-
Sometimes the sparse structure of a problem can change during non-linear optimization. It would be useful to be able to update the symbolic analysis of the matrix in-place, to avoid re-allocating all …
-
### Description
S3267 advises to use **LINQ To Objects** in loops, which is a sever performance penalty.
### Related information
The compliant code:
```csharp
foreach (var someValue in co…
-
https://github.com/llvm/llvm-project/commit/6dc1636815cb9321657f1cb7ac87a46553870dc7 aka implementation of `[dcl.fct]` changes introduced by https://wg21.link/P0929 causes the following example to tri…
-
I'm not going to argue here at all, but we should consider using `netcoreapp2.2` in the future, and defaulting to the non-allocating Span-based API:s.
_Originally posted by @haf in https://github.c…
-
I have mixed feelings about your proposed `Claim` operation. It feels to me like it's trying to do three things that could meaningfully be distinguished, and as a result, lacks clarity.
1. On the o…
-
#4 is fixed but there's a bunch of stuff I didn't have time for.
To test:
* [ ] allocating >1 page works
* [ ] pointers from one page to another keep referents alive
* [ ] gc collects cycles acr…
-
Consider a tuple of `NamedTuple`s of `f::Float64` and `i::Int64` (inhomogeneous types):
```julia-repl
julia> VERSION
v"1.9.3"
julia> s = ((f = 0.0, i = 128), (f = 1.0, i = 256))
((f = 0.0, i = …
-
According to our telemetry, in many cases users have a bytes-like type with width defined (i.e. something like `STRING(128)`). Our `coldata.Bytes` vectors can support values up to 30 bytes that are fu…
-
osm2pgsql version 0.96.0 (64 bit id space)
Allocating memory for dense node cache
Allocating dense node cache in one big chunk
Allocating memory for sparse node cache
Sharing dense sparse
Node-…
-
Whilst working on speedups & unit tests in #1392, I noticed the following behaviour in the model. Not sure if this is intended/accepted, but it struck me as odd so mentioning it here.
At [present](…