 Donato Capitella
|
d19875828c
|
add script to compare performance with/without forcing the hipblaslt path
|
2025-11-18 08:45:25 +00:00 |
|
|
kevinjohncolo
|
f631e45674
|
Create docker-compose-how-to.md (#28)
* Create docker-compose-how-to.md
Simple how-to on using docker compose instead of toolbox.
* Update docker-compose-how-to.md
|
2025-11-18 08:43:22 +00:00 |
|
|
Donato Capitella
|
5001413bcc
|
more typos
|
2025-11-18 08:40:49 +00:00 |
|
|
Donato Capitella
|
6dc6423034
|
typo
|
2025-11-18 08:35:18 +00:00 |
|
|
Donato Capitella
|
140ba2d035
|
Updated README
|
2025-11-18 08:33:54 +00:00 |
|
|
Donato Capitella
|
ccf29e6b22
|
fixed naming convention
|
2025-11-17 23:09:04 +00:00 |
|
|
Donato Capitella
|
1d6d48fae1
|
updated benchmarks
|
2025-11-17 23:02:56 +00:00 |
|
|
Donato Capitella
|
ad32126872
|
Updating retries for run)_benchmark
|
2025-11-17 17:53:53 +00:00 |
|
|
Donato Capitella
|
12f057612b
|
restoring correct llama-bench flags
|
2025-11-17 16:00:10 +00:00 |
|
|
Donato Capitella
|
de02a53d96
|
restored correct benchmark behaviour
|
2025-11-17 15:55:31 +00:00 |
|
|
Donato Capitella
|
f62c6e47c5
|
updated benchmark script to cover HBLASLT for all rocm backends
|
2025-11-17 15:30:19 +00:00 |
|
|
Donato Capitella
|
528923aa66
|
restore PR_15405 for Vulkan backends
|
2025-11-17 11:44:11 +00:00 |
|
|
Donato Capitella
|
eae357f9dd
|
disable PR_15405 for vulkan
|
2025-11-17 11:19:51 +00:00 |
|
|
Donato Capitella
|
1bb4c1f0cc
|
improve logic to check if a benchmakr as already been run
|
2025-11-17 11:01:19 +00:00 |
|
|
Donato Capitella
|
de49d65b3c
|
Ensure ROCBLAS_USE_HIPBLASLT is properly set remotely as well
|
2025-11-17 09:33:14 +00:00 |
|
|
Donato Capitella
|
17b1ec2825
|
run llama-bench INSIDe container (vibe coding is tiring)
|
2025-11-17 09:26:56 +00:00 |
|
|
Donato Capitella
|
6d8ac6d6f4
|
remove user from ssh script
|
2025-11-17 09:10:52 +00:00 |
|
|
Donato Capitella
|
1eade84757
|
fixed remote targets
|
2025-11-17 09:09:15 +00:00 |
|
|
Donato Capitella
|
1e184979df
|
remove user from default host
|
2025-11-17 09:05:13 +00:00 |
|
|
Donato Capitella
|
ecbe5c14c3
|
Fixed model path resolution
|
2025-11-17 08:57:42 +00:00 |
|
|
Donato Capitella
|
a50adb0c15
|
add benchmakr script for RPC
|
2025-11-17 08:27:54 +00:00 |
|
|
Donato Capitella
|
79a2438861
|
copy rpc-server binary to runtime container
|
2025-11-17 08:04:02 +00:00 |
|
|
Donato Capitella
|
9254f7b9e2
|
revert styatic library flag
|
2025-11-16 22:43:14 +00:00 |
|
|
kyuz0
|
c0e74afbb8
|
Disable RPC PR for .rocm-7alpha-rocwmma-improved
|
2025-11-16 10:32:54 +00:00 |
|
|
Donato Capitella
|
0c0835e64c
|
disable 15405 PR for rocmwmma-improved
|
2025-11-16 10:14:41 +00:00 |
|
|
Donato Capitella
|
6b08c48d91
|
adjust toolbox pruning script to remove 6.4.4 old ones
|
2025-11-16 10:06:19 +00:00 |
|
|
