Skip to content
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

Adds support for large number of items to DevicePartition::If with the ThreeWayPartition overload #2506

Open
wants to merge 40 commits into
base: main
Choose a base branch
from
Open

Adds support for large number of items to DevicePartition::If with the ThreeWayPartition overload #2506

wants to merge 40 commits into from
+1,442 −497

Conversation

elstehle
Copy link
Collaborator

@elstehle elstehle commented Oct 4, 2024 •
edited
Loading

Description

Closes #2442

Note: this is a chained PR building on #2400

Performance summary:

Diff i32 vs i32.main any num items Diff i32 vs i32.main 2^28 num items Diff i64 vs i32.main any num items Diff i64 vs i32.main 2^28 num items
min -4.39% 0.09% -8.46% 0.14%
max 3.91% 1.99% 19.53% 19.53%
avg -0.07% 0.79% 0.30% 3.58%
H100 three_way_partition
T{ct} Entropy Elements{io} I32 Diff i32 vs i32.main i64 Diff i64 vs. main.i32
I8 1 2^16 11.589 1.10% 11.235 -1.99%
I8 1 2^20 14.819 2.68% 13.909 -3.63%
I8 1 2^24 70.931 1.33% 70.711 1.02%
I8 1 2^28 1015 1.47% 1021 2.07%
I8 0.544 2^16 11.565 1.64% 10.898 -4.22%
I8 0.544 2^20 14.352 2.94% 13.563 -2.72%
I8 0.544 2^24 69.678 1.31% 69.484 1.03%
I8 0.544 2^28 998.363 1.37% 1003 1.84%
I8 0 2^16 10.897 0.33% 10.768 -0.86%
I8 0 2^20 13.8 1.34% 13.567 -0.37%
I8 0 2^24 68.644 1.03% 68.741 1.17%
I8 0 2^28 987.011 0.98% 991.846 1.47%
I16 1 2^16 11.857 -2.89% 11.338 -7.14%
I16 1 2^20 14.88 -2.56% 14.793 -3.13%
I16 1 2^24 78.842 0.97% 90.149 15.45%
I16 1 2^28 1095 1.86% 1285 19.53%
I16 0.544 2^16 11.453 -4.39% 11.237 -6.19%
I16 0.544 2^20 14.416 -2.39% 14.654 -0.78%
I16 0.544 2^24 77.864 1.10% 89.04 15.61%
I16 0.544 2^28 1079 1.56% 1267 19.26%
I16 0 2^16 11.125 -2.97% 10.972 -4.30%
I16 0 2^20 14.002 -3.00% 14.414 -0.15%
I16 0 2^24 76.452 0.51% 87.869 15.52%
I16 0 2^28 1064 1.27% 1248 18.78%
I32 1 2^16 11.718 -0.48% 11.113 -5.62%
I32 1 2^20 18.86 1.87% 18.143 -2.00%
I32 1 2^24 123.674 1.72% 123.868 1.88%
I32 1 2^28 1773 1.99% 1787 2.80%
I32 0.544 2^16 11.671 3.91% 11.139 -0.83%
I32 0.544 2^20 18.122 0.78% 18.008 0.15%
I32 0.544 2^24 121.747 1.65% 122.065 1.92%
I32 0.544 2^28 1740 1.74% 1749 2.27%
I32 0 2^16 11.142 3.44% 11.172 3.72%
I32 0 2^20 17.897 0.46% 17.522 -1.64%
I32 0 2^24 120.483 0.97% 120.853 1.28%
I32 0 2^28 1726 1.31% 1734 1.78%
I64 1 2^16 10.963 -0.45% 10.616 -3.60%
I64 1 2^20 20.237 -0.97% 19.995 -2.15%
I64 1 2^24 170.984 0.36% 170.828 0.27%
I64 1 2^28 2621 0.26% 2620 0.22%
I64 0.544 2^16 10.98 -1.24% 10.635 -4.34%
I64 0.544 2^20 20.299 -0.90% 19.967 -2.52%
I64 0.544 2^24 170.839 0.31% 170.451 0.08%
I64 0.544 2^28 2613 0.22% 2616 0.34%
I64 0 2^16 10.608 -4.30% 10.607 -4.31%
I64 0 2^20 19.703 -2.09% 19.724 -1.99%
I64 0 2^24 168.043 -0.10% 168.246 0.02%
I64 0 2^28 2584 0.09% 2588 0.24%
I128 1 2^16 12.05 -2.70% 11.337 -8.46%
I128 1 2^20 29.705 0.11% 29.355 -1.07%
I128 1 2^24 324.479 -0.17% 324.349 -0.21%
I128 1 2^28 5095 0.14% 5095 0.14%
I128 0.544 2^16 11.516 -1.27% 11.351 -2.68%
I128 0.544 2^20 29.736 1.01% 29.341 -0.33%
I128 0.544 2^24 324.803 0.24% 324.555 0.16%
I128 0.544 2^28 5085 0.16% 5084 0.14%
I128 0 2^16 11.392 -1.78% 11.508 -0.78%
I128 0 2^20 28.933 -0.57% 28.719 -1.31%
I128 0 2^24 322.584 -0.01% 321.669 -0.29%
I128 0 2^28 5053 0.17% 5053 0.17%
F32 1 2^16 11.828 -0.28% 10.945 -7.72%
F32 1 2^20 18.054 -4.21% 17.843 -5.33%
F32 1 2^24 122.669 -0.15% 123.065 0.17%
F32 1 2^28 1762 0.77% 1778 1.69%
F32 0.544 2^16 11.624 -1.42% 10.903 -7.53%
F32 0.544 2^20 17.732 -3.87% 17.557 -4.82%
F32 0.544 2^24 120.894 -0.10% 121.181 0.14%
F32 0.544 2^28 1733 0.60% 1742 1.12%
F32 0 2^16 10.869 -2.72% 10.936 -2.12%
F32 0 2^20 17.408 -3.00% 17.352 -3.31%
F32 0 2^24 119.708 -0.35% 120.096 -0.03%
F32 0 2^28 1715 0.15% 1722 0.56%
F64 1 2^16 10.532 -1.70% 10.565 -1.39%
F64 1 2^20 20.149 -0.40% 20.244 0.07%
F64 1 2^24 170.075 0.19% 170.476 0.43%
F64 1 2^28 2609 0.17% 2612 0.29%
F64 0.544 2^16 10.639 0.25% 10.776 1.54%
F64 0.544 2^20 20.318 -0.15% 20.117 -1.14%
F64 0.544 2^24 170.03 0.17% 170.433 0.41%
F64 0.544 2^28 2604 0.19% 2609 0.38%
F64 0 2^16 10.441 -1.67% 10.822 1.92%
F64 0 2^20 19.894 -0.58% 19.733 -1.38%
F64 0 2^24 167.27 -0.04% 167.655 0.19%
F64 0 2^28 2576 0.11% 2577 0.15%

