Skip to content

Issues: ROCm/AMDMIGraphX

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

342 Open 892 Closed
342 Open 892 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Investigate memory failures when workspace size is set to 0 in compile hipblaslt pass bug Something isn't working
#3622 opened Nov 14, 2024 by ahsan-ca
@ahsan-ca
Investigate llama2 int4 memory usage
#3617 opened Nov 13, 2024 by turneram
@turneram
Quantized distillgpt2 accuracy issue when quant params are inputs bug Something isn't working
#3612 opened Nov 11, 2024 by shivadbhavsar
@pfultz2 @shivadbhavsar
UAI llama2 script recompiles for each prompt even with prefill approach
#3601 opened Nov 8, 2024 by turneram
@turneram
Dangling quantizelinear from horizontal fusion, BERT and DistilGPT2 FP8 issues related to FP8 implemenation INT8 Perf Improve
#3598 opened Nov 7, 2024 by CharlieL7
Missing constant propagation: Literal -> Multibroadcast -> Quantizelinear FP8 issues related to FP8 implemenation INT8 Perf Improve
#3597 opened Nov 7, 2024 by CharlieL7
GroupQueryAttention produces incorrect results when loaded from mxr
#3596 opened Nov 6, 2024 by turneram
1
@turneram
4
Remove hipblaslt version check in hip_gemm_impl.cpp for OCP FP8 types
#3592 opened Nov 5, 2024 by CharlieL7
PagedAttention support in MIGraphX Under Investigation
#3588 opened Nov 5, 2024 by gyulaz-htec
Make check failures on Navi48
#3576 opened Oct 30, 2024 by ahsan-ca
@ahsan-ca
1
int4_unpack_conv test of test_gpu_mlir failing on Navi21
#3565 opened Oct 28, 2024 by CharlieL7
Use dynamic shapes for kv-cache
#3552 opened Oct 24, 2024 by pfultz2
@CharlieL7 @turneram
Improve fusions with dequantizelinear bug Something isn't working high priority A PR with high priority for review and merging.
#3551 opened Oct 24, 2024 by causten
@lakhinderwalia
7
Benchmark llama2 int4 performance
#3534 opened Oct 16, 2024 by turneram
@turneram
Support text export of fusion subgraphs
#3525 opened Oct 11, 2024 by hgaspar
[BF16] GPU Implementation
#3519 opened Oct 9, 2024 by richagadgil
@pfultz2 @richagadgil
1
Add SD3 end-to-end example
#3518 opened Oct 9, 2024 by kahmed10
2 of 3 tasks
1
@kahmed10
Add hipblaslt related env vars to env_vars.rst
#3499 opened Oct 2, 2024 by ahsan-ca
@ahsan-ca
[BUG] Runtime Error: No HIP GPUs Available when torch imported after migraphx
#3488 opened Sep 27, 2024 by richagadgil
@richagadgil
Reduce compile time for MIGraphX
#3485 opened Sep 27, 2024 by aarushjain29
@aarushjain29
Uncomment -bugprone-multi-level-implicit-pointer-conversion Under Investigation
#3472 opened Sep 23, 2024 by TedThemistokleous
3 tasks
1
@TedThemistokleous
1
Add llama2 with KV-cache example
#3458 opened Sep 18, 2024 by turneram
@turneram
Add llama2 with KV-cache to DLM
#3457 opened Sep 18, 2024 by turneram
@turneram
Add compile pass for hipblaslt enhancement New feature or request
#3455 opened Sep 18, 2024 by ahsan-ca
2
@ahsan-ca
3
Improve driver verify through bisection
#3447 opened Sep 16, 2024 by richagadgil
@richagadgil
ProTip! Adding no:label will show everything without a label.