Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Linalg: matprod_dest for Diagonal and adjvec #53611

Merged
merged 1 commit into from
Mar 11, 2024

Conversation

jishnub
Copy link
Contributor

@jishnub jishnub commented Mar 5, 2024

Fixes JuliaLang/LinearAlgebra.jl#1057

This restores the v1.10 behavior:

julia> Diagonal([1.0]) \ [1.0]'
1×1 adjoint(::Vector{Float64}) with eltype Float64:
 1.0

This interprets the Diagonal as the matrix representation of a linear operator in the adjoint vector space.

Something similar may also be done for other structured matrix types for consistency, but that may be a different PR as this one may need to be backported.

@jishnub jishnub added linear algebra Linear algebra arrays [a, r, r, a, y, s] backport 1.11 Change should be backported to release-1.11 labels Mar 5, 2024
@KristofferC KristofferC mentioned this pull request Mar 6, 2024
60 tasks
@jishnub jishnub requested a review from dkarrasch March 10, 2024 05:50
@dkarrasch dkarrasch merged commit f882c00 into master Mar 11, 2024
11 checks passed
@dkarrasch dkarrasch deleted the jishnub/matproddest_similar_diagadjvec branch March 11, 2024 14:15
KristofferC pushed a commit that referenced this pull request Mar 15, 2024
KristofferC added a commit that referenced this pull request Mar 17, 2024
Backported PRs:
- [x] #39071 <!-- Add a lazy `logrange` function and `LogRange` type -->
- [x] #51802 <!-- Allow AnnotatedStrings in log messages -->
- [x] #53369 <!-- Orthogonalize re-indexing for FastSubArrays -->
- [x] #48050 <!-- improve `--heap-size-hint` arg handling -->
- [x] #53482 <!-- add IR encoding for EnterNode -->
- [x] #53499 <!-- Avoid compiler warning about redefining jl_globalref_t
-->
- [x] #53507 <!-- update staled `Core.Compiler.Effects` documentation
-->
- [x] #53408 <!-- task splitting: change additive accumulation to
multiplicative -->
- [x] #53523 <!-- add back an alias for `check_top_bit` -->
- [x] #53377 <!-- add _readdirx for returning more object info gathered
during dir scan -->
- [x] #53525 <!-- fix InteractiveUtils call in Base.runtests on failure
-->
- [x] #53540 <!-- use more efficient `_readdirx` for tab completion -->
- [x] #53545 <!-- use `_readdirx` for `walkdir` -->
- [x] #53551 <!-- revert "Add @create_log_macro for making custom styled
logging macros (#52196)" -->
- [x] #53554 <!-- Always return a value in 1-d circshift! of
abstractarray.jl -->
- [x] #53424 <!-- yet more atomics & cache-line fixes on work-stealing
queue -->
- [x] #53571 <!-- Update Documenter to v1.3 for inventory writing -->
- [x] #53403 <!-- Move parallel precompilation to Base -->
- [x] #53589 <!-- add back `unsafe_convert` to pointer for arrays -->
- [x] #53596 <!-- build: remove extra .a file -->
- [x] #53606 <!-- fix error path in `precompilepkgs` -->
- [x] #53004 <!-- Unexport with, at_with, and ScopedValue from Base -->
- [x] #53629 <!-- typo fix in scoped values docs -->
- [x] #53630 <!-- sroa: Fix incorrect scope counting -->
- [x] #53598 <!-- Use Base parallel precompilation to build stdlibs -->
- [x] #53649 <!-- precompilepkgs: package in boths deps and weakdeps are
in fact only weak -->
- [x] #53671 <!-- Fix bootstrap Base precompile in cross compile
configuration -->
- [x] #52125 <!-- Load Pkg if not already to reinstate missing package
add prompt -->
- [x] #53602 <!-- Handle zero on arrays of unions of number types and
missings -->
- [x] #53516 <!-- permit NamedTuple{<:Any, Union{}} to be created -->
- [x] #53643 <!-- Bump CSL to 1.1.1 to fix libgomp bug -->
- [x] #53679 <!-- move precompile workload back from Base -->
- [x] #53663 <!-- add isassigned methods for reinterpretarray -->
- [x] #53662 <!-- [REPL] fix incorrectly cleared line after completions
accepted -->
- [x] #53611 <!-- Linalg: matprod_dest for Diagonal and adjvec -->
- [x] #53659 <!-- fix #52025, re-allow all implicit pointer casts in
cconvert for Array -->
- [x] #53631 <!-- LAPACK: validate input parameters to throw informative
errors -->
- [x] #53628 <!-- Make some improvements to the Scoped Values
documentation. -->
- [x] #53655 <!-- Change tbaa of ptr_phi to tbaa_value  -->
- [x] #53391 <!-- Default to the medium code model in x86 linux -->
- [x] #53699 <!-- Move `isexecutable, isreadable, iswritable` to
`filesystem.jl` -->
- [x] #41232 <!-- Fix linear indexing for ReshapedArray if the parent
has offset axes -->
- [x] #53527 <!-- Enable analyzegc checks for try catch and fix found
issues -->
- [x] #52092 
- [x] #53682 <!-- Increase build precompilation -->
- [x] #53720 
- [x] #53553 <!-- typeintersect: fix `UnionAll` unaliasing bug caused by
innervars. -->

