fix typo in Float32 random number generation #57338
Conversation
introduced in #56144
KristofferC
changed the title
stdlib/Random/test/runtests.jl
Outdated
| @testset "Float32 RNG typo" begin | ||
| @test length(unique(rand(Float32, 14))) > 10 | ||
| end |
There was a problem hiding this comment.
Extend to other types as well? This won't catch similar issues otherwise.
There was a problem hiding this comment.
BTW, there are two completely different codepaths for generating random numbers, one for less than 32 numbers and another one for larger vectors. It'd be good to exercise both, especially since the larger one involves llvmcalls which are harder to debug.
There was a problem hiding this comment.
The test here is pretty much just to check that the concrete issue is fixed. I don't think there is much value in extending the tests more than that.
There was a problem hiding this comment.
Well, it'd have caught #55997 (comment)
There was a problem hiding this comment.
Why wasn't the test added there? ;). Anyway, general RNG test improvements are nice but doesn't have to be mixed up with this specific bugfix.
There was a problem hiding this comment.
Why wasn't the test added there?
I was thinking of count iszero tests, but those are brittle at best, especially for Float16 (which generates 2048 unique numbers, so the probability of picking exactly 0 is non-negligible) and especially for small vectors (which was the broken case), I could hit test failures fairly easily.
| @@ -1250,6 +1250,12 @@ end | |||
| @test length(xs) == 3 | |||
| end | |||
|
|
|||
| @testset "Float32 RNG typo" begin | |||
There was a problem hiding this comment.
| @testset "Float32 RNG typo" begin | |
| @testset "Randomness of RNGs" begin |
There was a problem hiding this comment.
Well, that oversells it a bit maybe heh.
Co-authored-by: Mosè Giordano <[email protected]>
introduced in #56144 --------- Co-authored-by: Mosè Giordano <[email protected]> (cherry picked from commit dd13878)
Backported PRs: - [x] #57142 <!-- Add reference to time_ns in time --> - [x] #57241 <!-- Handle `waitpid` race condition when `SIGCHLD` is set to `SIG_IGN` --> - [x] #57249 <!-- restore non-freebsd-unix fix for profiling --> - [x] #57211 <!-- Ensure read/readavailable for BufferStream are threadsafe --> - [x] #57262 <!-- edit NEWS for v1.12 --> - [x] #57226 <!-- cfunction: reimplement, as originally planned, for reliable performance --> - [x] #57253 <!-- bpart: Fully switch to partitioned semantics --> - [x] #57273 <!-- fix "Right arrow autocompletes at line end" implementation --> - [x] #57280 <!-- dep: Update JuliaSyntax --> - [x] #57229 <!-- staticdata: Close data race after backedge insertion --> - [x] #57298 <!-- Updating binding version to fix MMTk CI --> - [x] #57248 <!-- improve concurrency safety for `Compiler.finish!` --> - [x] #57312 <!-- Profile.print: de-focus sleeping frames as gray --> - [x] #57289 <!-- Make `OncePerX` subtype `Function` --> - [x] #57310 <!-- Make ptls allocations at least 128 byte aligned --> - [x] #57311 <!-- Add a warning for auto-import of types --> - [x] #57338 <!-- fix typo in Float32 random number generation --> - [x] #57293 <!-- Fix getfield_tfunc when order or boundscheck is Vararg --> - [x] #57349 <!-- docs: fix-up world-age handling for META access --> - [x] #57344 <!-- Add missing type asserts when taking the queue out of the task struct --> - [x] #57348 <!-- 🤖 [master] Bump the SparseArrays stdlib from 212981b to 72c7cac --> - [x] #55040 <!-- Allow macrocall as function sig --> - [x] #57299 <!-- Add missing latestworld after parameterized type alias -->
KristofferC
removed
the
backport 1.12
introduced in #56144