Fix setindex!
on Bidiagonal
with ev
zero bug.
#32328
Merged
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Below is a MWE and explanation of the bug. The bug is that we can assign to structural zeros in
Bidiagonal
matrices when theB.ev
vector is zero, but they show up in the off diagonal. For example:Why it happens: In
setindex!
are checkingistriu(B)
which isjulia/stdlib/LinearAlgebra/src/bidiag.jl
Line 249 in 802e719
So, when the Bidiagonal matrix has
uplo = 'U'
butB.ev
is zero,setindex!
allows us to set values on the lower off diagonal.This fix has the added effect of improving access time in the (admittedly, artificial) pessimal case of the very last element in
B.ev
being the only non-zero one.