Skip to content

Commit

Permalink
SWDEV-458672 - Porting the Fix for dead links from staging
Browse files Browse the repository at this point in the history
Change-Id: Iea5b47d2db3b6e33fc18f1ed5bb79be688cce600
  • Loading branch information
jujiang-del committed May 10, 2024
1 parent ade7dae commit d95a074
Show file tree
Hide file tree
Showing 14 changed files with 53 additions and 48 deletions.
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -26,7 +26,7 @@ hip-tests releases are typically naming convention for each ROCM release to help

### Build HIP catch tests

For building HIP from sources, please check instructions on [HIP page] (https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/build.html)
For building HIP from sources, please check instructions on [HIP page](https://rocm.docs.amd.com/projects/HIP/en/latest/install/build.html)

HIP catch tests can be built via the following instructions,

Expand Down
9 changes: 5 additions & 4 deletions samples/2_Cookbook/0_MatrixTranspose/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ This tutorial shows how to get write simple HIP application. We will write the s
HIP is a C++ runtime API and kernel language that allows developers to create portable applications that can run on AMD and other GPU’s. Our goal was to rise above the lowest-common-denominator paths and deliver a solution that allows you, the developer, to use essential hardware features and maximize your application’s performance on GPU hardware.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -102,10 +102,11 @@ PASSED!


## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
18 changes: 9 additions & 9 deletions samples/2_Cookbook/10_inline_asm/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,16 +6,15 @@ This tutorial is about how to use inline GCN asm in kernel. In this tutorial, we

If you want to take advantage of the extra performance benefits of writing in assembly as well as take advantage of special GPU hardware features that were only available through assemby, then this tutorial is for you. In this tutorial we'll be explaining how to start writing inline asm in kernel.

For more insight Please read the following blogs by Ben Sander
[The Art of AMDGCN Assembly: How to Bend the Machine to Your Will](gpuopen.com/amdgcn-assembly)
[AMD GCN Assembly: Cross-Lane Operations](http://gpuopen.com/amd-gcn-assembly-cross-lane-operations/)
For more insight Please read the following blogs by Ben Sander:
[The Art of AMDGCN Assembly: How to Bend the Machine to Your Will](https://gpuopen.com/learn/amdgcn-assembly/) and [AMD GCN Assembly: Cross-Lane Operations](http://gpuopen.com/amd-gcn-assembly-cross-lane-operations/).

For more information:
[AMD GCN3 ISA Architecture Manual](http://gpuopen.com/compute-product/amd-gcn3-isa-architecture-manual/)
[User Guide for AMDGPU Back-end](llvm.org/docs/AMDGPUUsage.html)
[AMD GCN3 ISA Architecture Manual](https://gpuopen.com/amd-isa-documentation/),
[User Guide for AMDGPU Back-end](https://llvm.org/docs/AMDGPUUsage.html).

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -62,10 +61,11 @@ PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
7 changes: 4 additions & 3 deletions samples/2_Cookbook/12_cmake_hip_add_executable/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -50,10 +50,11 @@ PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)

- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
2 changes: 1 addition & 1 deletion samples/2_Cookbook/15_static_library/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -78,4 +78,4 @@ PASSED!

**Note:** Currently, defined arch is `gfx900`, `gfx906`, `gfx908`, `gfx1010`,`gfx1030`,`gfx1100`,`gfx1101`,`gfx1102` and `gfx1103`. Any undefined arch can be modified with make argument `GPU_ARCHxx`.

## For More Information, please refer to the HIP FAQ.
## For More Information, please refer to [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html).
2 changes: 1 addition & 1 deletion samples/2_Cookbook/16_assembly_to_executable/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -78,4 +78,4 @@ PASSED!

**Note:** Currently, defined arch is `gfx900`, `gfx906`, `gfx908`, `gfx1010`,`gfx1030`,`gfx1100`,`gfx1101`,`gfx1102` and `gfx1103`. Any undefined arch can be modified with make argument `GPU_ARCHxx`.

## For More Information, please refer to the HIP FAQ.
## For More Information, please refer to [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html).
2 changes: 1 addition & 1 deletion samples/2_Cookbook/17_llvm_ir_to_executable/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -106,4 +106,4 @@ PASSED!

