Skip to content

[TIR][VM] Revert a change to lower_tvm_builtin.cc from #6126 #8274

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
tmoreau89 merged 2 commits into from
Sep 9, 2021
Merged

[TIR][VM] Revert a change to lower_tvm_builtin.cc from #6126 #8274

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
3787e93
revert a change to lower_tvm_builtin.cc from #6126
Jun 17, 2021
56331a4
disable sim target on VTA tutorial
Jul 28, 2021
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
10 changes: 10 additions & 0 deletions src/tir/transforms/lower_tvm_builtin.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,6 +113,16 @@ class BuiltinLower : public StmtExprMutator {
op = stmt.as<AllocateNode>();
// Get constant allocation bound.
int64_t nbytes = GetVectorBytes(op->dtype);
if (device_type_.defined()) {
Copy link
Contributor

@manupak manupak Sep 13, 2021

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hi, @mbrookhart

Can you explain the rationale for not generating TVMBAW calls for static size allocates for DLCPU ?

cc: @grant-arm @Mousius @mbaret

if (const auto* dev_type = device_type_.as<IntImmNode>()) {
if (dev_type->value == kDLCPU) {
int32_t constant_size = op->constant_allocation_size();
if (constant_size > 0 && constant_size * nbytes < runtime::kMaxStackAlloca) {
return stmt;
}
}
}
}
PrimExpr total_bytes = make_const(op->extents[0].dtype(), nbytes);
for (size_t i = 0; i < op->extents.size(); ++i) {
total_bytes = total_bytes * op->extents[i];
Expand Down
6 changes: 3 additions & 3 deletions vta/tutorials/frontend/deploy_classification.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -191,7 +191,7 @@
env.WGT_WIDTH,
start_name=pack_dict[model][0],
stop_name=pack_dict[model][1],
device_annot=(env.TARGET == "intelfocl" or env.TARGET == "sim"),
device_annot=(env.TARGET == "intelfocl"),
)
else:
relay_prog = mod["main"]
Expand All @@ -203,7 +203,7 @@
relay_prog, target=target, params=params, target_host=env.target_host
)
else:
if env.TARGET == "intelfocl" or env.TARGET == "sim":
if env.TARGET == "intelfocl":
# multiple targets to run both on cpu and vta
target = {"cpu": env.target_vta_cpu, "ext_dev": target}
with vta.build_config(opt_level=3, disabled_pass={"AlterOpLayout"}):
Expand All @@ -221,7 +221,7 @@
remote.upload(temp.relpath("graphlib.tar"))
lib = remote.load_module("graphlib.tar")

if env.TARGET == "intelfocl" or env.TARGET == "sim":
if env.TARGET == "intelfocl":
ctxes = [remote.ext_dev(0), remote.cpu(0)]
m = graph_executor.create(graph, lib, ctxes)
else:
Expand Down