs baseline: +1.5%

Memory: ✅ 43.821MB (SLO: <46.000MB -4.7%) vs baseline: +5.1%

✅ translate_aspect

Time: ✅ 518.608µs (SLO: <700.000µs 📉 -25.9%) vs baseline: +2.8%

Memory: ✅ 43.824MB (SLO: <46.000MB -4.7%) vs baseline: +4.9%

✅ translate_noaspect

Time: ✅ 429.777µs (SLO: <500.000µs 📉 -14.0%) vs baseline: -1.1%

Memory: ✅ 43.873MB (SLO: <46.000MB -4.6%) vs baseline: +5.4%

✅ upper_aspect

Time: ✅ 300.824µs (SLO: <500.000µs 📉 -39.8%) vs baseline: -0.5%

Memory: ✅ 43.795MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

✅ upper_noaspect

Time: ✅ 246.773µs (SLO: <400.000µs 📉 -38.3%) vs baseline: +3.3%

Memory: ✅ 43.873MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

📈 iastaspectsospath - 24/24

✅ ospathbasename_aspect

Time: ✅ 524.510µs (SLO: <700.000µs 📉 -25.1%) vs baseline: 📈 +22.2%

Memory: ✅ 43.823MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ ospathbasename_noaspect

Time: ✅ 442.424µs (SLO: <700.000µs 📉 -36.8%) vs baseline: +1.4%

Memory: ✅ 43.868MB (SLO: <46.000MB -4.6%) vs baseline: +5.5%

✅ ospathjoin_aspect

Time: ✅ 643.100µs (SLO: <700.000µs -8.1%) vs baseline: +1.9%

Memory: ✅ 43.842MB (SLO: <46.000MB -4.7%) vs baseline: +5.3%

✅ ospathjoin_noaspect

Time: ✅ 650.010µs (SLO: <700.000µs -7.1%) vs baseline: +1.2%

Memory: ✅ 43.817MB (SLO: <46.000MB -4.7%) vs baseline: +5.1%

✅ ospathnormcase_aspect

Time: ✅ 360.487µs (SLO: <700.000µs 📉 -48.5%) vs baseline: +2.2%

Memory: ✅ 43.854MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ ospathnormcase_noaspect

Time: ✅ 364.782µs (SLO: <700.000µs 📉 -47.9%) vs baseline: +1.0%

Memory: ✅ 43.942MB (SLO: <46.000MB -4.5%) vs baseline: +5.7%

✅ ospathsplit_aspect

Time: ✅ 497.730µs (SLO: <700.000µs 📉 -28.9%) vs baseline: +0.9%

Memory: ✅ 43.762MB (SLO: <46.000MB -4.9%) vs baseline: +5.3%

✅ ospathsplit_noaspect

Time: ✅ 502.271µs (SLO: <700.000µs 📉 -28.2%) vs baseline: -0.2%

Memory: ✅ 43.821MB (SLO: <46.000MB -4.7%) vs baseline: +5.4%

✅ ospathsplitdrive_aspect

Time: ✅ 379.438µs (SLO: <700.000µs 📉 -45.8%) vs baseline: +0.2%

Memory: ✅ 43.885MB (SLO: <46.000MB -4.6%) vs baseline: +5.7%

✅ ospathsplitdrive_noaspect

Time: ✅ 72.239µs (SLO: <700.000µs 📉 -89.7%) vs baseline: -1.1%

Memory: ✅ 43.856MB (SLO: <46.000MB -4.7%) vs baseline: +5.5%

✅ ospathsplitext_aspect

Time: ✅ 469.585µs (SLO: <700.000µs 📉 -32.9%) vs baseline: +1.9%

Memory: ✅ 43.797MB (SLO: <46.000MB -4.8%) vs baseline: +5.5%

✅ ospathsplitext_noaspect

Time: ✅ 477.921µs (SLO: <700.000µs 📉 -31.7%) vs baseline: +1.7%

Memory: ✅ 43.800MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

📈 iastaspectssplit - 12/12

✅ rsplit_aspect

Time: ✅ 167.704µs (SLO: <250.000µs 📉 -32.9%) vs baseline: 📈 +10.6%

Memory: ✅ 43.804MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

✅ rsplit_noaspect

Time: ✅ 163.805µs (SLO: <250.000µs 📉 -34.5%) vs baseline: +1.8%

Memory: ✅ 43.952MB (SLO: <46.000MB -4.5%) vs baseline: +5.6%

✅ split_aspect

Time: ✅ 152.720µs (SLO: <250.000µs 📉 -38.9%) vs baseline: +0.7%

Memory: ✅ 43.790MB (SLO: <46.000MB -4.8%) vs baseline: +5.2%

✅ split_noaspect

Time: ✅ 160.364µs (SLO: <250.000µs 📉 -35.9%) vs baseline: -0.6%

Memory: ✅ 43.889MB (SLO: <46.000MB -4.6%) vs baseline: +5.3%

✅ splitlines_aspect

Time: ✅ 148.527µs (SLO: <250.000µs 📉 -40.6%) vs baseline: +2.0%

Memory: ✅ 43.787MB (SLO: <46.000MB -4.8%) vs baseline: +5.3%

✅ splitlines_noaspect

Time: ✅ 160.293µs (SLO: <250.000µs 📉 -35.9%) vs baseline: +4.0%

Memory: ✅ 43.859MB (SLO: <46.000MB -4.7%) vs baseline: +5.2%

🟡 Near SLO Breach (5 suites)

