Skip to content
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

Jump to bottom

Fix mixed precision bug #49239

Merged
merged 3 commits into from
Dec 22, 2022
Merged

Fix mixed precision bug #49239

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
4743cc8
[Release2.4] Revert python link prs (#48573)
chenwhql Dec 2, 2022
18f151e
Merge branch 'release/2.4' of https://github.com/PaddlePaddle/Paddle …
yuanlehome Dec 21, 2022
0741d1b
fix mixed precision inference
yuanlehome Dec 21, 2022
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
33 changes: 33 additions & 0 deletions paddle/fluid/framework/ir/auto_mixed_precision_pass.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -437,6 +437,20 @@ void AutoMixedPrecisionPass::UpdateOpPrecision() const {
vars_should_not_low_precision.insert(in_var_node->Var()->Name());
}
}

// when op_1 only support cpu kernel. if op_2's intput var is op_1's
// output var, then op_2 should not run half.
if (GetOpOriginalType(op_type) != "feed" &&
!GpuKernelSupportPrecision(GetOpOriginalType(op_type),
phi::DataType::FLOAT32)) {
for (auto* out_var_node : op_node->outputs) {
CHECK_EQ(out_var_node->IsVar(), true);
if (out_var_node->Var()->Persistable()) continue;
if (!VarNodeHasDtype(out_var_node)) continue;

vars_should_not_low_precision.insert(out_var_node->Var()->Name());
}
}
}
}
};
Expand All @@ -449,6 +463,25 @@ void AutoMixedPrecisionPass::UpdateOpPrecision() const {
for (auto* op_node : nodes) {
if (op_run_low_precision_.count(op_node->Op()->Type()) == 0) continue;

for (auto* in_var_node : op_node->inputs) {
CHECK_EQ(in_var_node->IsVar(), true);
if (!VarNodeHasDtype(in_var_node)) continue;

auto* real_in_var_node = real_vars_[in_var_node->Var()->Name()];
if (real_in_var_node->Var()->Persistable()) continue;

if (vars_should_not_low_precision.count(
real_in_var_node->Var()->Name())) {
op_run_low_precision_.erase(op_node->Op()->Type());
precision_updated = true;
VLOG(4) << op_node->Op()->Type()
<< " should not run at low precision.";
break;
}
}

if (op_run_low_precision_.count(op_node->Op()->Type()) == 0) continue;

for (auto* out_var_node : op_node->outputs) {
CHECK_EQ(out_var_node->IsVar(), true);
if (!VarNodeHasDtype(out_var_node)) continue;
Expand Down