Skip to content

Fix no-golden TS rows: feed golden a 0-dim tensor scalar #13

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
Aswinmcw merged 1 commit into from
Jul 3, 2026
Merged

Fix no-golden TS rows: feed golden a 0-dim tensor scalar #13

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: 5 additions & 1 deletion eltwise_support_probe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -424,7 +424,11 @@ def build_args(base, dtype, layout, mem, device, bcast="none", variant="tt"):
if variant == "ts" and base in BINARY:
s = _ts_scalar(dtype)
p = BPARAMS.get(base, ())
return [a, s, *p], {}, [a_t, s, *p]
# the device op takes a python scalar, but several torch goldens (maximum,
# minimum, logical_*, ldexp, logaddexp*) reject a python scalar 2nd operand
# -> feed the golden a 0-dim tensor instead (broadcasts identically).
s_ref = torch.tensor(s, dtype=(torch.int32 if dtype in _INT_DTYPES else torch.float32))
return [a, s, *p], {}, [a_t, s_ref, *p]
bsh = _bcast_shape(base, bcast)
fi = bcast != "none" # broadcast operands go interleaved (can't shard a sub-tile)
if base in TERNARY:
Expand Down
Loading