🟡 djangosimple - 30/30

✅ appsec

Time: ✅ 19.641ms (SLO: <22.300ms 📉 -11.9%) vs baseline: -0.3%

Memory: ✅ 69.383MB (SLO: <73.500MB -5.6%) vs baseline: +5.2%

✅ exception-replay-enabled

Time: ✅ 1.363ms (SLO: <1.450ms -6.0%) vs baseline: ~same

Memory: ✅ 67.603MB (SLO: <71.500MB -5.4%) vs baseline: +5.1%

✅ iast

Time: ✅ 19.684ms (SLO: <22.250ms 📉 -11.5%) vs baseline: -0.4%

Memory: ✅ 69.380MB (SLO: <75.000MB -7.5%) vs baseline: +5.0%

✅ profiler

Time: ✅ 15.105ms (SLO: <16.550ms -8.7%) vs baseline: -0.1%

Memory: ✅ 60.163MB (SLO: <61.000MB 🟡 -1.4%) vs baseline: +5.7%

✅ resource-renaming

Time: ✅ 19.508ms (SLO: <21.750ms 📉 -10.3%) vs baseline: -0.6%

Memory: ✅ 69.530MB (SLO: <73.500MB -5.4%) vs baseline: +5.3%

✅ span-code-origin

Time: ✅ 20.074ms (SLO: <28.200ms 📉 -28.8%) vs baseline: +1.0%

Memory: ✅ 69.363MB (SLO: <75.000MB -7.5%) vs baseline: +5.0%

✅ tracer

Time: ✅ 19.669ms (SLO: <21.750ms -9.6%) vs baseline: ~same

Memory: ✅ 69.422MB (SLO: <75.000MB -7.4%) vs baseline: +5.1%

✅ tracer-and-profiler

Time: ✅ 20.968ms (SLO: <23.500ms 📉 -10.8%) vs baseline: -0.2%

Memory: ✅ 71.339MB (SLO: <75.000MB -4.9%) vs baseline: +5.1%

✅ tracer-dont-create-db-spans

Time: ✅ 19.834ms (SLO: <21.500ms -7.7%) vs baseline: +0.3%

Memory: ✅ 69.392MB (SLO: <75.000MB -7.5%) vs baseline: +5.1%

✅ tracer-minimal

Time: ✅ 16.681ms (SLO: <17.500ms -4.7%) vs baseline: ~same

Memory: ✅ 69.396MB (SLO: <75.000MB -7.5%) vs baseline: +5.2%

✅ tracer-native

Time: ✅ 19.708ms (SLO: <21.750ms -9.4%) vs baseline: +1.1%

Memory: ✅ 69.404MB (SLO: <72.500MB -4.3%) vs baseline: +5.2%

✅ tracer-no-caches

Time: ✅ 17.489ms (SLO: <19.650ms 📉 -11.0%) vs baseline: +0.6%

Memory: ✅ 69.272MB (SLO: <75.000MB -7.6%) vs baseline: +4.9%

✅ tracer-no-databases

Time: ✅ 19.231ms (SLO: <20.100ms -4.3%) vs baseline: ~same

Memory: ✅ 69.344MB (SLO: <75.000MB -7.5%) vs baseline: +5.1%

✅ tracer-no-middleware

Time: ✅ 19.408ms (SLO: <21.500ms -9.7%) vs baseline: +0.1%

Memory: ✅ 69.383MB (SLO: <75.000MB -7.5%) vs baseline: +5.0%

✅ tracer-no-templates

Time: ✅ 19.528ms (SLO: <22.000ms 📉 -11.2%) vs baseline: +1.2%

Memory: ✅ 69.343MB (SLO: <73.500MB -5.7%) vs baseline: +5.0%

🟡 otelspan - 22/22

✅ add-event

Time: ✅ 40.731ms (SLO: <47.150ms 📉 -13.6%) vs baseline: -0.4%

Memory: ✅ 41.427MB (SLO: <47.000MB 📉 -11.9%) vs baseline: +5.9%

✅ add-metrics

Time: ✅ 236.157ms (SLO: <344.800ms 📉 -31.5%) vs baseline: +0.2%

Memory: ✅ 45.731MB (SLO: <47.500MB -3.7%) vs baseline: +5.6%

✅ add-tags

Time: ✅ 277.708ms (SLO: <330.000ms 📉 -15.8%) vs baseline: +2.0%

Memory: ✅ 45.767MB (SLO: <47.500MB -3.6%) vs baseline: +5.3%

✅ get-context

Time: ✅ 83.691ms (SLO: <92.350ms -9.4%) vs baseline: ~same

Memory: ✅ 41.583MB (SLO: <46.500MB 📉 -10.6%) vs baseline: +5.4%

✅ is-recording

Time: ✅ 39.098ms (SLO: <44.500ms 📉 -12.1%) vs baseline: -0.5%

Memory: ✅ 41.173MB (SLO: <47.500MB 📉 -13.3%) vs baseline: +5.3%

✅ record-exception

Time: ✅ 61.114ms (SLO: <67.650ms -9.7%) vs baseline: ~same

Memory: ✅ 41.719MB (SLO: <47.000MB 📉 -11.2%) vs baseline: +5.1%

✅ set-status

Time: ✅ 45.202ms (SLO: <50.400ms 📉 -10.3%) vs baseline: +0.1%

Memory: ✅ 41.131MB (SLO: <47.000MB 📉 -12.5%) vs baseline: +5.5%

✅ start

Time: ✅ 40.420ms (SLO: <44.500ms -9.2%) vs baseline: +4.3%

Memory: ✅ 41.136MB (SLO: <47.000MB 📉 -12.5%) vs baseline: +5.3%

✅ start-finish

Time: ✅ 90.396ms (SLO: <91.000ms 🟡 -0.7%) vs baseline: +0.3%

Memory: ✅ 38.751MB (SLO: <46.500MB 📉 -16.7%) vs baseline: +5.3%

✅ start-finish-telemetry

Time: ✅ 91.929ms (SLO: <92.000ms 🟡 ~same) vs baseline: +0.3%

Memory: ✅ 38.692MB (SLO: <46.500MB 📉 -16.8%) vs baseline: +5.2%

✅ update-name

Time: ✅ 40.211ms (SLO: <45.150ms 📉 -10.9%) vs baseline: -0.5%

Memory: ✅ 41.127MB (SLO: <47.000MB 📉 -12.5%) vs baseline: +5.4%

🟡 samplingrules - 8/8

✅ average_match

Time: ✅ 169.816µs (SLO: <290.000µs 📉 -41.4%) vs baseline: +0.7%

Memory: ✅ 36.156MB (SLO: <38.000MB -4.9%) vs baseline: +5.3%

✅ high_match

Time: ✅ 214.941µs (SLO: <480.000µs 📉 -55.2%) vs baseline: -0.7%

Memory: ✅ 36.176MB (SLO: <38.000MB -4.8%) vs baseline: +5.4%

✅ low_match

Time: ✅ 119.219µs (SLO: <120.000µs 🟡 -0.7%) vs baseline: +0.2%

Memory: ✅ 701.561MB (SLO: <780.000MB 📉 -10.1%) vs baseline: +4.9%

✅ very_low_match

Time: ✅ 3.099ms (SLO: <8.500ms 📉 -63.5%) vs baseline: -0.6%

Memory: ✅ 78.688MB (SLO: <85.000MB -7.4%) vs baseline: +5.0%

🟡 span - 26/26

✅ add-event

Time: ✅ 19.520ms (SLO: <22.500ms 📉 -13.2%) vs baseline: -1.6%

Memory: ✅ 38.456MB (SLO: <53.000MB 📉 -27.4%) vs baseline: +5.5%

✅ add-metrics

Time: ✅ 89.687ms (SLO: <93.500ms -4.1%) vs baseline: +0.8%

Memory: ✅ 42.850MB (SLO: <53.000MB 📉 -19.2%) vs baseline: +5.4%

✅ add-tags

Time: ✅ 148.101ms (SLO: <155.000ms -4.5%) vs baseline: ~same

Memory: ✅ 42.865MB (SLO: <53.000MB 📉 -19.1%) vs baseline: +5.7%

✅ get-context

Time: ✅ 18.789ms (SLO: <20.500ms -8.3%) vs baseline: -1.4%

Memory: ✅ 38.384MB (SLO: <53.000MB 📉 -27.6%) vs baseline: +5.6%

✅ is-recording

Time: ✅ 18.901ms (SLO: <20.500ms -7.8%) vs baseline: -1.0%

Memory: ✅ 38.313MB (SLO: <53.000MB 📉 -27.7%) vs baseline: +5.6%

✅ record-exception

Time: ✅ 38.522ms (SLO: <41.000ms -6.0%) vs baseline: -0.3%

Memory: ✅ 38.809MB (SLO: <53.000MB 📉 -26.8%) vs baseline: +5.5%

✅ set-status

Time: ✅ 20.552ms (SLO: <22.000ms -6.6%) vs baseline: -1.2%

Memory: ✅ 38.331MB (SLO: <53.000MB 📉 -27.7%) vs baseline: +5.6%

✅ start

Time: ✅ 19.715ms (SLO: <20.500ms -3.8%) vs baseline: +4.2%

Memory: ✅ 38.444MB (SLO: <53.000MB 📉 -27.5%) vs baseline: +5.8%

✅ start-finish

Time: ✅ 58.498ms (SLO: <58.500ms 🟡 ~same) vs baseline: -0.1%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.1%

