Ecosyste.ms: Timeline
Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.
KristofferC opened an issue on JuliaLang/julia
10.0^68 gives a different result on nightly vs 1.11 after llvm-muladd pass removal
Master: ``` julia> 10.0^68 1.0000000000000001e68 ``` 1.11: ``` julia> 10.0^68 1.0e68 ``` https://github.com/JuliaLang/julia/pull/55802 changed stuff related to `muladd` which `pow` uses at ht...knuesel created a comment on an issue on JuliaLang/julia
> The remaining 60% is probably dominated by invalidation-checks upon loading That seems right: running `julia +1.11.1 --startup-file=no -e "using Profile; using InteractiveUtils; @profile using R...
nanosoldier created a comment on a pull request on JuliaLang/julia
[Your benchmark job](https://github.com/JuliaLang/julia/pull/55002#issuecomment-2434595609) has completed - no performance regressions were detected. A full report can be found [here](https://githu...
aravindh-krishnamoorthy opened a pull request on JuliaLang/julia
Fix log_quasitriu for internal scaling s=0
This PR is a potential fix for #54833. ## Description The function https://github.com/JuliaLang/julia/blob/2a06376c18afd7ec875335070743dcebcd85dee7/stdlib/LinearAlgebra/src/triangular.jl#L2220 ...jishnub created a review comment on a pull request on JuliaLang/julia
The difference probably doesn't matter in this case as we use linear indexing, but for Cartesian indexing, I've recently found that putting the `@inbounds` on the loop can slow down the iterations ...
dkarrasch created a review comment on a pull request on JuliaLang/julia
Hm, that seems a bit weird to add another `begin ... end` group. I'd suggest to leave the annotation in front of the `for`-loop, since it's just as tight in scope as this one.
dkarrasch created a review comment on a pull request on JuliaLang/julia
Do you need the line `m, n = size(A)` anymore? Looks like dead code, so we could just as well remove it. Also, personally I prefer `for i in axes(A, 1)` over the matlabian `i =`, but that's just a ...
timholy created a comment on an issue on JuliaLang/julia
> but only 40% of that is compilation The remaining 60% is probably dominated by invalidation-checks upon loading. (When things are baked into the sysimage, you don't have to run those checks.) If...
dkarrasch created a comment on a pull request on JuliaLang/julia
@nanosoldier `runbenchmarks("linalg", vs=":master")`
KristofferC created a comment on a pull request on JuliaLang/julia
``` julia> using JuliaInterpreter, BracedErrors julia> bracederror(2.34567e70, 3.46e69) == @interpret bracederror(2.34567e70, 3.46e69) false ``` Maybe some fast math stuff?
aviatesk pushed 21 commits to jn/codeinfo-edges JuliaLang/julia
- REPL: fix closing quote on completing files in a ~ path (#56253) 133051f
- Implement faster `issubset` for `CartesianIndices{N}` (#56282) Co-authored-by: xili <[email protected]> 0b9fcb5
- Improve doc example: Extracting the type parameter from a super-type (#55983) Documentation describes the correct wa... 6c70bf7
- llvmpasses: force vector width for compatibility with non-x86 hosts. (#56300) The pipeline-prints test currently fai... 005608a
- Reduce generic matrix*vector latency (#56289) ```julia julia> using LinearAlgebra julia> A = rand(Int,4,4); x = ran... b9b4dfa
- Type `Base.is_interactive` as `Bool` (#56303) Before, typing `Base.is_interactive = 7` would cause weird internal RE... 28b0abd
- REPL: don't complete str and cmd macros when the input matches the internal name like `r_` to `r"` (#56254) 4236a33
- fix REPL test if a "juliadev" directory exists in home (#56218) 0a6277d
- Fix trampoline warning on x86 as well (#56280) 894296b
- typeintersect: more fastpath to skip intersect under circular env (#56304) fix #56040 53ffe56
- Preserve type in `first` for `OneTo` (#56263) With this PR, ```julia julia> first(Base.OneTo(10), 4) Base.OneTo(4) `... 2a06376
- add edges metadata field to CodeInfo/CodeInstance, prepare for using This records all invoke targets as edges as a f... 15e5227
- inference: compute edges more precisely in post-inference Start computing edges from stmt_info later (after CodeInst... 829ca4c
- compute edges post-inference, from info available there eb80a9f
- Use original computed edges during serialization instead of trying to guess them 26bd567
- make `add_edges!` a proper `CallInfo` interface 75776e1
- fix `NoinlineCallInfo` implementation 56aae3e
- propagate results of `_hasmethod`/`applicable` using `MethodResultPure` To avoid calling `add_edges!` directly. In f... f7e18fa
- remove no longer needed `edge` variables 6cda1eb
- make `CallInfo` propagate the `edges` list of `CodeInstance`s Remaining TODOs: - Finalize the format for `sv.edges`.... 27fe7f0
- and 1 more ...
aviatesk opened a pull request on JuliaLang/julia
inference: fix `[modifyfield!|replacefield!]_tfunc`s
Currently the following code snippet results in an internal error: ```julia julia> func(x) = @atomic :monotonic x[].count += 1; julia> let;Base.Experimental.@force_compile x = Ref(...aviatesk pushed 1 commit to avi/fix-atomic-modify-tfunc JuliaLang/julia
- inference: fix `[modifyfield!|replacefield!]_tfunc`s Currently the following code snippet results in an internal err... 98cd9bb
aviatesk created a branch on JuliaLang/julia
avi/fix-atomic-modify-tfunc - The Julia Programming Language
jishnub created a comment on a pull request on JuliaLang/julia
We should probably run the linalg benchmarks here as well
jishnub pushed 1 commit to master JuliaLang/julia
- Preserve type in `first` for `OneTo` (#56263) With this PR, ```julia julia> first(Base.OneTo(10), 4) Base.OneTo(4) `... 2a06376