Commit Graph

2717 Commits

Author SHA1 Message Date
Peter Park
d0c8ba0805 Add Wan2.1 to PyTorch inference Docker documentation (#4984)
* add wan2.1 to pyt inference models

* update group name

* fix container tag

* fix group name

* change documented data type to bfloat16

* fix col width
2025-07-02 09:58:37 -04:00
ammallya
73de8a3e46 Removing failing checkout step 2025-07-01 11:25:17 -07:00
Daniel Su
1fc312f90f [Ex CI] fix hardcoded gfx in MIOpen CK script (#4993) 2025-06-30 15:34:54 -04:00
Daniel Su
fde2647ccd [Ex CI] migrate rocBLAS to monorepo (#4987) 2025-06-30 15:16:58 -04:00
Daniel Su
798c8debb5 [Ex CI] consolidate artifact extraction and deletion in deps-rocm (#4961) 2025-06-30 14:12:52 -04:00
dependabot[bot]
393ba600c2 Build(deps): Bump sphinx-sitemap from 2.6.0 to 2.7.2 in /docs/sphinx (#4985)
Bumps [sphinx-sitemap](https://github.com/jdillard/sphinx-sitemap) from 2.6.0 to 2.7.2.
- [Release notes](https://github.com/jdillard/sphinx-sitemap/releases)
- [Changelog](https://github.com/jdillard/sphinx-sitemap/blob/master/CHANGELOG.rst)
- [Commits](https://github.com/jdillard/sphinx-sitemap/compare/v2.6.0...v2.7.2)

---
updated-dependencies:
- dependency-name: sphinx-sitemap
  dependency-version: 2.7.2
  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>
2025-06-30 09:33:28 -06:00
Daniel Su
c64c545b52 [Ex CI] hipBLASLt: build some archs on medium pool (#4986) 2025-06-30 11:32:35 -04:00
Daniel Su
76ee1d720f [Ex CI] rocAL: switch to medium pool (#4983) 2025-06-27 13:41:07 -04:00
Daniel Su
5adc040367 [Ex CI] migrate hipBLAS-common & hipBLASLt pipeline IDs (#4982) 2025-06-27 12:09:58 -04:00
Daniel Su
061da8f306 [Ex CI] enable almalinux8 and gfx1100 builds for hipBLASLt, rocBLAS, rocSOLVER (#4955) 2025-06-27 10:39:30 -04:00
Daniel Su
e26767bca6 [Ex CI] Tensile: add boost filesystem (#4980) 2025-06-27 10:38:31 -04:00
Daniel Su
7b6f1800d4 [Ex CI] fix miopen-get-ck for new artifact naming scheme (#4979) 2025-06-26 15:49:13 -04:00
Pratik Basyal
a6221937f2 KMD UMD support footnote update ROCm 640 (#4973) (#4976)
* KMD UMD support footnote update ROCm 640

* Histotical footnote
2025-06-26 15:34:21 -04:00
Daniel Su
ac2df2961d [Ex CI] add component name to artifact download filter (#4974) 2025-06-26 13:55:03 -04:00
Daniel Su
f20e8dec8b [Ex CI] revert PRIM default branch to develop (#4960) 2025-06-23 16:35:02 -04:00
Daniel Su
10e9157f39 [Ex CI] allow rerun jobs to upload artifacts (#4959) 2025-06-23 15:37:52 -04:00
Daniel Su
a2ce6021cb [Ex CI] add more OSs to nightly build (#4958) 2025-06-23 15:13:11 -04:00
Peter Park
2196fc9a2f Fix pytorch training 25.6 doc (#4956)
* fix pytorch-training history

* fix pytorch-training

fix
2025-06-23 13:45:50 -04:00
Daniel Su
925689f89e [Ex CI] enable gfx1100 builds (#4954) 2025-06-23 11:26:35 -04:00
Peter Park
91a541f8b9 Update PyTorch training benchmark doc for v25.6 (#4950)
* update pytorch-training docker details

