8000 `PanasonicV8Decompressor::decompressStrip()`: rewrite by LebedevRI · Pull Request #849 · darktable-org/rawspeed · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

PanasonicV8Decompressor::decompressStrip(): rewrite #849

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 14 commits into from
May 18, 2025
Merged

PanasonicV8Decompressor::decompressStrip(): rewrite #849

merged 14 commits into from
May 18, 2025

Conversation

LebedevRI
Copy link
Member
@LebedevRI LebedevRI commented May 18, 2025 •
edited
Loading

(as compared to without this PR)

$ /repositories/googlebenchmark/tools/compare.py -a benchmarks ~/rawspeed/build-{old,new}/src/utilities/rsbench/rsbench Panasonic/DC-S5M2/P1126458_mechanical.RW2 --benchmark_min_warmup_time=0.5 --benchmark_min_time=0s --benchmark_repetitions=99 --benchmark_counters_tabular=true -t --benchmark_filter="/threads:(1|32)/" 
RUNNING: /home/lebedevri/rawspeed/build-old/src/utilities/rsbench/rsbench Panasonic/DC-S5M2/P1126458_mechanical.RW2 --benchmark_min_warmup_time=0.5 --benchmark_min_time=0s --benchmark_repetitions=99 --benchmark_counters_tabular=true -t --benchmark_filter=/threads:(1|32)/ --benchmark_display_aggregates_only=true --benchmark_out=/tmp/tmpy_nco89y
2025-05-18T20:06:23+03:00
Running /home/lebedevri/rawspeed/build-old/src/utilities/rsbench/rsbench
Run on (32 X 3400 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x16)
  L1 Instruction 32 KiB (x16)
  L2 Unified 512 KiB (x16)
  L3 Unified 32768 KiB (x2)
Load Average: 8.45, 3.84, 2.61
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Benchmark                                                                                   Time             CPU   Iterations  CPUTime,s CPUTime/WallTime     Pixels Pixels/CPUTime Pixels/WallTime Raws/CPUTime Raws/WallTime WallTime,s
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_mean           243 ms          243 ms           99   0.242746         0.999852   24.0801M       99.1995M        99.1848M      4.11957       4.11896   0.242782
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_median         243 ms          243 ms           99   0.242616         0.999891   24.0801M       99.2518M        99.2357M      4.12174       4.12107   0.242655
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_stddev       0.694 ms        0.681 ms           99   681.374u         160.461u          0        276.32k        281.278k     0.011475     0.0116809   694.023u
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_cv            0.29 %          0.28 %            99      0.28%            0.02%      0.00%          0.28%           0.28%        0.28%         0.28%      0.29%
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_mean          128 ms          256 ms           99   0.255867          1.99978   24.0801M       94.1118M        188.203M      3.90829       7.81573   0.127947
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_median        128 ms          256 ms           99   0.255674          1.99987   24.0801M       94.1827M        188.345M      3.91123       7.82161   0.127851
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_stddev      0.196 ms        0.393 ms           99   392.892u         310.161u          0       144.295k        287.894k     5.99231m     0.0119557   196.012u
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_cv           0.15 %          0.15 %            99      0.15%            0.02%      0.00%          0.15%           0.15%        0.15%         0.15%      0.15%
RUNNING: /home/lebedevri/rawspeed/build-new/src/utilities/rsbench/rsbench Panasonic/DC-S5M2/P1126458_mechanical.RW2 --benchmark_min_warmup_time=0.5 --benchmark_min_time=0s --benchmark_repetitions=99 --benchmark_counters_tabular=true -t --benchmark_filter=/threads:(1|32)/ --benchmark_display_aggregates_only=true --benchmark_out=/tmp/tmpta85pklr
2025-05-18T20:07:01+03:00
Running /home/lebedevri/rawspeed/build-new/src/utilities/rsbench/rsbench
Run on (32 X 3400 MHz CPU s)
CPU Caches:
  L1 Data 32 KiB (x16)
  L1 Instruction 32 KiB (x16)
  L2 Unified 512 KiB (x16)
  L3 Unified 32768 KiB (x2)
Load Average: 5.12, 3.55, 2.56
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Benchmark                                                                                   Time             CPU   Iterations  CPUTime,s CPUTime/WallTime     Pixels Pixels/CPUTime Pixels/WallTime Raws/CPUTime Raws/WallTime WallTime,s
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_mean           220 ms          220 ms           99   0.220266         0.999832   24.0801M       109.323M        109.305M      4.53998       4.53922   0.220303
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_median         220 ms          220 ms           99    0.22044         0.999884   24.0801M       109.236M        109.226M      4.53638       4.53594   0.220462