✅ start-finish-telemetry

Time: ✅ 59.622ms (SLO: <60.000ms 🟡 -0.6%) vs baseline: -0.1%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.5%

✅ start-finish-traceid128

Time: ✅ 60.896ms (SLO: <62.000ms 🟡 -1.8%) vs baseline: -0.4%

Memory: ✅ 36.137MB (SLO: <38.000MB -4.9%) vs baseline: +5.0%

✅ start-traceid128

Time: ✅ 18.745ms (SLO: <22.500ms 📉 -16.7%) vs baseline: -1.0%

Memory: ✅ 38.333MB (SLO: <53.000MB 📉 -27.7%) vs baseline: +5.4%

✅ update-name

Time: ✅ 19.254ms (SLO: <22.000ms 📉 -12.5%) vs baseline: -1.8%

Memory: ✅ 38.296MB (SLO: <53.000MB 📉 -27.7%) vs baseline: +4.9%

🟡 tracer - 6/6

✅ large

Time: ✅ 33.081ms (SLO: <32.950ms +0.4%) vs baseline: ~same

Memory: ✅ 37.827MB (SLO: <39.250MB -3.6%) vs baseline: +6.1%

✅ medium

Time: ✅ 3.308ms (SLO: <3.500ms -5.5%) vs baseline: -1.1%

Memory: ✅ 36.294MB (SLO: <38.750MB -6.3%) vs baseline: +5.7%

✅ small

Time: ✅ 386.609µs (SLO: <390.000µs 🟡 -0.9%) vs baseline: +4.0%

Memory: ✅ 36.235MB (SLO: <38.750MB -6.5%) vs baseline: +5.7%

⚠️ Unstable Tests (2 suites)

⚠️

coreapiscenario - 10/10 (1 unstable)

⚠️ context_with_data_listeners

Time: ⚠️ 13.579µs (SLO: <20.000µs 📉 -32.1%) vs baseline: -1.0%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.5%

✅ context_with_data_no_listeners

Time: ✅ 3.615µs (SLO: <10.000µs 📉 -63.9%) vs baseline: +1.1%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.3%

✅ get_item_exists

Time: ✅ 0.584µs (SLO: <10.000µs 📉 -94.2%) vs baseline: +0.9%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ get_item_missing

Time: ✅ 0.635µs (SLO: <10.000µs 📉 -93.6%) vs baseline: +0.6%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ set_item

Time: ✅ 24.362µs (SLO: <30.000µs 📉 -18.8%) vs baseline: ~same

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.5%

⚠️

packagesupdateimporteddependencies - 24/24 (1 unstable)

✅ import_many

Time: ✅ 156.366µs (SLO: <170.000µs -8.0%) vs baseline: ~same

Memory: ✅ 41.088MB (SLO: <46.000MB 📉 -10.7%) vs baseline: +5.6%

✅ import_many_cached

Time: ✅ 120.460µs (SLO: <130.000µs -7.3%) vs baseline: -0.4%

Memory: ✅ 41.148MB (SLO: <46.000MB 📉 -10.5%) vs baseline: +5.9%

✅ import_many_stdlib

Time: ✅ 0.802ms (SLO: <1.750ms 📉 -54.2%) vs baseline: +0.5%

Memory: ✅ 41.095MB (SLO: <46.000MB 📉 -10.7%) vs baseline: +5.4%

⚠️ import_many_stdlib_cached

Time: ⚠️ 0.185ms (SLO: <1.100ms 📉 -83.1%) vs baseline: -0.9%

Memory: ✅ 41.107MB (SLO: <46.000MB 📉 -10.6%) vs baseline: +5.8%

✅ import_many_unknown

Time: ✅ 831.434µs (SLO: <890.000µs -6.6%) vs baseline: -0.2%

Memory: ✅ 41.272MB (SLO: <46.000MB 📉 -10.3%) vs baseline: +5.2%

✅ import_many_unknown_cached

Time: ✅ 788.813µs (SLO: <870.000µs -9.3%) vs baseline: +0.1%

Memory: ✅ 41.168MB (SLO: <46.000MB 📉 -10.5%) vs baseline: +5.1%

✅ import_one

Time: ✅ 19.916µs (SLO: <30.000µs 📉 -33.6%) vs baseline: +0.7%

Memory: ✅ 41.048MB (SLO: <46.000MB 📉 -10.8%) vs baseline: +5.4%

✅ import_one_cache

Time: ✅ 6.324µs (SLO: <10.000µs 📉 -36.8%) vs baseline: +0.8%

Memory: ✅ 41.068MB (SLO: <46.000MB 📉 -10.7%) vs baseline: +6.6%

✅ import_one_stdlib

Time: ✅ 18.757µs (SLO: <20.000µs -6.2%) vs baseline: -0.1%

Memory: ✅ 41.074MB (SLO: <46.000MB 📉 -10.7%) vs baseline: +5.9%

✅ import_one_stdlib_cache

Time: ✅ 6.270µs (SLO: <10.000µs 📉 -37.3%) vs baseline: -0.7%

Memory: ✅ 40.970MB (SLO: <46.000MB 📉 -10.9%) vs baseline: +6.3%

✅ import_one_unknown

Time: ✅ 45.893µs (SLO: <50.000µs -8.2%) vs baseline: +0.6%

Memory: ✅ 40.924MB (SLO: <46.000MB 📉 -11.0%) vs baseline: +4.8%

✅ import_one_unknown_cache

Time: ✅ 6.322µs (SLO: <10.000µs 📉 -36.8%) vs baseline: +0.7%

Memory: ✅ 41.003MB (SLO: <43.000MB -4.6%) vs baseline: +6.0%

✅ All Tests Passing (16 suites)