* add previous version

* add models data

* update models data id

* add models picker

* update data

* update fmt

fmt

* update data yaml

* update template

* update data

* fix

* fix vllm-0.6.4 broken link

* fix vllm history
2025-06-23 09:26:15 -04:00
Peter Park
34f8d57ece Organize version histories in ROCm for AI benchmark Docker docs (#4948)
* add vllm 0.8.3 20250415

update prev versions table

* add vllm previous versions page

* move index to vllm-history

* add standalone megatron-lm version history

* add pytorch training version history

* fix

* add vllm-0.4.3

* add vllm-0.6.4

* update vllm-history

* add vllm-0.7.3

* add vllm-0.6.6

* add notes

* fix vllm readme links

fix main page link

* add latest version to previous versions list

* add jax-maxtext history

* fix jax-maxtext history

* add pytorch-training history

* add link in jax-maxtext 25.4

* add megatron-lm history

* fix datatemplate path for vllm 0.8.3

* fix jax-maxtext history link

* update note about performance measurements

* add vllm 0.8.5_20250521 previous version

* consistency fixes
2025-06-20 15:01:38 -04:00
yugang-amd
55f95adc7c Update for vllm -06/10 (#4943) 2025-06-20 08:41:37 -04:00
Daniel Su
e05b1702d8 [Ex CI] fix experimental HIP to CLR triggers (#4946) 2025-06-19 12:56:53 -04:00
Daniel Su
4179042cf7 [Ex CI] add multi-OS support to copyHIP (#4945) 2025-06-19 12:15:22 -04:00
dependabot[bot]
ae2de81b79 Build(deps): Bump urllib3 from 2.4.0 to 2.5.0 in /docs/sphinx (#4942)
Bumps [urllib3](https://github.com/urllib3/urllib3) from 2.4.0 to 2.5.0.
- [Release notes](https://github.com/urllib3/urllib3/releases)
- [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst)
- [Commits](https://github.com/urllib3/urllib3/compare/2.4.0...2.5.0)

---
updated-dependencies:
- dependency-name: urllib3
  dependency-version: 2.5.0
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-06-19 09:03:29 -06:00
Daniel Su
efd6cec4a4 [Ex CI] disable downstream triggers for mathlibs not yet migrated (#4936) 2025-06-18 14:10:58 -04:00
Daniel Su
b65996587f [Ex CI] remove ALLOWED_PARTIAL_SUCCEED_BUILDS library variable (#4937) 2025-06-18 12:10:04 -04:00
yugang-amd
7b7eaf69f2 remove broken xref (#4939) 2025-06-18 10:15:53 -04:00
Daniel Su
4cfc8ddad2 [Ex CI] MIVisionX: add hipBLASLt to build deps (#4931) 2025-06-17 13:40:35 -04:00
Daniel Su
97ebbb227d [Ex CI] rocprof-sdk: add cmake, libsqlite3-dev (#4935) 2025-06-17 13:40:15 -04:00
Daniel Su
8c6a1726fe [Ex CI] remove old aqlprofile param in Pytorch (#4927) 2025-06-16 15:17:23 -04:00
Daniel Su
2656143c9e [Ex CI] fix ROCm versions (#4930) 2025-06-16 11:42:51 -04:00
Daniel Su
7910841c94 [Ex CI] rccl: use vendored gtest, use GPU_TARGETS flag (#4929) 2025-06-16 11:35:20 -04:00
Daniel Su
30fec8f74a [Ex CI] update ROCm versioning (#4928) 2025-06-16 11:31:19 -04:00
Daniel Su
1923f801e0 [Ex CI] fix hipRAND multi-OS tests, Tensile sparse dir (#4923) 2025-06-13 16:21:13 -04:00
Peter Park
d69037bfcc Fix Sphinx issue in vllm-benchmark 0.8.5-20250513 previous version (#4924)
* fix sphinx issue in vllm-benchmark 0.8.5-20250513 previous version

