Add _throw_dmrs device override for reshape of views

[Abdelrahman912]

Problem

reshape(@view(data[1:n*n]), (n, n)) fails to compile on the GPU. @view creates a SubArray, which has no specialized reshape method on the device, so it falls back to Base's generic _reshape. That path calls _throw_dmrs, which tries to construct a DimensionMismatch string — unsupported on the GPU.

Reason: unsupported call to an external C function
Stacktrace:
 [1] _string_n
   @ ./strings/string.jl:109
 [2] StringMemory
   @ ./iobuffer.jl:167
 [3] dec
   @ ./intfuncs.jl:918
 [4] #string#403
   @ ./intfuncs.jl:1000
 [5] multiple call sites
   @ unknown:0
   

function _reshape(parent::AbstractArray, dims::Dims)
  n = length(parent)
  prod(dims) == n || _throw_dmrs(n, "size", dims)
  __reshape((parent, IndexStyle(parent)), dims)
end

@noinline function _throw_dmrs(n, str, dims)
  throw(DimensionMismatch("parent has $n elements, which is incompatible with $str $dims")) ## THIS IS THE CULPRIT
end

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.42%. Comparing base (6ccd4b4) to head (774cffc).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #3095      +/-   ##
==========================================
- Coverage   90.43%   90.42%   -0.01%     
==========================================
  Files         141      141              
  Lines       12025    12025              
==========================================
- Hits        10875    10874       -1     
- Misses       1150     1151       +1     

