Moorko
@moorko.net
I like to make high performance [networking-related] systems with C++ (and probably Go?). I like aviation too. Want to become a scientist someday.
https://moorko.net
https://github.com/Moork0
https://moorko.net
https://github.com/Moork0
Pinned
Moorko
@moorko.net
· Nov 10
Adding TLS Fragmentation to OpenConnect
In this post, I discussed what it means to apply fragmentation at the TCP and TLS Record layers during the TLS handshake, and how we can use these techniques to evade Internet censorship.
moorko.net
Hi there! It's my first post here in bsky!
I recently published my first blog post (in English):
moorko.net/adding-tls-r...
I wrote about TLS fragmentation for evading Internet censorship.
I recently published my first blog post (in English):
moorko.net/adding-tls-r...
I wrote about TLS fragmentation for evading Internet censorship.
Man... Google search is broken af.
I remember a time when you could easily find what you want but nowadays it even fails to show you relevant results.
I remember a time when you could easily find what you want but nowadays it even fails to show you relevant results.
January 28, 2025 at 1:42 AM
Man... Google search is broken af.
I remember a time when you could easily find what you want but nowadays it even fails to show you relevant results.
I remember a time when you could easily find what you want but nowadays it even fails to show you relevant results.
I adore night. I basically wait all day for it. Wish it was a bit easier to handle the late night depression/anxiety though.
January 26, 2025 at 7:12 PM
I adore night. I basically wait all day for it. Wish it was a bit easier to handle the late night depression/anxiety though.
I'd love to have Github runners in my homelab. Would be fun experience to integrate that with Kubernetes ig.
January 25, 2025 at 12:09 AM
I'd love to have Github runners in my homelab. Would be fun experience to integrate that with Kubernetes ig.
I'm this 🤏 close to spend ~650 bucks buying this Synology DS723+
God help me.
God help me.
January 24, 2025 at 11:50 PM
I'm this 🤏 close to spend ~650 bucks buying this Synology DS723+
God help me.
God help me.
Linux kernel documentation says that if you find yourself only using the non-RMW operations of atomic_t (i.e. only store, load), you do not in fact need atomic_t at all and are doing it wrong.
January 3, 2025 at 10:21 PM
Linux kernel documentation says that if you find yourself only using the non-RMW operations of atomic_t (i.e. only store, load), you do not in fact need atomic_t at all and are doing it wrong.
Many CPUs speculate with loads: that is they see that they will need to load an item from memory, and they find a time where they're not using the bus for any other loads, and so do the load in advance - even though they haven't actually got to that point in the instruction execution flow yet.
December 31, 2024 at 3:34 AM
Many CPUs speculate with loads: that is they see that they will need to load an item from memory, and they find a time where they're not using the bus for any other loads, and so do the load in advance - even though they haven't actually got to that point in the instruction execution flow yet.
TIL that this is UB in C++:
n = ++i + i;
n = ++i + i;
December 30, 2024 at 8:02 PM
TIL that this is UB in C++:
n = ++i + i;
n = ++i + i;
I though I know memory orderings.... until I read the Linux kernel's memory barrier documentation.
Now I forgot everything I knew about memory orderings (in C++) and how to use them properly. Damn.
Now I forgot everything I knew about memory orderings (in C++) and how to use them properly. Damn.
December 29, 2024 at 12:44 AM
I though I know memory orderings.... until I read the Linux kernel's memory barrier documentation.
Now I forgot everything I knew about memory orderings (in C++) and how to use them properly. Damn.
Now I forgot everything I knew about memory orderings (in C++) and how to use them properly. Damn.
Reposted by Moorko
What should I spend my time on? The job that my visa and stay in the Netherlands (and effectively being alive) depends on or some random concept in computer architecture detail in the Intel software manual?
Dammit, of course the latter!
Dammit, of course the latter!
December 22, 2024 at 7:40 PM
What should I spend my time on? The job that my visa and stay in the Netherlands (and effectively being alive) depends on or some random concept in computer architecture detail in the Intel software manual?
Dammit, of course the latter!
Dammit, of course the latter!
