TIL that you cannot play #Bad #Apple on a MacBook Pro with a FALD display. Bad Apple indeed LOL!
(It gets extremely blurry and smeary blue shades around everything. Can’t deal with the quick full white / full black changes)
=> More informations about this toot | View the thread
@gparker I’m seeing a regression in libobjc.dylib on macOS 15 on x86_64 (14 unknown, but worked correclty in 13): objc_msgSend_fpret calls into the stret handler passed to objc_setForwardHandler instead of the regular one. This results in a crash (with custom handlers and Foundation’s handlers as well) when trying to forward something with a long double return on x86_64 for obvious reasons. This seems like a pretty bad regression, so I filed FB16316692, but given how feedbacks often never get looked at, I’d thought I’d ping you.
=> More informations about this toot | View the thread
In case someone else has an #AMD #notebook on #Linux and washed out colors when using it on battery, add the following to your kernel cmdline: amdgpu.abmlevel=0
Apparently this is an AMD antifeature called “Varibright” that is also hard to disable on Windows and they pushed this crap into the 6.9 kernel. It is supposed to make your battery last longer, but it’s bullshit. To the contrary, you’ll have to set your power profile to performance to have usable colors, needing much more power. And even worse, turning the fans on when trying to watch a movie.
Yes, they really did drink paint at AMD. No, I don’t know how much. No, I don’t know the color either. (Though probably grey, given how everything looks with their antifeature.)
=> More informations about this toot | View the thread
@Rairii Do you happen to know if there’s a MinGW port to NT4/PPC (or MIPS)?
=> More informations about this toot | View the thread
Hm, all the bad scrapers seem to connect via IPv4. Facebook is the only scraper connecting via IPv6 and identifying itself properly. Time to disable IPv4?
=> More informations about this toot | View the thread
AI scrapers now seem to employ botnets to scrape websites, effectively DDoSing them with multiple thousand requests per second from residential IPs all over the world. As they know they’ll get banned, they randomize the user agent, using stuff like Firefox on Windows 98. Yeah, sure, the last version of Firefox on Windows 98 wouldn’t even be able to do the TLS handshake.
They can still be identified by doing things no human would ever do. Like for a repository, requests every single commit as an archive. Or requesting blame on a file that nobody would ever be interested in.
In any case, has anyone found an effective way to ban this AI bullshit? I fear I need to employ some JavaScript to return the actual site contents, but that would reduce accessibility.
=> More informations about this toot | View the thread
@solene Hi! My last mail with an update seems to have gotten no attention and the next update after, my mail to ports@ never even showed up. Could you maybe apply attached update? Thanks!
=> More informations about this toot | View the thread
@GrapheneOS Are there plans to support the Pixel 9 Pro Fold?
=> More informations about this toot | View the thread
Got banned without comment from the #amiga-gcc GitHub for filing a ticket asking for the sources for the proprietary libamiga.a it contains in the repo: https://github.com/bebbo/amiga-gcc/issues/389
Well, and of course the author broke his compiler again for the 100ths or so time and now I cannot report a bug about it miscompiling stuff.
I guess it’s time to either fork an old version of it that isn’t complete garbage and actually works or drop Amiga support from #ObjFW altogether. amiga-gcc never worked well anyway, I always had to use -O0 so that it doesn’t miscompile ObjFW in so many places that it crashes before even finishing the constructors.
Note that this will NOT affect #MorphOS: MorphOS is able to not break their compiler every week and ObjFW runs there perfectly, even with optimizations. It’s utilizing the same code as AmigaOS did, so if there is ever a working compiler for AmigaOS again, the code will not have bitrotten as it’s still actively used by MorphOS.
=> More informations about this toot | View the thread
You know you’re too tired when SSH lags and you move the terminal window to the right side of the screen to move it closer to the WiFi access point…
=> More informations about this toot | View the thread
@ariadne 923800a0fcb2cbb90315afaf3f99aba8d139d3018a4594866f148254fcffc7b6
Do you have the the same?
=> More informations about this toot | View the thread
@ariadne Also, they used a debug build. Have you dumped the reference board yet? If that one is not debug, the XC-LS3A6M might be very interesting for porting coreboot ;).
=> More informations about this toot | View the thread
@ariadne Also, I now got definitive proof that the board is made by ASUS. I believe the leaked source paths starting with “/home/luka_lian/xc-fusion/Build/Ls3a60007a/RELEASE_GCC5/LOONGARCH64/AsusPkg” are very clear.
=> More informations about this toot | View the thread
Finally managed to dump the UEFI from the #LoongArch #XC-LS3A6M board. Found two flash chips that didn’t contain UEFI and was confused. Guess where the actual UEFI flash chip was? Yep. Below the CPU cooler!
Also, I now know that the UEFI consists of parts compiled by a Meng Xiangdong for the LoongArch UEFI parts, while the parts specific to the XC-LS3A6M are compiled by a Luka Lian.
I know that because they leaked source paths into the binary, of course.
=> More informations about this toot | View the thread
@ariadne OK, you were wrong about the XC-LS3A6M not being made by ASUS:
1.) There is a presentation of this board by the director of ASUS China.
2.) I just dumped one of the flashs and it only contains 6 bytes. The first of which are cc 28 aa. Which is the MAC address prefix for ASUStek.
Not sure who gave you the info it’s “definitely not ASUS”, but it seems very likely that it is ASUS.
=> More informations about this toot | View the thread
@ariadne Ping. You have the XC-LS3A6M too and dumped the flash, right? I found the pins next to the flash, however, I cannot figure out the pinout. In particular, I don’t find /WE, HOLD and VCC for the flash. How did you do it? Do you have the pinout?
=> More informations about this toot | View the thread
@lanodan Weirdly enough, old images load :/. So it’s only newly uploaded ones. The file exists in the uploads directory in the server just fine. Is it forgetting to add it to the DB or something?
=> More informations about this toot | View the thread
@lanodan It seems https://ap-sandbox.webkeks.org/48990c0fb04cf93c08074223c6ffbc34b8d2b2adf44773dbe89448d2e9c30bd0.png returns HTML and Akkoma is confused rather than serving the image? I guess this is since I updated Akkoma then, as this used to work.
=> More informations about this toot | View the thread
@lanodan I’m having a weird problem where every image I upload to Akkoma is broken and cannot be displayed. There is nothing in the logs at all. Any idea what could be causing it? Attaching a picture that should be a screenshot of this message.
=> More informations about this toot | View the thread
And my #LoongArch board also finally arrived. Time to dump the flash before even turning it on :)
=> More informations about this toot | View the thread
=> This profile with reblog | Go to js@nil.im account This content has been proxied by September (3851b).Proxy Information
text/gemini