Align up structs to IntPtr.Size if they have gc pointers by EgorBo · Pull Request #100289 · dotnet/runtime

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

EgorBo · 2024-03-26T13:36:15Z

Fixes#100220

I am not an expert in R2R/NAOT type system, but looks like it fixes the issue

…fix-structsize-vm

EgorBo · 2024-04-06T15:43:42Z

/azp list

EgorBo · 2024-04-06T15:44:29Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-06T15:44:38Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T03:10:52Z

/azp run runtime-nativeaot-outerloop

azure-pipelines · 2024-04-07T03:11:06Z

Azure Pipelines successfully started running 1 pipeline(s).

EgorBo · 2024-04-07T11:49:48Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

Otherwise should be ready. The rules turn out to be a bit complex, e.g.

[StructLayout(LayoutKind.Sequential,Size=1000)]classClassA{longA;}[StructLayout(LayoutKind.Explicit,Size=1000)]classClassB{[FieldOffset(0)]longA;}

Here, CoreCLR allocates 1016 bytes for an instance of ClassA and 24 bytes for ClassB - I am not sure whether it should be "fixed" in CoreCLR, but I aligned NAOT/R2R, so the rules are:

Classes:
- Layout has gc pointers - we ignore explicit Size
- LayoutKind.Explicit - we ignore explicit Size (bug-to-bug compat?)
Structs:
- LayoutKind.Sequential + gc pointers - we ignore explicit Size
- Layout has gc pointers - we align up size to target pointer

So for classes, the only case when we respect Size is LayoutKind.Sequential + no gc pointers.

Questionable behavior in CoreCLR (Main) for (boxed) object size:

For a class without gc refs, why do we ignore Size= for LayoutKind.Explicit, but not for LayoutKind.Sequential?
For a struct with gc ref(s), should we ignore Size= for even LayoutKind.Explicit ? (we do for LayoutKind.Sequential). Since gc references can easily "invalidate" explicit Size by AlignUp(8).

cc @jkotas

…fix-structsize-vm

jkotas · 2024-04-07T16:56:35Z

Probably needs #100618 since the GC.* trick is not reliable to get object size in tests.

I do not think it would help you. This API gives you the exact same result as sizeof.

jkotas · 2024-04-07T17:04:57Z

+ }
+
+ [MethodImpl(MethodImplOptions.NoInlining | MethodImplOptions.AggressiveOptimization)]
+ private static int GetManagedSize(Func<object> allocator)


This is not reliable as you have pointed out.
I think a reliable version of this check would be to inherit from the type you want to check, add a byte field, and verify offset of the byte field.

Would an UnsafeAccessor to RuntimeHelpers.GetRawObjectDataSize work better here instead maybe?
EDIT: Ah wait, that's a static class, nevermind.

Invoked GetRawObjectDataSize via reflection

jkotas · 2024-04-07T17:06:03Z

/azp run runtime-coreclr r2r

azure-pipelines · 2024-04-07T17:06:18Z

Azure Pipelines successfully started running 1 pipeline(s).

MichalPetryka · 2024-04-08T14:01:47Z

+ // Also, Mono needs RuntimeHelpers.GetRawObjectDataSize or equivalent to get an object size
+ public static int TestEntryPoint()
+ {
+ if (IntPtr.Size == 8)


Should this maybe not have the if and multiply the pointer size appropriately in each case?

I prefer expected value to be a constant rather than computed expression for Asserts

EgorBo · 2024-04-08T14:49:12Z

@jkotas anything else here? looks like R2R and NAOT outerloops passed

lewing · 2024-04-09T18:48:02Z

The maui team is waiting for a resolution on this

EgorBo · 2024-04-09T19:10:07Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

MichalStrehovsky · 2024-04-09T22:02:48Z

I assume Jan is OOF, so maybe @MichalStrehovsky can review this?

There is a reason I moved this to crossgen2 team (#100220 (comment)). Type layout needs to exactly match CoreCLR type layout for ReadyToRun to work. It has been a bug farm in the past. I don't see anything obviously wrong in this PR but that's not a high quality bar. I know we have modes of operation of crossgen2 where it serializes type layout into the image and validates it at runtime, but I don't know what CI legs that runs in and if this would cover it too. Can you get @davidwrighton to have a look? I really only have surface-level knowledge of R2R testing and very much hope I'm not the biggest expert.

jkotas

LGTM. Thank you for adding a lot of tests!

EgorBo · 2024-04-10T20:17:11Z

The maui team is waiting for a resolution on this

@lewing do we need to backport the fix to a specific release preview branch? if so, which?

