Skip to content

GAP9: -O3 hot kernels; make tile-control-table memory level configurable #199

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
Victor-Jung merged 5 commits into from
Aug 20, 2026
Merged

GAP9: -O3 hot kernels; make tile-control-table memory level configurable #199

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
57e7b82
perf(GAP9): -O3 on hot forward kernels (Conv / DWConv / Gemm)
runwangdl Jul 2, 2026
1e91ff7
fix(GAP9): hoist L2->L1 tile-control tables to L2
runwangdl Jul 2, 2026
9eb5515
Revert "fix(GAP9): hoist L2->L1 tile-control tables to L2"
runwangdl Aug 19, 2026
3fe7c12
Make the tile-control-table memory level configurable
runwangdl Aug 19, 2026
72fd24f
Clarify comment
Victor-Jung Aug 20, 2026
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
6 changes: 5 additions & 1 deletion Deeploy/TilingExtension/CodeTransformationPasses/TilingHoistingMixIn.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,10 @@ def dictOfArrays(arrayOfDicts: Sequence[Mapping[KT, VT]]) -> Mapping[KT, List[VT

class TilingHoistingMixIn:

# JUNGVI: This attributes is here to override manually where the tiling informations are stored.
# It is useful for debugging purposes but too nice to expose to the CLI.
tileControlTableMemoryLevel: Optional[str] = None

_DEFAULT_HOIST_PREFIX = "TILING_CODEGEN_"

def __init__(self, memory: str) -> None:
Expand Down Expand Up @@ -60,7 +64,7 @@ def _hoistValues(self,
else:
cb._type = PointerClass(BasicDataTypes.minimalIntegerType(values))
cb._instance = cb._type(cb.name, ctxt)
cb._memoryLevel = self.memory
cb._memoryLevel = self.tileControlTableMemoryLevel or self.memory
return cb

def _hoistReference(self,
Expand Down
10 changes: 10 additions & 0 deletions TargetLibraries/GAP9/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,16 @@ target_compile_options(deeploygap9 PRIVATE

target_link_libraries(deeploygap9 PUBLIC pmsis)

# Compile the hot forward kernels at -O3 (set last so it wins over the SDK's
# default -Os). Conv / depthwise-conv / Gemm dominate GAP9 inference cycles;
# -O3 turns on the RISC-V (XpulpV2) hardware loops on their tight inner loops.
Comment thread
Victor-Jung marked this conversation as resolved.
set(_KERNEL_O3_FILES
${CMAKE_CURRENT_LIST_DIR}/../PULPOpen/src/Convolution_fp32.c
${CMAKE_CURRENT_LIST_DIR}/../PULPOpen/src/DWConvolution_fp32.c
${CMAKE_CURRENT_LIST_DIR}/../PULPOpen/src/Gemm.c
)
set_source_files_properties(${_KERNEL_O3_FILES} PROPERTIES COMPILE_OPTIONS "-O3")

#RW: Link PULP-NN
#RW: Set PULP-NN version and bitwidth for pulp-nn-mixed
set(PULPNNVERSION XPULPV2)
Expand Down
Loading