Jordan Rife

I'm curious if [`tseries`](https://github.com/bpftrace/bpftrace/blob/master/man/adoc/bpftrace.adoc#map-functions-tseries) would help at all with your use case. It wouldn't be exactly the same effect as what you describe above, where you capture *every* sample, but...

* Override `print` vs `print_tseries` in general for different output options. > Unicode characters instead of the '*' for text output Mostly cosmetic, but this is something I'd like to...

[Surfacing my comment regarding print formatting and options](https://github.com/bpftrace/bpftrace/pull/3838#issuecomment-3025368242), since it's kind of buried now. > I guess more detailed discussion can wait until that follow-up PR, but on the surface...

Would `probe-filters` be a little more future-proof? I know we discussed in https://github.com/bpftrace/bpftrace/issues/4347 not allowing users to run tests, benchmarks, or probes at the same time for now, but even...

@jordalgo I added a bit of detail comparing various syntaxes here. As I went down the road of implementing what we discussed [here](https://github.com/bpftrace/bpftrace/issues/3839#issuecomment-2719341821) I found some rough edges I wasn't...

> I'll start reviewing this as-is anyway but feel free to make that update if you prefer. @jordalgo Sounds good, will do.

@jordalgo Alright, I went ahead and shifted to the other syntax (e.g. `tseries(5, "1s", 10, "max")`). Was able to drop two commits from the series along the way, which is...

> In the example about where you're tracking TCP congestion the timestamp on the left never changes. Ah, I copied this example from the original RFC. I think this was...

> I'm going to try to finish up my first review pass tomorrow - sorry it's taking me so long. Np, thanks for taking a look. Will follow up with...

> I'm torn but leaning toward the * instead of the x - I think I like both better than o I switched to `*` > When we print the...