Checklist

  • New or existing tests cover these changes.
  • The documentation is up to date with these changes.

@elstehle
adds streaming selection and partition
97f3eec
@elstehle
ensures policy lookup uses per-partition offset type
3085f86
@elstehle
mitigates perf degradation on select
7f264ec
@elstehle
makes device interfaces use i64 num_items
412ea3c
@elstehle
updates select::if large num_items tests
70d562b
@elstehle
fixes syntax
b274e1b
@elstehle
adds tests for large num_items for select::flagged
d6b21ba
@elstehle
adds tests for large num_items for partition::if
d26eb65
@elstehle
adds tests for large num_items for partition::flagged
1be152c
@elstehle
fixes redundant definition
466f915
@elstehle
fixes implicit conversion
d0f6e7f
@elstehle
fixes f32 select::if perf regression
cea8509
@elstehle
fixes perf regression for partition
96be1e8
@elstehle
fix feature macro
a117b45
@elstehle
fixes feature macro
480da39
@elstehle
fixes feature macro
aaed489
@elstehle
silences msvc constant conditional warning
3333738
@elstehle
add support for streamin ctx dummy for partition with small offset types
66ca8fd
@elstehle
removes superfluous template parameter
2781775
@elstehle
adds test for different offset types for partition::if & ::flagged
7373696
@elstehle
adds tests and support for streaming select::unique
687fb99
@elstehle
fixes msvc warning
0e3f602
@elstehle
fixes perf for partition
305b307
@elstehle
fixes format
db36ae9
@elstehle
fixes mixup for partition perf fix
5f38f5e
@elstehle
fixes syntax
16a2a75
@elstehle
fixes partition:flagged perf
be83ebe
@elstehle
fixes perf for partition::flagged
d4816b6
@elstehle
switches unique to always use i64 offset types
3c9788d
@elstehle
adds benchmark for partition with distinct iterators
49a50f5
@elstehle elstehle requested review from a team as code owners October 4, 2024 05:06
@github-actions GitHub Actions
Copy link
Contributor