✅ codeprovenancefork - 2/2

✅ fork-10

Time: ✅ 2.187s (SLO: <2.300s -4.9%) vs baseline: +3.3%

Memory: ✅ 17.400MB (SLO: <20.000MB 📉 -13.0%) vs baseline: +4.9%

✅ errortrackingdjangosimple - 6/6

✅ errortracking-enabled-all

Time: ✅ 16.180ms (SLO: <19.850ms 📉 -18.5%) vs baseline: +0.1%

Memory: ✅ 68.872MB (SLO: <75.000MB -8.2%) vs baseline: +5.0%

✅ errortracking-enabled-user

Time: ✅ 16.171ms (SLO: <19.400ms 📉 -16.6%) vs baseline: -0.2%

Memory: ✅ 68.891MB (SLO: <75.000MB -8.1%) vs baseline: +5.1%

✅ tracer-enabled

Time: ✅ 16.179ms (SLO: <19.450ms 📉 -16.8%) vs baseline: -0.3%

Memory: ✅ 68.889MB (SLO: <75.000MB -8.1%) vs baseline: +5.1%

✅ errortrackingflasksqli - 6/6

✅ errortracking-enabled-all

Time: ✅ 2.122ms (SLO: <2.300ms -7.7%) vs baseline: +0.2%

Memory: ✅ 56.381MB (SLO: <60.000MB -6.0%) vs baseline: +5.1%

✅ errortracking-enabled-user

Time: ✅ 2.135ms (SLO: <2.250ms -5.1%) vs baseline: +0.7%

Memory: ✅ 56.443MB (SLO: <60.000MB -5.9%) vs baseline: +5.3%

✅ tracer-enabled

Time: ✅ 2.127ms (SLO: <2.300ms -7.5%) vs baseline: +0.8%

Memory: ✅ 55.964MB (SLO: <60.000MB -6.7%) vs baseline: +4.3%

✅ flasksimple - 18/18

✅ appsec-get

Time: ✅ 3.414ms (SLO: <4.750ms 📉 -28.1%) vs baseline: +0.5%

Memory: ✅ 56.584MB (SLO: <66.500MB 📉 -14.9%) vs baseline: +5.6%

✅ appsec-post

Time: ✅ 2.906ms (SLO: <6.750ms 📉 -56.9%) vs baseline: +0.3%

Memory: ✅ 56.545MB (SLO: <66.500MB 📉 -15.0%) vs baseline: +5.5%

✅ appsec-telemetry

Time: ✅ 3.440ms (SLO: <4.750ms 📉 -27.6%) vs baseline: +0.9%

Memory: ✅ 56.543MB (SLO: <66.500MB 📉 -15.0%) vs baseline: +5.5%

✅ debugger

Time: ✅ 1.888ms (SLO: <2.000ms -5.6%) vs baseline: +0.3%

Memory: ✅ 49.436MB (SLO: <51.500MB -4.0%) vs baseline: +6.1%

✅ iast-get

Time: ✅ 1.871ms (SLO: <2.000ms -6.4%) vs baseline: +0.1%

Memory: ✅ 46.095MB (SLO: <49.000MB -5.9%) vs baseline: +5.8%

✅ profiler

Time: ✅ 1.912ms (SLO: <2.100ms -9.0%) vs baseline: ~same

Memory: ✅ 52.030MB (SLO: <53.500MB -2.7%) vs baseline: +5.7%

✅ resource-renaming

Time: ✅ 3.384ms (SLO: <3.650ms -7.3%) vs baseline: ~same

Memory: ✅ 56.543MB (SLO: <60.000MB -5.8%) vs baseline: +5.5%

✅ tracer

Time: ✅ 3.398ms (SLO: <3.650ms -6.9%) vs baseline: -0.1%

Memory: ✅ 56.573MB (SLO: <60.000MB -5.7%) vs baseline: +5.6%

✅ tracer-native

Time: ✅ 3.405ms (SLO: <3.650ms -6.7%) vs baseline: +0.2%

Memory: ✅ 56.604MB (SLO: <60.000MB -5.7%) vs baseline: +5.7%

✅ flasksqli - 6/6

✅ appsec-enabled

Time: ✅ 2.108ms (SLO: <4.200ms 📉 -49.8%) vs baseline: -0.3%

Memory: ✅ 56.330MB (SLO: <66.000MB 📉 -14.7%) vs baseline: +4.9%

✅ iast-enabled

Time: ✅ 2.123ms (SLO: <2.800ms 📉 -24.2%) vs baseline: +0.3%

Memory: ✅ 56.272MB (SLO: <62.500MB -10.0%) vs baseline: +4.8%

✅ tracer-enabled

Time: ✅ 2.107ms (SLO: <2.250ms -6.3%) vs baseline: ~same

Memory: ✅ 56.368MB (SLO: <60.000MB -6.1%) vs baseline: +5.0%

✅ forktime - 4/4

✅ baseline

Time: ✅ 1.953ms (SLO: <3.000ms 📉 -34.9%) vs baseline: +5.1%

Memory: ✅ 29.216MB (SLO: <33.000MB 📉 -11.5%) vs baseline: +4.9%

✅ configured

Time: ✅ 8.787ms (SLO: <13.000ms 📉 -32.4%) vs baseline: +0.2%

Memory: ✅ 56.448MB (SLO: <60.000MB -5.9%) vs baseline: +5.8%

✅ httppropagationextract - 60/60

✅ all_styles_all_headers

Time: ✅ 80.881µs (SLO: <100.000µs 📉 -19.1%) vs baseline: +5.5%

Memory: ✅ 36.294MB (SLO: <38.000MB -4.5%) vs baseline: +5.3%

✅ b3_headers

Time: ✅ 12.855µs (SLO: <20.000µs 📉 -35.7%) vs baseline: +0.7%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +4.9%

✅ b3_single_headers

Time: ✅ 11.896µs (SLO: <20.000µs 📉 -40.5%) vs baseline: +0.6%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.5%

✅ datadog_tracecontext_tracestate_not_propagated_on_trace_id_no_match

Time: ✅ 60.928µs (SLO: <80.000µs 📉 -23.8%) vs baseline: +1.3%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.4%

✅ datadog_tracecontext_tracestate_propagated_on_trace_id_match

Time: ✅ 64.138µs (SLO: <80.000µs 📉 -19.8%) vs baseline: +0.7%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ empty_headers

Time: ✅ 1.295µs (SLO: <10.000µs 📉 -87.1%) vs baseline: -0.4%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.3%

✅ full_t_id_datadog_headers

Time: ✅ 21.584µs (SLO: <30.000µs 📉 -28.1%) vs baseline: -0.1%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ invalid_priority_header

Time: ✅ 5.868µs (SLO: <10.000µs 📉 -41.3%) vs baseline: -1.3%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ invalid_span_id_header

Time: ✅ 5.897µs (SLO: <10.000µs 📉 -41.0%) vs baseline: -0.2%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.2%

✅ invalid_tags_header

Time: ✅ 5.870µs (SLO: <10.000µs 📉 -41.3%) vs baseline: -0.3%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.3%

✅ invalid_trace_id_header

Time: ✅ 5.938µs (SLO: <10.000µs 📉 -40.6%) vs baseline: +0.5%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.4%

✅ large_header_no_matches

Time: ✅ 26.998µs (SLO: <30.000µs 📉 -10.0%) vs baseline: +0.3%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.3%

✅ large_valid_headers_all

Time: ✅ 28.050µs (SLO: <40.000µs 📉 -29.9%) vs baseline: ~same

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ medium_header_no_matches

Time: ✅ 9.217µs (SLO: <20.000µs 📉 -53.9%) vs baseline: ~same

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.4%

✅ medium_valid_headers_all

Time: ✅ 10.664µs (SLO: <20.000µs 📉 -46.7%) vs baseline: -0.3%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.2%

✅ none_propagation_style

Time: ✅ 1.384µs (SLO: <10.000µs 📉 -86.2%) vs baseline: -1.7%

Memory: ✅ 36.313MB (SLO: <38.000MB -4.4%) vs baseline: +5.1%

✅ tracecontext_headers

Time: ✅ 32.825µs (SLO: <40.000µs 📉 -17.9%) vs baseline: ~same

Memory: ✅ 36.313MB (SLO: <38.000MB -4.4%) vs baseline: +5.3%

✅ valid_headers_all

Time: ✅ 5.886µs (SLO: <10.000µs 📉 -41.1%) vs baseline: -0.7%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ valid_headers_basic

Time: ✅ 5.471µs (SLO: <10.000µs 📉 -45.3%) vs baseline: ~same

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.1%

✅ wsgi_empty_headers

Time: ✅ 1.296µs (SLO: <10.000µs 📉 -87.0%) vs baseline: +0.3%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.0%

✅ wsgi_invalid_priority_header

Time: ✅ 5.959µs (SLO: <10.000µs 📉 -40.4%) vs baseline: +0.3%

Memory: ✅ 36.294MB (SLO: <38.000MB -4.5%) vs baseline: +5.4%

✅ wsgi_invalid_span_id_header

Time: ✅ 1.293µs (SLO: <10.000µs 📉 -87.1%) vs baseline: -0.4%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.3%

