Stream: git-wasmtime

Topic: wasmtime / PR #2219 Fix remaining failures and re-enable ...


view this post on Zulip Wasmtime GitHub notifications bot (Sep 22 2020 at 17:28):

bnjbvr opened PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 22 2020 at 17:36):

bnjbvr updated PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 22 2020 at 17:58):

abrown submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 22 2020 at 17:58):

abrown created PR Review Comment:

Why can't we use the new script?

view this post on Zulip Wasmtime GitHub notifications bot (Sep 22 2020 at 17:59):

abrown submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 08:23):

bnjbvr submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 08:23):

bnjbvr created PR Review Comment:

Because the new script uses cargo +nightly while we need cargo +nightly-2020-... here. I've just been a bit lazy and will implement retrieving the Cargo version with an env variable; with some luck, the other setup actions even do that for us.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 13:18):

bnjbvr updated PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 13:43):

bnjbvr updated PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 13:59):

bnjbvr updated PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 14:03):

bnjbvr updated PR #2219 from reenable-x64 to main:

See commit messages for details.

The interesting failure was actually in the implementation of nearest; the error was hidden in the old backend by the fact that the old backend uses native x86 instructions for nearest, and not the runtime intrinsic. There's a todo for this in the new backend, fwiw.

view this post on Zulip Wasmtime GitHub notifications bot (Sep 23 2020 at 14:42):

bnjbvr merged PR #2219.


Last updated: Nov 22 2024 at 16:03 UTC