social.noyu.me

did you know that there are Android phones with 64-bit CPUs and 64-bit kernels but which will only run 32-bit apps? did you know they are actually very common, still manufactured in 2024 by major brands, and the bane of Android console emulator developers everywhere? it sucks.

‪the reason for this is memory consumption. on Android, all apps are forked from the zygote, a process that is always running and has a preloaded set of libraries (libc, graphics drivers, presumably also Android Runtime, etc). but to support 64-bit and 32-bit apps you need two.‬

‪so… if you have a phone with a 64-bit CPU (every modern Android phone) and a 64-bit kernel (probably also every modern Android phone), but less than 4GB RAM… supporting both 32-bit and 64-bit applications is kind of painful memory-wise, so they just… use a 32-bit only userland‬

‪there is no destiny for these devices other than e-waste and it is fucking painful‬

‪the fact that 4GB is considered a small amount of memory is extremely jokerifying also. iPhones have delivered a premium 64-bit experience both with that much memory and with half that. Android is an absolutely wretchedly inefficient platform. skill issue skill issue skill iss—‬

‪the absolute worst thing, that pisses me off more than anything else, is that ofc making an Android phone that supports both 64-bit and 32-bit is an option when building the OS image, or even better, just 64-bit. but Google said they'd start mandating the opposite. i'm serious‬

‪obviously that mandate was not for phones with lots of RAM. they said they'd mandate only supporting 32-bit specifically on brand-new 64-bit low-memory phones. to reduce the number of configurations they had to worry about.‬

‪i do not know if they went ahead with it but SCREAMS‬

‪ARM GAVE YOU THE WORLD'S SMALLEST 64-BIT-CAPABLE CORE TWELVE YEARS AGO AND THIS IS WHAT YOU CHOOSE TO DO WITH IT? THIS????? DOES YOUR LUST FOR INFLICTED SUFFERING KNOW NO BOUNDS, GOOGLE? DO YOU NOT LOOK UPON YOUR CREATIONS AND WEEP FOR WHAT YOU HAVE DONE. WHY DO YOU CURSE US‬

‪if this does blow up then please do not follow me for technology opinions. you'd be enabling my worst tendencies. i am here to post about kanbaru suruga from the monogatari series and also how pretty the sky is‬

@hikari mot even doing anything cool like arm64_32?

@hikari ARM had 64-bit cores in 2012?

@saagar yes. in the ruins of our civilisation, they will know us by the Cortex-A53 chips

@hikari Ah ok so this is the typical ARM-brain at work. Yes it was “available” in 2012 (in the form of IP of course) and then it took at least two years for anyone to actually ship it

@saagar sure, that's the typical lead time. anyway that means we've been making cortex-a53 phones for an entire decade now

@hikari And we shall make them for a decade to come

@saagar at the end of the world, there will be famine, pestilence, and Cortex-A53

@hikari tbh Windows NT 4 delivered a premium 32-bit experience with 64 MB RAM, with noticeably lower interface latency

I will never stop being weirded out by how much RAM modern apps need

@mia @hikari don't forget Rhapsody DR2 also managed exactly that, but with Objects™®©

pretty sure that everything could be streamedlined and slimmed down to fit in, maybe not into 64MB (stuff like video is just too much, even for basic use), but 512MB-1GB should be fine

@xerz @hikari if we're talking non-Windows, take a look at QNX that fit an entire real time OS with a network stack, a GUI, and a browser all on a single 3.5" diskette.

@mia @hikari and ofc there’s MenuetOS too

@mia @hikari I played with Windows 7 running on a 533MHz Via C3 with 512 MB RAM (of which 8 MB was taken by integrated graphics), and despite running with default VESA graphic driver (meaning no graphic acceleration at all), the system was quite snappy. Opening Start Menu was so much faster than opening Windows 10 or 11 Start Menu on today's computers.

@jernej__s @hikari the last low UI latency Windows was, I think, 2003 Server. 7's Start menu is pretty much comparable to 11 23H2's; the primary difference is that 11 animates the slide-out — and this animation (among many others) is skipped with so-called basic display adaptor drivers. You can also disable them via troubleshooting options, I think.

@jernej__s @hikari and personally I would not run 7 on 512 MB; that is barely enough for the OS itself and I say this as someone who had to use 7 with exactly that amount of RAM on Intel Celeron D machines with iGPUs at work back in 2007-2009. Adding Firefox to the mix made the system swap. If you have 512 MB of RAM you want Server 2003 or maybe XP.

@mia @hikari I was mostly just trying to see if I can get Aero to work on such a low-end machine, not because I'd use this for anything serious (turned out that Nvidia WDDM drivers for GeForce 5200 PCI do not work on C3 and cause a blue screen due to illegal instruction; C3 is a Pentium-class CPU).

I did do something even stupider after I found 2x512 MB SDRAM to put in the machine:

@mia @hikari AFAIK both Win10 and 11 Start Menus are implemented as a separate application that's started every time you open the menu (and 11's is worse, because it seems to use some sort of web framework).

About a month ago Start Menu on client's RDS server (running 2016) broke somehow, and nothing I tried would fix is, and since the server's going away by the end of the year, I didn't want to rebuild it, so I installed StartIsBack as a workaround. The first thing I noticed was how much faster it opens.

