ChanServ changed the topic of #asahi-dev to: Asahi Linux: porting Linux to Apple Silicon macs | Non-development talk: #asahi | General development | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Logs: https://alx.sh/l/asahi-dev
jeisom_ has quit [Quit: Leaving]
jeisom has joined #asahi-dev
completenoob has joined #asahi-dev
completenoob has quit []
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-dev
completenoob has joined #asahi-dev
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
gabuscus has quit []
hightower3 has joined #asahi-dev
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-dev
hightower4 has quit [Ping timeout: 480 seconds]
crabbedhaloablut has joined #asahi-dev
iaguis_ has joined #asahi-dev
jeisom has quit [Ping timeout: 480 seconds]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
iaguis__ has joined #asahi-dev
iaguis has quit [Ping timeout: 480 seconds]
gabuscus has joined #asahi-dev
iaguis_ has quit [Ping timeout: 480 seconds]
tristan2_ has joined #asahi-dev
tristan2 has quit [Ping timeout: 480 seconds]
<marcan>
eiln: does it need to be bottom up?
<marcan>
you can tell the iommu stuff how many bits of address space you have, then it'll restrict itself to that range
<marcan>
presumably it doesn't matter where in that AS you allocate, right?
<marcan>
and yeah, for ane you're allowed to use the drm_mm stuff if necessary since that's in accel/ which is morally drm, so if it makes sense there I wouldn't worry about it, but for ISP it'd be nice (and probably simplify things quite a bit) to switch back to the regular iova allocator
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
JTL has quit []
JTL has joined #asahi-dev
completenoob has joined #asahi-dev
JTL has quit []
JTL has joined #asahi-dev
JTL has quit []
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
completenoob has joined #asahi-dev
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
pg12 has quit [Remote host closed the connection]
ellyq has joined #asahi-dev
pg12 has joined #asahi-dev
completenoob has joined #asahi-dev
iaguis__ is now known as iaguis
<sven>
dma_set_mask_and_coherent is the thing he's talking about fwiw
chadmed has joined #asahi-dev
<_jannau__>
we added dart dts properties to limit the aperture as well. basicly the same as vm-base, vm-size in the adt
<_jannau__>
not sure if we already use it somewhere
chadmed has quit [Quit: Konversation terminated!]
chadmed has joined #asahi-dev
<marcan>
IIRC I made DCP use that (which fixed a bunch of weird)
<marcan>
and MTP too I think
ellyq_ has joined #asahi-dev
<marcan>
sven: turned out that function is useless because by the time the driver probe() is called, the iommu aperture is already initialized
<marcan>
I spent way too long figuring this mess out
ellyq has quit [Read error: Connection reset by peer]
<marcan>
that's why we now have apple,dma-range = <0x100 0x0 0x10 0x0>;
<marcan>
and there's logic in the DART code that allows address space truncation (which is what the hardware does) as long as the range is smaller than the DART address space and does not straddle natural boundaries
<marcan>
we need to get rid of the remaining users of apple,asc-dram-mask, I think only t8103 dcp has anything nonzero there and should be changed to dma-range (and then this removes a bunch of code from dcp)
<jannau>
I think we have also a zero apple,asc-dram-mask in the t600x dtsi
givtrah has quit [Quit: leaving]
givtrah has joined #asahi-dev
<sven>
huh, dma set coherent should work independently of the iommu subsystem I thought
<sven>
iirc dma_set_mask_and_coherent is meant to express "this device can handle n bits of dma address" independently of whatever the iommu does
<sven>
but then again, dma-ranges is ignored when an iommu is present
<jannau>
it does, I think the issue was to limit dart and iopgtbl to 36 bits / 3 levels
<sven>
that makes more sense
<sven>
if that 36bit thing indeed is an iommu limitation and not just a limitation of the dma device i guess
<sven>
yeah, from what i can tell dma-iommu.c respects the dma mask set by dma_set_mask_and_coherent
<sven>
so dma_set_mask_and_coherent is for device limits (e.g. this device has only 36 address bits for dma) while the iommu domain aperture is for iommu limits (e.g. only 32bit of the device address lines are actually connected to the iommu)
ourdumbfuture has joined #asahi-dev
ourdumbfuture has quit []
<sven>
lol, my cd3217/8 "new from the factory" chips arrived from china and they are obviously reballed :D
<sven>
so i guess "new from the factory" actually means "stripped from some broken apple mainboard"
mps has quit [Quit: leaving]
ourdumbfuture has joined #asahi-dev
lynndotpy has quit [Quit: bye bye]
lynndotpy has joined #asahi-dev
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-dev
jeisom has joined #asahi-dev
dylanchapell has quit [Remote host closed the connection]
cylm has joined #asahi-dev
<marcan>
ah yeah maybe that was it, I was trying to use the dma_mask to limit iommu page table levels and failing miserably
<marcan>
sven: also dma-ranges is NOT ignored
<marcan>
there's a stupid 32bit limit that creeps in and was clamping everything to 4GiB, that I remember
<marcan>
without dma-ranges
<marcan>
we had to add this ridiculous hack: dma-ranges = <0 0 0 0 0xffffffff 0xffffc000>;
<marcan>
because end=0xffffffffffffffff will actually overflow into 0 in the code
<marcan>
to get it to (mostly) allow 64bit dma
<marcan>
somewhere in the OF/whatever code was defaulting to 32bit range if dma-ranges is not present
<sven>
hrm, maybe I remember it wrong then
<sven>
i thought with an iommu enabled the iommu would fully decide virt/phys mapping instead of dma-ranges which is used without an iommu
<sven>
or maybe that was just a bug that was fixed at some point
<marcan>
sven: see of_dma_configure_id
<marcan>
that is the broken code that runs before the probe() gets called, which means the DMA mask is unset, then defaults to 32bit
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
mps has joined #asahi-dev
dylanchapell has joined #asahi-dev
commandoline has quit [Quit: Bye!]
