[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Florian Hahn via Phabricator via cfe-commits]

fhahn added a comment.

In D107349#2941385 , @erichkeane 
wrote:

> In D107349#2941257 , @mehdi_amini 
> wrote:
>
>> Revert to unbreak bots (like this one : 
>> https://lab.llvm.org/buildbot/#/builders/13/builds/10930 )
>
> Looks like this should be a pretty trivial test update at least.

Thanks for the heads-up! Recommitted with an updated MLIR test: f999312872b8 



Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Erich Keane via Phabricator via cfe-commits]

erichkeane added a comment.

In D107349#2941257 , @mehdi_amini 
wrote:

> Revert to unbreak bots (like this one : 
> https://lab.llvm.org/buildbot/#/builders/13/builds/10930 )

Looks like this should be a pretty trivial test update at least.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Mehdi AMINI via Phabricator via cfe-commits]

mehdi_amini added a comment.

Revert to unbreak bots (like this one : 
https://lab.llvm.org/buildbot/#/builders/13/builds/10930 )


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Adrian Kuegel via Phabricator via cfe-commits]

akuegel added a comment.

It seems this patch caused a test failure in MLIR:
test/Target/LLVMIR/llvm-intrinsics.mlir


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Florian Hahn via Phabricator via cfe-commits]

fhahn added a comment.

In D107349#2922208 , @erichkeane 
wrote:

> This looks fine to me, and seems to fix my problem, thanks!  I didn't spot 
> anything obvious,and proof-read the LangRef and think it is all fine, but am 
> not really the expert here, so please don't commit without the others having 
> a day or two to comment.

Thanks! I landed the change as there have not been any further comments in a 
while.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Florian Hahn via Phabricator via cfe-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
Closed by commit rGa1ef81de35a4: [Matrix] Overload stride arg in 
matrix.columnwise.load/store. (authored by fhahn).

Changed prior to commit:
  https://reviews.llvm.org/D107349?vs=363723&id=365949#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

Files:
  clang/test/CodeGen/matrix-type-builtins.c
  clang/test/CodeGenCXX/matrix-type-builtins.cpp
  clang/test/CodeGenObjC/matrix-type-builtins.m
  llvm/docs/LangRef.rst
  llvm/include/llvm/IR/Intrinsics.td
  llvm/include/llvm/IR/MatrixBuilder.h
  llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
  llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
  llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
  llvm/test/Verifier/matrix-intrinsics.ll

Index: llvm/test/Verifier/matrix-intrinsics.ll
===
--- llvm/test/Verifier/matrix-intrinsics.ll
+++ llvm/test/Verifier/matrix-intrinsics.ll
@@ -39,11 +39,11 @@
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: immarg operand has non-immediate parameter
 ; CHECK-NEXT: i32 %arg
-; CHECK-NEXT:   %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 %arg)
-  %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32(float* %m, i64 0, i1 false, i32 0, i32 0)
-  %result.1 = call <4 x float> @llvm.matrix.column.major.load.v4f32(float* %m, i64 2, i1 false, i32 1, i32 2)
-  %result.2 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 3)
-  %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 %arg)
+; CHECK-NEXT:   %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 %arg)
+  %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32.i64(float* %m, i64 0, i1 false, i32 0, i32 0)
+  %result.1 = call <4 x float> @llvm.matrix.column.major.load.v4f32.i64(float* %m, i64 2, i1 false, i32 1, i32 2)
+  %result.2 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 3)
+  %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 %arg)
   ret <4 x float> %result.1
 }
 
