AmrDeveloper wrote:
> I have mixed feelings about this change. It mostly overlaps with what I am
> working on, which is class `LexicalScope`, which has a bunch of the code for
> handling the return value. I think I would prefer that this PR be dropped, as
> the changes will be part of the PR I
AmrDeveloper wrote:
There will be two upcoming patches.
1 - Initializing const array and zero initializing support (Depends on
upstreaming `ConstantEmitter` and `ZeroAttr`)
2 - Dynamic-size arrays, and accessing arrays using the `cir.ptr_stride`
operation (Depends on upstreaming `CastOp`)
ht
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/130502
This change adds the basic support for ArrayType
Issue #130197
>From 0b00b1b477f7d81220350669ecb43f87d2667a6d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sun, 9 Mar 2025 19:14:34 +0100
Subject: [PAT
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/129933
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/129285
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -33,6 +33,14 @@ class CIRGenBuilderTy : public cir::CIRBaseBuilderTy {
llvm_unreachable("NYI: PPC double-double format for long double");
llvm_unreachable("Unsupported format for long double");
}
+
+ bool isSized(mlir::Type ty) {
+if (mlir::isa(ty))
+ r
@@ -202,6 +202,18 @@ mlir::Type CIRGenTypes::convertType(QualType type) {
break;
}
+ case Type::ConstantArray: {
+const ConstantArrayType *arrTy = cast(ty);
+mlir::Type elemTy = convertTypeForMem(arrTy->getElementType());
+
+// FIXME: In LLVM, "lower arrays
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/129116
Fix structured binding shadows template parameter location
Fixes: #129060
>From ed0cf3f026e439288f8334f27555ad26825fd56a Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Thu, 27 Feb 2025 21:49:32 +0100
S
AmrDeveloper wrote:
@shafik I added the sample to our test cases, not sure if we can add source loc
in expected line, I will check that
https://github.com/llvm/llvm-project/pull/129116
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://li
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/129285
Improve the diagnostics for chained comparisons to report actual expressions
and operators
Fixes #129069
>From 66cd32d7c58976bf050f82c0b6af39bc2b838a74 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: F
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/129116
>From ed0cf3f026e439288f8334f27555ad26825fd56a Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Thu, 27 Feb 2025 21:49:32 +0100
Subject: [PATCH 1/2] [Clang][diagnostics] Fix structured binding shadows
tem
@@ -85,7 +85,7 @@ void f(int x, int y, int z) {
if ((ayy y < z' don't
have their mathematical meaning}}
AmrDeveloper wrote:
Yes i think the word `like` should be removed here, but I am thinking should we
prefer the previous diagnostics or the one with actual
@@ -57,3 +57,15 @@ bool boolfunc() { return true; }
// CHECK: %0 = cir.const #true
// CHECK: cir.return %0 : !cir.bool
// CHECK: }
+
+float floatfunc() { return 42.42f; }
+// CHECK: cir.func @floatfunc() -> !cir.float {
+// CHECK: %0 = cir.const #cir.fp<4.242000e+01> : !c
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/129304
>From d13f2d0d40987302243efad7304a1a006fdb2dde Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 28 Feb 2025 21:42:36 +0100
Subject: [PATCH 1/2] [CIR] Upstream floating point literal expressions
---
@@ -85,7 +85,7 @@ void f(int x, int y, int z) {
if ((ayy y < z' don't
have their mathematical meaning}}
AmrDeveloper wrote:
Yes, I got your point also it will not be great with large expressions :D, I am
thinking should we change the message to be different
https://github.com/AmrDeveloper edited
https://github.com/llvm/llvm-project/pull/130502
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130847
>From 337b23e5ddb8ad9ef470e9c62d6fc136ccb070ee Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Tue, 11 Mar 2025 22:42:28 +0100
Subject: [PATCH 1/4] [CIR] Upstream basic support for sizeof and alignof
---
AmrDeveloper wrote:
I updated the test to include fixed size ArrayType
https://github.com/llvm/llvm-project/pull/130847
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/130502
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
AmrDeveloper wrote:
@AaronBallman If releasenote is okey, i can merge now :D
https://github.com/llvm/llvm-project/pull/126793
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/129285
>From 7a61f41a3104dd0765dea9930e0cf6f94d656598 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 28 Feb 2025 19:27:41 +0100
Subject: [PATCH 1/3] [Clang][diagnostics] Improve the diagnostics for chained
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/126793
>From cfa6bfb6ae5834f251d8e0c2c0f447aa82c97ef2 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Tue, 11 Feb 2025 21:30:18 +0100
Subject: [PATCH 1/2] [Clang][ASTMatcher] Improve matching isDerivedFrom base
AmrDeveloper wrote:
> LGTM, but you should add a release note to `clang/docs/ReleaseNotes.rst` so
> users know about the change in behavior.
