Skip to content
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

Unconditional Lookup in Neighbor Table #1797

Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Unconditional Lookup in Neighbor Table #1797

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
fe05eac
Unconditional lookup in neighbor table
fthaler Jul 24, 2024
4602b32
Merge remote-tracking branch 'upstream/master' into uncond-lookup
fthaler Jul 24, 2024
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
4 changes: 2 additions & 2 deletions include/gridtools/fn/unstructured.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -86,9 +86,9 @@ namespace gridtools::fn {
template <class Tag, class Ptr, class Strides, class Domain, class Conn, class Offset>
GT_FUNCTION constexpr auto horizontal_shift(iterator<Tag, Ptr, Strides, Domain> const &it, Conn, Offset) {
auto const &table = host_device::at_key<Conn>(it.m_domain.m_tables);
auto new_index = it.m_index == -1 ? -1 : get<Offset::value>(neighbor_table::neighbors(table, it.m_index));
auto new_index = get<Offset::value>(neighbor_table::neighbors(table, std::max(it.m_index, 0)));
Copy link
Contributor

Choose a reason for hiding this comment

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

Hmm. For exotic cases the table might not have an entry at 0? E.g. after an origin shift. But maybe we can document in the neighbor_table concept that the 0 index need to exist.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

No idea if there are realistic cases where this happens. And no idea if it makes sense to include this workaround. It’s currently required to have reasonable performance on Daint due to the old CUDA versions. But once we move the performance tests to Alps (we probably will, right?), the GT_PROMISE thing should be good enough, I guess.

auto shifted = it;
shifted.m_index = new_index;
shifted.m_index = it.m_index == -1 ? -1 : new_index;
return shifted;
}

Expand Down
Loading