[RISC-V] Fix struct info value in crossgen2 by clamp03 · Pull Request #99602 · dotnet/runtime

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

clamp03 · 2024-03-12T12:55:53Z

Fixed assertions in fgMorphMultiregStructArg when crossgen2 compiles in Debug build with --verify-type-and-field-layout option

Error messages

/runtime/src/coreclr/jit/morph.cpp:3766
Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in ...

System.Private.CoreLib.dll: ValueType which contains two double fields
./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp: ValueType which contains FieldOffset.

Part of #84834
cc @dotnet/samsung

Asserts in `./Interop/StructMarshalling/PInvoke/MarshalStructAsLayoutExp/MarshalStructAsLayoutExp.sh` Error message is `Assertion failed 'roundUp(structSize, TARGET_POINTER_SIZE) == roundUp(loadExtent, TARGET_POINTER_SIZE)' in 'Managed:MarshalStructAsParam_AsExpByVal(int)' during 'Morph - Global' (IL size 2208; hash 0x9fd9734a; MinOpts)` Copied missed codes of GetRiscV64PassStructInRegiste in vm to crossgen2

clamp03 · 2024-03-19T07:49:56Z

Checked the same assertions in coreclr tests (with crossgened test dll) and fixed.
There are still other issues. These will be fixed in other PR.
@dotnet/samsung @jkotas Could you please review it?
@dotnet/loongarch64-contrib I think the same issue will be in LoongArch64 too.

jkotas · 2024-03-19T15:22:41Z

Could you please highlight the equivalent code in MethodTable::GetRiscV64PassStructInRegisterFlags? Would it make sense to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar so that they are easier to compare?

clamp03 · 2024-03-20T01:06:48Z

@jkotas
I referenced

runtime/src/coreclr/vm/methodtable.cpp

Lines 3759 to 3762 in f0117c9

if (pFieldStart->GetOffset() || !pFieldStart[1].GetOffset() || (pFieldStart[0].GetSize() > pFieldStart[1].GetOffset())) 
 { 
goto _End_arg; 
 } 

and

runtime/src/coreclr/vm/methodtable.cpp

Lines 3920 to 3930 in f0117c9

if ((GetRiscV64PassStructInRegisterFlags((CORINFO_CLASS_HANDLE)pMethodTable2) & STRUCT_FLOAT_FIELD_ONLY_ONE) != 0) 
 { 
if (pFieldStart[1].GetSize() == 4) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND) : (size | STRUCT_FLOAT_FIELD_SECOND); 
 } 
elseif (pFieldStart[1].GetSize() == 8) 
 { 
 size = size & STRUCT_FLOAT_FIELD_FIRST ? (size ^ STRUCT_MERGE_FIRST_SECOND_8) : (size | STRUCT_SECOND_FIELD_DOUBLE); 
 } 
 } 

Right. These looks so different. AndMethodTable::GetRiscV64PassStructInRegisterFlags is too complex.
I agree that it is good to make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler.

@bartlomiejko Today, I am off. Could you refactor both codes to make to be more similar and simpler in your team (including this patch)? If then, I will close this PR. If no one in your team is available, I will handle from tomorrow. Thank you.

tomeksowi · 2024-03-20T08:14:58Z

Could you refactor both codes to make to be more similar and simpler in your team (including this patch)?

I'll take this, I worked on these flags recently.

clamp03 · 2024-03-20T09:02:28Z

@tomeksowi Thank you.

tomeksowi · 2024-03-20T11:40:19Z

(...) make the code structure of MethodTable::GetRiscV64PassStructInRegisterFlags to be more similar to RISCV64PassStructInRegister:GetRISCV64PassStructInRegisterFlags and simpler

Speaking of simpler, I noticed TypeHandle::IsBlittable always returns true for native value types:

runtime/src/coreclr/vm/typehandle.cpp

Lines 483 to 492 in 589beb0

if (!IsTypeDesc()) 
 { 
// This is a simple type (not an array, ptr or byref) so if
// simply check to see if the type is blittable.
returnAsMethodTable()->IsBlittable(); 
 } 
elseif (AsTypeDesc()->IsNativeValueType()) 
 { 
returnTRUE; 
 } 

So given blittable types have the same layout as managed, I think we could remove the whole NativeLayoutInfo branch from MethodTable::GetRiscV64PassStructInRegisterFlags. If anyone sees a fault in this reasoning, let me know.

jkotas · 2024-03-20T14:01:25Z

NativeValueType is unmanaged marshalled view of the managed type. For example, if you have struct MyStruct { [MarshalAs(UnmanagedType.Bool)] bool myField; }, the NativeValueType is going to be equivalent to struct MyStruct { int myField; }.

The NativeLayoutInfo branch is only needed for built-in runtime marshalling that we have been moving away from. If you were to remove the NativeLayoutInfo branch, you would be effectively forcing DisableRuntimeMarshalling(true). User code that depends on built-in runtime marshalling for interop would be broken some of the time.

tomeksowi · 2024-03-20T14:15:58Z

OK, for the time being I'll keep support for NativeLayoutInfo.

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

[RISC-V] Fix struct info value in crossgen2
684c2e7

clamp03 self-assigned this Mar 12, 2024

ghost added the area-ReadyToRun label Mar 12, 2024

clamp03 added the arch-riscv Related to the RISC-V architecture label Mar 12, 2024

dotnet-policy-serviceBot added the community-contribution Indicates that the PR has been added by a community member label Mar 12, 2024

tomeksowi approved these changes Mar 15, 2024
View reviewed changes

clamp03 marked this pull request as ready for review March 19, 2024 07:36

clamp03 requested a review from MichalStrehovsky as a code owner March 19, 2024 07:36

sirntar approved these changes Mar 19, 2024
View reviewed changes

Bajtazar approved these changes Mar 19, 2024
View reviewed changes

clamp03 requested a review from jkotas March 19, 2024 11:48

clamp03 closed this Mar 20, 2024

tomeksowi mentioned this pull request Mar 21, 2024
[RISC-V] Simplify flags for passing struct in registers #100080
Merged

github-actionsBot locked and limited conversation to collaborators Apr 20, 2024

Uh oh!

Conversation

clamp03 commented Mar 12, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clamp03 commented Mar 19, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jkotas commented Mar 19, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

jkotas commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Uh oh!

Conversation

clamp03 commented Mar 12, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clamp03 commented Mar 19, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jkotas commented Mar 19, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

jkotas commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

5 participants

Uh oh!

Conversation

clamp03 commented Mar 12, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

clamp03 commented Mar 19, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jkotas commented Mar 19, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

clamp03 commented Mar 20, 2024

Uh oh!

tomeksowi commented Mar 20, 2024

Uh oh!

jkotas commented Mar 20, 2024

Uh oh!

clamp03 commented Mar 12, 2024•
edited
Loading

clamp03 commented Mar 19, 2024•
edited
Loading

clamp03 commented Mar 12, 2024•
edited
Loading

clamp03 commented Mar 19, 2024•
edited
Loading

clamp03 commented Mar 12, 2024•
edited
Loading

clamp03 commented Mar 19, 2024•
edited
Loading

clamp03 commented Mar 12, 2024•
edited
Loading

clamp03 commented Mar 19, 2024•
edited
Loading

clamp03 commented Mar 12, 2024•
edited
Loading

clamp03 commented Mar 19, 2024•
edited
Loading