Stream: git-wasmtime

Topic: wasmtime / PR #1528 arm64: Support less-than-64-bit integ...


view this post on Zulip Wasmtime GitHub notifications bot (Apr 16 2020 at 21:40):

cfallin opened PR #1528 from aarch64-bit-ops-fix to master:

This PR adds support for 8- and 16-bit cases of bitrev, clz, cls and popcnt.

Also includes a temporary bugfix for popcnt with 32-bit operand. The popcnt
issue was initially identified by Benjamin Bouvier \<public@benj.me>\, and
the root cause was debugged by Joey Gouly \<joey.gouly@arm.com>`. This
patch is simply a quick fix that zero-extends the operand to 64 bits;
Joey plans to contribute a more permanent fix shortly.

Rebased from commits 7e915fa4f443373d7483392637905dc299ab1c67, bb47c80dbe4aae7d631e9878ad7808cd96f50299 and 7a2eda3f54593bc2249a63bfb1f17a0dd18f8eaf on the arm64 development branch, which landed there after our initial merge.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 16 2020 at 21:41):

cfallin edited PR #1528 from aarch64-bit-ops-fix to master:

This PR adds support for 8- and 16-bit cases of bitrev, clz, cls and popcnt.

Also includes a temporary bugfix for popcnt with 32-bit operand. The popcnt
issue was initially identified by Benjamin Bouvier \<public@benj.me>\, and
the root cause was debugged by Joey Gouly \<joey.gouly@arm.com>`. The
part of this patch related to 32-bit popcnt is simply a quick fix that zero-extends
the operand to 64 bits; Joey plans to contribute a more permanent fix shortly.

Rebased from commits 7e915fa4f443373d7483392637905dc299ab1c67, bb47c80dbe4aae7d631e9878ad7808cd96f50299 and 7a2eda3f54593bc2249a63bfb1f17a0dd18f8eaf on the arm64 development branch, which landed there after our initial merge.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr created PR Review Comment:

nit: either uncomment, or can you open an issue to fix this later + mention the issue number here, please?

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr created PR Review Comment:

nit: Can you replace has_rev by its definition here? (since it's used only once)

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr created PR Review Comment:

nit: can you move this definition just before it's used, please?

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 09:52):

bnjbvr created PR Review Comment:

For what it's worth, I think there's a way to start catching from this line, and stopping at the line that does the lsr. Unless you explicitly wanted to test that there was no stack space allocated and no other instructions generated for this? (which is nice for completeness)

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin created PR Review Comment:

Done!

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin created PR Review Comment:

Done!

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:05):

cfallin created PR Review Comment:

Yep, opened #1537 for this.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:07):

cfallin submitted PR Review.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:07):

cfallin created PR Review Comment:

This is something I've wanted to clean up in our vcode filetests -- there should be a better way than matching a whole golden output. The reason for this for now is to ensure nothing else is generated, yes; I'll create an issue to track this desired improvement though.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 17:10):

cfallin updated PR #1528 from aarch64-bit-ops-fix to master:

This PR adds support for 8- and 16-bit cases of bitrev, clz, cls and popcnt.

Also includes a temporary bugfix for popcnt with 32-bit operand. The popcnt
issue was initially identified by Benjamin Bouvier \<public@benj.me>\, and
the root cause was debugged by Joey Gouly \<joey.gouly@arm.com>`. The
part of this patch related to 32-bit popcnt is simply a quick fix that zero-extends
the operand to 64 bits; Joey plans to contribute a more permanent fix shortly.

Rebased from commits 7e915fa4f443373d7483392637905dc299ab1c67, bb47c80dbe4aae7d631e9878ad7808cd96f50299 and 7a2eda3f54593bc2249a63bfb1f17a0dd18f8eaf on the arm64 development branch, which landed there after our initial merge.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 17 2020 at 23:43):

cfallin updated PR #1528 from aarch64-bit-ops-fix to master:

This PR adds support for 8- and 16-bit cases of bitrev, clz, cls and popcnt.

Also includes a temporary bugfix for popcnt with 32-bit operand. The popcnt
issue was initially identified by Benjamin Bouvier \<public@benj.me>\, and
the root cause was debugged by Joey Gouly \<joey.gouly@arm.com>`. The
part of this patch related to 32-bit popcnt is simply a quick fix that zero-extends
the operand to 64 bits; Joey plans to contribute a more permanent fix shortly.

Rebased from commits 7e915fa4f443373d7483392637905dc299ab1c67, bb47c80dbe4aae7d631e9878ad7808cd96f50299 and 7a2eda3f54593bc2249a63bfb1f17a0dd18f8eaf on the arm64 development branch, which landed there after our initial merge.

view this post on Zulip Wasmtime GitHub notifications bot (Apr 18 2020 at 00:16):

cfallin merged PR #1528.


Last updated: Dec 23 2024 at 12:05 UTC