Contains multiple commits, manual intervention needed:
- [ ] #53305 <!-- Propagate inbounds in isassigned with CartesianIndex
indices -->

Non-merged PRs with backport label:
- [ ] #53736 <!-- fix literal-pow to return the right type when the base
is -1 -->
- [ ] #53707 <!-- Make ScopedValue public -->
- [ ] #53696 <!-- add invokelatest to on_done callback in bracketed
paste -->
- [ ] #53660 <!-- put Logging back in default sysimage -->
- [ ] #53509 <!-- revert moving "creating packages" from Pkg.jl -->
- [ ] #53452 <!-- RFC: allow Tuple{Union{}}, returning Union{} -->
- [ ] #53402 <!-- Add `jl_getaffinity` and `jl_setaffinity` -->
- [ ] #52694 <!-- Reinstate similar for AbstractQ for backward
compatibility -->
- [ ] #51928 <!-- Styled markdown, with a few tweaks -->
- [ ] #51816 <!-- User-themable stacktraces -->
- [ ] #51811 <!-- Make banner size depend on terminal size -->
- [ ] #51479 <!-- prevent code loading from lookin in the versioned
environment when building Julia -->
@KristofferC KristofferC removed the backport 1.11 Change should be backported to release-1.11 label Mar 18, 2024
@maleadt
Copy link
Member

maleadt commented Apr 9, 2024

This broke CUDA.jl, where we relied on Diagonal{CuArray} * Adjoint{CuMatrix} being implemented with lmul! -> mul!, as we don't have a direct mul! for these types. We actually do have a mostly generic implementation of mul! for Diagonal, https://github.com/JuliaGPU/GPUArrays.jl/blob/228fb9e8f878333cc0eda8678c8e7c2d6b5a6df2/src/host/linalg.jl#L260-L318, however those aren't matched because:

julia> LinearAlgebra.Adjoint{Float64, Matrix{Float64}} <: AbstractVecOrMat
true

julia> LinearAlgebra.Adjoint{Float64, CuMatrix{Float64}} <: GPUArrays.AbstractGPUVecOrMat
false

(and simply widening the AbstractGPUArray inputs to the methods linked above creates a bunch of ambiguities.)

So I guess the root of the problem is #51910 again. Any suggested alternatives?

@dkarrasch
Copy link
Member

But this seems to touch AdjOrTransAbsVec only, not adjoint(::AbstractMatrix). Perhaps the breaking change is something else/earlier?

@maleadt
Copy link
Member

maleadt commented Apr 9, 2024

I guess #52389 then.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
arrays [a, r, r, a, y, s] linear algebra Linear algebra
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Change in backslash between adjoined arrays.
4 participants