Donato Capitella
|
7e583193d0
|
migrated to fedora 43 from rawhide to fix build issues
|
2025-11-16 10:04:39 +00:00 |
|
|
Donato Capitella
|
a164b2308b
|
switching from rawhide to 43
|
2025-11-16 09:44:29 +00:00 |
|
|
Donato Capitella
|
5253e1143b
|
tryign DBUILD_SHARED_LIBS to check if it fixes HIP backend build issues
|
2025-11-16 09:37:53 +00:00 |
|
|
Donato Capitella
|
8cea1363f3
|
remove dangling
|
2025-11-16 08:39:39 +00:00 |
|
|
Donato Capitella
|
bf0d083975
|
Dropping PR 15405 (llama.cpp RPC experimental improvement) due to compile issue
|
2025-11-16 08:25:25 +00:00 |
|
|
Donato Capitella
|
9de07b1d25
|
Enable RPC builds and merge PR 15405 across Dockerfiles
|
2025-11-16 07:54:49 +00:00 |
|
|
Donato Capitella
|
67fb3a002b
|
Updated benchmarks
|
2025-11-15 08:36:25 +00:00 |
|
|
Donato Capitella
|
1d945f2c21
|
change llama-bench retries to 3
|
2025-11-12 14:19:47 +00:00 |
|
|
kyuz0
|
bc8fb4b531
|
Merge pull request #25 from kyuz0/pr-20
add benchmakrs
|
2025-11-12 14:07:28 +00:00 |
|
|
Donato Capitella
|
79479ec596
|
adding rocm7alpha to the benchmarks
|
2025-11-12 14:06:00 +00:00 |
|
|
Donato Capitella
|
f93c88b792
|
limit llama-bench to 1 try to support longer context
|
2025-11-12 13:57:24 +00:00 |
|
|
Donato Capitella
|
11048c22f2
|
update benchmark script
|
2025-11-12 13:48:42 +00:00 |
|
|
kyuz0
|
6a50844302
|
Merge pull request #24 from kyuz0/pr-20
Pr 20
|
2025-11-12 13:20:58 +00:00 |
|
|
Donato Capitella
|
40a47116a9
|
Merge remote-tracking branch 'origin/main' into pr-20
|
2025-11-12 13:19:56 +00:00 |
|
|
Donato Capitella
|
9529c03e61
|
Fix Docker build parallelism flag by removing extra quoting around -j$(nproc)
|
2025-11-12 12:32:09 +00:00 |
|
|
Donato Capitella
|
a044056534
|
Use absolute include path for HIP shuffle shim to fix CMake compiler detection
|
2025-11-12 11:54:52 +00:00 |
|
|
Donato Capitella
|
0fc19e1849
|
moving folder to the right place
|
2025-11-12 11:12:07 +00:00 |
|
|
Donato Capitella
|
e9ed0bac22
|
Copy local HIP shuffle shim into build image to restore __shfl_sync support on gfx1151
|
2025-11-12 10:44:26 +00:00 |
|
|
Donato Capitella
|
48ba7c43a1
|
Add HIP shuffle compatibility shim header for gfx1151
|
2025-11-12 10:42:37 +00:00 |
|
|
Donato Capitella
|
42bbc2301e
|
Force-include HIP shuffle shim to fix missing __shfl_sync on gfx1151 builds
|
2025-11-12 10:16:34 +00:00 |
|
|
Donato Capitella
|
b6de7881dd
|
fix hredoc synatx in rocm-6.4.4-rocwmma
|
2025-11-12 09:53:55 +00:00 |
|
|
Donato Capitella
|
ff0ef125cc
|
Add HIP shuffle macro shim to restore __shfl_sync support on gfx1151
|
2025-11-12 09:43:12 +00:00 |
|
|
Donato Capitella
|
be28cb2ad5
|
Add HIP shuffle compatibility shim for gfx1151 builds
|
2025-11-12 08:28:03 +00:00 |
|
|
Donato Capitella
|
e36bd3e4ec
|
trying another fix for rocm-6.4.4-rocwmma
|
2025-11-12 07:55:03 +00:00 |
|