Thank you, I added a release note, once you confirm it's approved I can land
after CI is green
https://github.com/llvm/llvm-project/pull/126793
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/126793
>From cfa6bfb6ae5834f251d8e0c2c0f447aa82c97ef2 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Tue, 11 Feb 2025 21:30:18 +0100
Subject: [PATCH 1/3] [Clang][ASTMatcher] Improve matching isDerivedFrom base
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/126793
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/129933
This change adds support for the basic case of emitAndUpdateRetAlloca
>From 1db2baafe9ed0ab360517810144300db0e7410f6 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Wed, 5 Mar 2025 18:18:07 +0100
Subject
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130502
>From 0b00b1b477f7d81220350669ecb43f87d2667a6d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sun, 9 Mar 2025 19:14:34 +0100
Subject: [PATCH 1/4] [CIR] Upstream basic support for ArrayType
---
clang/in
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/130847
This change adds the essential support for sizeof and alignof operators
- Support for VariableArrayType can be added after closing #130197
>From 337b23e5ddb8ad9ef470e9c62d6fc136ccb070ee Mon Sep 17 00:00:0
@@ -369,6 +369,22 @@ BoolType::getABIAlignment(const ::mlir::DataLayout
&dataLayout,
return 1;
}
+//===--===//
+// Definitions
+//===--==
AmrDeveloper wrote:
> My concern is still not fixed, and Andy/Bruno still have comments here, so I
> don't believe this to be ready yet.
I already addressed all comments except the test for Size and it's need sizeof
PR to merge, and the assert line which we need to agree on, other than that
e
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130847
>From 337b23e5ddb8ad9ef470e9c62d6fc136ccb070ee Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Tue, 11 Mar 2025 22:42:28 +0100
Subject: [PATCH 1/3] [CIR] Upstream basic support for sizeof and alignof
---
@@ -148,3 +150,27 @@ mlir::Value ScalarExprEmitter::VisitCastExpr(CastExpr *ce)
{
}
return {};
}
+
+/// Return the size or alignment of the type of argument of the sizeof
+/// expression as an integer.
+mlir::Value ScalarExprEmitter::VisitUnaryExprOrTypeTraitExpr(
+con
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130502
>From 0b00b1b477f7d81220350669ecb43f87d2667a6d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sun, 9 Mar 2025 19:14:34 +0100
Subject: [PATCH 1/6] [CIR] Upstream basic support for ArrayType
---
clang/in
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130502
>From 0b00b1b477f7d81220350669ecb43f87d2667a6d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sun, 9 Mar 2025 19:14:34 +0100
Subject: [PATCH 1/6] [CIR] Upstream basic support for ArrayType
---
clang/in
@@ -33,6 +33,15 @@ class CIRGenBuilderTy : public cir::CIRBaseBuilderTy {
llvm_unreachable("NYI: PPC double-double format for long double");
llvm_unreachable("Unsupported format for long double");
}
+
+ bool isSized(mlir::Type ty) {
+if (mlir::isa(ty))
+ r
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/130502
>From 0b00b1b477f7d81220350669ecb43f87d2667a6d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sun, 9 Mar 2025 19:14:34 +0100
Subject: [PATCH 1/5] [CIR] Upstream basic support for ArrayType
---
clang/in
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/133100
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/132974
>From 4dc1e77299c71b8f01fb73f7fba5f14e0fbe3edd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 21 Mar 2025 21:07:11 +0100
Subject: [PATCH 1/7] [CIR] [Upstream local initialization for ArrayType
---
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/132974
