Antoine Pitrou comments

Results 823 comments of


                                            Antoine Pitrou

GH-41301: [C++] Extract the kernel loops used for PrimitiveTakeExec and generalize to any fixed-width type

> Do you want me to copy and paste the code so that `gcc` can inline modular code correctly? No, it's not important. I was just posting the results for...

GH-41035: [C++] Add a grouper benchmark for preventing performance regression

I would really prefer items/s over bytes/s as bytes/s is highly misleading. In the end, we are most of the time interested in performance over logical items, not in how...

GH-41035: [C++] Add a grouper benchmark for preventing performance regression

(that said, it's not forbidden to add both)

GH-41035: [C++] Add a grouper benchmark for preventing performance regression

I find it interesting that some benchmark variations fall to abysmal speeds, for example for {int32, int64}: ``` GrouperWithMultiTypes/"{int32, int64}"/1024/100 12616 us 12614 us 55 items_per_second=81.1817k/s null_percent=1 size=1.024k GrouperWithMultiTypes/"{int32, int64}"/1024/10...

GH-41035: [C++] Add a grouper benchmark for preventing performance regression

@ZhangHuiGui I have lost track of what this PR's status is. Does it need to wait for other PRs? Do you intend to bring further changes to it?

GH-39858: [C++][Device] Add Copy/View slice functions to MemoryManager

I don't really understand what the purpose of this PR is. It claims that slicing a Buffer is overkill, but I'm a bit skeptical: are there really situations where the...

GH-39858: [C++][Device] Add Copy/View slice functions to MemoryManager

@github-actions crossbow submit test-cuda*

Update Glibc Version in Conda Images

> The need for this old version comes from the need to build binaries that can run on a large set of machines and updating the global minimum comes at...

GH-39565: [C++] Do not concatenate chunked values of fixed-width types to run "array_take"

Hmm, some of the CI failures are definitely related to this PR (there are crashes when invoking a chunked array take). Could you take a look?

GH-39565: [C++] Do not concatenate chunked values of fixed-width types to run "array_take"

Ok, a first issue is that taking a chunked dictionary array crashes (are there no tests for this?): ```python >>> a1 = pa.array(["foo", "bar", "foo"]).dictionary_encode() >>> a2 = pa.array([None, "bar",...