✅ wsgi_invalid_tags_header

Time: ✅ 5.961µs (SLO: <10.000µs 📉 -40.4%) vs baseline: ~same

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.3%

✅ wsgi_invalid_trace_id_header

Time: ✅ 5.966µs (SLO: <10.000µs 📉 -40.3%) vs baseline: -0.4%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.3%

✅ wsgi_large_header_no_matches

Time: ✅ 28.159µs (SLO: <40.000µs 📉 -29.6%) vs baseline: +0.2%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ wsgi_large_valid_headers_all

Time: ✅ 29.153µs (SLO: <40.000µs 📉 -27.1%) vs baseline: -0.4%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.1%

✅ wsgi_medium_header_no_matches

Time: ✅ 9.530µs (SLO: <20.000µs 📉 -52.4%) vs baseline: +0.1%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.2%

✅ wsgi_medium_valid_headers_all

Time: ✅ 11.008µs (SLO: <20.000µs 📉 -45.0%) vs baseline: +0.2%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.3%

✅ wsgi_valid_headers_all

Time: ✅ 5.922µs (SLO: <10.000µs 📉 -40.8%) vs baseline: -0.5%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.1%

✅ wsgi_valid_headers_basic

Time: ✅ 5.525µs (SLO: <10.000µs 📉 -44.7%) vs baseline: -0.2%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.0%

✅ httppropagationinject - 16/16

✅ ids_only

Time: ✅ 20.861µs (SLO: <30.000µs 📉 -30.5%) vs baseline: +4.0%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ with_all

Time: ✅ 27.234µs (SLO: <40.000µs 📉 -31.9%) vs baseline: +1.0%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.1%

✅ with_dd_origin

Time: ✅ 23.970µs (SLO: <30.000µs 📉 -20.1%) vs baseline: ~same

Memory: ✅ 36.176MB (SLO: <38.000MB -4.8%) vs baseline: +4.7%

✅ with_priority_and_origin

Time: ✅ 23.379µs (SLO: <40.000µs 📉 -41.6%) vs baseline: +0.4%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ with_sampling_priority

Time: ✅ 20.236µs (SLO: <30.000µs 📉 -32.5%) vs baseline: ~same

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ with_tags

Time: ✅ 25.246µs (SLO: <40.000µs 📉 -36.9%) vs baseline: -0.3%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ with_tags_invalid

Time: ✅ 26.646µs (SLO: <40.000µs 📉 -33.4%) vs baseline: +0.4%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ with_tags_max_size

Time: ✅ 25.765µs (SLO: <40.000µs 📉 -35.6%) vs baseline: -0.1%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.1%

