Skip to content

bulk-copy fast path for List(u8, _) deserialize #66

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
gballet merged 2 commits into from
Jul 16, 2026
Merged

bulk-copy fast path for List(u8, _) deserialize #66

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
39d2eb3
perf: bulk-copy fast path for List(u8, _) deserialize
omerfirmak Jun 14, 2026
a9d6dbe
skip call to deserialize altogether to remove ambiguity
gballet Jul 15, 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
7 changes: 7 additions & 0 deletions src/utils.zig
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,6 +72,13 @@ pub fn List(T: type, comptime N: usize) type {
const alloc = allocator orelse return error.AllocatorRequired;
out.* = try init(alloc);

if (comptime Self.Item == u8) {
// route through deserialize's existing []u8 fast path
if (serialized.len > N) return error.OffsetExceedsSize;
try out.inner.resize(alloc, serialized.len);

@gballet gballet Jun 19, 2026

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

this is tricky, because if the data isn't already allocated in the slice, then the resizing will incur a copy of everything that came before the deserialization. There are therefore cases in which allocating the data in deserialize by passing an allocator is the better move.

@omerfirmak omerfirmak Jun 26, 2026

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

copying over from TG.

out is initialized as empty. so resize will be a fresh new allocation. Also if we let deserialize do the allocation, ArrayList will be weirdly initialized. capacity field will be not set properly, no?

@gballet gballet Jul 15, 2026

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

so the problem was really that there is an allocation site that is optional, and because the array list has its own allocator, it was really redundant. But if you look at it, the "fast path" is to just call the @memcopy function. Might as well save a lot of clutter and just do the copy directly here. LMK what you think

return deserialize([]u8, serialized, &out.inner.items, null);
}

// FastSSZ-style capacity optimization: pre-allocate based on input size
// TODO: replace this with the definite value, taken from the list
if (serialized.len > 0) {
Expand Down
Loading