RPP Tensor Support - Fisheye on HOST and HIP#346
Closed
sampath1117 wants to merge 25 commits intor-abishek:developfrom
Closed
RPP Tensor Support - Fisheye on HOST and HIP#346sampath1117 wants to merge 25 commits intor-abishek:developfrom
sampath1117 wants to merge 25 commits intor-abishek:developfrom
Conversation
sampath1117
commented
Sep 25, 2024
- Adds tensor support for Fog function optimized on AVX2 and HIP
- Adds test suite support
consolidated the code
01e5007 to
2113c51
Compare
Srihari-mcw
reviewed
Jan 3, 2025
CHANGELOG.md
Outdated
|
|
||
| Full documentation for RPP is available at [https://rocm.docs.amd.com/projects/rpp/en/latest](https://rocm.docs.amd.com/projects/rpp/en/latest) | ||
|
|
||
| ## RPP 1.18.0 (unreleased) |
Collaborator
There was a problem hiding this comment.
Add the change under 1.9.10
Srihari-mcw
reviewed
Jan 3, 2025
CMakeLists.txt
Outdated
|
|
||
| # RPP Version | ||
| set(VERSION "1.9.10") | ||
| set(VERSION "1.18.0") |
Collaborator
There was a problem hiding this comment.
Set version here to 1.9.10
Srihari-mcw
reviewed
Jan 3, 2025
| #endif | ||
| // NOTE: IMPORTANT: Match the version with CMakelists.txt version | ||
| #define RPP_VERSION_MAJOR 1 | ||
| #define RPP_VERSION_MINOR 9 |
Collaborator
There was a problem hiding this comment.
Here also retain version to 1.9.10
Srihari-mcw
reviewed
Jan 5, 2025
src/include/cpu/rpp_cpu_simd.hpp
Outdated
| const __m256 avx_p1op255 = _mm256_set1_ps(1.0f / 255.0f); | ||
| const __m256 avx_p1op3 = _mm256_set1_ps(1.0f / 3.0f); | ||
| const __m256 avx_p2op3 = _mm256_set1_ps(2.0f / 3.0f); | ||
| const __m256 avx_pMinus1 = _mm256_set1_ps(-1); |
Collaborator
There was a problem hiding this comment.
Can we make this -1.0f?
Srihari-mcw
reviewed
Jan 5, 2025
|
|
||
| inline void compute_fisheye_src_loc_avx(__m256 &pDstY, __m256 &pDstX, __m256 &pSrcY, __m256 &pSrcX, __m256 &pHeight, __m256 &pWidth) | ||
| { | ||
| __m256 pNormX, pNormY, pDist; |
Collaborator
There was a problem hiding this comment.
Add the AVX helper functions overall inside #if AVX2
Srihari-mcw
reviewed
Jan 5, 2025
src/modules/cpu/kernel/fisheye.hpp
Outdated
| } | ||
|
|
||
| // fisheye without fused output-layout toggle (NHWC -> NHWC) | ||
| if ((srcDescPtr->c == 3) && (srcDescPtr->layout == RpptLayout::NHWC) && (dstDescPtr->layout == RpptLayout::NHWC)) |
Srihari-mcw
reviewed
Jan 5, 2025
src/modules/cpu/kernel/fisheye.hpp
Outdated
| { | ||
| __m256i pRow; | ||
| rpp_simd_load(rpp_generic_nn_load_i8pln1_avx, srcPtrTempChn, srcLocArray, invalidLoad, pRow); | ||
| rpp_storeu_si64((__m128i *)(dstPtrTempChn), _mm256_castsi256_si128(pRow)); |
Collaborator
There was a problem hiding this comment.
Can probably use reinterpret_cast<__m128i *>(dstPtrTempChn) here
Srihari-mcw
reviewed
Jan 5, 2025
| int id_z = hipBlockIdx_z * hipBlockDim_z + hipThreadIdx_z; | ||
| int4 srcRoi_i4 = *(int4 *)&roiTensorPtrSrc[id_z]; | ||
| int width = (srcRoi_i4.z - srcRoi_i4.x) + 1; | ||
| int height = (srcRoi_i4.w - srcRoi_i4.y) + 1; |
Srihari-mcw
requested changes
Jan 5, 2025
Collaborator
Srihari-mcw
left a comment
There was a problem hiding this comment.
Please address review comments. Thanks
b25e8ec to
8b40e87
Compare
…ocs/sphinx (r-abishek#515) Bumps [rocm-docs-core[api_reference]](https://github.com/ROCm/rocm-docs-core) from 1.14.1 to 1.15.0. - [Release notes](https://github.com/ROCm/rocm-docs-core/releases) - [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md) - [Commits](ROCm/rocm-docs-core@v1.14.1...v1.15.0) --- updated-dependencies: - dependency-name: rocm-docs-core[api_reference] dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Collaborator
|
Moved to different PR - Closed |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.