More tests and a matmatmul fix

[kshyatt]

Need the copyto! for the COO method to correctly do the in-place modification.

[github-actions]

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

Click here to view the suggested changes.

diff --git a/test/libraries/cusparse.jl b/test/libraries/cusparse.jl
index 4e4a46995..e5b2bb9f0 100644
--- a/test/libraries/cusparse.jl
+++ b/test/libraries/cusparse.jl
@@ -108,7 +108,7 @@ blockdim = 5
     @test CuSparseMatrixCOO(d_x) === d_x
     @test length(d_x) == m*n
     @test size(d_x)   == (m,n)
-    @test size(d_rx)  == (n,m)
+    @test size(d_rx) == (n, m)
     @test size(d_x,1) == m
     @test size(d_x,2) == n
     @test size(d_x,3) == 1
@@ -124,10 +124,10 @@ blockdim = 5
         @test d_x[end]             == x[end]
         @test d_tx[:, 1]           == transpose(x)[:, 1]
         @test d_ax[1, :]           == adjoint(x)[1, :]
-        @test d_tcx[:, 1]          == transpose(x)[:, 1]
-        @test d_acx[1, :]          == adjoint(x)[1, :]
-        @test d_rx[:, 1]           == transpose(x)[:, 1]
-        @test d_rx[1, :]           == transpose(x)[1, :]
+        @test d_tcx[:, 1] == transpose(x)[:, 1]
+        @test d_acx[1, :] == adjoint(x)[1, :]
+        @test d_rx[:, 1] == transpose(x)[:, 1]
+        @test d_rx[1, :] == transpose(x)[1, :]
         @test d_x[firstindex(d_x), firstindex(d_x)] == x[firstindex(x), firstindex(x)]
         @test d_x[div(end, 2), div(end, 2)]         == x[div(end, 2), div(end, 2)]
         @test d_x[end, end]        == x[end, end]

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 83.38%. Comparing base (4df814c) to head (d5b842f).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2697      +/-   ##
==========================================
+ Coverage   83.19%   83.38%   +0.18%     
==========================================
  Files         153      153              
  Lines       13605    13606       +1     
==========================================
+ Hits        11319    11345      +26     
+ Misses       2286     2261      -25     

☔ 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.

[github-actions]

CUDA.jl Benchmarks

Details

Benchmark suite	Current: d5b842f	Previous: 4df814c	Ratio
latency/precompile	46553118469 ns	46259331296 ns	1.01
latency/ttfp	7047011971 ns	7101267034 ns	0.99
latency/import	3715679733 ns	3747845800 ns	0.99
integration/volumerhs	9614349 ns	9616966 ns	1.00
integration/byval/slices=1	146659 ns	147004 ns	1.00
integration/byval/slices=3	425056 ns	425132.5 ns	1.00
integration/byval/reference	144834 ns	144971 ns	1.00
integration/byval/slices=2	286054 ns	286030 ns	1.00
integration/cudadevrt	103139 ns	103437 ns	1.00
kernel/indexing	13984 ns	14052 ns	1.00
kernel/indexing_checked	14356 ns	14687 ns	0.98
kernel/occupancy	686.825806451613 ns	642.1705882352941 ns	1.07
kernel/launch	2042.3 ns	2082.1 ns	0.98
kernel/rand	17678 ns	15366 ns	1.15
array/reverse/1d	19559 ns	19836 ns	0.99
array/reverse/2d	23734 ns	23708 ns	1.00
array/reverse/1d_inplace	10128 ns	10304 ns	0.98
array/reverse/2d_inplace	11179 ns	11633 ns	0.96
array/copy	20907 ns	21384.5 ns	0.98
array/iteration/findall/int	158087 ns	157650 ns	1.00
array/iteration/findall/bool	138836 ns	138713 ns	1.00
array/iteration/findfirst/int	154356 ns	153306 ns	1.01
array/iteration/findfirst/bool	155281.5 ns	154985 ns	1.00
array/iteration/scalar	71239 ns	71844 ns	0.99
array/iteration/logical	212894 ns	213346 ns	1.00
array/iteration/findmin/1d	41807 ns	41486 ns	1.01
array/iteration/findmin/2d	93973.5 ns	94252 ns	1.00
array/reductions/reduce/1d	40096 ns	36083 ns	1.11
array/reductions/reduce/2d	43757.5 ns	41306 ns	1.06
array/reductions/mapreduce/1d	37973.5 ns	33401 ns	1.14
array/reductions/mapreduce/2d	51146 ns	41277 ns	1.24
array/broadcast	20672 ns	20684 ns	1.00
array/copyto!/gpu_to_gpu	11969 ns	13946 ns	0.86
array/copyto!/cpu_to_gpu	208444 ns	208068 ns	1.00
array/copyto!/gpu_to_cpu	242747 ns	243583 ns	1.00
array/accumulate/1d	108900 ns	108492 ns	1.00
array/accumulate/2d	80163 ns	80106 ns	1.00
array/construct	1318.15 ns	1292 ns	1.02
array/random/randn/Float32	49605 ns	44847 ns	1.11
array/random/randn!/Float32	26512 ns	26564 ns	1.00
array/random/rand!/Int64	26984 ns	27174 ns	0.99
array/random/rand!/Float32	8614 ns	8731.666666666666 ns	0.99
array/random/rand/Int64	35749 ns	35317 ns	1.01
array/random/rand/Float32	13023 ns	13279 ns	0.98
array/permutedims/4d	61009 ns	61551 ns	0.99
array/permutedims/2d	55341 ns	55398 ns	1.00
array/permutedims/3d	56143.5 ns	56402 ns	1.00
array/sorting/1d	2777530 ns	2777475.5 ns	1.00
array/sorting/by	3368261 ns	3367837 ns	1.00
array/sorting/2d	1086222 ns	1085345 ns	1.00
cuda/synchronization/stream/auto	1030.1 ns	1021.2 ns	1.01
cuda/synchronization/stream/nonblocking	6492.6 ns	6526.2 ns	0.99
cuda/synchronization/stream/blocking	856.2933333333333 ns	830.5833333333334 ns	1.03
cuda/synchronization/context/auto	1203.8 ns	1191.2 ns	1.01
cuda/synchronization/context/nonblocking	6719.2 ns	6765.8 ns	0.99
cuda/synchronization/context/blocking	947.8709677419355 ns	958.9655172413793 ns	0.99

This comment was automatically generated by workflow using github-action-benchmark.