github-actions bot commented Oct 4, 2024

🟨 CI finished in 52m 14s: Pass: 97%/208 | Total: 1d 06h | Avg: 8m 51s | Max: 41m 07s | Hits: 99%/11150

  • 🟨 cub: Pass: 94%/104 | Total: 20h 12m | Avg: 11m 39s | Max: 41m 07s

     
    🔍 cpu: amd64 🔍
  🔍 amd64              Pass:  93%/96  | Total: 18h 41m | Avg: 11m 41s | Max: 41m 07s
  🟩 arm64              Pass: 100%/8   | Total:  1h 30m | Avg: 11m 19s | Max: 13m 07s
🔍 cudacxx_family: nvcc 🔍
  🟩 ClangCUDA          Pass: 100%/2   | Total: 14m 08s | Avg:  7m 04s | Max:  7m 10s
  🔍 nvcc               Pass:  94%/102 | Total: 19h 58m | Avg: 11m 44s | Max: 41m 07s
🟨 ctk
  🟨 11.1               Pass:  93%/15  | Total:  3h 05m | Avg: 12m 23s | Max: 41m 07s
  🟩 11.8               Pass: 100%/3   | Total: 43m 52s | Avg: 14m 37s | Max: 15m 53s
  🟨 12.6               Pass:  94%/86  | Total: 16h 22m | Avg: 11m 25s | Max: 33m 39s
🟨 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total: 14m 08s | Avg:  7m 04s | Max:  7m 10s
  🟨 nvcc11.1           Pass:  93%/15  | Total:  3h 05m | Avg: 12m 23s | Max: 41m 07s
  🟩 nvcc11.8           Pass: 100%/3   | Total: 43m 52s | Avg: 14m 37s | Max: 15m 53s
  🟨 nvcc12.6           Pass:  94%/84  | Total: 16h 08m | Avg: 11m 31s | Max: 33m 39s
🟨 cxx
  🟩 Clang9             Pass: 100%/6   | Total:  1h 03m | Avg: 10m 35s | Max: 11m 51s
  🟩 Clang10            Pass: 100%/3   | Total: 31m 57s | Avg: 10m 39s | Max: 11m 19s
  🟩 Clang11            Pass: 100%/4   | Total: 38m 18s | Avg:  9m 34s | Max: 10m 38s
  🟩 Clang12            Pass: 100%/4   | Total: 39m 52s | Avg:  9m 58s | Max: 11m 16s
  🟩 Clang13            Pass: 100%/4   | Total: 38m 39s | Avg:  9m 39s | Max: 10m 37s
  🟩 Clang14            Pass: 100%/4   | Total: 40m 04s | Avg: 10m 01s | Max: 10m 15s
  🟩 Clang15            Pass: 100%/4   | Total: 38m 23s | Avg:  9m 35s | Max: 10m 19s
  🟩 Clang16            Pass: 100%/4   | Total: 41m 41s | Avg: 10m 25s | Max: 11m 38s
  🟩 Clang17            Pass: 100%/4   | Total: 40m 06s | Avg: 10m 01s | Max: 12m 01s
  🟩 Clang18            Pass: 100%/9   | Total:  1h 48m | Avg: 12m 00s | Max: 23m 27s
  🟩 GCC6               Pass: 100%/2   | Total: 23m 23s | Avg: 11m 41s | Max: 11m 49s
  🟨 GCC7               Pass:  83%/6   | Total: 59m 03s | Avg:  9m 50s | Max: 11m 14s
  🟩 GCC8               Pass: 100%/6   | Total: 57m 26s | Avg:  9m 34s | Max: 10m 22s