* update article_info in conf.py

* update rocm/vllm
2025-06-13 15:03:51 -04:00
Daniel Su
7ac6aa4084 [Ex CI] add OS support to monorepo downstream triggers (#4920) 2025-06-13 12:26:05 -04:00
Daniel Su
14f3c42320 [Ex CI] Tensile almalinux8 builds (#4915) 2025-06-12 16:43:55 -04:00
Daniel Su
67be6f6249 [Ex CI] migrate roc/hipRAND pipelines, change migrated mathlibs's default branch to rocm-rel-7.0 (#4918)
* [Ex CI] migrate roc/hipRAND pipeline IDs to monorepo

* [Ex CI] change migrated mathlibs's default branch to rocm-rel-7.0
2025-06-12 15:39:41 -04:00
powderluv
2502fc5bcf Update README.md to point to TheRock (#4907)
* Update README.md to point to TheRock

Point to TheRock build system to build ROCm

* Update README.md

---------

Co-authored-by: David Galiffi <dgaliffi@amd.com>
Co-authored-by: alexxu-amd <159800977+alexxu-amd@users.noreply.github.com>
2025-06-12 10:44:34 -04:00
Pratik Basyal
61c6749a10 Link to 6.4.1 updated from internal to public (#4913) 2025-06-10 16:59:52 -04:00
Daniel Su
8e8104c811 [Ex CI] add new rocprof-compute pip packages (#4905) 2025-06-10 16:06:51 -04:00
Peter Park
cfb3504d77 Add Mochi Video to pytorch-inference-benchmark-models.yaml
Add Mochi Video to pytorch-inference-benchmark-models.yaml
2025-06-10 13:18:41 -04:00
Joseph Macaranas
3602bc5142 [External CI] Revert hip/clr workaround (#4908)
- Undo workaround for hip/clr.
- Build latest hip/clr and deal with fallout.
- Issues need to be caught and escalated.
2025-06-10 11:13:14 -04:00
Scott Todd
cf4a8ecf28 Advertise TheRock near the top of README.md. (#4906) 2025-06-09 20:20:55 -07:00
Daniel Su
a5aae151b7 [Ex CI] shard rocPRIM tests across 3 runners (#4895) 2025-06-09 15:52:35 -04:00
Alex Xu
685457834a upgrade rocm-docs-core to 1.20.1 2025-06-09 14:53:09 -04:00
Joseph Macaranas
1c715356b6 [External CI] More AlmaLinux 8 Pipelines (#4898)
Added AlmaLinux 8 Pipeline Support
- aomp
- HIPIFY
- rocDecode
- ROCgdb
- rocJPEG
- rocprofiler
- aqlprofile dependency template
- build autotools template
- download latest cmake template

Pipeline Changes
- More gfx build targets.
- Copying llvm-lit to the llvm-project published artifacts.
- HIPIFY now uses our built version of llvm-project for its pipeline.
- Disable testing in HIPIFY pipeline due to low value provided. Revisit in the future.
- aomp's ROCm dependency list reduced.
- aomp's openmp build had issues with ninja on AlmaLinux 8.
2025-06-09 14:16:59 -04:00
Joseph Macaranas
ca8df59ba8 [External CI] Fix pytorch nightly build errors (#4901)
- Add hipSPARSELt dependency.
- Add hipBLASLt test dependency for rocroller shared library.
- Update pip dependency versions.
- Install another typing_extensions at a specific folder for one of the builds we do not control to work.
- Wheel renaming no longer works, so we need to find another mechanism if we start doing builds for different branches and gfx architectures.
2025-06-09 13:15:41 -04:00
Joseph Macaranas
ad1ac5a4e8 [External CI] AlmaLinux 8 builds for more libraries (#4897)
- Fixed rocprim pipeline to not rebuild during install step.
- Updates to hipblas-common, hipcub, hiprand, and rocthrust pipelines to build on AlmaLinux8 and more gfx architectures.
- Include rocm-cmake dependency when CMake setup mentions it.
2025-06-09 10:35:52 -04:00