>From 4dc1e77299c71b8f01fb73f7fba5f14e0fbe3edd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 21 Mar 2025 21:07:11 +0100
Subject: [PATCH 01/10] [CIR] [Upstream local initialization for ArrayType
--
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/132974
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -110,6 +110,8 @@ class CIRGenFunction : public CIRGenTypeCache {
public:
mlir::Value createDummyValue(mlir::Location loc, clang::QualType qt);
+ void emitNullInitialization(mlir::Location loc, Address destPtr, QualType
ty);
AmrDeveloper wrote:
Thank yo
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/133100
>From b4aafe8bdb2754597d440e256c33b9f8b9a92a91 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Wed, 26 Mar 2025 05:52:17 -0700
Subject: [PATCH 1/2] [CIR] Upstream zero init for global variables
---
.../
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/132974
>From 4dc1e77299c71b8f01fb73f7fba5f14e0fbe3edd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 21 Mar 2025 21:07:11 +0100
Subject: [PATCH 1/6] [CIR] [Upstream local initialization for ArrayType
---
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/132974
>From 4dc1e77299c71b8f01fb73f7fba5f14e0fbe3edd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 21 Mar 2025 21:07:11 +0100
Subject: [PATCH 1/6] [CIR] [Upstream local initialization for ArrayType
---
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/132974
>From 4dc1e77299c71b8f01fb73f7fba5f14e0fbe3edd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri, 21 Mar 2025 21:07:11 +0100
Subject: [PATCH 1/8] [CIR] [Upstream local initialization for ArrayType
---
AmrDeveloper wrote:
@bcardosolopes, I think we can finish this PR, and I will continue searching
for a test case for folding without blocking upstreaming other Vector Ops.
Also, I think the same idea can be applied to Arrays, what do you think?
https://github.com/llvm/llvm-project/pull/138413
https://github.com/AmrDeveloper edited
https://github.com/llvm/llvm-project/pull/139304
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/139146
>From 11de7d363ae9f06e98f15fc5c492e74a660693f8 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Thu, 8 May 2025 22:10:39 +0200
Subject: [PATCH 1/3] [CIR] Upstream insert op for VectorType
---
clang/inclu
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/139304
This change adds a folder for the VecExtractOp
Issue https://github.com/llvm/llvm-project/issues/136487
>From 6a21dc2a983094f61a54755454d12c45cf074c9d Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Fri
@@ -15901,6 +15901,13 @@ ExprResult
SemaOpenMP::VerifyPositiveIntegerConstantInClause(
<< E->getSourceRange();
return ExprError();
}
+
+ if (!Result.isRepresentableByInt64()) {
AmrDeveloper wrote:
In our implementation, yes, because we const ev
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138592
>From 793bd974b779450a84c72d54237067ab56a59cdc Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Mon, 5 May 2025 23:01:02 +0200
Subject: [PATCH 1/5] [clang][OpenMP] Add error for large expr in collapse
---
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138592
>From 793bd974b779450a84c72d54237067ab56a59cdc Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Mon, 5 May 2025 23:01:02 +0200
Subject: [PATCH 1/4] [clang][OpenMP] Add error for large expr in collapse
---
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138413
>From d0a6c4fa94d387d8c9afec10deae150d6bce54ee Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sat, 3 May 2025 13:43:04 +0200
Subject: [PATCH 1/3] [CIR] Upstream extract op for VectorType
---
clang/incl
AmrDeveloper wrote:
> > > Overall looks good. While here, can you please implement a folder for
> > > this operation? It should kick-in if both idx and input vector are
> > > constants.