  🟩 GCC9               Pass: 100%/6   | Total:  1h 24m | Avg: 14m 07s | Max: 33m 39s
  🟩 GCC10              Pass: 100%/4   | Total: 39m 15s | Avg:  9m 48s | Max: 10m 59s
  🟩 GCC11              Pass: 100%/7   | Total:  1h 23m | Avg: 11m 55s | Max: 15m 53s
  🟩 GCC12              Pass: 100%/4   | Total: 39m 32s | Avg:  9m 53s | Max: 10m 45s
  🟨 GCC13              Pass:  93%/16  | Total:  3h 26m | Avg: 12m 55s | Max: 25m 09s
  🟩 Intel2023.2.0      Pass: 100%/3   | Total: 37m 21s | Avg: 12m 27s | Max: 13m 44s
  🟥 MSVC14.16          Pass:   0%/1   | Total: 41m 07s | Avg: 41m 07s | Max: 41m 07s
  🟥 MSVC14.29          Pass:   0%/2   | Total: 38m 01s | Avg: 19m 00s | Max: 20m 01s
  🟥 MSVC14.39          Pass:   0%/1   | Total: 21m 36s | Avg: 21m 36s | Max: 21m 36s
🟨 cxx_family
  🟩 Clang              Pass: 100%/46  | Total:  8h 00m | Avg: 10m 26s | Max: 23m 27s
  🟨 GCC                Pass:  96%/51  | Total:  9h 53m | Avg: 11m 38s | Max: 33m 39s
  🟩 Intel              Pass: 100%/3   | Total: 37m 21s | Avg: 12m 27s | Max: 13m 44s
  🟥 MSVC               Pass:   0%/4   | Total:  1h 40m | Avg: 25m 11s | Max: 41m 07s
🟨 jobs
  🟨 Build              Pass:  94%/96  | Total: 17h 34m | Avg: 10m 59s | Max: 41m 07s
  🟥 DeviceLaunch       Pass:   0%/1   | Total: 18m 02s | Avg: 18m 02s | Max: 18m 02s
  🟩 GraphCapture       Pass: 100%/1   | Total: 15m 47s | Avg: 15m 47s | Max: 15m 47s
  🟩 HostLaunch         Pass: 100%/3   | Total: 50m 34s | Avg: 16m 51s | Max: 17m 39s
  🟩 TestGPU            Pass: 100%/3   | Total:  1h 13m | Avg: 24m 28s | Max: 25m 09s
🟨 gpu
  🟨 v100               Pass:  94%/104 | Total: 20h 12m | Avg: 11m 39s | Max: 41m 07s
🟩 sm
  🟩 60;70;80;90        Pass: 100%/3   | Total: 43m 52s | Avg: 14m 37s | Max: 15m 53s
  🟩 90a                Pass: 100%/4   | Total: 23m 37s | Avg:  5m 54s | Max:  6m 30s
🟨 std
  🟨 11                 Pass:  96%/28  | Total:  5h 53m | Avg: 12m 37s | Max: 33m 39s
  🟨 14                 Pass:  92%/27  | Total:  5h 35m | Avg: 12m 26s | Max: 41m 07s
  🟨 17                 Pass:  96%/26  | Total:  4h 05m | Avg:  9m 25s | Max: 18m 00s
  🟨 20                 Pass:  91%/23  | Total:  4h 38m | Avg: 12m 05s | Max: 24m 48s

  • 🟩 thrust: Pass: 100%/103 | Total: 10h 15m | Avg: 5m 58s | Max: 26m 05s | Hits: 99%/11150 

