[robert3005]

VTable::validate has no context, but FSST validation executed the last codes offset to check it fits in the codes bytes, using legacy_session(). Validation is now split:

• FSSTData::validate (called by VTable::validate) runs only the structural checks and needs no context.
• The codes-bounds check moves to validate_codes_bounds, which takes an ExecutionCtx. It runs in FSST::try_new, FSST::try_new_with_symbol_table and both deserialize paths. Deserialize builds its context from the session.

part of #9446

[codspeed]

Merging this PR will degrade performance by 14.27%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ 1 benchmark measured no execution time

Nothing ran under measurement, usually because the compiler removed the code under test. This result is not comparable, so it counts as unchanged.

Preventing compiler optimizations

• bench_compare_primitive[(10000, 2)]

⚠️ 18 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• test_scan_filter[int]
• random_i128[0.01]
• test_scan_filter[int_str]
• compress_fsst[(500, 16, 4)]
• decompress[u32, (1000, 256)]
• decompress[u16, (4000, 256)]
• test_scan_filter[int_float]
• test_scan_filter[int_str_float]
• pcodec_compress_f64
• zstd_compress_string
• …and 8 more

ℹ️ 40 benchmarks have no flame graph

The flamegraph could not be generated because we failed to resolve the symbols in the profiling trace. This happens when we fail to harvest information about language frames.

• test_repeated_scan_scalar_at[int_str]
• test_scan_filter_to_arrow_array[int_float]
• test_scan_filter[int]
• test_repeated_scan_filter[int_float]
• test_repeated_scan_filter[int]
• test_scan_filter_to_arrow_array[int]
• test_repeated_scan_scalar_at[int]
• test_scan_filter[int_str]
• test_repeated_scan_scalar_at[int_float]
• test_scan_scalar_at[int_str_float]
• …and 30 more

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 6 improved benchmarks
❌ 13 regressed benchmarks
✅ 2389 untouched benchmarks
⏩ 12 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	filter_neon[8]	933 ns	1,835 ns	-49.16%
❌ ℹ️	Simulation	test_repeated_scan_scalar_at[int_str]	442.9 µs	855.8 µs	-48.24%
❌	WallTime	threshold_neon[8]	1.5 µs	2.6 µs	-42.62%
❌	WallTime	filter_neon[64]	1.8 µs	2.9 µs	-38.18%
❌	WallTime	threshold_neon[64]	2.6 µs	4.1 µs	-36.41%
❌	WallTime	filter_neon[80]	2.8 µs	3.5 µs	-20.71%
❌	WallTime	threshold_neon[80]	3.9 µs	4.6 µs	-13.87%
❌ ℹ️	Simulation	test_scan_filter_to_arrow_array[int_float]	3.5 ms	4.1 ms	-13.42%
❌	Simulation	or_true_constant	17.8 µs	20.5 µs	-13.22%
❌ ⚠️	Simulation	test_scan_filter[int]	3.2 ms	3.6 ms	-12.96%
❌	WallTime	bitpack_blocked_compress_avx2	6.8 µs	7.8 µs	-12.49%
❌ ℹ️	Simulation	test_repeated_scan_filter[int_float]	3.4 ms	3.9 ms	-11.04%
❌ ℹ️	Simulation	test_repeated_scan_filter[int]	2.4 ms	2.7 ms	-10.51%
⚡	WallTime	mul_u64_nonnull_neon	39.8 µs	28.7 µs	+38.83%
⚡	WallTime	multiply_shapes_neon[(32768, PerRowPerRow)]	38.5 µs	32.5 µs	+18.44%
⚡	WallTime	mul_i64_nonnull_neon	38.4 µs	32.5 µs	+18.25%
⚡	Simulation	sum_i64	225.1 µs	194.6 µs	+15.71%
⚡	Simulation	sum_v2_i64	224.7 µs	194.2 µs	+15.68%
⚡	WallTime	mul_u16_nonnull_avx2	12.2 µs	10.9 µs	+11.62%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing rk/legacy-session-fsst (4c25064) with develop (31e0724)²}

Footnotes

1. 12 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on develop (f5f222a) during the generation of this report, so 31e0724 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