Make some use of `CoalescingOutputIterator` by LebedevRI · Pull Request #675 · darktable-org/rawspeed

LebedevRI · 2024-02-20T03:43:47Z

No description provided.

codecov · 2024-02-20T22:37:17Z

Codecov Report

Attention: 8 lines in your changes are missing coverage. Please review.

Comparison is base (6f1a5bf) 60.94% compared to head (ae1a159) 60.86%.

Files	Patch %	Lines
src/librawspeed/bitstreams/BitVacuumerJPEG.h	0.00%	6 Missing ⚠️
src/librawspeed/adt/CoalescingOutputIterator.h	71.42%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #675      +/-   ##
===========================================
- Coverage    60.94%   60.86%   -0.08%     
===========================================
  Files          271      271              
  Lines        16327    16342      +15     
  Branches      2075     2077       +2     
===========================================
- Hits          9950     9947       -3     
- Misses        6249     6268      +19     
+ Partials       128      127       -1

Flag	Coverage Δ
benchmarks	`11.82% <75.00%> (-0.02%)`	⬇️
integration	`45.11% <0.00%> (-0.06%)`	⬇️
linux	`57.07% <67.74%> (-0.09%)`	⬇️
macOS	`25.07% <79.31%> (+0.01%)`	⬆️
rpu_u	`45.11% <0.00%> (-0.06%)`	⬇️
unittests	`21.46% <11.76%> (-0.08%)`	⬇️
windows	`∅ <ø> (∅)`

