Skip to content

Disable ZeroOutBuffer for CK Grouped Conv Fwd #3756

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

Open
wants to merge 1 commit into
base: develop
Choose a base branch
from
+9 −5
Open

Disable ZeroOutBuffer for CK Grouped Conv Fwd #3756

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 9 additions & 5 deletions src/include/miopen/solver/implicitgemm_ck_util.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -869,11 +869,15 @@ ConvSolution InitInvokerFactoryNCHW(const ExecutionContext& ctx,
output_init_tr_inst.ConvertFrom(handle, kernels, conv_tensors);

/// \todo: Will need SetTensor() to properly zero out non-packed tensors
/// Note: Need to clear buffer memory for output since all values may not be set.
elapsed = handle.IsProfilingEnabled() ? handle.GetKernelTime() : 0.0f;
output_tr_inst.ZeroOutBuffer(handle);
if(handle.IsProfilingEnabled())
elapsed += handle.GetKernelTime();
/// Note: Need to clear buffer memory for BWD and WRW since all values may not be set.
if(output_tr_inst.GetConvOperandTag() == internal::ConvOperandTag::Weights ||
output_tr_inst.GetConvOperandTag() == internal::ConvOperandTag::Input)
{
elapsed = handle.IsProfilingEnabled() ? handle.GetKernelTime() : 0.0f;
output_tr_inst.ZeroOutBuffer(handle);
if(handle.IsProfilingEnabled())
elapsed += handle.GetKernelTime();
}

std::array<internal::TransposeInstanceTagged*, 3> tr_ptrs = {
&input1_tr_inst, &input2_tr_inst, &output_tr_inst};
Expand Down
Loading