Update performant_matmul.jl

[XingyuZhang2018]

debug when N and M are not an integer multiple of TILE_DIM

[github-actions]

Your PR requires formatting changes to meet the project's style guidelines.
Please consider running Runic (git runic main) to apply these changes.

Click here to view the suggested changes.

diff --git a/examples/performant_matmul.jl b/examples/performant_matmul.jl
index 02b25523..be72e76f 100644
--- a/examples/performant_matmul.jl
+++ b/examples/performant_matmul.jl
@@ -37,7 +37,7 @@ const TILE_DIM = 32
         else
             @inbounds tile1[i, j] = 0.0
         end
-        
+
         if t * TILE_DIM + i <= R && J <= M
             @inbounds tile2[i, j] = input2[t * TILE_DIM + i, J]
         else
@@ -63,18 +63,18 @@ const TILE_DIM = 32
     end
 end
 
-@testset "dims for $N, $R, $M" for (N,R,M) in [rand(500:1000,3) for _ in 1:10]
+@testset "dims for $N, $R, $M" for (N, R, M) in [rand(500:1000, 3) for _ in 1:10]
     A = rand!(allocate(backend, Float32, N, R))
     B = rand!(allocate(backend, Float32, R, M))
     C = KernelAbstractions.zeros(backend, Float32, N, M)
-    
+
     kern = coalesced_matmul_kernel!(backend, (TILE_DIM, TILE_DIM))
-    
+
     group_size_x = cld(N, TILE_DIM)
     group_size_y = cld(M, TILE_DIM)
-    
+
     kern(C, A, B, N, R, M, ndrange = (group_size_x * TILE_DIM, group_size_y * TILE_DIM))
     KernelAbstractions.synchronize(backend)
-    
+
     @test isapprox(A * B, C)
 end

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 0.00%. Comparing base (de5dc99) to head (b398435).

❗ There is a different number of reports uploaded between BASE (de5dc99) and HEAD (b398435). Click for more details.

HEAD has 2 uploads less than BASE

Flag	BASE (de5dc99)	HEAD (b398435)
	12	10

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #590       +/-   ##
==========================================
- Coverage   47.52%   0.00%   -47.53%     
==========================================
  Files          22      21        -1     
  Lines        1715    1571      -144     
==========================================
- Hits          815       0      -815     
- Misses        900    1571      +671     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[christiangnrd]

Have you seen JuliaGPU/GPUArrays.jl#590?

[XingyuZhang2018]

Have you seen JuliaGPU/GPUArrays.jl#590?

I’m sorry for not noticing that you had already found the same bug.

[christiangnrd]

Have you seen JuliaGPU/GPUArrays.jl#590?

I’m sorry for not noticing that you had already found the same bug.

No worries! It made me realize that my workarounds might not be necessary once 0.10 gets released.