Comments (7)
Only milestone3_prerelease
branch has install
and package
targets. We a pretty sure they should work without problems, we have CI with docker images testing both of those targets.
from rocprim.
Thanks @jszuppe , that helps.
A separate question, do you have a script to run the benchmarks ?
from rocprim.
A separate question, do you have a script to run the benchmarks ?
No, we don't.
from rocprim.
Are you planning to provide such scripts or benchmarks suites?
It's fine for now though as we'll mainly focus on functionality.
from rocprim.
I don't know. It wasn't planned. Why do you need one [script to run all benchmarks]?
from rocprim.
Mainly to compare the performance of rocPRIM verse our target NV GPUs.
Please refer to how rocBLAs supplied such needs:
https://github.com/ROCmSoftwarePlatform/rocBLAS/wiki/Running#benchmarks
from rocprim.
rocPRIM provides benchmarks for its warp, block and device level function, so there is no problem with comparison. In situations where performance of certain algorithm does depend on input values (like reduce-by-key, or segmented reduction), our benchmarks usually test various distributions of input values. Benchmarks are implemented using Google Benchmark library, so it should be easy to export results to csv and compare with other results you have. rocPRIM does not provide benchmarks for CUB or any other similar CUDA library.
from rocprim.
Related Issues (20)
- Block_reduce fails to distribute correct answer to all lanes when hipBlockDim > 64 HOT 5
- Invalid use of inline assembly for GFX1010 target HOT 14
- Compiler Build issue HOT 1
- ROCm 3.9.1 fails to compile benchmark_warp_scan HOT 2
- error: reference to __host__ function 'inclusive_scan<rocprim::default_config, double *, double *, thrust::plus<double>>' in __host__ __device__ function HOT 2
- Ignoring return value warning HOT 2
- error: invalid operands to binary expression HOT 6
- Please enable two factor authentication in your github account
- Alternative documentation HOT 2
- error: no matching function for call to 'ceiling_div' HOT 2
- Add support for NAVI22 and NAVI23 i.e. gfx1031 and gfx1032 HOT 1
- rocPRIM reduction (block_reduce_int) issue HOT 2
- Follow the example of rocPRIM custom types example but compile failed. HOT 2
- rocPRIM on NAVI22 = gfx1031 only one test fails - can we fix this? HOT 6
- rocPRIM -- rocThrust dependency: error: no member named 'init_offset_scan_state_kernel' in rocprim::detail HOT 2
- rocPRIM 5.4.3 failed device_adjacent_difference on gfx906 HOT 4
- rocPRIM selects wrong code paths for warp reductions and scas on gfx1036 HOT 3
- Page fault during `rocprim::select` for no obvious reason. HOT 2
- `rocprim::block_load` fails to instantiate HOT 8
- first installation failed
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from rocprim.