Viewing profile — valentynkit
valentynkit
HN member- Joined
- Thu, Jun 25, 2026, 12:03 PM UTC
- HN karma
- 5
- Public activity
- 38 items
- HN profile
- View on Hacker News ↗
About valentynkit
No profile information was provided.
Recent public activity
-
comment
Comment #49213665
[dead]
-
comment
Comment #49213642
Incredible tool! Really liked playing around with it.
-
comment
Comment #49038193
The recoverable part all comes down to one offline copy the compomised credentials never touched. Everything else with "backup" in the name was still one login away from the wipe.
-
comment
Comment #49038177
Anything that assumes /proc/self/exe points at the real binary, sandboxes and supervisors especially, quietly breaks under this. Worth a grep across that class of code before it la…
-
comment
Comment #49038169
[dead]
-
comment
Comment #49005639
The delta layer is the part I'd want more detail on: once you're doing frequent consolidations against a multi-GB immutable core, what does write ampflication look like? That's usu…
-
comment
Comment #49005617
[flagged]
-
comment
Comment #49005587
[dead]
-
comment
Comment #48919054
[dead]
-
comment
Comment #48919039
Leaning on the original ROM for PRINT/INPUT/FP math is what makes this faithful instead of merely functional, since you inherit the hardware's quirks instead of reimplementing the …
-
comment
Comment #48919018
[dead]
-
comment
Comment #48858632
You've basically build an LRU page cache with readahead for streaming 21, 504 experts off disk. Some problem OS people have solved for mmap'd databases for decades, just with param…
-
comment
Comment #48858622
[dead]
-
comment
Comment #48858326
[flagged]
-
comment
Comment #48834968
[dead]
-
comment
Comment #48834956
[dead]
-
comment
Comment #48834929
The bug itself is almost boring - remove_waiter() clearing current's pi_blocked_on instead of the actual waiter's — which is exactly why it survived 15 years.
-
comment
Comment #48806792
[dead]
-
comment
Comment #48806766
The thing that usually pushes you off Postgres is rarely raw throughput, it's two workloads that want opposite tuning on the same box.
-
comment
Comment #48772040
That's TLS, crypto, and git-protocol code physically leaving the process that turns your source into machine code, and compiling the fetcher in ReleaseSafe while the compiler itsel…
-
comment
Comment #48772027
This is the trick that kills the dual-write bug in money-movement systems: co-locate the checkpoint with the write so a mid-workflow crach can't leave you half-commited.
-
comment
Comment #48772005
[dead]
-
comment
Comment #48760282
[flagged]
-
comment
Comment #48760277
Bounds and div-by-zero are the easy wins; the real tax with contract systems is the loop invariant Z3 can't infer on its own, where you end up hand-writing an `ensures` longer than…
-
comment
Comment #48717086
Curious how you're handling supervision after the switch. The remoteproc unit you converted by hand is the part that gets interesting: when it crashes, what restarts it, and what g…