OSSForks/tensorflow
1
0
Fork 0
mirror of https://github.com/zebrajr/tensorflow.git synced 2025-12-06 12:20:11 +01:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

Merge pull request #88548 from ROCm:gpublaslt_refactoring2_matmul_plan

Browse Source 
PiperOrigin-RevId: 753162409
This commit is contained in:
TensorFlower Gardener 2025-04-30 08:06:01 -07:00
commit cfd4e4d85e
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tensorflow/core/kernels/matmul_util.cc
View File

@ -198,6 +198,7 @@ Status PlanAndAlgorithms::ExecuteOnStream(
if (!plan || algorithm_idx >= algorithms.size()) { if (!plan || algorithm_idx >= algorithms.size()) {
return errors::Internal("MatmulPlan or algorithms are not initialized!"); return errors::Internal("MatmulPlan or algorithms are not initialized!");
} }
TF_RETURN_IF_ERROR(plan->SetAlgorithm(algorithms[algorithm_idx]));
return plan->ExecuteOnStream(stream, a, b, c, c, return plan->ExecuteOnStream(stream, a, b, c, c,
bias, // bias_buffer bias, // bias_buffer
se::DeviceMemoryBase{}, // aux_buffer se::DeviceMemoryBase{}, // aux_buffer
@ -206,8 +207,7 @@ Status PlanAndAlgorithms::ExecuteOnStream(
se::DeviceMemoryBase{}, // c_scale_buffer se::DeviceMemoryBase{}, // c_scale_buffer
se::DeviceMemoryBase{}, // d_scale_buffer se::DeviceMemoryBase{}, // d_scale_buffer
se::DeviceMemoryBase{}, // d_amax_buffer se::DeviceMemoryBase{}, // d_amax_buffer
algorithms[algorithm_idx], scratch_allocator, scratch_allocator, profile_result);
profile_result);
} }
} // namespace tensorflow } // namespace tensorflow