✅ iastpropagation - 8/8

✅ no-propagation

Time: ✅ 48.778µs (SLO: <60.000µs 📉 -18.7%) vs baseline: +1.0%

Memory: ✅ 40.088MB (SLO: <42.000MB -4.6%) vs baseline: +5.6%

✅ propagation_enabled

Time: ✅ 135.195µs (SLO: <190.000µs 📉 -28.8%) vs baseline: +0.9%

Memory: ✅ 40.187MB (SLO: <42.000MB -4.3%) vs baseline: +5.8%

✅ propagation_enabled_100

Time: ✅ 1.550ms (SLO: <2.300ms 📉 -32.6%) vs baseline: +0.3%

Memory: ✅ 39.931MB (SLO: <42.000MB -4.9%) vs baseline: +4.8%

✅ propagation_enabled_1000

Time: ✅ 28.908ms (SLO: <34.550ms 📉 -16.3%) vs baseline: -0.6%

Memory: ✅ 40.029MB (SLO: <42.000MB -4.7%) vs baseline: +5.4%

✅ otelsdkspan - 24/24

✅ add-event

Time: ✅ 40.706ms (SLO: <42.000ms -3.1%) vs baseline: ~same

Memory: ✅ 39.007MB (SLO: <40.750MB -4.3%) vs baseline: +5.9%

✅ add-link

Time: ✅ 36.844ms (SLO: <38.550ms -4.4%) vs baseline: +0.6%

Memory: ✅ 39.145MB (SLO: <40.750MB -3.9%) vs baseline: +6.3%

✅ add-metrics

Time: ✅ 221.147ms (SLO: <232.000ms -4.7%) vs baseline: +0.8%

Memory: ✅ 39.066MB (SLO: <40.750MB -4.1%) vs baseline: +6.0%

✅ add-tags

Time: ✅ 212.489ms (SLO: <221.600ms -4.1%) vs baseline: +0.7%

Memory: ✅ 39.007MB (SLO: <40.750MB -4.3%) vs baseline: +6.2%

✅ get-context

Time: ✅ 29.205ms (SLO: <31.300ms -6.7%) vs baseline: +0.2%

Memory: ✅ 38.771MB (SLO: <40.750MB -4.9%) vs baseline: +5.6%

✅ is-recording

Time: ✅ 29.246ms (SLO: <31.000ms -5.7%) vs baseline: +0.7%

Memory: ✅ 38.712MB (SLO: <40.750MB -5.0%) vs baseline: +5.3%

✅ record-exception

Time: ✅ 63.651ms (SLO: <65.850ms -3.3%) vs baseline: -0.2%

Memory: ✅ 39.007MB (SLO: <40.750MB -4.3%) vs baseline: +5.9%

✅ set-status

Time: ✅ 32.128ms (SLO: <34.150ms -5.9%) vs baseline: -0.2%

Memory: ✅ 39.007MB (SLO: <40.750MB -4.3%) vs baseline: +6.0%

✅ start

Time: ✅ 29.408ms (SLO: <30.150ms -2.5%) vs baseline: +2.0%

Memory: ✅ 38.732MB (SLO: <40.750MB -5.0%) vs baseline: +5.0%

✅ start-finish

Time: ✅ 33.934ms (SLO: <35.350ms -4.0%) vs baseline: -0.5%

Memory: ✅ 38.987MB (SLO: <40.750MB -4.3%) vs baseline: +5.7%

✅ start-finish-telemetry

Time: ✅ 34.143ms (SLO: <35.450ms -3.7%) vs baseline: -0.4%

Memory: ✅ 39.046MB (SLO: <40.750MB -4.2%) vs baseline: +6.4%

✅ update-name

Time: ✅ 31.196ms (SLO: <33.400ms -6.6%) vs baseline: +0.2%

Memory: ✅ 39.086MB (SLO: <40.750MB -4.1%) vs baseline: +6.2%

✅ packagespackageforrootmodulemapping - 4/4

✅ cache_off

Time: ✅ 345.597ms (SLO: <354.300ms -2.5%) vs baseline: +0.5%

Memory: ✅ 41.954MB (SLO: <46.000MB -8.8%) vs baseline: +5.3%

✅ cache_on

Time: ✅ 0.381µs (SLO: <10.000µs 📉 -96.2%) vs baseline: -1.2%

Memory: ✅ 39.967MB (SLO: <46.000MB 📉 -13.1%) vs baseline: +5.1%

✅ rand - 2/2

✅ rand128bits

Time: ✅ 0.182µs (SLO: <21.000µs 📉 -99.1%) vs baseline: -2.0%

✅ rand64bits

Time: ✅ 0.122µs (SLO: <15.000µs 📉 -99.2%) vs baseline: +3.7%

✅ ratelimiter - 11/11

✅ defaults

Time: ✅ 2.343µs (SLO: <10.000µs 📉 -76.6%) vs baseline: -0.9%

Memory: ✅ 36.549MB (SLO: <38.000MB -3.8%) vs baseline: +6.1%

✅ high_rate_limit

Time: ✅ 2.401µs (SLO: <10.000µs 📉 -76.0%) vs baseline: -0.4%

Memory: ✅ 36.549MB (SLO: <38.000MB -3.8%) vs baseline: +6.2%

✅ long_window

Time: ✅ 2.338µs (SLO: <10.000µs 📉 -76.6%) vs baseline: -1.3%

✅ low_rate_limit

Time: ✅ 2.361µs (SLO: <10.000µs 📉 -76.4%) vs baseline: ~same

Memory: ✅ 36.549MB (SLO: <38.000MB -3.8%) vs baseline: +6.5%

✅ no_rate_limit

Time: ✅ 0.826µs (SLO: <10.000µs 📉 -91.7%) vs baseline: -1.6%

Memory: ✅ 36.530MB (SLO: <38.000MB -3.9%) vs baseline: +6.2%

✅ short_window

Time: ✅ 2.481µs (SLO: <10.000µs 📉 -75.2%) vs baseline: -0.3%

Memory: ✅ 36.569MB (SLO: <38.000MB -3.8%) vs baseline: +6.2%

✅ recursivecomputation - 8/8

✅ deep

Time: ✅ 313.256ms (SLO: <320.950ms -2.4%) vs baseline: +0.2%

Memory: ✅ 37.375MB (SLO: <38.750MB -3.5%) vs baseline: +5.0%

✅ deep-profiled

Time: ✅ 327.473ms (SLO: <359.150ms -8.8%) vs baseline: -0.5%

Memory: ✅ 43.686MB (SLO: <46.000MB -5.0%) vs baseline: +5.3%

✅ medium

Time: ✅ 7.408ms (SLO: <7.400ms +0.1%) vs baseline: -0.3%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.1%

✅ shallow

Time: ✅ 1.053ms (SLO: <1.050ms +0.3%) vs baseline: +2.0%

Memory: ✅ 36.176MB (SLO: <38.000MB -4.8%) vs baseline: +5.1%

✅ sethttpmeta - 32/32

✅ all-disabled

Time: ✅ 10.589µs (SLO: <20.000µs 📉 -47.1%) vs baseline: -0.1%

Memory: ✅ 37.120MB (SLO: <38.750MB -4.2%) vs baseline: +5.5%

✅ all-enabled

Time: ✅ 39.858µs (SLO: <50.000µs 📉 -20.3%) vs baseline: +1.9%

Memory: ✅ 37.080MB (SLO: <38.750MB -4.3%) vs baseline: +5.2%

✅ collectipvariant_exists

Time: ✅ 39.793µs (SLO: <50.000µs 📉 -20.4%) vs baseline: +0.3%

Memory: ✅ 37.080MB (SLO: <38.750MB -4.3%) vs baseline: +5.6%

✅ no-collectipvariant

Time: ✅ 38.946µs (SLO: <50.000µs 📉 -22.1%) vs baseline: -0.1%

Memory: ✅ 37.002MB (SLO: <38.750MB -4.5%) vs baseline: +5.0%

✅ no-useragentvariant

Time: ✅ 37.678µs (SLO: <50.000µs 📉 -24.6%) vs baseline: ~same

Memory: ✅ 37.002MB (SLO: <38.750MB -4.5%) vs baseline: +5.1%

✅ obfuscation-no-query

Time: ✅ 39.409µs (SLO: <50.000µs 📉 -21.2%) vs baseline: -0.2%

Memory: ✅ 37.120MB (SLO: <38.750MB -4.2%) vs baseline: +5.6%

✅ obfuscation-regular-case-explicit-query

Time: ✅ 76.039µs (SLO: <90.000µs 📉 -15.5%) vs baseline: ~same

Memory: ✅ 37.454MB (SLO: <38.750MB -3.3%) vs baseline: +6.0%

✅ obfuscation-regular-case-implicit-query

Time: ✅ 76.484µs (SLO: <90.000µs 📉 -15.0%) vs baseline: ~same

Memory: ✅ 37.434MB (SLO: <38.750MB -3.4%) vs baseline: +5.9%

✅ obfuscation-send-querystring-disabled

Time: ✅ 154.519µs (SLO: <170.000µs -9.1%) vs baseline: ~same

Memory: ✅ 37.415MB (SLO: <38.750MB -3.4%) vs baseline: +5.1%

✅ obfuscation-worst-case-explicit-query

Time: ✅ 149.016µs (SLO: <160.000µs -6.9%) vs baseline: ~same

Memory: ✅ 37.415MB (SLO: <38.750MB -3.4%) vs baseline: +5.2%

✅ obfuscation-worst-case-implicit-query

Time: ✅ 155.334µs (SLO: <170.000µs -8.6%) vs baseline: -0.1%

Memory: ✅ 37.415MB (SLO: <38.750MB -3.4%) vs baseline: +5.0%

✅ useragentvariant_exists_1

Time: ✅ 38.485µs (SLO: <50.000µs 📉 -23.0%) vs baseline: +0.4%

Memory: ✅ 37.041MB (SLO: <38.750MB -4.4%) vs baseline: +5.2%

✅ useragentvariant_exists_2

Time: ✅ 39.479µs (SLO: <50.000µs 📉 -21.0%) vs baseline: +0.3%

Memory: ✅ 37.080MB (SLO: <38.750MB -4.3%) vs baseline: +5.1%

✅ useragentvariant_exists_3

Time: ✅ 39.027µs (SLO: <50.000µs 📉 -21.9%) vs baseline: -0.3%

Memory: ✅ 37.080MB (SLO: <38.750MB -4.3%) vs baseline: +5.2%

✅ useragentvariant_not_exists_1

Time: ✅ 38.410µs (SLO: <50.000µs 📉 -23.2%) vs baseline: -0.1%

Memory: ✅ 37.002MB (SLO: <38.750MB -4.5%) vs baseline: +5.2%

✅ useragentvariant_not_exists_2

Time: ✅ 38.389µs (SLO: <50.000µs 📉 -23.2%) vs baseline: -0.3%

Memory: ✅ 37.120MB (SLO: <38.750MB -4.2%) vs baseline: +5.4%

✅ telemetryaddmetric - 29/29

✅ 1-count-metric-1-times

Time: ✅ 2.320µs (SLO: <20.000µs 📉 -88.4%) vs baseline: +9.5%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.4%

✅ 1-count-metrics-100-times

Time: ✅ 154.578µs (SLO: <220.000µs 📉 -29.7%) vs baseline: +1.4%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.4%

✅ 1-distribution-metric-1-times

Time: ✅ 2.425µs (SLO: <20.000µs 📉 -87.9%) vs baseline: -3.1%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.4%

✅ 1-distribution-metrics-100-times

Time: ✅ 163.452µs (SLO: <230.000µs 📉 -28.9%) vs baseline: -0.7%

Memory: ✅ 36.274MB (SLO: <38.000MB -4.5%) vs baseline: +5.7%

✅ 1-gauge-metric-1-times

Time: ✅ 1.982µs (SLO: <20.000µs 📉 -90.1%) vs baseline: -2.6%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +4.9%

✅ 1-gauge-metrics-100-times

Time: ✅ 136.597µs (SLO: <150.000µs -8.9%) vs baseline: -0.6%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.3%

