Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Magic-trace – High-resolution traces of what a process is doing

github.com

Post reply on HN
Print view View original

41–50 of 146 posts

Re: Magic-trace – High-resolution traces of what a process is doing

#41

Post by DonaldPShimoda »

dqpb wrote: post #32
DonaldPShimoda wrote: post #17

Earlier quoted context omitted.

They are among the most notable users of OCaml in the world, and have taken significant stewardship of the language. They sponsor, attend, and publish at multiple academic conferences, especially in the realm of programming languages. I don't think I can say I support their company's primary mission, but their commitment to improving the world of software through various means (language influence, academic publicatio…

Do you take issue with the market maker business model?

I think just the idea of a company that does not directly produce things, and instead spends its efforts turning its money into more money via investment, is something that... doesn't entirely sit well with me.

There is, of course, something to be said about the by-products of their work. Jane Street is far from an evil company, and I would not be entirely morally opposed to working for or with them. They do a lot of good in academic research in areas I care about. I just wish that that was their primary purpose instead of direct money-making, if that makes sense.

Re: Magic-trace – High-resolution traces of what a process is doing

#42

Post by kolbe »

bitcharmer wrote: post #19
kolbe wrote: post #6

Nice. And it satisfies my curiosity about whether trading firms are switching to AMD or not

In HFT single-threaded performance is king so that's why we're all still on Intel. AMD is making progress but not just quite there yet.

Huh. My experience has been that AMD wins that unless your application is so small that it can fit into Intel's smaller cache. And the new 3D architecture from AMD I thought would make your developers drool, allowing them to actually inline everything instead of being scared of building apps that are too big to fit into cache

Re: Magic-trace – High-resolution traces of what a process is doing

#43

Post by kangda123 »

DonaldPShimoda wrote: post #41
dqpb wrote: post #32

Earlier quoted context omitted.

Do you take issue with the market maker business model?

I think just the idea of a company that does not directly produce things, and instead spends its efforts turning its money into more money via investment, is something that... doesn't entirely sit well with me. There is, of course, something to be said about the by-products of their work. Jane Street is far from an evil company, and I would not be entirely morally opposed to working for or with them. They do a lot of…

Doesn't that pretty much characterise all for-profit companies?

Re: Magic-trace – High-resolution traces of what a process is doing

#44

Post by cgaebel »

DSingularity wrote: post #37

I thought perf can also exploit Intel processor trace?

It can! You can read more about how to do that yourself here: https://perf.wiki.kernel.org/index.php/Perf_tools_support_fo....

magic-trace uses perf. If you want, you can think of it as a mere "alternative frontend" for the Intel PT decoding offered by perf.

Re: Magic-trace – High-resolution traces of what a process is doing

#45

Post by cgaebel »

bitcharmer wrote: post #19
kolbe wrote: post #6

Nice. And it satisfies my curiosity about whether trading firms are switching to AMD or not

In HFT single-threaded performance is king so that's why we're all still on Intel. AMD is making progress but not just quite there yet.

To be clear: bitcharmer says "we" to mean "fellow HFTs" not "Jane Street".

Re: Magic-trace – High-resolution traces of what a process is doing

#46

Post by tbr1 »

TooSmugToFail wrote: post #40
cgaebel wrote: post #2

Hi HN! I'm Clark, one of the maintainers of magic-trace. magic-trace was submitted before, our first announcement was this blog post: https://blog.janestreet.com/magic-trace/ . Since then, we've worked hard at making magic-trace more accessible to outside users. We've heard stories of people thinking this was cool before but being unable to even find a download link. I'm posting this here because we just released "ve…

Awesome work Clark! Any plans to support Arm in the future? Thanks!

We don't have plans to add ARM support largely because we have no in-house expertise with ARM. That said, ARM has CoreSight which sounds like it could support something like magic-trace in some form, and we'd definitely be open to community contributions for CoreSight support in magic-trace.

Re: Magic-trace – High-resolution traces of what a process is doing

#47

Post by tbr1 »

rrss wrote: post #26
Scaevolus wrote: post #23

If you just want calls and returns, can't you use one of the other PMUs for that? Or is sampling at the "1 sample per event" level higher overhead than IPT?

do you mean configuring the other PMUs to interrupt the core every function call / return? If yes, then yes that is much much higher overhead than processor trace.

It's worth noting that aside from the overhead, function call / returns are not quite enough to reconstruct the callstack: tailcalls are just regular branch instructions.

Re: Magic-trace – High-resolution traces of what a process is doing

#50

Post by DSingularity »

cgaebel wrote: post #44
DSingularity wrote: post #37

I thought perf can also exploit Intel processor trace?

It can! You can read more about how to do that yourself here: https://perf.wiki.kernel.org/index.php/Perf_tools_support_fo... . magic-trace uses perf. If you want, you can think of it as a mere "alternative frontend" for the Intel PT decoding offered by perf.

Ah okay, I misunderstood magic-trace to be an alternative to perf.
Post reply on HN