Reposted by Moorko
How do you spot Noob Programmers: they are too enthusiastic about Tech.
December 17, 2024 at 1:10 AM
How do you spot Noob Programmers: they are too enthusiastic about Tech.
Reposted by Moorko
Security pro tip: if you find a USB key on the ground somewhere, DO NOT plug it in to your home computer.
Take it to the office and plug it in there. They have way better software that will mitigate any threats.
Take it to the office and plug it in there. They have way better software that will mitigate any threats.
November 27, 2024 at 6:09 PM
Security pro tip: if you find a USB key on the ground somewhere, DO NOT plug it in to your home computer.
Take it to the office and plug it in there. They have way better software that will mitigate any threats.
Take it to the office and plug it in there. They have way better software that will mitigate any threats.
I implemented a simple interruptible thread.
github.com/Moork0/Concu...
Now I'll move to Golang and eBPF
github.com/Moork0/Concu...
Now I'll move to Golang and eBPF
github.com
November 28, 2024 at 2:41 PM
I implemented a simple interruptible thread.
github.com/Moork0/Concu...
Now I'll move to Golang and eBPF
github.com/Moork0/Concu...
Now I'll move to Golang and eBPF
Reposted by Moorko
I want to be so many things all at the same time
November 23, 2024 at 12:37 AM
I want to be so many things all at the same time
Reposted by Moorko
I asked the electrician:
- How bad is the previous installation?
He replied:
- This isn’t that bad, things change over time, things get old. You measure a good implementation on how easy it is to replace. This one is easy to replace.
I love that way of thinking, it relates to programming as well.
- How bad is the previous installation?
He replied:
- This isn’t that bad, things change over time, things get old. You measure a good implementation on how easy it is to replace. This one is easy to replace.
I love that way of thinking, it relates to programming as well.
November 22, 2024 at 10:11 AM
I asked the electrician:
- How bad is the previous installation?
He replied:
- This isn’t that bad, things change over time, things get old. You measure a good implementation on how easy it is to replace. This one is easy to replace.
I love that way of thinking, it relates to programming as well.
- How bad is the previous installation?
He replied:
- This isn’t that bad, things change over time, things get old. You measure a good implementation on how easy it is to replace. This one is easy to replace.
I love that way of thinking, it relates to programming as well.
Well, actually implementing a cancel/interrupt feature isn't difficult especially since std::jthread is in fact an interruptible thread. But I want to implement it myself so...
November 23, 2024 at 1:20 AM
Well, actually implementing a cancel/interrupt feature isn't difficult especially since std::jthread is in fact an interruptible thread. But I want to implement it myself so...
The first two features are now implemented.
I'm going to implement the third one soon.
After that, I want to switch over Golang and eBPF and do some experiments there. My main goal for them is to implement a L4 load balancer like the one @cloudflare.social unimog team has.
I'm going to implement the third one soon.
After that, I want to switch over Golang and eBPF and do some experiments there. My main goal for them is to implement a L4 load balancer like the one @cloudflare.social unimog team has.
I'm writing a simple thread pool and some concurrent algorithms in C++. It's fun to review some of concurrency techniques.
3 more things to implement and then I guess it would be done:
1- Local task queues for each thread to reduce cache contention on the global queue
2- Simple form of work stealing
3 more things to implement and then I guess it would be done:
1- Local task queues for each thread to reduce cache contention on the global queue
2- Simple form of work stealing
November 22, 2024 at 12:24 AM
The first two features are now implemented.
I'm going to implement the third one soon.
After that, I want to switch over Golang and eBPF and do some experiments there. My main goal for them is to implement a L4 load balancer like the one @cloudflare.social unimog team has.
I'm going to implement the third one soon.
After that, I want to switch over Golang and eBPF and do some experiments there. My main goal for them is to implement a L4 load balancer like the one @cloudflare.social unimog team has.
Reposted by Moorko
working on this unix redirects diagram today
November 21, 2024 at 6:56 PM
working on this unix redirects diagram today
Reposted by Moorko
For further analysis on this incident, be sure to check out the great work by Emile Aben of RIPE NCC and @dbelson.com of @cloudflare.social:
labs.ripe.net/author/emile...
blog.cloudflare.com/resilient-in...
labs.ripe.net/author/emile...
blog.cloudflare.com/resilient-in...