☔ 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: 774cffc	Previous: 6ccd4b4	Ratio
array/accumulate/Float32/1d	101843 ns	101723 ns	1.00
array/accumulate/Float32/dims=1	77138 ns	76608 ns	1.01
array/accumulate/Float32/dims=1L	1585554 ns	1585294 ns	1.00
array/accumulate/Float32/dims=2	144144 ns	143948 ns	1.00
array/accumulate/Float32/dims=2L	657967.5 ns	657945 ns	1.00
array/accumulate/Int64/1d	118678 ns	118967 ns	1.00
array/accumulate/Int64/dims=1	80133 ns	79956 ns	1.00
array/accumulate/Int64/dims=1L	1706215 ns	1694445 ns	1.01
array/accumulate/Int64/dims=2	156239.5 ns	156040 ns	1.00
array/accumulate/Int64/dims=2L	962068 ns	961840 ns	1.00
array/broadcast	20708 ns	20347 ns	1.02
array/construct	1330.4 ns	1311.9 ns	1.01
array/copy	19016 ns	18931 ns	1.00
array/copyto!/cpu_to_gpu	215947 ns	215113 ns	1.00
array/copyto!/gpu_to_cpu	284326 ns	283517 ns	1.00
array/copyto!/gpu_to_gpu	11431.5 ns	11647 ns	0.98
array/iteration/findall/bool	131568 ns	132615 ns	0.99
array/iteration/findall/int	149234 ns	149623 ns	1.00
array/iteration/findfirst/bool	81883.5 ns	82175 ns	1.00
array/iteration/findfirst/int	83535.5 ns	84437 ns	0.99
array/iteration/findmin/1d	89031.5 ns	87647 ns	1.02
array/iteration/findmin/2d	117635 ns	117309 ns	1.00
array/iteration/logical	200232.5 ns	203627.5 ns	0.98
array/iteration/scalar	67840 ns	68729 ns	0.99
array/permutedims/2d	52486 ns	52820 ns	0.99
array/permutedims/3d	53326 ns	52914 ns	1.01
array/permutedims/4d	52208 ns	51983 ns	1.00
array/random/rand/Float32	13239 ns	13104 ns	1.01
array/random/rand/Int64	37304 ns	37312 ns	1.00
array/random/rand!/Float32	8615 ns	8603.333333333334 ns	1.00
array/random/rand!/Int64	34462 ns	34156 ns	1.01
array/random/randn/Float32	44189.5 ns	38723.5 ns	1.14
array/random/randn!/Float32	31115 ns	31520 ns	0.99
array/reductions/mapreduce/Float32/1d	34760 ns	35427 ns	0.98
array/reductions/mapreduce/Float32/dims=1	49713 ns	49562 ns	1.00
array/reductions/mapreduce/Float32/dims=1L	52089 ns	51766 ns	1.01
array/reductions/mapreduce/Float32/dims=2	57176 ns	56838 ns	1.01
array/reductions/mapreduce/Float32/dims=2L	70359 ns	69604 ns	1.01
array/reductions/mapreduce/Int64/1d	43204 ns	43423 ns	0.99
array/reductions/mapreduce/Int64/dims=1	42820 ns	44694.5 ns	0.96
array/reductions/mapreduce/Int64/dims=1L	87995 ns	87805 ns	1.00
array/reductions/mapreduce/Int64/dims=2	59706 ns	60051.5 ns	0.99
array/reductions/mapreduce/Int64/dims=2L	85032 ns	85186 ns	1.00
array/reductions/reduce/Float32/1d	35128 ns	35458 ns	0.99
array/reductions/reduce/Float32/dims=1	40488 ns	46307.5 ns	0.87
array/reductions/reduce/Float32/dims=1L	52249 ns	52046 ns	1.00
array/reductions/reduce/Float32/dims=2	57013 ns	57117 ns	1.00
array/reductions/reduce/Float32/dims=2L	69997 ns	70127.5 ns	1.00
array/reductions/reduce/Int64/1d	43669 ns	41220 ns	1.06
array/reductions/reduce/Int64/dims=1	44428.5 ns	51895 ns	0.86
array/reductions/reduce/Int64/dims=1L	87860 ns	87739 ns	1.00
array/reductions/reduce/Int64/dims=2	59799 ns	59630 ns	1.00
array/reductions/reduce/Int64/dims=2L	84792 ns	84743.5 ns	1.00
array/reverse/1d	18598 ns	18349 ns	1.01
array/reverse/1dL	69158 ns	68960 ns	1.00
array/reverse/1dL_inplace	65919 ns	65909 ns	1.00
array/reverse/1d_inplace	10278 ns	8540.833333333332 ns	1.20
array/reverse/2d	20844 ns	20881 ns	1.00
array/reverse/2dL	72918.5 ns	72996 ns	1.00
array/reverse/2dL_inplace	65969 ns	65926 ns	1.00
array/reverse/2d_inplace	11211 ns	10076 ns	1.11
array/sorting/1d	2733057 ns	2735188.5 ns	1.00
array/sorting/2d	1074904 ns	1069206 ns	1.01
array/sorting/by	3301743 ns	3304125.5 ns	1.00
cuda/synchronization/context/auto	1167.8 ns	1176.2 ns	0.99
cuda/synchronization/context/blocking	921.0277777777778 ns	924.5869565217391 ns	1.00
cuda/synchronization/context/nonblocking	8133.8 ns	6942.1 ns	1.17
cuda/synchronization/stream/auto	997.7333333333333 ns	999.9375 ns	1.00
cuda/synchronization/stream/blocking	802.5567010309278 ns	787.7961165048544 ns	1.02
cuda/synchronization/stream/nonblocking	7348.2 ns	7168.6 ns	1.03
integration/byval/reference	144066 ns	143982 ns	1.00
integration/byval/slices=1	146060 ns	145868 ns	1.00
integration/byval/slices=2	284760 ns	284528 ns	1.00
integration/byval/slices=3	423409 ns	422970 ns	1.00
integration/cudadevrt	102664 ns	102612 ns	1.00
integration/volumerhs	9442040.5 ns	9440461 ns	1.00
kernel/indexing	13355 ns	13181 ns	1.01
kernel/indexing_checked	14001 ns	14081 ns	0.99
kernel/launch	2113 ns	2150.777777777778 ns	0.98
kernel/occupancy	670.566037735849 ns	672 ns	1.00
kernel/rand	14516 ns	14396 ns	1.01
latency/import	3809768988.5 ns	3814290062.5 ns	1.00
latency/precompile	4591409363 ns	4590207670.5 ns	1.00
latency/ttfp	4386579833 ns	4409319020 ns	0.99

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