> >
> >
> > @bcardosolopes I implement it like this snippet
> > ```
> > OpFoldResult cir::VecExtractOp::fol
@@ -213,3 +213,68 @@ void foo4() {
// OGCG: %[[TMP2:.*]] = load i32, ptr %[[IDX]], align 4
// OGCG: %[[ELE:.*]] = extractelement <4 x i32> %[[TMP1]], i32 %[[TMP2]]
// OGCG: store i32 %[[ELE]], ptr %[[INIT]], align 4
+
+void foo9() {
AmrDeveloper wrote:
Both of
@@ -1399,8 +1400,10 @@ mlir::LogicalResult
CIRToLLVMShiftOpLowering::matchAndRewrite(
if (op.getIsShiftleft()) {
rewriter.replaceOpWithNewOp(op, llvmTy, val, amt);
} else {
-assert(!cir::MissingFeatures::vectorType());
-bool isUnsigned = !cirValTy.isSigned();
+
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/138592
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/139146
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -1399,8 +1400,10 @@ mlir::LogicalResult
CIRToLLVMShiftOpLowering::matchAndRewrite(
if (op.getIsShiftleft()) {
rewriter.replaceOpWithNewOp(op, llvmTy, val, amt);
} else {
-assert(!cir::MissingFeatures::vectorType());
-bool isUnsigned = !cirValTy.isSigned();
+
https://github.com/AmrDeveloper deleted
https://github.com/llvm/llvm-project/pull/139465
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper edited
https://github.com/llvm/llvm-project/pull/139465
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper edited
https://github.com/llvm/llvm-project/pull/139465
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/139465
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
@@ -1399,8 +1400,10 @@ mlir::LogicalResult
CIRToLLVMShiftOpLowering::matchAndRewrite(
if (op.getIsShiftleft()) {
rewriter.replaceOpWithNewOp(op, llvmTy, val, amt);
} else {
-assert(!cir::MissingFeatures::vectorType());
-bool isUnsigned = !cirValTy.isSigned();
+
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138592
>From af3e49f1290c61d5856a5ddf600cfd19ef485e25 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Mon, 5 May 2025 23:01:02 +0200
Subject: [PATCH 1/3] [clang][OpenMP] Add error for large expr in collapse
---
AmrDeveloper wrote:
> In addition to the collapse clause, the ordered clause also seems to have the
> same problem. For example, does your patch also fix the following crash? If
> so, you also need tests regarding the ordered clause.
> https://godbolt.org/z/96sa5jxff
>
> ```c++
> void f(void)
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138592
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/139465
This change adds support for shift ops for VectorType
Issue https://github.com/llvm/llvm-project/issues/136487
Rate limit · GitHub
body {
background-color: #
@@ -1969,6 +1969,43 @@ def VecCreateOp : CIR_Op<"vec.create", [Pure]> {
let hasVerifier = 1;
}
+//===--===//
+// VecInsertOp
+//===--===//
AmrDeveloper wrote:
> Thank you for the fix! I've added a few more reviewers. This is a pretty
> general problem, so I think we may want additional test coverage for
> basically any of the OpenMP clauses which accept an integer argument. For
> example, this is another related issue: #139268 --
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/139146
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/139444
This change adds support for unary ops for VectorType
Issue https://github.com/llvm/llvm-project/issues/136487
>From 9fd2f92f9122c8c79c4977e00dab93bfaa468787 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Da
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/139146
This change adds an insert op for VectorType
Issue https://github.com/llvm/llvm-project/issues/136487
Rate limit · GitHub
body {
background-color: #f6f8fa;
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/138592
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/139465
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/139444
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/139827
This change adds support for splat op for VectorType
Issue https://github.com/llvm/llvm-project/issues/136487
>From e9b7f553f7ad8a9167a9aa4a11c86ccdb8472cdd Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Dat
AmrDeveloper wrote:
More test cases will be added when Bin and Comp operators are upstreamed for
Vector
https://github.com/llvm/llvm-project/pull/139827
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/lis
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/140099
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/140099
This change adds support for binary ops for VectorType
Issue https://github.com/llvm/llvm-project/issues/136487
Rate limit · GitHub
body {
background-color:
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/140125
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
@@ -54,10 +54,12 @@ namespace direct {
namespace {
/// If the given type is a vector type, return the vector's element type.
/// Otherwise return the given type unchanged.
-// TODO(cir): Return the vector element type once we have support for vectors
-// instead of the identity
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/140177
Fix for elementTypeIfVector
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFon
AmrDeveloper wrote:
I will quickly merge it as a fix for the previous elementTypeIfVector PR to
make it buildable
https://github.com/llvm/llvm-project/pull/140177
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/
https://github.com/AmrDeveloper closed
https://github.com/llvm/llvm-project/pull/140177
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/139444
>From 46d3b466df954eee87a5e5e7df5cb03802d468c7 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Sat, 10 May 2025 20:37:05 +0200
Subject: [PATCH 1/2] [CIR] Upstream unary operators for VectorType
---
.../
@@ -54,10 +54,12 @@ namespace direct {
namespace {
/// If the given type is a vector type, return the vector's element type.