Does the Internet Route Around Damage? - Baltic Sea Cable Cuts
This week's Internet cable cuts in the Baltic Sea have been widely reported, even as attempts to understand their cause and impact continue. We turn to RIPE Atlas to provide a preliminary analysis of ...
labs.ripe.net
November 21, 2024 at 10:03 PM
For further analysis on this incident, be sure to check out the great work by Emile Aben of RIPE NCC and @dbelson.com of @cloudflare.social:
labs.ripe.net/author/emile...
blog.cloudflare.com/resilient-in...
labs.ripe.net/author/emile...
blog.cloudflare.com/resilient-in...
It's been months since I submitted my PR to the openconnect project but not only the maintainers don't approve, but also they don't even respond 😕
November 19, 2024 at 2:20 PM
It's been months since I submitted my PR to the openconnect project but not only the maintainers don't approve, but also they don't even respond 😕
Reposted by Moorko
“I'm now convinced that async/await is, in fact, a bad abstraction for most languages, and we should be aiming for something better instead and that I believe to be thread.”
lucumr.pocoo.org/2024/11/18/t...
lucumr.pocoo.org/2024/11/18/t...
Playground Wisdom: Threads Beat Async/Await
Musings about async await again and why I think virtual threads are a better model.
lucumr.pocoo.org
November 18, 2024 at 12:04 PM
“I'm now convinced that async/await is, in fact, a bad abstraction for most languages, and we should be aiming for something better instead and that I believe to be thread.”
lucumr.pocoo.org/2024/11/18/t...
lucumr.pocoo.org/2024/11/18/t...
Reposted by Moorko
I made a thing!
I was playing around with the AT protocol and as a little experiment I made a website that visualises activity around Bluesky: nightsky.hctr.dev
It listens to all new posts and shows them as little stars across a night sky 🌃
Every star is someone, somewhere, posting something
I was playing around with the AT protocol and as a little experiment I made a website that visualises activity around Bluesky: nightsky.hctr.dev
It listens to all new posts and shows them as little stars across a night sky 🌃
Every star is someone, somewhere, posting something
Nightsky | hctr.dev
See live conversations from all over Bluesky as a dynamic night sky
nightsky.hctr.dev
November 16, 2024 at 11:05 PM
I made a thing!
I was playing around with the AT protocol and as a little experiment I made a website that visualises activity around Bluesky: nightsky.hctr.dev
It listens to all new posts and shows them as little stars across a night sky 🌃
Every star is someone, somewhere, posting something
I was playing around with the AT protocol and as a little experiment I made a website that visualises activity around Bluesky: nightsky.hctr.dev
It listens to all new posts and shows them as little stars across a night sky 🌃
Every star is someone, somewhere, posting something
Someone asked me why do we need local queues?
Well, every time a thread calls submit() on a particular instance of the thread pool, it has to push a new item onto the single shared work queue. Likewise, the worker threads are continually popping items off the queue in order to run the tasks.
Well, every time a thread calls submit() on a particular instance of the thread pool, it has to push a new item onto the single shared work queue. Likewise, the worker threads are continually popping items off the queue in order to run the tasks.
I'm writing a simple thread pool and some concurrent algorithms in C++. It's fun to review some of concurrency techniques.
3 more things to implement and then I guess it would be done:
1- Local task queues for each thread to reduce cache contention on the global queue
2- Simple form of work stealing
3 more things to implement and then I guess it would be done:
1- Local task queues for each thread to reduce cache contention on the global queue
2- Simple form of work stealing
November 18, 2024 at 1:30 AM
Someone asked me why do we need local queues?
Well, every time a thread calls submit() on a particular instance of the thread pool, it has to push a new item onto the single shared work queue. Likewise, the worker threads are continually popping items off the queue in order to run the tasks.
Well, every time a thread calls submit() on a particular instance of the thread pool, it has to push a new item onto the single shared work queue. Likewise, the worker threads are continually popping items off the queue in order to run the tasks.