@@ -52,10 +52,10 @@
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
-  call void @llvm.matrix.column.major.store.v4f32(<4 x float> zeroinitializer, float* %m, i64 0, i1 false, i32 0, i32 0)
-  call void @llvm.matrix.column.major.store.v4f32(<4 x float> zeroinitializer, float* %m, i64 2, i1 false, i32 1, i32 2)
-  call void @llvm.matrix.column.major.store.v6f32(<6 x float> zeroinitializer, float* %n, i64 2, i1 false, i32 3, i32 3)
-  call void @llvm.matrix.column.major.store.v6f32(<6 x float> zeroinitializer, float* %n, i64 %arg, i1 false, i32 3, i32 3)
+  call void @llvm.matrix.column.major.store.v4f32.i64(<4 x float> zeroinitializer, float* %m, i64 0, i1 false, i32 0, i32 0)
+  call void @llvm.matrix.column.major.store.v4f32.i64(<4 x float> zeroinitializer, float* %m, i64 2, i1 false, i32 1, i32 2)
+  call void @llvm.matrix.column.major.store.v6f32.i64(<6 x float> zeroinitializer, float* %n, i64 2, i1 false, i32 3, i32 3)
+  call void @llvm.matrix.column.major.store.v6f32.i64(<6 x float> zeroinitializer, float* %n, i64 %arg, i1 false, i32 3, i32 3)
   ret void
 }
 
@@ -94,18 +94,18 @@
 ; CHECK-NEXT: Intrinsic has incorrect argument type!
 ; CHECK-NEXT: <4 x float> (i32*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4f32.pi32
 ; CHECK-NEXT: Intrinsic has incorrect argument type!
-; CHECK-NEXT: <4 x i32> (float*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4i32
+; CHECK-NEXT: <4 x i32> (float*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4i32.i64
 ;
   %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32.pi32(i32* %m, i64 2, i1 false, i32 2, i32 2)
-  %result.1 = call <4 x i32> @llvm.matrix.column.major.load.v4i32(float* %n, i64 2, i1 false, i32 2, i32 2)
+  %result.1 = call <4 x i32> @llvm.matrix.column.major.load.v4i32.i64(float* %n, i64 2, i1 false, i32 2, i32 2)
   ret <4 x float> %result.0
 }
 
 define void @column.major_store_mixed_types(float* %m, i32* %n, i64 %arg) {
 ;
-; CHECK-NEXT: Intrinsic has incorrect argument type! 
+; CHECK-NEXT: Intrinsic has incorrect argument type!
 ; CHECK-NEXT: void (<4 x i32>, float*, i64, i1, i32, i32)* @llvm.matrix.column.major.store.v4i32.vi32
-; CHECK-NEXT: Intrinsic has incorrect argument type! 
+; CHECK-NEXT: Intrinsic has incorrect argument typ

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Erich Keane via Phabricator via cfe-commits]

erichkeane accepted this revision.
erichkeane added a comment.
This revision is now accepted and ready to land.

This looks fine to me, and seems to fix my problem, thanks!  I didn't spot 
anything obvious,and proof-read the LangRef and think it is all fine, but am 
not really the expert here, so please don't commit without the others having a 
day or two to comment.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D107349/new/

https://reviews.llvm.org/D107349

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D107349: [Matrix] Overload stride arg in matrix.columnwise.load/store.

[Florian Hahn via Phabricator via cfe-commits]

fhahn created this revision.
fhahn added reviewers: rjmccall, anemet, thegameg, erichkeane.
Herald added subscribers: dexonsmith, jdoerfert, tschuett, hiraditya.
fhahn requested review of this revision.
Herald added projects: clang, LLVM.

This patch adjusts the intrinsics definition of
llvm.matrix.column.major.load and llvm.matrix.column.major.store to
allow overloading the type of the stride. The bitwidth of the stride is
used to perform the offset computation.

This fixes a crash when using __builtin_matrix_column_major_load or
__builtin_matrix_column_major_store on 32 bit platforms. The stride argument
of the builtins are defined as `size_t`, which is 32 bits wide on 32 bit
platforms.

Note that we still perform offset computations with 64 bit width on 32
bit platforms for accesses that do not take a user-specified stride.
This can be fixed separately.