/// Otherwise return the given type unchanged.
-// TODO(cir): Return the vector element type once we have support for vectors
-// instead of the identity
@@ -400,4 +400,165 @@ void foo9() {
// OGCG: %[[TMP_A:.*]] = load <4 x i32>, ptr %[[VEC_A]], align 16
// OGCG: %[[TMP_B:.*]] = load <4 x i32>, ptr %[[VEC_B]], align 16
// OGCG: %[[SHR:.*]] = ashr <4 x i32> %[[TMP_A]], %[[TMP_B]]
-// OGCG: store <4 x i32> %[[SHR]], ptr %[[SHR_RE
@@ -1076,20 +1074,41 @@ mlir::LogicalResult
CIRToLLVMUnaryOpLowering::matchAndRewrite(
rewriter.replaceOp(op, adaptor.getInput());
return mlir::success();
case cir::UnaryOpKind::Minus: {
- assert(!isVector &&
- "Add vector handling when vector
@@ -57,7 +57,11 @@ namespace {
// TODO(cir): Return the vector element type once we have support for vectors
// instead of the identity type.
mlir::Type elementTypeIfVector(mlir::Type type) {
- assert(!cir::MissingFeatures::vectorType());
+ if (auto vecType = mlir::dyn_cast(t
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/140125
Upstream vector support in the element type if vector as a required change for
upstreaming other Vec Ops
Issue https://github.com/llvm/llvm-project/issues/136487
>From b9f624e508db8349516f909f0c8e9cb85c04
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/140099
>From de3e5aa6af44a082060ad0edc8ba9bf213b0fba5 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Wed, 14 May 2025 22:25:09 +0200
Subject: [PATCH 1/2] [CIR][LLVMLowering] Upstream binary operators for
Vecto
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/140125
Rate limit · GitHub
body {
background-color: #f6f8fa;
color: #24292e;
font-family: -apple-system,BlinkMacSystemFont,Segoe
UI,Helvetica,Arial,s
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/14
>From 2a56a5d8db006bd466db00989806d548ce042b37 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Thu, 22 May 2025 19:38:08 +0200
Subject: [PATCH 1/2] [CIR] Allow use different Int types together in Vec Shif
@@ -1427,12 +1427,12 @@ OpFoldResult cir::SelectOp::fold(FoldAdaptor adaptor) {
//===--===//
LogicalResult cir::ShiftOp::verify() {
mlir::Operation *op = getOperation();
- mlir::Type resType = getResult().g
https://github.com/AmrDeveloper created
https://github.com/llvm/llvm-project/pull/14
Update the verification of ShiftOp for Vector to allow performing shift op
between signed and unsigned integers, similar to LLVM IR
Issue https://github.com/llvm/llvm-project/issues/136487
>From 2a56a5d8d
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/123806
>From 38031e9aed519f4fd3b21dff7e79c69a0150efd9 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Tue, 21 Jan 2025 20:11:21 +0100
Subject: [PATCH] [Clang] Improve testing of non std functions with name
infi
@@ -1427,13 +1427,13 @@ OpFoldResult cir::SelectOp::fold(FoldAdaptor adaptor) {
//===--===//
LogicalResult cir::ShiftOp::verify() {
mlir::Operation *op = getOperation();
- mlir::Type resType = getResult().g
https://github.com/AmrDeveloper updated
https://github.com/llvm/llvm-project/pull/14
>From 2a56a5d8db006bd466db00989806d548ce042b37 Mon Sep 17 00:00:00 2001
From: AmrDeveloper
Date: Thu, 22 May 2025 19:38:08 +0200
Subject: [PATCH 1/3] [CIR] Allow use different Int types together in Vec Shif
@@ -1427,13 +1427,13 @@ OpFoldResult cir::SelectOp::fold(FoldAdaptor adaptor) {
//===--===//
LogicalResult cir::ShiftOp::verify() {
mlir::Operation *op = getOperation();
- mlir::Type resType = getResult().g
301 - 400 of 426 matches
Mail list logo
