This is a collection of various resources from #archived-resources that don't deserver their own post, or haven't been categorized yet.
Feel free to discuss in this channel. Conversational posts will be removed everywhere else.
#(META)
shut rune
(meta)
shut rune
(META)
hidden siren
absolute gigachad
burnt creek
wouldnt it make more sense to merge these into single post titled
introduction to assembly with additional links about getting into it from #asm-arch-osdev ?
shut rune
wouldnt it make more sense to merge these into single post titled
`introduction ...
tbf there's a lot of merging that could be done here, yeah; I've kept the TCCPPCon videos all separate so far
I'm not really sure which one should win in the merger. The second on is a whole playlist.
burnt creek
tbf there's a lot of merging that could be done here, yeah; I've kept the TCCPPC...
Wdym by win
shut rune
Wdym by win
one of the two would have to be merged into the other, so one of them wins and becomes the "main post". The #1124619767542718524 posts alway have a primary resource which also has its link in the title. I'm not really sure whether these should be merged at all.
burnt creek
one of the two would have to be merged into the other, so one of them wins and b...
wouldnt it be better to have post named introduction to assembly without single "main" resource?
I understand the approach but this will lead to duplication
similarly here, these two could be merged into one GDB post
in general i think post for topics rather than specific resources would work much better
shut rune
similarly here, these two could be merged into one `GDB` post
I've merged this one
burnt creek
shut rune
#1124619767542718524 Style Guide
Title
- if it's a simple resource with only one link, put that link in the title
- if there are multiple links and one is most important, but the dominant link the title
- if there are multiple but no link is dominant, or if the resource is an infographic, omit the link
- use https://tinyurl.com to shorten the link if necessary
Body
- first, the non-shortened link first
- short description of the resource (use blockquotes if it's quoted from the resource)
- "See Also" sections and other useful links
Comment Section
- remove purely conversational comments, e.g. "thanks, great resource"
- if someone asks for clarifications and you resolve it, consider merging the info into the main post, or creating a Q&A section, and removing the question comment
shut rune
## <#1124619767542718524> Style Guide
### Title
- if it's a simple resource wit...
<@&331719468440879105>
hidden siren
hidden siren
shut rune
https://google.github.io/styleguide/cppguide.html
I think that could go into a larger resource that includes clang-format and collects style guides
shut rune
https://build-bench.com/#
red fox
merged into https://discord.com/channels/331718482485837825/1124634753098858577
we should put a huge caveat there that this tool isn't necessarily very reliable and results should not be taken at face value. ime quick-bench is unfortunately pretty much unusable for actual microbenchmarking. you'll often see stuff like if you just change the order in which you run benchmarks, the results are reversed and stuff. i suspect it's because of the vm it runs in. in fact, i was reminded of that just yesterday night over on tph when i got tricked into trying to explain what turned out to be completely bogus results before questioning them despite me knowing better ^^
shut rune
we should put a huge caveat there that this tool isn't necessarily very reliable...
hmm, I've added a note recommending to run these things locally, not online
that may be enough
red fox
the other problem is that quickbench only shows you an average, which greatly diminishes its usefulness, especially for microbenchmarking.
it's nice if you want to show some quick macro benchmarks but that's about it
shut rune
but it's a bit fishy in general sometimes; e.g. results can be massively changed by what gets inlined and what doesn't
red fox
but one needs to be really careful with using that tool
shut rune
you have to look into the assembly
red fox
yeah
red fox
i would maybe turn it into
Note
Creating a meaningful benchmark is hard. Make sure that what you're measuring is actually what you wanted to measure. This will typically require checking the generated assembly to make sure the behavior you were interested in is actually present in the machine code. There's a button that takes you to compiler explorer to look at the benchmark there.
Results from these online tools can be unreliable, likely as a result of running in a VM. Sanity check the results you get, make sure that, e.g., simply changing the order of benchmarks does not lead to completely different outcome. You can set up a local instance for more reliable results (see button on the website). Also, be aware that these sites only present averages, the plots do not show the variance of results across multiple runs. Thus, it may be advisable to run the comparison multiple times (beware of results caching!) to ensure the outcome is reproducible.
hidden siren
red fox
nice 🙂
hidden siren
https://www.youtube.com/watch?v=_sSFtJwgVYQ for #1124650885155405845 maybe
burnt creek
post for computer architecture? could include this: https://youtube.com/playlist?list=PL5Q2soXY2Zi_uej3aY39YB5pfW4SJ7LlN and few more links i may send if the post is created
oak smelt
vestal delta
## <#1124619767542718524> Style Guide
### Title
- if it's a simple resource wit...
tinyurls can be used to send virusues
latent mica
dreamy sable
!ban 996791514195820644 racism
drowsy kernelBOT
User is already banned
sleek ferry
Pls pls, some modern c features would be nice
feral token
similarly here, these two could be merged into one `GDB` post
im not finding the link to this learn gdb video