#asahi · 2021-05-17 10:26
marcancertainly on this kind of machine#asahi · 2021-05-17 10:25
marcanI wouldn't expect boot to be significantly CPU-limited anyway; other than uncompressing xnu (which I don't think takes that long), it's probably largely limited by hardware bring-up delays and such#asahi · 2021-05-17 10:24
marcan(they do have more of them)#asahi · 2021-05-17 10:24
marcanbecause the little core is core 0, and probably because big cores are more likely to need chicken bits to not explode#asahi · 2021-05-17 10:19
marcanprobably upclocked by iBoot to speed up boot#asahi · 2021-05-17 10:08
marcanI'm very much looking forward to not having to deal with DisplayPort training, hopefully :-)#asahi · 2021-05-17 10:08
marcanmight be pretty high level, which would make our DRM/KMS driver for this largely a shim in that case#asahi · 2021-05-17 10:08
marcanI'm very curious as to what the OS level interface to it is#asahi · 2021-05-17 10:08
marcandeals with a lot of the details of display pipe setup and such#asahi · 2021-05-17 10:07
marcandisplay controller processor (guess)#asahi · 2021-05-17 10:06
marcanyay, less things to care about#asahi · 2021-05-17 10:06
marcanyup, DCP definitely talks to PMP#asahi · 2021-05-17 10:06
marcanstrings on DCP suggests there is in fact some doorbell or interface to take care of this exchange, presumably behind the scenes#asahi · 2021-05-17 10:05
marcandisplay is managed by another engine though, so they might just communicate among themselves for that#asahi · 2021-05-17 10:04
marcanshould be interesting to see what the interface is, if this is implicit or if we actually need to issue commands to it to tell it e.g. what the display pipe isochronous bandwidth requirement is#asahi · 2021-05-17 10:04
marcanClock Req.#asahi · 2021-05-17 10:04
marcanRT BW Req.#asahi · 2021-05-17 10:04
marcanDRAM energy greater than threshold energy = %u nj#asahi · 2021-05-17 10:04
marcanDRAM power greater than threshold power = %u mw#asahi · 2021-05-17 10:04
marcanjudging by strings#asahi · 2021-05-17 10:03
marcanseems PMP indeed manages DRAM bandwidth and presumably clocks#asahi · 2021-05-17 10:01
marcanso it's possibly that is implicitly managed or something like that#asahi · 2021-05-17 10:01
marcanhowever, remember there is a power manager sub-cpu that might take care of a lot of this for us#asahi · 2021-05-17 10:01
marcanthey should be independent, but yeah there should be some control for it#asahi · 2021-05-17 09:56
marcananything I can easily work out with mmiotrace, I'd much rather do that way than use the corellium code as a reference#asahi · 2021-05-17 09:56
marcanbgb: if they do I'm happy to use it, if they don't, well, I'm sure I'll have information on them soon from mmiotrace#asahi · 2021-05-17 09:46
marcanbgb: cpufreq management will be done in linux, but in a pinch I'm sure we can add some minimal support to m1n1#asahi · 2021-05-16 20:47
marcanso spam like this is manageable#asahi · 2021-05-16 20:46
marcanall on 32bit boundaries#asahi · 2021-05-16 20:46
marcanor choose to trace only writes, not reads#asahi · 2021-05-16 20:46
marcanthe good thing is the tracing granularity is 32 bits, so you can punch single-register holes you ignore#asahi · 2021-05-16 20:46
marcandisable sync mode for even more throughput :)#asahi · 2021-05-16 20:46
marcanand this is without packet coalescing!#asahi · 2021-05-16 20:45
marcanI have a huge scrollback on my konsole and it still eats through it in seconds#asahi · 2021-05-16 20:45
marcangiven CPU speculation, probably very tight#asahi · 2021-05-16 20:45
marcanit's a tight poll loop#asahi · 2021-05-16 20:45
marcanyes that would spam like that :)#asahi · 2021-05-16 20:45
marcanoh you mean m1n1 on m1n1#asahi · 2021-05-16 20:45
marcan40c is the event reg, isn't it?#asahi · 2021-05-16 20:44
marcanoh, secondary USB device?#asahi · 2021-05-16 20:42
marcanjannau: you're tracing *memory*, not just MMIO? that's... not exactly supported, I don't support anything but the basic load/stores expected to be used for MMIO :)#asahi · 2021-05-16 20:40
marcansvenpeter: that was my idea with numpy, except precomputing a chunk of powers and doing it in blocks#asahi · 2021-05-16 20:39
marcan(if you press it twice, it goes into an alt recovery that doesn't qualify as true 1TR)#asahi · 2021-05-16 20:39
marcanand if kmutil or bputil have been complaining, you might not have been holding down the power button properly :-)#asahi · 2021-05-16 20:38
marcanyou always need to run csrutil second#asahi · 2021-05-16 20:38
marcanamw: if you do csrutil then bputil, the bputil will *override* csrutil.#asahi · 2021-05-16 13:04
marcanjn__: USB has flow control#asahi · 2021-05-16 10:19
marcanthen feel free to dig deeper into those I guess#asahi · 2021-05-16 10:19
marcanjannau: maybe dump the data received inside the inner loop of readfull(); if the dupes show up there I'm inclined to blame pyserial and/or macos#asahi · 2021-05-16 10:14
marcananyway, off to do other stuff; more tomorrow :)