**Note:** Any undefined arch can be modified with make argument `GPU_ARCHxx`.

## For More Information, please refer to the HIP FAQ.
## For More Information, please refer to [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html).
8 changes: 4 additions & 4 deletions samples/2_Cookbook/1_hipEvent/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ This tutorial is follow-up of the previous one where we learn how to write our f
Memory transfer and kernel execution are the most important parameter in parallel computing (specially HPC and machine learning). Memory bottlenecks is the main problem why we are not able to get the highest performance, therefore obtaining the memory transfer timing and kernel execution timing plays key role in application optimization.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -83,10 +83,10 @@ PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
8 changes: 4 additions & 4 deletions samples/2_Cookbook/3_shared_memory/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ Earlier we learned how to write our first hip program, in which we compute Matri
As we mentioned earlier that Memory bottlenecks is the main problem why we are not able to get the highest performance, therefore minimizing the latency for memory access plays prominent role in application optimization. In this tutorial, we'll learn how to use static shared memory and will explain the dynamic one latter.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -42,10 +42,10 @@ PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
8 changes: 4 additions & 4 deletions samples/2_Cookbook/4_shfl/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ Let's talk about Warp first. The kernel code is executed in groups of fixed numb
```

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -52,10 +52,10 @@ PASSED!
please make sure you have a 3.0 or higher compute capable device in order to use warp shfl operations and add `-gencode arch=compute=30, code=sm_30` nvcc flag in the Makefile while using this application.

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
9 changes: 5 additions & 4 deletions samples/2_Cookbook/5_2dshfl/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ Let's talk about Warp first. The kernel code is executed in groups of fixed numb
```

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -54,10 +54,11 @@ PASSED!
please make sure you have a 3.0 or higher compute capable device in order to use warp shfl operations and add `-gencode arch=compute=30, code=sm_30` nvcc flag in the Makefile while using this application.

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)

- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
10 changes: 6 additions & 4 deletions samples/2_Cookbook/6_dynamic_shared/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ Earlier we learned how to use static shared memory. In this tutorial, we'll expl
As we mentioned earlier that Memory bottlenecks is the main problem why we are not able to get the highest performance, therefore minimizing the latency for memory access plays prominent role in application optimization. In this tutorial, we'll learn how to use dynamic shared memory.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -50,11 +50,13 @@ $ ./dynamic_shared
Device name Navi 14 [Radeon Pro W5500]
dynamic_shared PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)

- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
8 changes: 4 additions & 4 deletions samples/2_Cookbook/7_streams/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ In all Earlier tutorial we used single stream, In this tutorial, we'll explain h
The various instances of kernel to be executed on device in exact launch order defined by Host are called streams. We can launch multiple streams on a single device. We will learn how to learn two streams which can we scaled with ease.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -62,10 +62,10 @@ stream PASSED!
```

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)
8 changes: 4 additions & 4 deletions samples/2_Cookbook/9_unroll/Readme.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ Loop unrolling optimization hints can be specified with #pragma unroll and #prag
Specifying #pragma unroll without a parameter directs the loop unroller to attempt to fully unroll the loop if the trip count is known at compile time and attempt to partially unroll the loop if the trip count is not known at compile time.

## Requirement:
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/how_to_guides/install.html)
For hardware requirement and software installation [Installation](https://rocm.docs.amd.com/projects/HIP/en/latest/install/install.html)

## prerequiste knowledge:

Expand Down Expand Up @@ -47,10 +47,10 @@ PASSED
please make sure you have a 3.0 or higher compute capable device in order to use warp shfl operations and add `-gencode arch=compute=30, code=sm_30` nvcc flag in the Makefile while using this application.

## More Info:
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/faq.html)
- [HIP FAQ](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/faq.html)
- [HIP Kernel Language](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/kernel_language.html)
- [HIP Runtime API (Doxygen)](https://rocm.docs.amd.com/projects/HIP/en/latest/doxygen/html/index.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/user_guide/hip_porting_guide.html)
- [HIP Porting Guide](https://rocm.docs.amd.com/projects/HIP/en/latest/how-to/hip_porting_guide.html)
- [HIP Terminology](https://rocm.docs.amd.com/projects/HIP/en/latest/reference/terms.html) (including comparing syntax for different compute terms across CUDA/HIP/OpenL)
- [HIPIFY](https://rocm.docs.amd.com/projects/HIPIFY/en/latest/index.html)
- [Developer/CONTRIBUTING Info](https://rocm.docs.amd.com/projects/HIP/en/latest/developer_guide/contributing.html)
- [Developer/CONTRIBUTING Info](https://github.com/ROCm/HIP/blob/develop/CONTRIBUTING.md)

0 comments on commit d95a074

Please sign in to comment.