✅ 1-rate-metric-1-times

Time: ✅ 2.313µs (SLO: <20.000µs 📉 -88.4%) vs baseline: +2.3%

Memory: ✅ 36.215MB (SLO: <38.000MB -4.7%) vs baseline: +5.1%

✅ 1-rate-metrics-100-times

Time: ✅ 168.556µs (SLO: <250.000µs 📉 -32.6%) vs baseline: +2.3%

Memory: ✅ 36.255MB (SLO: <38.000MB -4.6%) vs baseline: +5.5%

✅ 100-count-metrics-100-times

Time: ✅ 15.482ms (SLO: <22.000ms 📉 -29.6%) vs baseline: +1.4%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.3%

✅ 100-distribution-metrics-100-times

Time: ✅ 1.728ms (SLO: <2.550ms 📉 -32.2%) vs baseline: -1.3%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.2%

✅ 100-gauge-metrics-100-times

Time: ✅ 1.417ms (SLO: <1.550ms -8.6%) vs baseline: +1.0%

Memory: ✅ 36.235MB (SLO: <38.000MB -4.6%) vs baseline: +5.5%

✅ 100-rate-metrics-100-times

Time: ✅ 1.744ms (SLO: <2.550ms 📉 -31.6%) vs baseline: +1.4%

Memory: ✅ 36.196MB (SLO: <38.000MB -4.7%) vs baseline: +5.0%

✅ flush-1-metric

Time: ✅ 3.641µs (SLO: <20.000µs 📉 -81.8%) vs baseline: +0.8%

✅ flush-100-metrics

Time: ✅ 175.845µs (SLO: <250.000µs 📉 -29.7%) vs baseline: ~same

Memory: ✅ 36.628MB (SLO: <38.000MB -3.6%) vs baseline: +6.6%

✅ flush-1000-metrics

Time: ✅ 2.204ms (SLO: <2.500ms 📉 -11.8%) vs baseline: ~same

Memory: ✅ 37.002MB (SLO: <38.750MB -4.5%) vs baseline: +5.4%

ℹ️ Scenarios Missing SLO Configuration (46 scenarios)

The following scenarios exist in candidate data but have no SLO thresholds configured:

coreapiscenario-core_dispatch_listeners
coreapiscenario-core_dispatch_no_listeners
coreapiscenario-core_dispatch_with_results_listeners
coreapiscenario-core_dispatch_with_results_no_listeners
djangosimple-baseline
errortrackingdjangosimple-baseline
errortrackingflasksqli-baseline
flasksimple-baseline
flasksqli-baseline
iast_aspects-re_expand_aspect
iast_aspects-re_expand_noaspect
iast_aspects-re_findall_aspect
iast_aspects-re_findall_noaspect
iast_aspects-re_finditer_aspect
iast_aspects-re_finditer_noaspect
iast_aspects-re_fullmatch_aspect
iast_aspects-re_fullmatch_noaspect
iast_aspects-re_group_aspect
iast_aspects-re_group_noaspect
iast_aspects-re_groups_aspect
iast_aspects-re_groups_noaspect
iast_aspects-re_match_aspect
iast_aspects-re_match_noaspect
iast_aspects-re_search_aspect
iast_aspects-re_search_noaspect
iast_aspects-re_sub_aspect
iast_aspects-re_sub_noaspect
iast_aspects-re_subn_aspect
iast_aspects-re_subn_noaspect
sethttpmeta-obfuscation-disabled
startup-baseline
startup-baseline_django
startup-baseline_flask
startup-ddtrace_run
startup-ddtrace_run_appsec
startup-ddtrace_run_profiling
startup-ddtrace_run_runtime_metrics
startup-ddtrace_run_send_span
startup-ddtrace_run_telemetry_disabled
startup-ddtrace_run_telemetry_enabled
startup-import_ddtrace
startup-import_ddtrace_auto
startup-import_ddtrace_auto_django
startup-import_ddtrace_auto_flask
startup-import_ddtrace_django
startup-import_ddtrace_flask

cit-pr-commenter-54b7da · 2026-04-10T15:26:37Z

Codeowners resolved as

ddtrace/internal/datadog/profiling/stack/echion/echion/stacks.h         @DataDog/profiling-python
ddtrace/internal/datadog/profiling/stack/src/echion/stacks.cc           @DataDog/profiling-python
ddtrace/internal/datadog/profiling/stack/src/echion/tasks.cc            @DataDog/profiling-python
ddtrace/internal/datadog/profiling/stack/src/echion/threads.cc          @DataDog/profiling-python
releasenotes/notes/profiling-fix-asyncio-unwinding-crash-b192aef728433e7d.yaml  @DataDog/apm-python

KowalskiThomas · 2026-04-13T14:38:45Z

@codex review

chatgpt-codex-connector · 2026-04-13T14:42:04Z

Codex Review: Didn't find any major issues. Swish!

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

vlad-scherbich · 2026-04-13T14:54:50Z

asyncio/FastAPI (n=3)

Have you tried a higher n, e.g. 10, which gives high confidence in the results?

KowalskiThomas · 2026-04-13T14:55:48Z

Have you tried a higher n, e.g. 10, which gives high confidence in the results?

Variance was already extremely low so no, I didn't. I can of course, but I honestly don't expect this to change anything.

vlad-scherbich · 2026-04-13T14:56:00Z

Do you have the numbers for memory usage from DoE, since this is what we're potentially affecting? It's not in the summary table.

KowalskiThomas · 2026-04-13T15:01:17Z

Do you have the numbers for memory usage from DoE, since this is what we're potentially affecting? It's not in the summary table.

Added to the description! Approximately +1% on average in the three runs but it might be from the outlier.

