Skip to content

[RNN] Dynamic algo optimization #3470

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 11 commits into from
Feb 7, 2025
Merged

[RNN] Dynamic algo optimization #3470

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
882850c
log2len
shurale-nkn Jan 7, 2025
db49050
tidy
shurale-nkn Jan 8, 2025
ff56f0f
driver update
shurale-nkn Jan 13, 2025
a4db508
post merge fix
shurale-nkn Jan 14, 2025
c46b371
driver speedup for benchmarking
shurale-nkn Jan 17, 2025
7c3fa95
Merge remote-tracking branch 'origin/develop' into dlstm_log2SeqLen
shurale-nkn Jan 17, 2025
a184191
tidy
shurale-nkn Jan 17, 2025
da178ce
tidy + separate class
shurale-nkn Jan 24, 2025
b4d00ff
cleanup
shurale-nkn Jan 27, 2025
c22fa88
Merge remote-tracking branch 'origin/develop' into dlstm_log2SeqLen
shurale-nkn Jan 27, 2025
8607915
Merge branch 'develop' into dlstm_log2SeqLen
shurale-nkn Feb 6, 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
12 changes: 10 additions & 2 deletions driver/rnn_driver.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -470,6 +470,10 @@ int RNNDriver<Tgpu, Tref>::SetRNNDescriptorFromCmdLineArgs()
{
algo = miopenRNNfundamental;
}
else if((inflags.GetValueInt("rnnalgo")) == 2)
{
algo = miopenRNNroundedDynamic;
}
else
{
printf("Incorrect RNN algorithm\n");
Expand Down Expand Up @@ -616,7 +620,9 @@ int RNNDriver<Tgpu, Tref>::AllocateBuffersAndCopy()
workspace = std::vector<Tgpu>(workSpace_sz, static_cast<Tgpu>(0));
reservespace = std::vector<Tgpu>(reserveSpace_sz, static_cast<Tgpu>(0));
outhost = std::vector<Tref>(out_sz, static_cast<Tref>(0));
workspace_host = std::vector<Tref>(workSpace_sz, static_cast<Tref>(0));
workspace_host = (inflags.GetValueInt("verify") == 1)
? std::vector<Tref>(workSpace_sz, static_cast<Tref>(0))
: std::vector<Tref>{};

int nseq = inflags.GetValueInt("seq_len");
std::vector<int> in_n = GetInputTensorLengthsFromCmdLine();
Expand All @@ -641,7 +647,9 @@ int RNNDriver<Tgpu, Tref>::AllocateBuffersAndCopy()
reserveSpaceHost_sz += (layer - 1) * inputBatchLenSum * hid_h * (bidir + 1);
reserveSpaceHost_sz = (reserveSpaceHost_sz + sizeof(Tref) - 1) / sizeof(Tref);
}
reservespace_host = std::vector<Tref>(reserveSpaceHost_sz, static_cast<Tref>(0));
reservespace_host = (inflags.GetValueInt("verify") == 1)
? std::vector<Tref>(reserveSpaceHost_sz, static_cast<Tref>(0))
: std::vector<Tref>{};

if(inflags.GetValueInt("forw") != 1)
{
Expand Down
19 changes: 13 additions & 6 deletions driver/rnn_seq_driver.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -700,6 +700,10 @@ int RNNSeqDriver<Tgpu, Tref>::SetRNNDescriptorFromCmdLineArgs()
{
algo = miopenRNNfundamental;
}
else if((inflags.GetValueInt("rnnalgo")) == 2)
{
algo = miopenRNNroundedDynamic;
}
else
{
printf("Incorrect RNN algorithm\n");
Expand Down Expand Up @@ -906,7 +910,9 @@ int RNNSeqDriver<Tgpu, Tref>::AllocateBuffersAndCopy()
workspace = std::vector<Tgpu>(workSpace_sz, static_cast<Tgpu>(0));
reservespace = std::vector<Tgpu>(reserveSpace_sz, static_cast<Tgpu>(0));
outhost = std::vector<Tref>(out_host_sz, static_cast<Tref>(0));
workspace_host = std::vector<Tref>(workSpace_sz, static_cast<Tref>(0));
workspace_host = (inflags.GetValueInt("verify") == 1)
? std::vector<Tref>(workSpace_sz, static_cast<Tref>(0))
: std::vector<Tref>{};

/// dropout legacy format
const std::size_t inputBatchLenSum = vectors_cnt_host;
Expand All @@ -927,7 +933,9 @@ int RNNSeqDriver<Tgpu, Tref>::AllocateBuffersAndCopy()
reserveSpaceHost_sz += (layer - 1) * inputBatchLenSum * hid_h * (bidir + 1);
reserveSpaceHost_sz = (reserveSpaceHost_sz + sizeof(Tref) - 1) / sizeof(Tref);
}
reservespace_host = std::vector<Tref>(reserveSpaceHost_sz, static_cast<Tref>(0));
reservespace_host = (inflags.GetValueInt("verify") == 1)
? std::vector<Tref>(reserveSpaceHost_sz, static_cast<Tref>(0))
: std::vector<Tref>{};
// end dropout

if(inflags.GetValueInt("forw") != 1)
Expand Down Expand Up @@ -1078,13 +1086,12 @@ int RNNSeqDriver<Tgpu, Tref>::RunForwardGPU()
RNNCombTimeLoger t(GetStream(), inflags.GetValueInt("iter"), inflags.GetValueInt("wall"));

from_gpu_out = std::vector<Tgpu>(out_dev->GetSize() / sizeof(Tgpu), static_cast<Tgpu>(0));
out_dev->ToGPU(q, from_gpu_out.data());
workspace_dev->ToGPU(q, workspace.data());
reservespace_dev->ToGPU(q, reservespace.data());

for(int i = 0; i < inflags.GetValueInt("iter"); i++)
{
out_dev->ToGPU(q, from_gpu_out.data());
workspace_dev->ToGPU(q, workspace.data());
reservespace_dev->ToGPU(q, reservespace.data());

t.Start();
miopenRNNForward(GetHandle(),
rnnDesc,
Expand Down
Loading
Loading