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_stddev       0.484 ms        0.479 ms           99   479.441u         179.485u          0       238.195k        240.159k     9.89181m      9.97333m   483.568u
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_cv            0.22 %          0.22 %            99      0.22%            0.02%      0.00%          0.22%           0.22%        0.22%         0.22%      0.22%
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_mean          120 ms          240 ms           99   0.239875          1.99936   24.0801M       100.387M        200.709M      4.16886       8.33505   0.119976
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_median        120 ms          240 ms           99   0.240076          1.99976   24.0801M       100.302M        200.544M      4.16535       8.32822   0.120074
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_stddev      0.247 ms        0.526 ms           99   526.522u         638.899u          0       220.596k        413.887k     9.16092m     0.0171879   247.158u
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_cv           0.21 %          0.22 %            99      0.22%            0.03%      0.00%          0.22%           0.21%        0.22%         0.21%      0.21%
Comparing /home/lebedevri/rawspeed/build-old/src/utilities/rsbench/rsbench to /home/lebedevri/rawspeed/build-new/src/utilities/rsbench/rsbench
Benchmark                                                                                            Time             CPU      Time Old      Time New       CPU Old       CPU New
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_pvalue                  0.0000          0.0000      U Test, Repetitions: 99 vs 99
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_mean                   -0.0926         -0.0926           243           220           243           220
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_median                 -0.0915         -0.0914           243           220           243           220
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_stddev                 -0.3031         -0.2963             1             0             1             0
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:1/process_time/real_time_cv                     -0.2320         -0.2245             0             0             0             0
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_pvalue                 0.0000          0.0000      U Test, Repetitions: 99 vs 99
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_mean                  -0.0623         -0.0625           128           120           256           240
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_median                -0.0608         -0.0610           128           120           256           240
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_stddev                +0.2620         +0.3377             0             0             0             1
Panasonic/DC-S5M2/P1126458_mechanical.RW2/threads:32/process_time/real_time_cv                    +0.3458         +0.4269             0             0             0             0
OVERALL_GEOMEAN                                                                                   -0.0776         -0.0777             0             0             0             0

LebedevRI added 14 commits May 18, 2025 16:40
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): use CroppedArray2DRef
7535893
@LebedevRI
PanasonicV8Decompressor: take array of output tiles
fa73f4b
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): make copyback block more …
18a1ebd 
…obvious
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): LSR loop
ddf851f
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): one more Array2DRef in co…
81b4831 
…pyback loop
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): de-unroll last loop
6f8cd51
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): split decode loop into in…
db50f1e 
…ner and outer loops
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): more Array2DRef
08fddd3
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): eliminate copyback loop
22ea670
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): update pred inplace
36ba99f
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): eliminate tmp lineBuffer
f836bd2
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): LSR outer loop
82aa1f1
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): transpose pred
36288e9
@LebedevRI
PanasonicV8Decompressor::decompressStrip(): shorten pred var name
675e467
@LebedevRI LebedevRI merged commit 8154697 into darktable-org:develop May 18, 2025
66 of 67 checks passed
@LebedevRI LebedevRI deleted the panasonicv8-rewrite-decompressstrip branch May 18, 2025 18:10
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@LebedevRI
0