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

…rsal reference

…_type`

…cept

``` Comparing bench/librawspeed/bitstreams/BitVacuumerJPEGBenchmark-old to bench/librawspeed/bitstreams/BitVacuumerJPEGBenchmark Benchmark Time CPU Time Old Time New CPU Old CPU New -------------------------------------------------------------------------------------------------------------------------------------------------------------------- BM<BitstreamFlavorJPEG,NoCoalescing>/Stuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,NoCoalescing>/Stuffed/2097152_mean +0.0125 +0.0126 6205 6282 6204 6282 BM<BitstreamFlavorJPEG,NoCoalescing>/Stuffed/2097152_median +0.0125 +0.0127 6204 6282 6203 6282 BM<BitstreamFlavorJPEG,NoCoalescing>/Stuffed/2097152_stddev +3.9793 +3.9137 2 10 2 10 BM<BitstreamFlavorJPEG,NoCoalescing>/Stuffed/2097152_cv +3.9180 +3.8527 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Stuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Stuffed/2097152_mean -0.0349 -0.0348 6453 6228 6452 6227 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Stuffed/2097152_median -0.0338 -0.0338 6447 6229 6446 6228 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Stuffed/2097152_stddev -0.8191 -0.8192 41 7 41 7 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Stuffed/2097152_cv -0.8125 -0.8126 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Stuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Stuffed/2097152_mean -0.1021 -0.1020 5819 5225 5819 5225 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Stuffed/2097152_median -0.1015 -0.1015 5816 5225 5815 5225 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Stuffed/2097152_stddev -0.6999 -0.6951 10 3 10 3 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Stuffed/2097152_cv -0.6658 -0.6605 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Stuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Stuffed/2097152_mean -0.0482 -0.0482 5681 5408 5681 5407 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Stuffed/2097152_median -0.0482 -0.0482 5682 5408 5682 5408 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Stuffed/2097152_stddev +0.3751 +0.3802 3 4 3 4 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Stuffed/2097152_cv +0.4448 +0.4500 0 0 0 0 BM<BitstreamFlavorJPEG,NoCoalescing>/Unstuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,NoCoalescing>/Unstuffed/2097152_mean -0.0402 -0.0401 6508 6247 6508 6246 BM<BitstreamFlavorJPEG,NoCoalescing>/Unstuffed/2097152_median -0.0401 -0.0400 6508 6247 6508 6247 BM<BitstreamFlavorJPEG,NoCoalescing>/Unstuffed/2097152_stddev +15.3546 +21.8853 1 9 0 9 BM<BitstreamFlavorJPEG,NoCoalescing>/Unstuffed/2097152_cv +16.0395 +22.8421 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Unstuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Unstuffed/2097152_mean -0.0328 -0.0328 6414 6203 6413 6203 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Unstuffed/2097152_median -0.0317 -0.0318 6409 6206 6409 6205 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Unstuffed/2097152_stddev -0.7829 -0.7804 38 8 38 8 BM<BitstreamFlavorJPEG,CoalesceTo<uint16_t>>/Unstuffed/2097152_cv -0.7756 -0.7729 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Unstuffed/2097152_pvalue 0.0004 0.0004 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Unstuffed/2097152_mean -0.0982 -0.0983 5939 5356 5939 5355 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Unstuffed/2097152_median -0.0984 -0.0985 5939 5355 5939 5354 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Unstuffed/2097152_stddev +0.0997 +0.1317 3 4 3 4 BM<BitstreamFlavorJPEG,CoalesceTo<uint32_t>>/Unstuffed/2097152_cv +0.2194 +0.2550 0 0 0 0 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Unstuffed/2097152_pvalue 0.0062 0.0062 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Unstuffed/2097152_mean +0.0041 +0.0041 5545 5568 5545 5568 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Unstuffed/2097152_median +0.0044 +0.0045 5544 5569 5543 5568 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Unstuffed/2097152_stddev +5.5353 +5.1423 3 17 3 17 BM<BitstreamFlavorJPEG,CoalesceTo<uint64_t>>/Unstuffed/2097152_cv +5.5087 +5.1171 0 0 0 0 BM<BitstreamFlavorMSB,NoCoalescing>/Unstuffed/2097152_pvalue 0.0062 0.0062 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorMSB,NoCoalescing>/Unstuffed/2097152_mean +0.0048 +0.0048 6243 6273 6242 6272 BM<BitstreamFlavorMSB,NoCoalescing>/Unstuffed/2097152_median +0.0061 +0.0061 6235 6273 6234 6272 BM<BitstreamFlavorMSB,NoCoalescing>/Unstuffed/2097152_stddev -0.5808 -0.5713 18 7 18 8 BM<BitstreamFlavorMSB,NoCoalescing>/Unstuffed/2097152_cv -0.5827 -0.5733 0 0 0 0 BM<BitstreamFlavorMSB,CoalesceTo<uint16_t>>/Unstuffed/2097152_pvalue 0.0423 0.0423 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorMSB,CoalesceTo<uint16_t>>/Unstuffed/2097152_mean -0.0028 -0.0028 6017 6000 6016 5999 BM<BitstreamFlavorMSB,CoalesceTo<uint16_t>>/Unstuffed/2097152_median -0.0034 -0.0033 6022 6001 6021 6001 BM<BitstreamFlavorMSB,CoalesceTo<uint16_t>>/Unstuffed/2097152_stddev -0.7949 -0.7906 22 4 22 5 BM<BitstreamFlavorMSB,CoalesceTo<uint16_t>>/Unstuffed/2097152_cv -0.7944 -0.7900 0 0 0 0 BM<BitstreamFlavorMSB,CoalesceTo<uint32_t>>/Unstuffed/2097152_pvalue 0.0036 0.0036 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorMSB,CoalesceTo<uint32_t>>/Unstuffed/2097152_mean +0.0005 +0.0006 5109 5112 5108 5111 BM<BitstreamFlavorMSB,CoalesceTo<uint32_t>>/Unstuffed/2097152_median +0.0006 +0.0006 5108 5111 5108 5111 BM<BitstreamFlavorMSB,CoalesceTo<uint32_t>>/Unstuffed/2097152_stddev -0.3388 -0.2929 2 1 1 1 BM<BitstreamFlavorMSB,CoalesceTo<uint32_t>>/Unstuffed/2097152_cv -0.3391 -0.2933 0 0 0 0 BM<BitstreamFlavorMSB,CoalesceTo<uint64_t>>/Unstuffed/2097152_pvalue 0.0047 0.0047 U Test, Repetitions: 9 vs 9 BM<BitstreamFlavorMSB,CoalesceTo<uint64_t>>/Unstuffed/2097152_mean -0.0023 -0.0022 5055 5043 5054 5043 BM<BitstreamFlavorMSB,CoalesceTo<uint64_t>>/Unstuffed/2097152_median -0.0024 -0.0023 5056 5044 5056 5044 BM<BitstreamFlavorMSB,CoalesceTo<uint64_t>>/Unstuffed/2097152_stddev +0.4571 +0.4766 6 9 6 9 BM<BitstreamFlavorMSB,CoalesceTo<uint64_t>>/Unstuffed/2097152_cv +0.4604 +0.4798 0 0 0 0 OVERALL_GEOMEAN -0.0290 -0.0290 0 0 0 0 ```

LebedevRI force-pushed the bitvacuumer-perf branch from 55572c5 to ae1a159 Compare February 20, 2024 22:06

LebedevRI changed the title ~~Make use of CoalescingOutputIterator~~ Make some use of CoalescingOutputIterator Feb 20, 2024

LebedevRI added 8 commits February 21, 2024 02:14

CoalescingOutputIteratorTest: add missing include

720d629

BitVacuumerBenchmark: just don't deal with JPEG here

9c48eb6

BitVacuumer: actually require that the output iterator takes bytes

d2f93c6

BitVacuumerJPEG: actually require that the output iterator takes bytes

ea87bf9

BitVacuumer: disallow type conversion in ctor, use universal reference

5f452a7

CoalescingOutputIterator: disallow type conversion in ctor, use unive…

dc93b7f

…rsal reference

CoalescingOutputIterator: provide dummy container_type, with `value…

0f76c92

…_type`

CoalescingOutputIterator: a few more invariants

d87468c

LebedevRI force-pushed the bitvacuumer-perf branch from ae1a159 to 9ac5693 Compare February 20, 2024 23:16

LebedevRI added 5 commits February 21, 2024 02:34

CoalescingOutputIterator: actually conform to the output iterator con…

3fceb92

…cept

CoalescingOutputIteratorBenchmark: parametrize coalescization

4aa40d7

BitVacuumerJPEGBenchmark: parametrize some more

517e535

BitVacuumerJPEGBenchmark: benchmarks with-coalescing too

b23c9c9

LebedevRI force-pushed the bitvacuumer-perf branch from 9ac5693 to c249c6c Compare February 20, 2024 23:34

LebedevRI merged commit d3e7824 into darktable-org:develop Feb 20, 2024

LebedevRI deleted the bitvacuumer-perf branch February 20, 2024 23:35

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Make some use of `CoalescingOutputIterator`#675

Make some use of `CoalescingOutputIterator`#675
LebedevRI merged 13 commits into
darktable-org:developfrom
LebedevRI:bitvacuumer-perf

LebedevRI commented Feb 20, 2024

Uh oh!

codecov Bot commented Feb 20, 2024 •

edited

Loading

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

LebedevRI commented Feb 20, 2024

Uh oh!

codecov Bot commented Feb 20, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

codecov Bot commented Feb 20, 2024 •

edited

Loading