Stream: git-wasmtime

Topic: wasmtime / PR #8207 PCC: x64: insertlane instructions rea...

Wasmtime GitHub notifications bot (Mar 21 2024 at 15:48):

cfallin opened PR #8207 from cfallin:fix-pcc-insertlane to bytecodealliance:main:

Fixes https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=67538.

<!--
Please make sure you include the following information:

• If this work has been discussed elsewhere, please include a link to that
  conversation. If it was discussed in an issue, just mention "issue #...".

• Explain why this change is needed. If the details are in an issue already,
  this can be brief.

Our development process is documented in the Wasmtime book:
https://docs.wasmtime.dev/contributing-development-process.html

Please ensure all communication follows the code of conduct:
https://github.com/bytecodealliance/wasmtime/blob/main/CODE_OF_CONDUCT.md
-->

Wasmtime GitHub notifications bot (Mar 21 2024 at 15:48):

cfallin requested wasmtime-compiler-reviewers for a review on PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 15:48):

cfallin requested wasmtime-core-reviewers for a review on PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 15:48):

cfallin requested fitzgen for a review on PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:06):

alexcrichton commented on PR #8207:

Mind adding more variants for loading floats, in addition to the variants for storing everything?

(module
  (memory 1)
  (func (param v128 i32) (result v128)
      local.get 0
      local.get 1
      f32.load
      f32x4.replace_lane 0)

  (func (param v128 i32) (result v128)
      local.get 0
      local.get 1
      f64.load
      f64x2.replace_lane 1)

  (func (param v128 i32) (result v128)
      local.get 0
      local.get 1
      f64.load
      f64x2.replace_lane 0)

  (func (param v128 i32)
      local.get 1
      local.get 0
      f64x2.extract_lane 1
      f64.store)

  (func (param v128 i32)
      local.get 1
      local.get 0
      f32x4.extract_lane 1
      f32.store)

  (func (param v128 i32)
      local.get 1
      local.get 0
      i8x16.extract_lane_s 1
      i32.store8)

  (func (param v128 i32)
      local.get 1
      local.get 0
      i16x8.extract_lane_s 1
      i32.store16)

  (func (param v128 i32)
      local.get 1
      local.get 0
      i32x4.extract_lane 1
      i32.store)

  (func (param v128 i32)
      local.get 1
      local.get 0
      i64x2.extract_lane 1
      i64.store)
)

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:07):

alexcrichton commented on PR #8207:

Also I'd recommend running the test with/without AVX to confirm that that all works too

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:37):

cfallin updated PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:38):

cfallin commented on PR #8207:

Sure, added additional cases, thanks! This surfaced another issue with the range-fact clamping on 128-bit values, now fixed.

The test cases already had distinct AVX and non-AVX variants; added your new cases to both.

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:39):

fitzgen submitted PR review:

r=me with Alex's suggestion

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:39):

fitzgen commented on PR #8207:

race

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:40):

cfallin commented on PR #8207:

Yep, I imagine you'll want to review the new changes just pushed, sorry!

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:41):

fitzgen submitted PR review.

Wasmtime GitHub notifications bot (Mar 21 2024 at 16:41):

fitzgen has enabled auto merge for PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 18:40):

cfallin updated PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 18:40):

cfallin has enabled auto merge for PR #8207.

Wasmtime GitHub notifications bot (Mar 21 2024 at 19:27):

cfallin merged PR #8207.

Last updated: Dec 23 2024 at 13:07 UTC