    🟩 cpu
  🟩 amd64              Pass: 100%/95  | Total:  9h 38m | Avg:  6m 05s | Max: 26m 05s | Hits:  99%/11150 
  🟩 arm64              Pass: 100%/8   | Total: 36m 57s | Avg:  4m 37s | Max:  5m 14s
🟩 ctk
  🟩 11.1               Pass: 100%/15  | Total:  1h 33m | Avg:  6m 12s | Max: 21m 34s | Hits:  99%/2230  
  🟩 11.8               Pass: 100%/3   | Total: 14m 42s | Avg:  4m 54s | Max:  5m 28s
  🟩 12.6               Pass: 100%/85  | Total:  8h 28m | Avg:  5m 58s | Max: 26m 05s | Hits:  99%/8920  
🟩 cudacxx
  🟩 ClangCUDA18        Pass: 100%/2   | Total:  9m 55s | Avg:  4m 57s | Max:  4m 59s
  🟩 nvcc11.1           Pass: 100%/15  | Total:  1h 33m | Avg:  6m 12s | Max: 21m 34s | Hits:  99%/2230  
  🟩 nvcc11.8           Pass: 100%/3   | Total: 14m 42s | Avg:  4m 54s | Max:  5m 28s
  🟩 nvcc12.6           Pass: 100%/83  | Total:  8h 18m | Avg:  6m 00s | Max: 26m 05s | Hits:  99%/8920  
🟩 cudacxx_family
  🟩 ClangCUDA          Pass: 100%/2   | Total:  9m 55s | Avg:  4m 57s | Max:  4m 59s
  🟩 nvcc               Pass: 100%/101 | Total: 10h 05m | Avg:  5m 59s | Max: 26m 05s | Hits:  99%/11150 
🟩 cxx
  🟩 Clang9             Pass: 100%/6   | Total: 30m 05s | Avg:  5m 00s | Max:  5m 58s
  🟩 Clang10            Pass: 100%/3   | Total: 16m 51s | Avg:  5m 37s | Max:  5m 56s
  🟩 Clang11            Pass: 100%/4   | Total: 19m 11s | Avg:  4m 47s | Max:  5m 10s
  🟩 Clang12            Pass: 100%/4   | Total: 18m 22s | Avg:  4m 35s | Max:  4m 46s
  🟩 Clang13            Pass: 100%/4   | Total: 18m 47s | Avg:  4m 41s | Max:  4m 59s
  🟩 Clang14            Pass: 100%/4   | Total: 18m 42s | Avg:  4m 40s | Max:  4m 58s
  🟩 Clang15            Pass: 100%/4   | Total: 20m 16s | Avg:  5m 04s | Max:  5m 32s
  🟩 Clang16            Pass: 100%/4   | Total: 19m 06s | Avg:  4m 46s | Max:  4m 56s
  🟩 Clang17            Pass: 100%/4   | Total: 19m 35s | Avg:  4m 53s | Max:  5m 19s
  🟩 Clang18            Pass: 100%/9   | Total: 54m 14s | Avg:  6m 01s | Max: 15m 18s
  🟩 GCC6               Pass: 100%/2   | Total: 25m 22s | Avg: 12m 41s | Max: 21m 34s
  🟩 GCC7               Pass: 100%/6   | Total: 24m 29s | Avg:  4m 04s | Max:  4m 46s
  🟩 GCC8               Pass: 100%/6   | Total: 25m 00s | Avg:  4m 10s | Max:  5m 04s
  🟩 GCC9               Pass: 100%/6   | Total: 26m 39s | Avg:  4m 26s | Max:  4m 54s
  🟩 GCC10              Pass: 100%/4   | Total: 19m 10s | Avg:  4m 47s | Max:  5m 00s
  🟩 GCC11              Pass: 100%/7   | Total: 34m 45s | Avg:  4m 57s | Max:  5m 32s
  🟩 GCC12              Pass: 100%/4   | Total: 20m 19s | Avg:  5m 04s | Max:  5m 30s
  🟩 GCC13              Pass: 100%/14  | Total:  1h 27m | Avg:  6m 15s | Max: 15m 11s
  🟩 Intel2023.2.0      Pass: 100%/3   | Total: 18m 45s | Avg:  6m 15s | Max:  6m 31s
  🟩 MSVC14.16          Pass: 100%/1   | Total: 19m 11s | Avg: 19m 11s | Max: 19m 11s | Hits:  99%/2230  
  🟩 MSVC14.29          Pass: 100%/2   | Total: 33m 54s | Avg: 16m 57s | Max: 17m 17s | Hits:  99%/4460  
  🟩 MSVC14.39          Pass: 100%/2   | Total: 45m 20s | Avg: 22m 40s | Max: 26m 05s | Hits:  99%/4460  
🟩 cxx_family
  🟩 Clang              Pass: 100%/46  | Total:  3h 55m | Avg:  5m 06s | Max: 15m 18s
  🟩 GCC                Pass: 100%/49  | Total:  4h 23m | Avg:  5m 22s | Max: 21m 34s
  🟩 Intel              Pass: 100%/3   | Total: 18m 45s | Avg:  6m 15s | Max:  6m 31s
  🟩 MSVC               Pass: 100%/5   | Total:  1h 38m | Avg: 19m 41s | Max: 26m 05s | Hits:  99%/11150 
🟩 gpu
  🟩 v100               Pass: 100%/103 | Total: 10h 15m | Avg:  5m 58s | Max: 26m 05s | Hits:  99%/11150 
🟩 jobs
  🟩 Build              Pass: 100%/96  | Total:  8h 46m | Avg:  5m 28s | Max: 21m 34s | Hits:  99%/8920  
  🟩 TestCPU            Pass: 100%/4   | Total: 47m 25s | Avg: 11m 51s | Max: 26m 05s | Hits:  99%/2230  
  🟩 TestGPU            Pass: 100%/3   | Total: 41m 55s | Avg: 13m 58s | Max: 15m 18s
🟩 sm
  🟩 60;70;80;90        Pass: 100%/3   | Total: 14m 42s | Avg:  4m 54s | Max:  5m 28s
  🟩 90a                Pass: 100%/4   | Total: 16m 07s | Avg:  4m 01s | Max:  4m 16s
🟩 std
  🟩 11                 Pass: 100%/28  | Total:  2h 29m | Avg:  5m 20s | Max: 21m 34s
  🟩 14                 Pass: 100%/27  | Total:  2h 34m | Avg:  5m 43s | Max: 19m 11s | Hits:  99%/4460  
  🟩 17                 Pass: 100%/26  | Total:  2h 21m | Avg:  5m 26s | Max: 16m 37s | Hits:  99%/2230  
  🟩 20                 Pass: 100%/22  | Total:  2h 50m | Avg:  7m 44s | Max: 26m 05s | Hits:  99%/4460

