Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
crest
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
27 ms
·
391.
▲
by
crest
2y ago
>95% of the Internet Archive is a useful service to society even if it pisses of a few companies. It should be legal even if they sometimes act as "holy fools".
392.
▲
by
crest
2y ago
Bhyve (FreeBSD's native hypervisor) already contains a 9P over VirtIO "server". The protocols allows virtual machines and their host to share a directory instead of block storage without having to configure and use a network
393.
▲
by
crest
2y ago
For a read-only filesystem, as a shared directory between guest and host OS for a specific guest (and host) users, etc. It's a significant, potentially dangerous limitation, but it's still worth having in its current form. Touchin
394.
▲
by
crest
2y ago
Sure, but you get to enjoy your bounty payout. Having $2M legally vs. having to become a money launderer?
395.
▲
by
crest
2y ago
I've seen a chain of satellites flare with the naked eye once, but as I'm not much into astronomy it was just a curiosity I looked up after the fact. I suspect astronomers don't care for Starlink "photo bombing" the
396.
▲
by
crest
2y ago
OpenBSD was stagnant when it came to hardening for a while, but the last few releases really added new defense in depth technologies e.g. whitelisting all syscall entry points at link time -> if your process doesn't contain fork() a
397.
▲
by
crest
2y ago
Do you have a source for prior art you can share?
398.
▲
by
crest
2y ago
Can you name a single RISC-V core that has been taped-out with enough single-threaded throughput to enable an enjoyable desktop experience? A 1.5GHz D1 core is a toy and architectural dead end.
399.
▲
by
crest
2y ago
The few cents for larger flash or the one-time cost to write less bloated firmware would cut into their budget for gambling, drugs and hookers.
400.
▲
by
crest
2y ago
Three resistors and a parallel port?
401.
▲
by
crest
2y ago
Or Intel's repeated failures to roll out a working <14nm processes forced them to cripple even the P-core AVX-512 implementation (one less ALU port compared to Xeon Gold), rip out whatever compatibility feature they wanted to put in
402.
▲
by
crest
2y ago
It's also going to be fun when it comes to communication of medical providers or lawyers. Both groups can be expected to be legally privileged and commonly come into contact with material they're legally required to process that w
403.
▲
by
crest
2y ago
I also expect my everyday tools to be responsive e.g. if a "desktop" application lags while typing I'm uninstalling that shit (if there is an alternative sigh ).
404.
▲
by
crest
2y ago
If only the the PPC 970 was able to sustain an IPC of eight... it would've been a true miracle to run at 2GHz in 130nm (later 90nm). An IPC of ≥8 for more than a few cycles is something you'll only see during unrealistic micro-ben
405.
▲
by
crest
2y ago
It better be otherwise it would be a failed design (attempt).
406.
▲
by
crest
2y ago
The teaser reminds me a lot of other *failed* high performance/high efficiency architecture redesigns that failed because of the unreasonable effort required to squeeze out a useful fraction of the promised gains e.g. Transputer and Ce
407.
▲
by
crest
2y ago
Unless you can break the SHA1 hashes for the torrent chunks and the global hash torrents also authenticate the content. If you have a valid "info" dictionary at least.
408.
▲
by
crest
2y ago
I would love my package manager to support a simple fetch helper protocol to plug-in distributed caches from torrents (with webseeds), to mDNS, heck just a Redis service tracking which systems should have which packages cached per site, to
409.
▲
by
crest
2y ago
Start putting ever harsher rate limits on their IP ranges in place until AWS has an actual human reaches out bypassing their so-called "support" channels by making their problem?
410.
▲
by
crest
2y ago
The advantage to 2.5Gb/s is that you can run it over reasonable lengths of existing crappy cable jobs about as well as 1Gb/s Ethernet without angry customers yelling blaming you because it doesn't work. sigh
411.
▲
by
crest
2y ago
Maybe they should keep interviewing for jobs that require nothing more complex, important, or dangerous than that.
412.
▲
by
crest
2y ago
Probably biased memory and better weather radar allowing planes to avoid turbulence more often.
413.
▲
by
crest
2y ago
This LWN article could be a good starting point: https://lwn.net/Articles/959562/ The best source for the history of the work are probably the OpenBSD mailing list archives and commit logs, but digging through tha
414.
▲
by
crest
2y ago
Yes and no. At least one process needs to run as root to be able to become any (other) user. It doesn't have to be the one accepting incoming connections, or the one handling user authentication and authorisation. OpenBSD already conta
415.
▲
by
crest
2y ago
OpenBSD recently added a bunch of new security mitigations most hardening processes against remote code execution. As of the latest release (OpenBSD 7.4) the runtime linker registers all system call entry points per memory mapping with the
416.
▲
by
crest
2y ago
Maybe it would be better to say it was a common enough interface to hook up devices that were essentially a bunch of potis and not much more?
417.
▲
by
crest
2y ago
Given that PowerPC supports indexed load/store instructions with reversed endianness it's doesn't cost extra cycles or code size to access little-endian peripherals. The real problem is how fragile drivers are these days.
418.
▲
by
crest
2y ago
The "attack" is just a clever use of DHCP option 121. It's a valid attack against a badly broken client configuration. Just replacing the default route (or overwriting it with two /1 routes) and adding a host route to th
419.
▲
by
crest
2y ago
Afaik those unified memory architectures are mostly neither cache coherent nor do they support virtual addresses efficiently (you have to trap into privileged code to pin/unpin the mappings) which means that the relative cost is lower
420.
▲
by
crest
2y ago
Okay further down it explains that it always goes through a fresh pseudo-tty (at least for interactive commands?). That solves the file descriptor passing problem but not the reliable signal handling for job control since there're sign
More ›