[mono][llvm][jit] Adding support for Vector128::ExtractMostSignificantBits intrinsic on ARM64 with LLVM + refactoring miniJIT opcodes by matouskozak · Pull Request #89431 · dotnet/runtime

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

matouskozak · 2023-07-25T11:36:49Z

Adding support for Vector128.ExtractMostSignificantBits intrinsic on ARM64 LLVM.

Implementation

The implementation for LLVM follows the same approach as with miniJIT introduced in #84345.

To unify LLVM and miniJIT approaches, the previously introduced opcodes OP_ARM64_USHL and OP_ARM64_EXT_IMM were removed and replaced with INTRINS_AARCH64_ADV_SIMD_USHL and OP_XLOWER/OP_XUPPER as these were already supported on LLVM.

This change introduces a slight change in generated assembly by miniJIT. Specifically, in byte/sbyte element types scenario the code changes from:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**ext.16b v2, v1, v0, #0x8** addv.16b b2, v2 umov.b w1, v2[0] ext.16b v0, v0, v1, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

to:

and.16b v0, v0, v1 ldr q1,0x60 ushl.16b v0, v0, v1 eor.16b v1, v1, v1**mov.8b v1, v0** addv.16b b1, v1 umov.b w1, v1[0]**eor.16b v16, v16, v16** ext.16b v0, v0, v16, #0x8 addv.16b b0, v0 umov.b w0, v0[0]lsl x0, x0, #8 orr w0, w0, w1

The change removed 1xext.16b instruction and added two new ones (mov.8b and eor.16b). If this change introduces any significant regressions we can revert the change for miniJIT.

An extract of generated code for byte element type using LLVM fullAOT:

and.16b v0, v1, v0 ushl.16b v0, v0, v2 ext.16b v1, v0, v0, #0x8 addv.8b b0, v0 addv.8b b1, v1 fmov w9, s0 fmov w8, s1 orr w0, w9, w8,lsl #8

Contributes to #76025.

/cc: @ivanpovazan

matouskozak · 2023-07-25T19:29:51Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:30:35Z

Azure Pipelines failed to run 1 pipeline(s).

matouskozak · 2023-07-25T19:32:37Z

/azp run runtime-llvm

azure-pipelines · 2023-07-25T19:32:42Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T06:27:10Z

/azp run runtime-llvm

azure-pipelines · 2023-07-26T06:27:15Z

No commit pushedDate could be found for PR 89431 in repo dotnet/runtime

matouskozak · 2023-07-26T15:35:21Z

/azp run runtime-extra-platforms

azure-pipelines · 2023-07-26T15:35:46Z

Azure Pipelines successfully started running 1 pipeline(s).

fanyang-mono

LGTM! Well done :)

matouskozak · 2023-07-27T09:27:20Z

The failing CI lanes look unrelated or are already tracked.

ivanpovazan

Looks good!
Thank you for verifying that the performance is not affected with this change (any further optimization can be addressed later - in a separate PR).

ExtractMSB arm64 LLVM + refactor mini JIT op codes
cba0e09

ghost assigned matouskozak Jul 25, 2023

ghost added the area-Codegen-JIT-mono label Jul 25, 2023

ivanpovazan reviewed Jul 25, 2023
View reviewed changes

Comment threadsrc/mono/mono/mini/simd-intrinsics.c

This was referenced Jul 25, 2023
Tracking issue for CI build timeouts #76454
Closed
[WASM][Debugger]Test timed out: DebuggerTests.BreakpointTests.DebuggerTests.BreakpointTests.CreateGoodBreakpoint #89453
Closed

vargaz approved these changes Jul 25, 2023
View reviewed changes

build-analysisBot mentioned this pull request Jul 25, 2023
[wasm][debugger] Debugger tests disabled #89409
Open

matouskozak requested a review from fanyang-mono July 26, 2023 14:58

matouskozak marked this pull request as ready for review July 26, 2023 15:51

matouskozak requested review from SamMonoRT and lambdageek as code owners July 26, 2023 15:51

fanyang-mono approved these changes Jul 26, 2023
View reviewed changes

ivanpovazan approved these changes Jul 27, 2023
View reviewed changes

matouskozak merged commit e928f16 into dotnet:mainJul 27, 2023

build-analysisBot mentioned this pull request Jul 27, 2023
[Android] Build failure: no previous prototype for function 'register_aot_modules' #89566
Closed

ghost locked as resolved and limited conversation to collaborators Aug 26, 2023

matouskozak deleted the extract-msb-arm64-llvm branch October 3, 2024 13:15