#asahi-gpu · 2021-04-03 04:29
marcanwill prioritize it#asahi-gpu · 2021-04-03 04:29
marcanyup#asahi-gpu · 2021-04-03 04:29
bloomfor your freedom#asahi-gpu · 2021-04-03 04:29
bloomdefinitely worth an engineering investigation#asahi-gpu · 2021-04-03 04:29
bloomoh, that's a good question#asahi-gpu · 2021-04-03 04:29
marcan:-)#asahi-gpu · 2021-04-03 04:29
marcanwonder if there's a way to destroy it with e.g. firmware misconfiguring power, then we wouldn't even need to take the things apart#asahi-gpu · 2021-04-03 04:28
marcan*registers a company to re-sell Apple M1 laptops with the wifi chip nerfed#asahi-gpu · 2021-04-03 04:27
bloom--ninja'd#asahi-gpu · 2021-04-03 04:27
bloomWell, as long as we pull out the wifi chi#asahi-gpu · 2021-04-03 04:27
marcanexcept for the wifi#asahi-gpu · 2021-04-03 04:27
marcanI guess!#asahi-gpu · 2021-04-03 04:27
bloomWe can apply for RYF for the Mac Mini, then?#asahi-gpu · 2021-04-03 04:27
marcanyup, exactly#asahi-gpu · 2021-04-03 04:27
bloommarcan: Hmm, so, the GNU/Linux side doesn't need to know about firmware blobs, so it's like they're not really there.#asahi-gpu · 2021-04-03 04:27
marcanso I do not expect us to ever have any redistribution issues, unless Apple starts having some hardware that 1TR lacks drivers for but needs firmware#asahi-gpu · 2021-04-03 04:26
marcanand 1TR *is* a full macOS install with all necessary firmware (currently, wifi as far as we know)#asahi-gpu · 2021-04-03 04:26
marcaneither way, we can assume 1TR will be there, deleting 1TR is not a supported use case for us#asahi-gpu · 2021-04-03 04:24
marcanall those pre-loaded: 1 nodes are things iBoot2 loads firmware for us for#asahi-gpu · 2021-04-03 04:24
marcanwhich is under sgx (yes, sgx, hi powervr)#asahi-gpu · 2021-04-03 04:24
marcanthat's under iop-gfx-nub#asahi-gpu · 2021-04-03 04:23
marcan firmware-name: GFX#asahi-gpu · 2021-04-03 04:23
marcanso the trick is to look for pre-loaded: 1 in the ADT#asahi-gpu · 2021-04-03 04:23
marcanoh actually GFX.img4 I believe#asahi-gpu · 2021-04-03 04:21
marcanand yes, RTKit is their RTOS#asahi-gpu · 2021-04-03 04:20
marcanor something like that#asahi-gpu · 2021-04-03 04:20
marcanand then on boot provisioning that should end up in AGX.img4 next to iBoot2#asahi-gpu · 2021-04-03 04:20
bloomsee 2 lines later :p#asahi-gpu · 2021-04-03 04:19
marcanit should get copied into ./Firmware/Preboot/C1EB0F38-C67D-43AB-BA6F-3CFA88FC9E6F/restore/Firmware/agx/armfw_g13g.im4p for installation/recovery purposes#asahi-gpu · 2021-04-03 04:19
bloommarcan: noted.#asahi-gpu · 2021-04-03 04:18
marcanbloom: that firmware should be loaded by iBoot2, before m1n1 runs#asahi-gpu · 2021-04-03 04:10
bloom- the thing at offset 0xe4 of my compute trace is the Encoder ID... probably just used to match up with Instruments, buy it's much too late here to test that now.#asahi-gpu · 2021-04-03 04:09
bloom - unk6 grabs a LOT of IDs (as returned), within the massive interval I think userspace allocates the IDs it actually needs for things like command buffers?#asahi-gpu · 2021-04-03 04:08
bloom - the word at offset 0x28 of the response to ALLOCATE_MEM is the resource ID#asahi-gpu · 2021-04-03 04:07
bloomAnyway, the big take aways are:#asahi-gpu · 2021-04-03 04:05
bloomCuriously Instruments at the low level involes a ton of S-expressions#asahi-gpu · 2021-04-03 04:03
bloommetal-io-surface-access shows WindowServer making use of a "83b&" pixel format for the display (4k if it matters) .. guessing that's compressed tiled something or other#asahi-gpu · 2021-04-03 04:01
bloomLooks like a _lot_ of these magic numbers solely exist to support profiling...#asahi-gpu · 2021-04-03 04:00
bloomthis is getting.. more complicated#asahi-gpu · 2021-04-03 04:00
bloommatched up an ID in the command stream to what instrmuents calls "Encoder ID"#asahi-gpu · 2021-04-03 04:00
bloomAh ah!#asahi-gpu · 2021-04-03 03:30
bloomInteresting columns '[resource] ID' and '[parent] IDs', explains so much of the weirdness.#asahi-gpu · 2021-04-03 03:27
bloomI like Instruments :'o#asahi-gpu · 2021-04-03 03:03
bloomA quick Google search indicates "signposts" are a macOS tracing thing?#asahi-gpu · 2021-04-03 03:02
bloomAlso, I'm not sure what /System/PrivateFrameworks/IOGPU.framework/Versions/A/Resources/IOGPUCommonSignposts.plist is but it might be interesting.#asahi-gpu · 2021-04-03 02:43
bloomgave in and started doing development on the mac#asahi-gpu · 2021-04-02 22:11
opticronlike you said, I've yet to see any that omit it#asahi-gpu · 2021-04-02 22:10
opticronwell, some parts of the spec do seem to mandate that it be present, but like I said, there are caveats elsewhere#asahi-gpu · 2021-04-02 22:08
Yuzuiirc no licensable core is FPless or even FP-optional, and i don't think any of the architectural licensees have gone FPU-free either#asahi-gpu · 2021-04-02 22:08
Yuzui'm surprised the spec doesn't mandate FP to be present