Skip to content

[https://nvbugs/5545522][fix] Correct Cutlass with PDL support #9335

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

Merged
merged 2 commits into from
Nov 22, 2025
+6 −2
Merged

[https://nvbugs/5545522][fix] Correct Cutlass with PDL support #9335

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
af6de8c
[https://nvbugs/5545522][fix] Correct Cutlass with PDL support
liji-nv Nov 20, 2025
fa7abd5
Limit thread to avoid build OOM
liji-nv Nov 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion 3rdparty/cutlass
Open in desktop
Submodule cutlass updated 1 files
+3 −1 include/cutlass/gemm/kernel/sm90_tile_scheduler_stream_k.hpp
4 changes: 4 additions & 0 deletions cpp/tensorrt_llm/kernels/cutlass_kernels/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,8 @@ function(process_target target_name enable_hopper enable_blackwell)
if(${enable_hopper} AND "90" IN_LIST CMAKE_CUDA_ARCHITECTURES_ORIG)
# No kernels should be parsed, unless hopper is specified. This is a build
# time improvement
target_compile_options(${target_name}
PRIVATE "-DCUTLASS_ENABLE_GDC_FOR_SM90=1")
target_compile_definitions(${target_name} PUBLIC COMPILE_HOPPER_TMA_GEMMS)
target_compile_definitions(${target_name}
PUBLIC COMPILE_HOPPER_TMA_GROUPED_GEMMS)
Expand All @@ -77,6 +79,8 @@ function(process_target target_name enable_hopper enable_blackwell)
OR "121" IN_LIST CMAKE_CUDA_ARCHITECTURES_ORIG
))

target_compile_options(${target_name}
PRIVATE "-DCUTLASS_ENABLE_GDC_FOR_SM100=1")
# Both 100 and 103 support these kernels
if("100" IN_LIST CMAKE_CUDA_ARCHITECTURES_ORIG
OR "103" IN_LIST CMAKE_CUDA_ARCHITECTURES_ORIG)
Expand Down
2 changes: 1 addition & 1 deletion jenkins/L0_Test.groovy
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,7 +82,7 @@ BUILD_CORES_REQUEST = "8"
BUILD_CORES_LIMIT = "8"
BUILD_MEMORY_REQUEST = "48Gi"
BUILD_MEMORY_LIMIT = "64Gi"
BUILD_JOBS = "8"
BUILD_JOBS = "4"

SLURM_CORES_REQUEST = "1"
SLURM_CORES_LIMIT = "1"
Expand Down