Who called it the USB-IF and not the USB Port Authority?
Remote
Matt Mullins
@mokomull@macaw.social
mastodon 4.7.2139 Followers
147 Following
23 Posts
Joined November 18, 2022
Replying to on nondeterministic.computer
@mjg59@nondeterministic.computer 10 years ago 's LEDs last approximately forever, but for batches since 2020-ish ... we're on our second replacements already.
Red Solo CUPS: can not print cyan, yellow, or black.
Replying to
@regehr definitely doable with some cgroup management; setting memory.oom.group to 1 means the OOM granularity would be one entire command execution and any children it might've forked.
or even by polling the memory stats and taking action faster than the kernel's oom-killer, like oomd does.
Replying to
@regehr @whitequark @dotstdy @tedmielczarek now this description is starting to feel like using CRIU to checkpoint a gcc process to a file in userspace
Replying to on tiny.tilde.website
@astrid@tiny.tilde.website @MLE_online@social.afront.org @selectric@tacobelllabs.net the trick is to drink enough that I don't mind the pain of the hills, but not so much as to render me unable to balance :)
unexpected side benefit of testing stuff on a v6-only network: stuff can't call home because the vendor is v4-only
(no, I don't know why it's ECONNREFUSED and not ENETUNREACH but I would not be surprised if podman's networking magic is to blame)
Replying to
@alisynthesis @selectric well, IIRC, it *was* the video's thumbnail for about seventeen whole minutes until someone at Apple realized what they'd done
Replying to on tacobelllabs.net
@selectric@tacobelllabs.net we've just been having the daily "all deployments are halted because Github shat the bed and now our CI is broken"
Replying to
@selectric@tacobelllabs.net well yeah if the circuit is grounded that's why the breaker is tripping
@xan@xantronix.social you're my on-call vertical development enjoyer
Replying to
@whitequark @regehr @dotstdy @tedmielczarek I wrote off portability when working-around-the-oomkiller's-behavior was mentioned ... and when I've had this same frustration has been on shockingly-modern Linux kernels at $last_job.
Replying to on techhub.social
@erikosterholm@techhub.social nah I found that in Osaka last summer :)
@whitequark@social.treehouse.systems thanks a bunch for the probe-rs glasgow applet — saved me a couple hours of work and maybe even buying a new debug probe since probe-rs doesn't think the Black Magic can speak APv2.
@astrid@cloudisland.nz for your Fuck Computers collection
Replying to on hachyderm.io
@petrillic@hachyderm.io Not sure what your constraints actually are, but both UJ20-C-V-C-2-SMT-TR and LUB-C131-0-007-G-19 seem to be 2.0mm (warning on the latter, though, -1- also exists and is 1.2mm)
Replying to on chaos.social
@gsuberland@chaos.social I appreciate how much your writing style matched my thought process looking at this ... my brain was asking "why two separate 118Ω resistors" at precisely the time my eyes caught up to
reduces the power dissipation per resistor a bunch
Who called it a "cherry picker" and not a folklift?
Replying to
@whitequark@mastodon.social this *almost* feels like NaCl ("native client", not the cryptography library) that required explicitly jumping forward so that no instruction would straddle a 32-byte boundary.
Except that isn't even a problem by definition in MIPS instruction encoding...
Replying to
@whitequark@mastodon.social ah I don't yet have experience with RISC-V, but I believe the same argument can be made unless RVC is supported. Not that it *matters*, NaCl died long before RISC-V took off 😁