Note that I disable most UI animations on my systems, since they might be cute the first few times, but then they just take time without adding any benefit, so the above comparisons were made with settings where Start menu pops up without any animation.

@jernej__s @hikari then I really can't say what is up, because in my 11 23H2 install with web search disabled via group policy the start menu appears immediately (and start menu search is practically instant, faster than on 7)

It did have noticeable latency in 11 20H2, that I admit

@hikari such is the case for android phones

@hikari if i understand it correctly, apple also plays a similar game on the watches. They do run aarch64, but with 32 bit pointers. Though this is most likely irrelevant for emulator devs. :P

@Kahanis i have much more respect for that choice! 64-bit with 32-bit pointers still lets you take full advantage of the hardware

@hikari agree, but still annoying if you decided to use swift for some reason and care about your app size (lol who does that in $CURRENT_YEAR). Can't share libraries with the main app, since that's a different arch.

@hikari @Kahanis With what, userspace PAE?

@saagar @Kahanis @hikari are these 32-bit-only Android devices still using the armv7 instruction set? arm64_32 at least lets you still use all thirty-one 64-bit GPRs

@joe @Kahanis @hikari PC is a GPR if you try hard enough

@saagar @Kahanis @joe @hikari that’s right

@saagar @joe @Kahanis @hikari PC is not addressable on arm64. It can only be indirectly modified, unlike arm32 that can use it as the source or destination of most instructions.

@shac @joe @Kahanis @hikari Guess 64-bit doesn’t really let you take full advantage of the hardware after all

@saagar @Kahanis @shac @hikari but the biggest crime aarch64 committed was taking away conditional everything

@joe @Kahanis @shac @hikari What about conditional conditions

@saagar @Kahanis @shac @hikari what if you use a thumb IT instruction over conditionalized ARM instructions

@joe @saagar @Kahanis @hikari An architecture where any instruction can be a conditional branch is objectively insane. But also the most fun assembly to write.

@shac @joe @Kahanis @hikari That’s every architecture that supports breakpoints

@shac @Kahanis @hikari @joe The extra-fun architectures are where you can branch halfway through an instruction

@saagar @Kahanis @joe @shac @hikari tryna watchpoint a CPY?

@siguza @Kahanis @joe @shac @hikari This is actually one of the things I think ARM is completely out of their depth on, x86 actually thinks about what happens in situations like these and for a lot of traps ARM goes “why don’t you figure out how to reconstruct state ¯\_(ツ)_/¯”

@saagar @Kahanis @joe @shac @hikari which is actually a more sane thing to present to software, just abysmal to figure out how to do... while x86 looks at all the options, picks the worst one, and then everyone is stuck with that forever.

@saagar huh, does ARMv8 (or later) have non-restartable instructions? Is this like vector stuff or something? I thought it was pretty clean in this regard in that you can always just ERET back after resolving the fault and the program will keep going.

Or, hmm, maybe things like non-FPAC IABORTs are non-restartable since PC is trashed and you can't (confidently) know if it's PAC or just a bogus address? Though if you're trying to recover from a PAC fail you kinda deserve it.

@ezhes_ Not sure, this was mostly a watchpoint hate session (which you may have seen a previous iteration of: https://federated.saagarjha.com/notice/ATdDMNXtw4QuTBZ1jU)

@saagar in theory, doesn't FEAT_Debugv8p2 solve this whole mess? ESR.ISS.WPT reports the matched watchpoint number for a trap and is mandatory. Though, they kinda ruin it with WPTV which can make WPT invalid, but that goes away in 8p9 where WPT cannot be invalid.

https://arm.jonpalmisc.com/2023_09_sysreg/AArch64-esr_el2#fieldset_0-24_0_24-23_18-1

@ezhes_ > Watchpoint might be false-positive.

lol

@saagar @Kahanis @joe @shac @hikari i can't find it now but i saw someone post a month or two ago a snippet that used the fact an instruction would mutate register state before faulting to tell if a page wasn't resident in memory and the kernel was reattempting it after a fault 🙂

@sickeningsprawl @saagar @Kahanis @joe @shac @hikari illegalize this

@sickeningsprawl @Kahanis @joe @shac @hikari https://infosec.exchange/@jann/112983458910799814

@saagar @Kahanis @joe @shac @hikari yes thank you 🙏

@shac @saagar @Kahanis @hikari it would've been nice to keep conditional calls in arm64 at least, there are some nice things you can do with them

@hikari@social.noyu.me

Apple didn't want to play this game and gave 32-bit a sudden, swift, merciful death.

I'd rather call it an execution or orbital strike, when half your apps stop working overnight.
Imagine the amount of support tickets that came out of iOS 11, especially once downgrade became impossible.

@hikari@social.noyu.me also, my last phone (redmi 6) had a lineageOS build that did exactly that. 64-bit cpu, (probably) 64-bit kernel, 32-bit programs.
I wouldn't call it e-waste for that. It is though for the severely damaged screen and questionable MMC health though.

@iska believe me i know (see https://hikari.noyu.me/blog/2023-02-06-touchhle-anouncement-thread-tech-games-me-and-passion-projects.html)

there must be some happy medium

@hikari every single Raspberry Pi question on Stack Overflow... "binary says no such file or directory but uname -m says arm64" aaaaaaargh!!