Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
nopurpose
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
91.
▲
by
nopurpose
1y ago
Then io_uring AIO mode underperformance is even more curious.
92.
▲
by
nopurpose
1y ago
> Google et al. were at one point pinnacle technologies too, but this was 20 years ago. In 2017 Google literally gave us transformer architecture all current AI boom is based on.
93.
▲
CamoLeak: Critical GitHub Copilot Vulnerability Leaks Private Source Code
(legitsecurity.com)
1 points
by
nopurpose
1y ago
|
0 comments
94.
▲
by
nopurpose
1y ago
When I left my phone (out of battery) on a plane, I went to the flightradar and checked all airports the airplane was visiting after. Then contacted lost&found at each of them individually and eventually got my phone back. It was found
95.
▲
by
nopurpose
1y ago
All Linux ISOs collectors in the world wholeheartedly agree.
96.
▲
by
nopurpose
1y ago
Usually their revenue is from affiliate links. YouTube channels have ad spots, merch sales etc. There are many shills of course, but some are doing their best to give accurate reviews and don't shy away from bashing manufacturers for t
97.
▲
by
nopurpose
1y ago
It always puzzles me how apparently well known flaws never mentioned by product reviewers, even genuinely pro-consumer and generally well respected, like rtings or notebookchecks. You buy product after stellar review, encounter problem, s
98.
▲
by
nopurpose
1y ago
How maps are copyrightable then?
99.
▲
by
nopurpose
1y ago
"Makefile" they have written and copyrighting is very non trivial and there are many man-months of effort. Configuring all sorts of software just with env vars and make it usable is not an easy feat. Have a look at https:/&#
100.
▲
by
nopurpose
1y ago
hash won't change on PG version upgrade, because that would cause a massive reshuffling by pg_upgrade: something PG managed to avoid so far.
101.
▲
by
nopurpose
1y ago
This sounds really interesting. Query planner acquires lightweight lock on every table partition and its indexes. There can be only 16 fast locks, exceed that number and locking becomes significantly more expensive slowing down queries sign
102.
▲
Gamersnexus smuggling documentary was DMCAed because it quoted the President [video]
(youtube.com)
3 points
by
nopurpose
1y ago
|
0 comments
103.
▲
by
nopurpose
1y ago
So much written about relation between objects and data, but not a single mention of Lisp and derivatives?
104.
▲
by
nopurpose
2y ago
I enjoyed your detailed response, it adds value to this discussion, but I feel you missed the point of my comment. I am against blanket statements "reduce allocations to reduce GC pressure", which lead people wrong way: they comp
105.
▲
by
nopurpose
2y ago
What do you mean? Pool content can't be GCed , because there are references to it: pool itself. What people do is what this article suggested, pool.Get/pool.Put, which makes it only grow in size even if load profile changes. App l
106.
▲
by
nopurpose
2y ago
From that perspective it boils down to "do less", which is what any perf guide already includes, allocations is just no different from anything else what app do. My comment is more about "reduce allocations to reduce GC press
107.
▲
by
nopurpose
2y ago
also no one GCs sync.Pool. After a spike in utilization, live with increased memory usage until program restart.
108.
▲
by
nopurpose
2y ago
Every perf guide recommends to minimize allocations to reduce GC times, but if you look at pprof of a Go app, GC mark phase is what takes time, not GC sweep. GC mark always starts with known live roots (goroutine stacks, globals, etc) and t
109.
▲
by
nopurpose
2y ago
> It didn’t correspond to a Google Docs release. The stack trace added very little information. There wasn’t an associated spike in user complaints Where mere mortals can complain about Google product?
110.
▲
by
nopurpose
2y ago
It tried to convince me that it is possible to break out of outer loop in C++ with `break 'label` statement placed in nested loop. No such syntax exists.
111.
▲
by
nopurpose
2y ago
you are comparing what is possible but can be slow with blobless with what is impossible with shallow, which is unfair.
112.
▲
by
nopurpose
2y ago
blobless still better than shallow because at least commit history is preserved.
113.
▲
by
nopurpose
2y ago
`git clone --filter blob:none` FTW
114.
▲
by
nopurpose
2y ago
It reminds me of an effort to improve docker image format and make it move away from being just a tar file. I can't find links anymore, but it was a pretty clever design, which still couldn't beat dumb tar in efficiency.
115.
▲
by
nopurpose
2y ago
With about half of the screen is black, can it also boost FPS by not spending GPU time on pixels in those areas if integrated deep into engine?
116.
▲
by
nopurpose
2y ago
How is it "expansion invasion" if peace terms offered by Russia, which are now public knowledge with recent documents leak, shortly after invastion didn't include any new territories for Russia? Ukraine walked away from that
117.
▲
by
nopurpose
2y ago
similar to how we have language servers for code analysis and syntax highlighting, we need to have AI assist server, so that any IDE can pull instruction what to do with what file
118.
▲
by
nopurpose
2y ago
Which tool can actually help coding and refactoring, not just autocomplete? Copilot plugin for Jetbrains IDE can only suggest source to copy paste or at most replace single snippet I selected. What I'd like to do is to ask "write
119.
▲
by
nopurpose
2y ago
Balmer launched Get the facts smearing campaign. It says all about who Balmer is.
120.
▲
by
nopurpose
2y ago
Speaking of rope. Right this moment GKE clusters can't provision large volumes (~4TiB) because their CSI driver receive OOMKill when formatting volume. Problem was reported back in Apr and still not fixed.
More ›