  • 🟩 pycuda: Pass: 100%/1 | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s

     
    🟩 cpu
  🟩 amd64              Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 ctk
  🟩 12.5               Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 cudacxx
  🟩 nvcc12.5           Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 cudacxx_family
  🟩 nvcc               Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 cxx
  🟩 GCC13              Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 cxx_family
  🟩 GCC                Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 gpu
  🟩 v100               Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s
🟩 jobs
  🟩 Test               Pass: 100%/1   | Total: 16m 12s | Avg: 16m 12s | Max: 16m 12s

👃 Inspect Changes

Modifications in project?

Project
CCCL Infrastructure
libcu++
+/- CUB
+/- Thrust
CUDA Experimental
pycuda
CUDA C Core Library

Modifications in project or dependencies?

Project
CCCL Infrastructure
libcu++
+/- CUB
+/- Thrust
CUDA Experimental
+/- pycuda
+/- CUDA C Core Library

🏃‍ Runner counts (total jobs: 208)
# Runner
171 linux-amd64-cpu16
16 linux-arm64-cpu16
12 linux-amd64-gpu-v100-latest-1
9 windows-amd64-cpu16

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@gevtushenko gevtushenko Awaiting requested review from gevtushenko gevtushenko is a code owner automatically assigned from NVIDIA/cccl-cub-codeowners

@fbusato fbusato Awaiting requested review from fbusato fbusato is a code owner automatically assigned from NVIDIA/cccl-codeowners

At least 1 approving review is required to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
CCCL
Status: In Review
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

Add support for large num_items to DevicePartition::ThreeWayPartition
1 participant
@elstehle