I assume this would break observability through existing methods, right? If you were to strace a process that has been patched, would you see regular syscall data (as if it wasnt patched) or would your syscall replacement appear along the way?
You mentioned SECCOMP_RET_TRACE, but there is also SECCOMP_RET_TRAP[1] which appears to perform better. There is also KVM. Both of these are options for gVisor: <https://github.com/google/gvisor>
This might be a very dumb question, but if the process is being run under KVM to catch `int 0x03` then couldn't you also use KVM to catch `syscall` and execute the original binary as-is? I don't understand what value the instruction rewriting is providing here.
This is the kind of foundation that I would feel comfortable running agents on. It’s not the whole solution of course (yes agent, you’re allowed to delete this email but not that email can’t be solved at this level)… let me know when you tackle that next :-)
You either have a writing style that is uncannily similar to what an LLM generates, or this article was substantially written by an LLM. I don't know what it is about the style, but I just find it a bit exhausting, like an overfit on "engaging writing" that strips away sincerity.
I think it's better to just adapt to this. A lot of people write the content their own way, and get AI to rewrite it so that it is more readable, and free from errors. Content over appearance and all. I think the problem is you consider this auto-completion tool insincere. many do as well, because they anthropomorphize LLMs, it feels like a different sentient entity wrote it than the person posting it. but in reality, that isn't the case; it's more like a spellchecker that helped the person communicate their idea.
The purpose of language is to communicate meaning and intent, not to sound or feel a particular way, unless you're reading for entertainment or enjoyment.
This is the second post I'm commenting on within a span of like 30 minutes where someone did some really good work and shared it, but the top comments are complaining about AI usage.
Either LLM-assisted content needs to be banned entirely (might be), or complaining about it should be considered a breach of etiquette at sites like HN that are tech-centric.
16 comments
[ 2.8 ms ] story [ 46.9 ms ] thread[1] <https://github.com/google/gvisor/blob/master/pkg/sentry/plat...>
How secure does this make a binary? For example would you be able to run untrusted binary code inside a browser using a method like this?
Then can websites just use C++ instead of javascript for example?
This is the kind of foundation that I would feel comfortable running agents on. It’s not the whole solution of course (yes agent, you’re allowed to delete this email but not that email can’t be solved at this level)… let me know when you tackle that next :-)
The purpose of language is to communicate meaning and intent, not to sound or feel a particular way, unless you're reading for entertainment or enjoyment.
This is the second post I'm commenting on within a span of like 30 minutes where someone did some really good work and shared it, but the top comments are complaining about AI usage.
Either LLM-assisted content needs to be banned entirely (might be), or complaining about it should be considered a breach of etiquette at sites like HN that are tech-centric.
What's stopping the process from reading its own memory and seeing that the syscall was patched?
https://gvisor.dev/
example how we used it in early 2000s to implement pre linux namespace containerization.
https://www.usenix.org/legacy/publications/library/proceedin... (note the shepherd and where kubernetes arguably got the pod name from).
and security policies on top of it
https://www.usenix.org/legacy/event/lisa07/tech/full_papers/...