Stream: git-wasmtime

Topic: wasmtime / PR #2324 Further compress the in-memory repres...

Wasmtime GitHub notifications bot (Oct 26 2020 at 22:50):

alexcrichton opened PR #2324 from compress-again to main:

This commit reduces the size of InstructionAddressMap from 24 bytes to
8 bytes by dropping the code_len field and reducing code_offset to
u32 instead of usize. The intention is to primarily make the
in-memory version take up less space, and the hunch is that the
code_len is largely not necessary since most entries in this map are
always adjacent to one another. The code_len field is now implied by
the code_offset field of the next entry in the map.

This isn't as big of an improvement to serialized module size as #2321
or #2322, primarily because of the switch to variable-length encoding.
Despite this though it shaves about 10MB off the encoded size of the
module from #2318

Wasmtime GitHub notifications bot (Oct 27 2020 at 00:14):

alexcrichton updated PR #2324 from compress-again to main:

This commit reduces the size of InstructionAddressMap from 24 bytes to
8 bytes by dropping the code_len field and reducing code_offset to
u32 instead of usize. The intention is to primarily make the
in-memory version take up less space, and the hunch is that the
code_len is largely not necessary since most entries in this map are
always adjacent to one another. The code_len field is now implied by
the code_offset field of the next entry in the map.

This isn't as big of an improvement to serialized module size as #2321
or #2322, primarily because of the switch to variable-length encoding.
Despite this though it shaves about 10MB off the encoded size of the
module from #2318

Wasmtime GitHub notifications bot (Nov 02 2020 at 23:56):

peterhuene requested peterhuene for a review on PR #2324.

Wasmtime GitHub notifications bot (Nov 03 2020 at 00:32):

peterhuene submitted PR Review.

Wasmtime GitHub notifications bot (Nov 03 2020 at 00:32):

peterhuene created PR Review Comment:

Should this have a trailing default range starting at 26 so that the below assert (e.g. assert_eq!(26, positions[0].gen_end);) remains unchanged?

Wasmtime GitHub notifications bot (Nov 03 2020 at 01:43):

peterhuene submitted PR Review.

Wasmtime GitHub notifications bot (Nov 03 2020 at 02:05):

alexcrichton updated PR #2324 from compress-again to main:

This commit reduces the size of InstructionAddressMap from 24 bytes to
8 bytes by dropping the code_len field and reducing code_offset to
u32 instead of usize. The intention is to primarily make the
in-memory version take up less space, and the hunch is that the
code_len is largely not necessary since most entries in this map are
always adjacent to one another. The code_len field is now implied by
the code_offset field of the next entry in the map.

This isn't as big of an improvement to serialized module size as #2321
or #2322, primarily because of the switch to variable-length encoding.
Despite this though it shaves about 10MB off the encoded size of the
module from #2318

Wasmtime GitHub notifications bot (Nov 03 2020 at 02:05):

alexcrichton submitted PR Review.

Wasmtime GitHub notifications bot (Nov 03 2020 at 02:05):

alexcrichton created PR Review Comment:

Makes sense to me!

Wasmtime GitHub notifications bot (Nov 03 2020 at 02:37):

alexcrichton merged PR #2324.

Last updated: Nov 22 2024 at 17:03 UTC