Skip to content

[TOPI] Expose mem_scope from generic conv2d variants to be more reusable #13680

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
masahi merged 1 commit into from
Jan 3, 2023
Merged

[TOPI] Expose mem_scope from generic conv2d variants to be more reusable #13680

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
6 changes: 4 additions & 2 deletions python/tvm/topi/generic/conv2d.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,7 @@ def schedule_conv_NCHWc_cpu_common_int8(
int8_elems=4,
intrin=None,
inline_fused=True,
mem_scope="global",
):
"""
Defines the schedule for INT8 for Intel and ARM machines
Expand Down Expand Up @@ -186,7 +187,7 @@ def schedule_conv_NCHWc_cpu_common_int8(

# schedule 5-D NCHW[x]c conv
C, O = conv_out, last
CC = s.cache_write(C, "global")
CC = s.cache_write(C, mem_scope)

batch, oc_chunk, oh, ow, oc_block = s[C].op.axis
ow_chunk, ow_block = s[C].split(ow, factor=reg_n)
Expand Down Expand Up @@ -279,6 +280,7 @@ def schedule_conv_NCHWc_cpu_1x1_int8(
int8_elems=4,
intrin=None,
inline_fused=False,
mem_scope="global",
):
"""
Defines the 1x1 conv schedule for INT8 for Intel and ARM machines
Expand Down Expand Up @@ -323,7 +325,7 @@ def schedule_conv_NCHWc_cpu_1x1_int8(
s[kernel_vec].parallel(parallel_axis)

C, O = conv_out, last
CC = s.cache_write(C, "global")
CC = s.cache_write(C, mem_scope)

batch, oc_chunk, oh, ow, oc_block = s[C].op.axis
oh_outer, oh_inner = s[C].split(oh, factor=oh_factor)
Expand Down