Ivan Babrou

@[email protected]
145 Followers
121 Following
180 Posts
Doing performance @ Cloudflare.
Websitehttps://ivan.computer
GitHubhttps://github.com/bobrik
Show threadIvan BabrouFeb 26

I would like to extend my sincere apologies to geniuses at Atlassian.

The call was coming from inside the house (there was a custom css in our instance).

Ivan BabrouFeb 25
Some genius at Atlassian decided that code blocks in their wiki product no longer need monospace fonts.
Show threadIvan BabrouFeb 10

Looking at devmapper ioctl failures shows limited utility of just 16 frames of LBR. You mostly capture cleanup if you attach to high level functions.

Having bpf accounting enabled doesn't help either.

Ivan BabrouFeb 6

Sneak peek: LBR based stacks in tracing spans for failed syscalls through ebpf_exporter.

Think "retsnoop, but more structured". It be attached to existing traces too.

Also pictured: SRSO sadness.

#ebpf #kernel #ebpf_exporter

Ivan BabrouFeb 1
Molly WhiteJan 29

As ICE terrorizes communities, crypto executives who spent years posting about freedom have gone conspicuously silent. But behind the scenes, they’ve contributed at least $315 million — more than double their record 2024 spending — to elect more lawmakers to enable this administration.

https://www.citationneeded.news/issue-100/

#crypto #cryptocurrency #USpol #USpolitics #CitationNeededNewsletter

Issue 100 – Freedom of all kinds is worth fighting for

As masked agents execute people and terrorize communities, crypto executives who spent years posting about freedom fall conspicuously silent — except when writing checks for the politicians enabling it

Citation Needed
Ivan BabrouJan 30

If you have a lot of ebpf programs and you have any sort of monitoring that looks at /proc/kallsyms regularly, you might want to have my patch applied: https://lore.kernel.org/bpf/20260129-i[email protected]/T/#u

Upstream kernel does a silly thing with quadratic pointer chasing under RCU, which is not great.

Testing the patch in production on v6.12 series (with ~10k bpf ksyms):

* On AMD EPYC 9684X (Zen4): ~870ms -> ~100ms
* On Ampere Altra Max M128-30: ~4650ms -> ~70ms

#kernel #ebpf

Ivan BabrouJan 26
404 MediaJan 21

"My local community is being systematically liquidated in what I can only describe as a targeted intellectual genocide."

https://www.404media.co/ham-radio-operators-in-belarus-arrested-face-the-death-penalty/

HAM Radio Operators in Belarus Arrested, Face the Death Penalty

"My local community is being systematically liquidated in what I can only describe as a targeted intellectual genocide."

404 Media
Show threadIvan BabrouJan 8

... and Rust symbols are in fact shorter demangled if you drop the hashes.

I'm seeing 2.65x longer symbols for ClickHouse and ~0.7-0.9x shorter Rust symbols for our production code.

Show threadIvan BabrouJan 3

627k for a ptr::drop_in_place 🫣

It is 753k mangled! Yes, it is longer mangled.

Take that C++, which tops at measly 167k.

Show threadIvan BabrouJan 3

Apparently even that is not the limit, because llvm-cxxfilt choked on the biggest ones.

Here's a symbol of 64,039 bytes! Well, a part of it, because it does not fit in the screenshot. It's a part of a single stack trace that is 184,838 bytes long.

Perhaps it's a good idea to do aggregation first and demangling second to deal with this nonsense.