v1.8.0

What's Changed

• Support new AbstractFFTs.AdjointStyle trait for FFT and DCT plans by @gaurav-arya in #249
• Allow MKL_jll v2024 by @ChrisRackauckas in #283
• Update Project.toml by @ChrisRackauckas in #284
• CompatHelper: make PRs to bump JLL compat entries by @DilumAluthge in #285
• CI: for the macOS CI jobs with the MKL provider, force MKL_jll 2023 to be installed by @DilumAluthge in #286
• enable dependabot for GitHub actions by @ranocha in #267
• Increase CI timeout length by @gaurav-arya in #276
• Bump actions/checkout from 3 to 4 by @dependabot in #288
• Bump codecov/codecov-action from 1 to 3 by @dependabot in #289
• Test on the latest stable release by @jishnub in #258
• Bump to 1.8 by @stevengj in #291

New Contributors

• @gaurav-arya made their first contribution in #249
• @DilumAluthge made their first contribution in #285
• @dependabot made their first contribution in #288

Full Changelog: v1.7.1...v1.8.0

v1.7.1

FFTW v1.7.1

Diff since v1.7.0

Merged pull requests:

• Support MKL_jll 2023 (#270) (@dlfivefifty)

v1.7.0

FFTW v1.7.0

Diff since v1.6.0

Closed issues:

• Normalization of FFTW.r2r for FFTW.RODFT00 (#268)

Merged pull requests:

• fix typos (#265) (@spaette)
• remove tuple specialization of fix_kinds (#266) (@jishnub)
• reduce allocations in dims_howmany (#269) (@jishnub)

v1.6.0

FFTW v1.6.0

Diff since v1.5.0

Closed issues:

• FFTW could not create plan for fft(n × m × 1 array, 2) (#142)
• why mkl can not get fft(x,1)? (#237)
• I didn't ask for plan_bfft, but got MethodError: no method matching plan_bfft(::Vector{Number}, ::UnitRange{Int64}) (#244)
• FFTW.fft() fails to calculate spectrum (#250)
• MKL: FFTW could not create plan (#252)

Merged pull requests:

• Create Invalidations.yml (#251) (@ranocha)
• Improve type inference in plan_r2r (#253) (@jishnub)
• inference + allocation in fix_kinds (#254) (@jishnub)
• remove test_broken on master (#257) (@jishnub)
• improve type inference in DCTPlan (#259) (@jishnub)
• Update GitHub Actions setup and move to Cirrus for other platforms (#262) (@giordano)
• Bump version to v1.5.1 (#263) (@jishnub)
• bump to 1.6 since new type params were added (#264) (@stevengj)

v1.5.0

FFTW v1.5.0

Diff since v1.4.6

Closed issues:

• get version of FFTW.set_num_threads()? (#117)
• Please tag a new version (with support for MKL 2022) (#230)

Merged pull requests:

• Add get_num_threads (#171) (@galenlynch)

v1.4.6

FFTW v1.4.6

Diff since v1.4.5

Closed issues:

• Migrate to travis-ci.com (#166)
• Darwin ARM: libfftw3 not defined (#187)
• Unable to install latest version of FFTW on Julia 1.6 (#206)
• Problem with FFTW=1.4.3 ignoring DEPOT_PATH variable at runtime (#208)
• Latest FFTW release causes portability issues (#210)
• Support applying plans by column? (Or is it really 300x slower??) (#216)
• fft(x,dim) fails on Julia-1.7.0-rc1 (#220)
• First call to plan_rfft trashes the input (#226)

Merged pull requests:

• allow MKL 2022 (#228) (@KristofferC)
• Improve inferrability of dims_howmany (#231) (@timholy)

v1.4.5

FFTW v1.4.5

Diff since v1.4.4

Merged pull requests:

• remove asserts from precompile (#214) (@KristofferC)

v1.4.4

FFTW v1.4.4

Diff since v1.4.3

Merged pull requests:

• make FFTW relocatable, take 2 (#212) (@KristofferC)
• bump version (#213) (@KristofferC)

v1.4.3

FFTW v1.4.3

Diff since v1.4.2

Merged pull requests:

• Make FFTW relocatable for PackageCompiler (#207) (@staticfloat)

v1.4.2

FFTW v1.4.2

Diff since v1.4.1

Closed issues:

• Deadlock bug in FFTW 1.2.4 (#163)
• Omit the alignment check when using MKL's fftw3 wrapper? (#202)

Merged pull requests:

• Add a safepoint in the maybe_destroy_plan spin lock (#205) (@Octogonapus)