Co-authored-by: Vlad Scherbich <vlad.scherbich@datadoghq.com>

https://datadoghq.atlassian.net/browse/PROF-13112 This fixes a crash in the Profiler codebase. The crash was caused by `ThreadInfo::unwind_task` trying to use `Frame` objects that it had a reference to (`Frame::Ref`) while those same Frames were being evicted from the cache, which can happen under high load. ``` Error UnixSignal: Process terminated with SI_KERNEL (SIGSEGV) ``` The proposed fix is to change `FrameStack` to extend `std::deque<Frame>` instead of `std::deque<Frame::Ref>`, which means pushing frames to `FrameStack` actually copies them and eliminates that race condition. This should be OK performance-wise because `Frame` is tiny and only has trivial fields to copy (no memory allocations, only integral types). Another possible fix would be to "lock" certain `Frame` objects in the cache (to prevent them from being evicted while any Task using them is being unwound) but this would be much more complicated to get right and although it would lead to less copies, the performance picture is blurry for more subtle reasons. So I suggest we don't go ahead with the alternative unless we absolutely need to (which we don't seem to). Running DoE with this change on the Enterprise archetype with asyncio/FastAPI (n=10) shows no significant difference (either in memory or CPU usage / latency), so I think this is safe to merge performance-wise. <details> | Commit | Run | p50 (ms) | p99 (ms) | CPU% | Mem (MB) | |--------|-----|----------|----------|------|----------| | `90de7913` (after) | 1 | 150.43 | 152.88 | 172.68 | 274.9 | | `90de7913` (after) | 2 | 150.45 | 152.38 | 172.17 | 266.2 | | `90de7913` (after) | 3 | 150.43 | 153.91 | 172.92 | 266.2 | | `90de7913` (after) | 4 | 150.46 | 154.16 | 172.83 | 265.8 | | `90de7913` (after) | 5 | 150.44 | 153.01 | 172.80 | 267.4 | | `90de7913` (after) | 6 | 150.43 | 153.28 | 172.70 | 265.8 | | `90de7913` (after) | 7 | 150.54 | 157.48 | 173.32 | 265.8 | | `90de7913` (after) | 8 | 150.51 | 156.61 | 172.18 | 265.5 | | `90de7913` (after) | 9 | 150.53 | 157.05 | 172.18 | 265.8 | | `90de7913` (after) | 10 | 150.45 | 152.84 | 172.08 | 265.5 | | `c26e6181` (before) | 1 | 150.51 | 152.80 | 173.63 | 265.5 | | `c26e6181` (before) | 2 | 150.48 | 152.41 | 173.49 | 265.4 | | `c26e6181` (before) | 3 | 150.47 | 156.31 | 172.53 | 265.9 | | `c26e6181` (before) | 4 | 150.49 | 152.60 | 172.86 | 265.5 | | `c26e6181` (before) | 5 | 150.40 | 152.86 | 172.99 | 266.0 | | `c26e6181` (before) | 6 | 150.46 | 152.66 | 172.53 | 265.5 | | `c26e6181` (before) | 7 | 150.49 | 154.91 | 172.74 | 265.3 | | `c26e6181` (before) | 8 | 150.46 | 152.46 | 171.93 | 265.7 | | `c26e6181` (before) | 9 | 150.52 | 156.70 | 172.96 | 265.7 | | `c26e6181` (before) | 10 | 150.43 | 153.02 | 172.52 | 265.8 | | Version | p50 (ms) | p99 (ms) | CPU% | Mem (MB) | |---------|----------|----------|------|----------| | `c26e6181` (before) | 150.470 ± 0.034 | 153.67 ± 1.66 | 172.82 ± 0.50 | 265.6 ± 0.2 | | `90de7913` (after) | 150.465 ± 0.043 | 154.36 ± 1.94 | 172.59 ± 0.41 | 266.9 ± 2.9 | | **Δ (after − before)** | **−0.005ms (0.00%)** | **+0.69ms (+0.45%)** | **−0.23pp (−0.13%)** | **+1.2MB (+0.47%)** | </details> Co-authored-by: thomas.kowalski <thomas.kowalski@datadoghq.com>

…ort 4.7] (#17600) https://datadoghq.atlassian.net/browse/PROF-13112 Backport of #17456. Co-authored-by: Thomas Kowalski <thomas.kowalski@datadoghq.com>

…17599) https://datadoghq.atlassian.net/browse/PROF-13112 Backport of #17456. Co-authored-by: Thomas Kowalski <thomas.kowalski@datadoghq.com>

KowalskiThomas added Profiling Continous Profling identified-by:crashtracking Identified by Crash Tracking labels Apr 10, 2026

KowalskiThomas force-pushed the kowalski/fix-profiling-fix-crash-due-to-frame-cache-eviction branch 2 times, most recently from 50c3976 to 1113136 Compare April 10, 2026 15:25

This comment has been minimized.

Sign in to view

KowalskiThomas force-pushed the kowalski/fix-profiling-fix-crash-due-to-frame-cache-eviction branch from 1113136 to 730ee23 Compare April 13, 2026 14:38

KowalskiThomas marked this pull request as ready for review April 13, 2026 14:51

KowalskiThomas requested review from a team as code owners April 13, 2026 14:52

KowalskiThomas requested review from taegyunkim and vlad-scherbich April 13, 2026 14:52

vlad-scherbich reviewed Apr 13, 2026

View reviewed changes

Comment thread releasenotes/notes/profiling-fix-asyncio-unwinding-crash-b192aef728433e7d.yaml Outdated

fix(profiling): fix crash due to Frame cache eviction

bb2d1ca

KowalskiThomas force-pushed the kowalski/fix-profiling-fix-crash-due-to-frame-cache-eviction branch from 730ee23 to bb2d1ca Compare April 13, 2026 15:14

KowalskiThomas requested a review from vlad-scherbich April 13, 2026 15:14

vlad-scherbich reviewed Apr 13, 2026

View reviewed changes

Comment thread ddtrace/internal/datadog/profiling/stack/src/echion/threads.cc Outdated

review: apply Vlad's redundant code suggestion

721b181

Co-authored-by: Vlad Scherbich <vlad.scherbich@datadoghq.com>

KowalskiThomas requested a review from vlad-scherbich April 13, 2026 15:19

vlad-scherbich approved these changes Apr 13, 2026

View reviewed changes

gh-worker-dd-mergequeue-cf854d Bot merged commit ac76f69 into main Apr 13, 2026
428 checks passed

gh-worker-dd-mergequeue-cf854d Bot deleted the kowalski/fix-profiling-fix-crash-due-to-frame-cache-eviction branch April 13, 2026 20:26

This was referenced Apr 17, 2026

fix(profiling): fix crash due to Frame cache eviction [backport 4.6] #17599

Merged

fix(profiling): fix crash due to Frame cache eviction (#17456) [backport 4.7] #17600

Merged

Conversation

KowalskiThomas commented Apr 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Crash

Proposed fix

Performance

Raw runs

Aggregated (mean ± stdev, n=10)

Uh oh!

pr-commenter Bot commented Apr 10, 2026

Performance SLOs

✅ add_aspect

✅ add_inplace_aspect

✅ add_inplace_noaspect

✅ add_noaspect

✅ bytearray_aspect

✅ bytearray_extend_aspect

✅ bytearray_extend_noaspect

✅ bytearray_noaspect

✅ bytes_aspect

✅ bytes_noaspect

✅ bytesio_aspect

✅ bytesio_noaspect

✅ capitalize_aspect

✅ capitalize_noaspect

✅ casefold_aspect

✅ casefold_noaspect

✅ decode_aspect

✅ decode_noaspect

✅ encode_aspect

✅ encode_noaspect

✅ format_aspect

✅ format_map_aspect

✅ format_map_noaspect

✅ format_noaspect

✅ index_aspect

✅ index_noaspect

✅ join_aspect

✅ join_noaspect

✅ ljust_aspect

✅ ljust_noaspect

✅ lower_aspect

✅ lower_noaspect

✅ lstrip_aspect

✅ lstrip_noaspect

✅ modulo_aspect

✅ modulo_aspect_for_bytearray_bytearray

✅ modulo_aspect_for_bytes

✅ modulo_aspect_for_bytes_bytearray

✅ modulo_noaspect

✅ replace_aspect

✅ replace_noaspect

✅ repr_aspect

✅ repr_noaspect

✅ rstrip_aspect

✅ rstrip_noaspect

✅ slice_aspect

✅ slice_noaspect

✅ stringio_aspect

✅ stringio_noaspect

✅ strip_aspect

✅ strip_noaspect

✅ swapcase_aspect

✅ swapcase_noaspect

✅ title_aspect

✅ title_noaspect

✅ translate_aspect

✅ translate_noaspect

✅ upper_aspect

✅ upper_noaspect

✅ ospathbasename_aspect

✅ ospathbasename_noaspect

✅ ospathjoin_aspect

✅ ospathjoin_noaspect

✅ ospathnormcase_aspect

✅ ospathnormcase_noaspect

✅ ospathsplit_aspect

✅ ospathsplit_noaspect

✅ ospathsplitdrive_aspect

KowalskiThomas commented Apr 10, 2026 •

edited

Loading