#asahi-dev · 2021-01-23 15:30
modwizcodeI wonder if anyone else has made that besides these chips (TSO that is)#asahi-dev · 2021-01-23 15:30
modwizcodeI guess I do have the pi#asahi-dev · 2021-01-23 15:30
modwizcodeIf I had other hardware I wouldn't use qemu c;#asahi-dev · 2021-01-23 15:29
modwizcodeIt would be fun to have a model in qemu anyway even if it would be hilariously slow#asahi-dev · 2021-01-23 14:59
modwizcodeOh cool you're working on that too sven?#asahi-dev · 2021-01-22 14:42
modwizcodei was having trouble understanding from twitter why we need this specific cable#asahi-dev · 2021-01-21 19:12
modwizcodeyeah#asahi-dev · 2021-01-21 18:39
modwizcodehmm 7.3#asahi-dev · 2021-01-21 18:38
modwizcodenope neither copy I have does#asahi-dev · 2021-01-21 18:37
modwizcodein my case the issue is my gcc doesn't seem to accept -static-pie I think I will try again though#asahi-dev · 2021-01-21 18:36
modwizcodeRight but I think that's because the linker was modified to handle -static -pie different#asahi-dev · 2021-01-21 18:34
modwizcodeYeah that's fine. Discarding the .interp section I think does the thing that's needed.#asahi-dev · 2021-01-21 18:34
modwizcodedoes -static -pie work correctly though? I feel like my copy of gcc doesn't do what we want that way.#asahi-dev · 2021-01-21 18:29
modwizcodeyrlf: keeping -pie and removing -shared seems wrong from my research into how things are supposed to work. I think you actually know more than me here so I'm mostly trying to learn/document what's going on and how the behaviors differ. So far the issue is that using different option doesn't lead to different results that I see#asahi-dev · 2021-01-21 15:05
modwizcodeI think the option change for clang will need to be solved a different way though#asahi-dev · 2021-01-21 15:04
modwizcodeYeah I think the reason I was confused on relocations is I wasn't dealing with any addresses on global scope and I didn't really understand how it worked which I've since resolved.#asahi-dev · 2021-01-20 23:40
modwizcodeI've been playing with it too and I was having things go very wrong but it looks like you're right -static-pie or the equivilient functionality is what we want here, the current flags are basically what (at least one definition of) -static-pie does#asahi-dev · 2021-01-20 22:46
modwizcode"our innovative technology features edge level interrupts for greater performance" ;)#asahi-dev · 2021-01-20 22:45
modwizcodeThat's interesting#asahi-dev · 2021-01-20 22:45
modwizcodeah#asahi-dev · 2021-01-20 22:39
modwizcodeAh that's what I was getting at, maybe I'll notice it if I rewatch from beginning at some point. The bit I watched was like "oh that's it" and then was like "okay I'm done" maybe next time he'll explain XD#asahi-dev · 2021-01-20 22:38
modwizcodeI thought the interrupt controller was level triggered?#asahi-dev · 2021-01-20 22:37
modwizcoderight but I was trying to figure out the specific#asahi-dev · 2021-01-20 22:37
modwizcodeWhat was the issue in the end?#asahi-dev · 2021-01-20 22:00
modwizcodefor -static-pie what does that do?#asahi-dev · 2021-01-20 21:59
modwizcodeAnd the compiler flags too#asahi-dev · 2021-01-20 21:59
modwizcodeYeah#asahi-dev · 2021-01-20 21:59
modwizcodeNo I think this is relevant I was asking because I think the flags being used for m1n1 should probably be paired down (apparently the existing ones break clang) and the fix for that didn't seem... quite right.#asahi-dev · 2021-01-20 21:54
modwizcodeah of course#asahi-dev · 2021-01-20 21:52
modwizcodeRight that part made sense to me#asahi-dev · 2021-01-20 21:51
modwizcodeAhhh I didn't actually know that#asahi-dev · 2021-01-20 21:51
modwizcodedamn it made sense too#asahi-dev · 2021-01-20 21:50
modwizcodeUhh it was a random blog post: https://eklitzke.org/position-independent-executables it was literally like the one thing I could find that gave a specific difference. I was going to follow up by checking my own code then the compiler source but my build system was broken.#asahi-dev · 2021-01-20 21:34
modwizcode"-Bsymbolic" I really want to know if this flag is there for a reason because looking it up suggests it does not#asahi-dev · 2021-01-20 21:33
modwizcodebut don't quote me on that#asahi-dev · 2021-01-20 21:33
modwizcode--no-apply-dynamic-relocs I think this might do it#asahi-dev · 2021-01-20 21:33
modwizcodeI think there's a flag on to do it anyway#asahi-dev · 2021-01-20 21:33
modwizcodewtf is a "NONE" relocation#asahi-dev · 2021-01-20 21:33
modwizcodeI don't think m1n1 should be big enough (I think it's not the overall size but internal to function size or something) to matter for that#asahi-dev · 2021-01-20 21:32
modwizcodeBut "correctness" suggest it's not#asahi-dev · 2021-01-20 21:32
modwizcodeAT least I don't think it's what we want it might not matter#asahi-dev · 2021-01-20 21:32
modwizcodewhich is not what we want#asahi-dev · 2021-01-20 21:32
modwizcodeAfaik (according to one place, next route is checking the gcc/ld source) the difference is the PIE has a fixed entry point#asahi-dev · 2021-01-20 21:31
modwizcodeAfter a bit of research I think what would probably be desired is -fpic when doing codegen and -shared when linking.#asahi-dev · 2021-01-20 19:39
modwizcodeI assume maybe using larger offsets but I don't see why I have to tell it to do that???#asahi-dev · 2021-01-20 19:39
modwizcodeI'm pedantic about these things and I have not figured out what the big flags are. the docs say large mode but wtf does that mean#asahi-dev · 2021-01-20 19:39
modwizcodeyep#asahi-dev · 2021-01-20 19:35
modwizcodeWhich doesn't tell you how that differs from -fpic, nor is the difference between -fPIE and -pie explained (I think I'm remembering the flag names right)#asahi-dev · 2021-01-20 19:34
modwizcodeProbably also because the most documentation you get is other projects with a bunch of those flags/the individual flag documentation which is very unclear about what things actually do. Like -pie is defined as something like "enables position independent executable" or something.#asahi-dev · 2021-01-20 19:33
modwizcodeYeah