commandoline has joined #asahi-dev
commandoline has quit [Quit: Bye!]
ourdumbfuture has joined #asahi-dev
commandoline has joined #asahi-dev
os has quit [Ping timeout: 480 seconds]
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
commandoline has quit []
commandoline has joined #asahi-dev
completenoob has joined #asahi-dev
completenoob has quit []
gladiac has joined #asahi-dev
gladiac has quit [Quit: k thx bye]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
axboe has quit [Remote host closed the connection]
axboe has joined #asahi-dev
<kevans91>
noting at a different time of this day, and probably the last time: I have three of the v3.1 central scrutinizer PCBs unpopulated showing up what looks like tomorrow from OSHPark if anyone's interested in them (for just the cost of shipping to you)
compassion1785 has quit [Quit: lounge quit]
cylm has quit [Ping timeout: 480 seconds]
cylm has joined #asahi-dev
compassion1785 has joined #asahi-dev
completenoob has joined #asahi-dev
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<maz>
kevans91: don't throw them away -- if someone asks, we can always direct them to you.
completenoob has joined #asahi-dev
ourdumbfuture has joined #asahi-dev
<kevans91>
maz: oh yeah, sorry, didnt mean to give that impression; I'll store them somewhere safe just in case a need arises, I just won't keep broadcasting it here. that works for me, thanks
<maz>
kevans91: did you get a stencil as well, or just the boards?
<sven>
i wouldn't want to solder some of those chips without a stencil :D
<maz>
sven: I know a few mad people... :D
<sven>
:D
<kevans91>
I hadn't ordered a stencil for that batch, but I'm not averse to ordering one to throw in and sweeten the pot a bit
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<maz>
I've been secretly thinking of buying a hot-plate to rework the handful of v3 that were butchered by jlcpcb... total nonsense from a cost perspective, but still...
rappet has joined #asahi-dev
cylm has quit [Ping timeout: 480 seconds]
<sven>
how did they mess them up? unaligned ics?
<maz>
yup. the switches (which are absolutely tiny) are either off by about 1mm, or have some bad bridges. the only change I made in v3.1 was to pick hummer-sized ICs, and it's been pretty solid since.
elvishjerricco has quit [Remote host closed the connection]
elvishjerricco has joined #asahi-dev
kettenis_ has joined #asahi-dev
kettenis has quit [Ping timeout: 480 seconds]
kettenis has joined #asahi-dev
kettenis_ has quit [Ping timeout: 480 seconds]
roxfan has joined #asahi-dev
delroth has quit [Remote host closed the connection]
delroth has joined #asahi-dev
completenoob has joined #asahi-dev
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
chadmed has quit [Ping timeout: 480 seconds]
roxfan has quit [Read error: Connection reset by peer]
maximbaz has quit [Quit: bye]
maximbaz has joined #asahi-dev
eiln has quit [Ping timeout: 480 seconds]
crabbedhaloablut has quit []
completenoob has joined #asahi-dev
completenoob has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
nela has quit [Quit: bye!]
nela has joined #asahi-dev
yayu has quit [Read error: Connection reset by peer]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
ourdumbfuture has joined #asahi-dev
Cyrinux9474 has quit []
Cyrinux9474 has joined #asahi-dev
ellyq has quit [Ping timeout: 480 seconds]
ourdumbfuture has quit [Quit: My Mac has gone to sleep. ZZZzzz…]