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in #100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Test introduced in dotnet#100289 failed the build of CLR tests for e.g. RISC-V: ``` /runtime/src/tests/Loader/classloader/explicitlayout/Regressions/100220/Runtime_100220.cs(15,23): error XUW1002: Tests should not unconditionally return 100. Convert to a void return. ```

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Align struct's size to pointer's size if it has gc fields
5ef2ca5

ghost added the area-ReadyToRun label Mar 26, 2024

dotnet-policy-serviceBot assigned EgorBo Mar 26, 2024

EgorBo added 2 commits March 26, 2024 15:47

Update LayoutClassTest.cs
dd42a8e

fix test
a05098b

tannergooding reviewed Mar 26, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

EgorBo added 2 commits March 26, 2024 19:03

Update TypeTests.StructLayoutAttribute.cs
cf3f416

Update TypeTests.StructLayoutAttribute.cs
b1bb5b0

jkotas reviewed Mar 26, 2024
View reviewed changes

Comment thread...ystem.Reflection.MetadataLoadContext/tests/src/Tests/Type/TypeTests.StructLayoutAttribute.cs Outdated

runfoappBot mentioned this pull request Mar 26, 2024
Methodical_others test JIT/Methodical/Coverage/copy_prop_byref_to_native_int crashing #69832
Open

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs Outdated

jkotas reviewed Mar 27, 2024
View reviewed changes

Comment threadsrc/coreclr/tools/Common/TypeSystem/Common/MetadataFieldLayoutAlgorithm.cs

EgorBo added 9 commits April 2, 2024 01:46

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
6149238

Address feedback
7a04cd1

Address feedback
94864cd

Merge branch 'main' of github.com:dotnet/runtime into fix-structsize-vm
02e8ed3

Update Runtime_100220.cs
e816157

Fix algorithm
894c3de

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
fa0e6e1
…fix-structsize-vm

Add more tests, fix 32bit
d10c560

fix test
66fdbff

This comment was marked as resolved.
Sign in to view

EgorBo added 2 commits April 6, 2024 17:48

Update InstanceFieldLayoutTests.cs
f235812

Update InstanceFieldLayoutTests.cs
e9931d6

Update InstanceFieldLayoutTests.cs
033163c

build-analysisBot mentioned this pull request Apr 6, 2024
slow macOS - "##[error]The job running on agent Azure Pipelines 9 ran longer than the maximum time of 60 minutes." dotnet/dnceng#1883
Open
3 tasks

EgorBo marked this pull request as ready for review April 7, 2024 12:08

EgorBo requested a review from MichalStrehovsky as a code owner April 7, 2024 12:08

EgorBo added 2 commits April 7, 2024 17:20

replace long with nint, to avoid Align8 artifacts on 32bit
38c5f16

Merge branch 'fix-structsize-vm' of github.com:EgorBo/runtime-1 into …
c804f61
…fix-structsize-vm

jkotas reviewed Apr 7, 2024
View reviewed changes

build-analysisBot mentioned this pull request Apr 7, 2024
System.Text.RegularExpressions.Tests on Mono_Minijit_Debug-Ubuntu #100715
Closed

use GetRawObjectDataSize
9a935d3

MichalPetryka reviewed Apr 8, 2024
View reviewed changes

EgorBo mentioned this pull request Apr 8, 2024
ILC crashes on x64, works on arm64 #99611
Closed

jkotas approved these changes Apr 10, 2024
View reviewed changes

EgorBo merged commit 4d5e5ff into dotnet:mainApr 10, 2024

tomeksowi mentioned this pull request Apr 11, 2024
Fix unconditional success in CLR tests #100905
Merged

matouskozak pushed a commit to matouskozak/runtime that referenced this pull request Apr 30, 2024
Align up structs to IntPtr.Size if they have gc pointers (dotnet#100289)
a382ac8

github-actionsBot locked and limited conversation to collaborators May 12, 2024

Uh oh!

Conversation

EgorBo commented Mar 26, 2024

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

EgorBo commented Apr 6, 2024

Uh oh!

This comment was marked as resolved.

EgorBo commented Apr 6, 2024

Uh oh!

azure-pipelinesBot commented Apr 6, 2024

Uh oh!

EgorBo commented Apr 7, 2024

Uh oh!

azure-pipelinesBot commented Apr 7, 2024

Uh oh!

EgorBo commented Apr 7, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

jkotas commented Apr 7, 2024

Uh oh!

jkotasApr 7, 2024

Choose a reason for hiding this comment

Uh oh!

MichalPetrykaApr 7, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

EgorBoApr 8, 2024

Choose a reason for hiding this comment

Uh oh!

jkotas commented Apr 7, 2024

Uh oh!

azure-pipelinesBot commented Apr 7, 2024

Uh oh!

MichalPetrykaApr 8, 2024

Choose a reason for hiding this comment

Uh oh!

EgorBoApr 8, 2024• edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

EgorBo commented Apr 8, 2024

Uh oh!

lewing commented Apr 9, 2024

Uh oh!

EgorBo commented Apr 9, 2024

Uh oh!

MichalStrehovsky commented Apr 9, 2024

Uh oh!

jkotas left a comment

Choose a reason for hiding this comment

Uh oh!

EgorBo commented Apr 10, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

Uh oh!

Conversation

EgorBo commented Mar 26, 2024

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

EgorBo commented Apr 6, 2024

Uh oh!

This comment was marked as resolved.

EgorBo commented Apr 6, 2024

Uh oh!

azure-pipelinesBot commented Apr 6, 2024

EgorBo commented Apr 7, 2024•
edited
Loading

MichalPetrykaApr 7, 2024•
edited
Loading

EgorBoApr 8, 2024•
edited
Loading

EgorBo commented Apr 7, 2024•
edited
Loading

MichalPetrykaApr 7, 2024•
edited
Loading

EgorBoApr 8, 2024•
edited
Loading