Skip to content

[Experimental] Enable kleidi AI examples to run on graviton3 #1721

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: main
Choose a base branch
from
+10 −5
Open

[Experimental] Enable kleidi AI examples to run on graviton3 #1721

Show file tree
Hide file tree
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
8 changes: 6 additions & 2 deletions torchao/experimental/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,13 +29,17 @@ if(TORCHAO_BUILD_KLEIDIAI)
endif()
include(CMakePrintHelpers)

add_compile_options("-Wall" "-Werror" "-Wno-deprecated")
if(CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
add_compile_options("-Wall" "-Werror" "-Wno-deprecated" "-march=armv8.2-a+dotprod" "-fPIC" "-Wno-error=unknown-pragmas")
else()
add_compile_options("-Wall" "-Werror" "-Wno-deprecated")
endif()

include(CMakePrintHelpers)
message("TORCHAO_INCLUDE_DIRS: ${TORCHAO_INCLUDE_DIRS}")
include_directories(${TORCHAO_INCLUDE_DIRS})

if(CMAKE_SYSTEM_PROCESSOR STREQUAL "arm64")
if(CMAKE_SYSTEM_PROCESSOR STREQUAL "arm64" OR CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
if(TORCHAO_BUILD_KLEIDIAI)
message(STATUS "Building with Arm KleidiAI library")
add_compile_definitions(TORCHAO_ENABLE_KLEIDI=1)
Expand Down
1 change: 1 addition & 0 deletions torchao/experimental/kernels/cpu/aarch64/valpacking/interleave.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@
#include <torchao/experimental/kernels/cpu/aarch64/valpacking/valpack.h>
#include <cassert>
#include <cstring>
#include <cstdint>

// Interleaves data across channels (row/column) and groups.
// Each channel is the same size (vals_per_channel) and is
Expand Down
4 changes: 2 additions & 2 deletions torchao/experimental/ops/packed_weights_header.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -51,7 +51,7 @@ class PackedWeightsHeader {
auto header = reinterpret_cast<const int*>(packed_weights);
assert(header[0] == PackedWeightsHeader::magic);
params_type params;
for (int i = 0; i < params.size(); i++) {
for (size_t i = 0; i < params.size(); i++) {
params[i] = header[i + 2];
}
return PackedWeightsHeader(
Expand All @@ -62,7 +62,7 @@ class PackedWeightsHeader {
if (type != other.type) {
return false;
}
for (int i = 0; i < params.size(); i++) {
for (size_t i = 0; i < params.size(); i++) {
if (params[i] != other.params[i]) {
return false;
}
Expand Down
2 changes: 1 addition & 1 deletion torchao/experimental/ops/parallel-aten-impl.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,7 +5,7 @@
// LICENSE file in the root directory of this source tree.

#pragma once
#include <Aten/Parallel.h>
#include <torchao/experimental/ops/parallel.h>
#include <torch/library.h>
#include <torch/torch.h>

Expand Down