Fixes PR51304.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D107349

Files:
  clang/test/CodeGen/matrix-type-builtins.c
  clang/test/CodeGenCXX/matrix-type-builtins.cpp
  clang/test/CodeGenObjC/matrix-type-builtins.m
  llvm/docs/LangRef.rst
  llvm/include/llvm/IR/Intrinsics.td
  llvm/include/llvm/IR/MatrixBuilder.h
  llvm/lib/Transforms/Scalar/LowerMatrixIntrinsics.cpp
  llvm/test/Transforms/LowerMatrixIntrinsics/strided-load-double.ll
  llvm/test/Transforms/LowerMatrixIntrinsics/strided-store-double.ll
  llvm/test/Verifier/matrix-intrinsics.ll

Index: llvm/test/Verifier/matrix-intrinsics.ll
===
--- llvm/test/Verifier/matrix-intrinsics.ll
+++ llvm/test/Verifier/matrix-intrinsics.ll
@@ -39,11 +39,11 @@
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: immarg operand has non-immediate parameter
 ; CHECK-NEXT: i32 %arg
-; CHECK-NEXT:   %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 %arg)
-  %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32(float* %m, i64 0, i1 false, i32 0, i32 0)
-  %result.1 = call <4 x float> @llvm.matrix.column.major.load.v4f32(float* %m, i64 2, i1 false, i32 1, i32 2)
-  %result.2 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 3)
-  %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32(float* %n, i64 2, i1 true, i32 3, i32 %arg)
+; CHECK-NEXT:   %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 %arg)
+  %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32.i64(float* %m, i64 0, i1 false, i32 0, i32 0)
+  %result.1 = call <4 x float> @llvm.matrix.column.major.load.v4f32.i64(float* %m, i64 2, i1 false, i32 1, i32 2)
+  %result.2 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 3)
+  %result.3 = call <6 x float> @llvm.matrix.column.major.load.v6f32.i64(float* %n, i64 2, i1 true, i32 3, i32 %arg)
   ret <4 x float> %result.1
 }
 
@@ -52,10 +52,10 @@
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
 ; CHECK-NEXT: Result of a matrix operation does not fit in the returned vector!
-  call void @llvm.matrix.column.major.store.v4f32(<4 x float> zeroinitializer, float* %m, i64 0, i1 false, i32 0, i32 0)
-  call void @llvm.matrix.column.major.store.v4f32(<4 x float> zeroinitializer, float* %m, i64 2, i1 false, i32 1, i32 2)
-  call void @llvm.matrix.column.major.store.v6f32(<6 x float> zeroinitializer, float* %n, i64 2, i1 false, i32 3, i32 3)
-  call void @llvm.matrix.column.major.store.v6f32(<6 x float> zeroinitializer, float* %n, i64 %arg, i1 false, i32 3, i32 3)
+  call void @llvm.matrix.column.major.store.v4f32.i64(<4 x float> zeroinitializer, float* %m, i64 0, i1 false, i32 0, i32 0)
+  call void @llvm.matrix.column.major.store.v4f32.i64(<4 x float> zeroinitializer, float* %m, i64 2, i1 false, i32 1, i32 2)
+  call void @llvm.matrix.column.major.store.v6f32.i64(<6 x float> zeroinitializer, float* %n, i64 2, i1 false, i32 3, i32 3)
+  call void @llvm.matrix.column.major.store.v6f32.i64(<6 x float> zeroinitializer, float* %n, i64 %arg, i1 false, i32 3, i32 3)
   ret void
 }
 
@@ -94,18 +94,18 @@
 ; CHECK-NEXT: Intrinsic has incorrect argument type!
 ; CHECK-NEXT: <4 x float> (i32*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4f32.pi32
 ; CHECK-NEXT: Intrinsic has incorrect argument type!
-; CHECK-NEXT: <4 x i32> (float*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4i32
+; CHECK-NEXT: <4 x i32> (float*, i64, i1, i32, i32)* @llvm.matrix.column.major.load.v4i32.i64
 ;
   %result.0 = call <4 x float> @llvm.matrix.column.major.load.v4f32.pi32(i32* %m, i64 2, i1 false, i32 2, i32 2)
-  %result.1 = call <4 x i32> @llvm.matrix.column.major.load.v4i32(float* %n, i64 2, i1 false, i32 2, i32 2)
+  %result.1 = call <4 x i32> @llvm.matrix.column.major.lo