From 166d937045f44cbedbe4a2316dddf2b5ec49174e Mon Sep 17 00:00:00 2001
From: Evelyne <110474206+evelyne-ringoot@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:53:12 -0400
Subject: [PATCH 1/3] Input/output convention in naive_transpose example

Putting output before input in example to match convention
---
 examples/naive_transpose.jl | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/examples/naive_transpose.jl b/examples/naive_transpose.jl
index de346be54..bca184fa1 100644
--- a/examples/naive_transpose.jl
+++ b/examples/naive_transpose.jl
@@ -3,7 +3,7 @@ include(joinpath(dirname(pathof(KernelAbstractions)), "../examples/utils.jl")) #
 
 @kernel function naive_transpose_kernel!(a, b)
     i, j = @index(Global, NTuple)
-    @inbounds b[i, j] = a[j, i]
+    @inbounds a[i, j] = b[j, i]
 end
 
 # create wrapper function to check inputs
@@ -24,8 +24,8 @@ end
 res = 1024
 
 # creating initial arrays
-a = rand!(allocate(backend, Float32, res, res))
-b = KernelAbstractions.zeros(backend, Float32, res, res)
+b = rand!(allocate(backend, Float32, res, res))
+a = KernelAbstractions.zeros(backend, Float32, res, res)
 
 naive_transpose!(a,b)
 KernelAbstractions.synchronize(backend)

From 2b91be5d70d34f361acf7064ecb9b4ddc58ec493 Mon Sep 17 00:00:00 2001
From: Evelyne <110474206+evelyne-ringoot@users.noreply.github.com>
Date: Wed, 24 Apr 2024 19:55:54 -0400
Subject: [PATCH 2/3] Output/input order convention in matmul example

Putting output before input in example to match convention
---
 examples/matmul.jl | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/examples/matmul.jl b/examples/matmul.jl
index f28f91000..305436a7a 100644
--- a/examples/matmul.jl
+++ b/examples/matmul.jl
@@ -2,34 +2,34 @@ using KernelAbstractions, Test, Random
 include(joinpath(dirname(pathof(KernelAbstractions)), "../examples/utils.jl")) # Load backend
 
 # Simple kernel for matrix multiplication
-@kernel function matmul_kernel!(a, b, c)
+@kernel function matmul_kernel!(output, a, b)
     i, j = @index(Global, NTuple)
 
     # creating a temporary sum variable for matrix multiplication
-    tmp_sum = zero(eltype(c))
+    tmp_sum = zero(eltype(output))
     for k = 1:size(a)[2]
         tmp_sum += a[i,k] * b[k, j]
     end
 
-    c[i,j] = tmp_sum
+    output[i,j] = tmp_sum
 end
 
 # Creating a wrapper kernel for launching with error checks
-function matmul!(a, b, c)
+function matmul!(output, a, b)
     if size(a)[2] != size(b)[1]
         println("Matrix size mismatch!")
         return nothing
     end
     backend = KernelAbstractions.get_backend(a)
     kernel! = matmul_kernel!(backend)
-    kernel!(a, b, c, ndrange=size(c)) 
+    kernel!(output, a, b, ndrange=size(c)) 
 end
 
 a = rand!(allocate(backend, Float32, 256, 123))
 b = rand!(allocate(backend, Float32, 123, 45))
-c = KernelAbstractions.zeros(backend, Float32, 256, 45)
+output = KernelAbstractions.zeros(backend, Float32, 256, 45)
 
-matmul!(a,b,c)
+matmul!(output, a,b)
 KernelAbstractions.synchronize(backend)
 
-@test isapprox(c, a*b)
+@test isapprox(output, a*b)

From 89fddbce498c5c1aaef6c66341c418549c5dac7e Mon Sep 17 00:00:00 2001
From: Evelyne <110474206+evelyne-ringoot@users.noreply.github.com>
Date: Wed, 24 Apr 2024 20:50:48 -0400
Subject: [PATCH 3/3] Update matmul.jl

---
 examples/matmul.jl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/matmul.jl b/examples/matmul.jl
index 305436a7a..11d5c1520 100644
--- a/examples/matmul.jl
+++ b/examples/matmul.jl
@@ -22,7 +22,7 @@ function matmul!(output, a, b)
     end
     backend = KernelAbstractions.get_backend(a)
     kernel! = matmul_kernel!(backend)
-    kernel!(output, a, b, ndrange=size(c)) 
+    kernel!(output, a, b, ndrange=size(output)) 
 end
 
 a = rand!(allocate(backend, Float32, 256, 123))