[AMD] Fix TP2 DeepSeek-R1 nhead=64 MLA decode crash and add nightly coverage (#27188)

Co-authored-by: clintg6 <7388379+clintg6@users.noreply.github.com>
2026-06-03 16:56:05 -07:00
parent c9ca56da8c
commit cfb7fb4fad
6 changed files with 518 additions and 1 deletions
@@ -114,6 +114,8 @@ NIGHTLY_SUITES = {
        "nightly-amd-1-gpu",
        "nightly-amd-1-gpu-mi35x",
        "nightly-amd-1-gpu-zimage-turbo",
+        "nightly-amd-2-gpu-mi35x-deepseek-r1-mxfp4-tp2",
+        "nightly-amd-8-gpu-mi35x-deepseek-r1-mxfp4-tp4",
        "nightly-amd-4-gpu",
        "nightly-amd-8-gpu",
        "nightly-amd-vlm",