#
✕ nick:jannau
#asahi-dev · 2023-02-23 19:30
jannaupovik: do those regions have a "no-map" property?
#asahi-dev · 2023-02-22 19:35
jannauI did, but pasted the filename instad of the link: https://www.jannau.net/asahi/tbt_samsung_970_evo_port0_dump.txt
#asahi-dev · 2023-02-22 18:01
jannaulet me format/mount the ssd
#asahi-dev · 2023-02-22 18:00
jannaua little surprised that APPCLK_CGDIS is set in the macos dump
#asahi-dev · 2023-02-22 17:59
jannaudoesn't look like there's a significant difference if we ignore the first 36 bytes: tbt_samsung_970_evo_port0_dump.txt
#asahi-dev · 2023-02-22 16:45
jannauBAR is readable in the HV with macos booted, mem address is identical
#asahi-dev · 2023-02-22 15:34
jannaunot sure how it works with the extended PCIe config space, dumped via lspci -xxxx
#asahi-dev · 2023-02-22 15:33
jannauit seems to parse lspci -x, shouldn't be hard to replicate
#asahi-dev · 2023-02-22 14:04
jannauwith the x2 enclosure it is 1.5 vs. 1.2 GB/s
#asahi-dev · 2023-02-22 14:03
jannauwith the x4 nvme it's 2.8 GB/s vs. 1.9 GB/s
#asahi-dev · 2023-02-22 14:01
jannauthe thunderbolt nvmes are considerably slower on macos. more than I would attribute to macos is bad compared to linux
#asahi-dev · 2023-02-22 13:59
jannaudoes someone know if `dd bs=8M if=/dev/rdiskN of=/dev/null` is a valid sequential read spped test on macos?
#asahi-dev · 2023-02-22 13:33
jannauwith the device enabled I read all 0xffffffff in the HV
#asahi-dev · 2023-02-22 13:07
jannauI'm still confused by the serror in the HV when reading the BAR, the nvme should be powered
#asahi-dev · 2023-02-22 12:25
jannau23.5gbps netto. samsung nvme doesn't work with titan ridge either. so running out of easy tests
#asahi-dev · 2023-02-22 12:07
jannausven: "Timing buffered disk reads: 8824 MB in 3.00 seconds = 2941.07 MB/sec" with a Titan Ridge based nvme enclosure with a PCIev3 x4 link
#asahi-dev · 2023-02-22 08:45
jannaureproducible with the samsung 970 evo (sans Plus, afaik just different nand) from my thinkpad
#asahi-dev · 2023-02-22 08:43
jannausven: it's 100% with the samsung nvme both connected directly and through the dock.
#asahi-dev · 2023-02-21 16:30
jannauthe samsung ssd issue looks like a device issue. pci resource allocation is identical with the working nvme
#asahi-dev · 2023-02-21 16:25
jannauand apple wasn't even in the search term
#asahi-dev · 2023-02-21 16:23
jannauwhat one finds when searching for thunderbolt/pcie issue: https://jobs.apple.com/en-gb/details/200452035/software-engineer-linux-driver-developer-thunderbolt-usb-pcie-displays
#asahi-dev · 2023-02-20 21:44
jannaunot even sure if that is correct, high packet loss, out of order and duplicate replies as well
#asahi-dev · 2023-02-20 21:42
jannauthis kind of broken: "408 bytes from 192.169.1.10: icmp_seq=183 ttl=64 time=541905 ms"
#asahi-dev · 2023-02-20 20:51
jannauthe pcie ethernet adapter is mostly busted. I shouldn't have assumed that it works just because config via dhcp works
#asahi-dev · 2023-02-20 20:15
jannaubtw, I see for for the root port and the thunderbolt bridges in the dock "BAR 13: no space for [io size 0x1000/0x3000]
#asahi-dev · 2023-02-20 20:10
jannauyes, it fails to activate the tunnel. I wasn't 100% certain that tunnel terminates at dwc3
#asahi-dev · 2023-02-20 20:04
jannauusb3 ports on my thunderbolt3 (titan ridge) work. do you understand how usb3 tunneling is supposed to work?
#asahi-dev · 2023-02-20 17:02
jannauthe thunderbolt controller is the same though
#asahi-dev · 2023-02-20 16:58
jannaumain selling point was the integrated 2.5gbps nic
#asahi-dev · 2023-02-20 16:54
jannauno, I have dock with 1 thunderbolt downstream port: hp thunderbolt dock g4
#asahi-dev · 2023-02-20 16:01
jannausigh, this seems to be ok as the enclosure is only a 20gbps / x2 thunderbolt 3 device
#asahi-dev · 2023-02-20 15:52
jannausysfs reports that the nvme runs with a PCIe v3 x2 link. so 1.5 GB/s seems to be ok/expected
#asahi-dev · 2023-02-20 15:40
jannaussd is slower hwen connected through the dock though, 1 GB/s instead of 1.5 GB/s
#asahi-dev · 2023-02-20 15:40
jannaualso hotplug of the dock worked on on the first try
#asahi-dev · 2023-02-20 15:39
jannauhotplug on the dock/hub's downstream port appears to be relieable. 10 out of 10 tries
#asahi-dev · 2023-02-20 15:19
jannauno, for hotplug on the mac port
#asahi-dev · 2023-02-20 15:18
jannausometimes it doesn't work at all, sometimes it takes a long time and sometimes it get into hotplug hot-unplug loop
#asahi-dev · 2023-02-20 15:17
jannauhaven't tried that yet as hotplug is not relieable in general
#asahi-dev · 2023-02-20 15:00
jannauthe intel ssd should be good for 2000 MB/s. so there might be an issue with the enclosure
#asahi-dev · 2023-02-20 14:58
jannauyes
#asahi-dev · 2023-02-20 14:56
jannausamsung ssd is not faster on the intel system but it should be capable of ~3500 MB/s
#asahi-dev · 2023-02-20 14:50
jannausame PCIe bandwidth message
#asahi-dev · 2023-02-20 14:50
jannaudd reports just 1.2 GB/s (intel) instead of 1.5 GB/s (mac) with a block size of 8M
#asahi-dev · 2023-02-20 14:49
jannauit's faster than on my intel thinkpad with Alpine Ridge tbt controller: 1070 MB/sec
#asahi-dev · 2023-02-20 14:39
jannauit's faster than the same ssd with a 10 gbps usb3 link though
#asahi-dev · 2023-02-20 14:37
jannauerr probably not
#asahi-dev · 2023-02-20 14:37
jannaui.e. it's only using PCIe1 x4
#asahi-dev · 2023-02-20 14:36
jannauprobably: "pci 0000:03:00.0: 8.000 Gb/s available PCIe bandwidth, limited by 2.5 GT/s PCIe x4 link at 0000:01:00.0 (capable of 31.504 Gb/s with 8.0 GT/s PCIe x4 link)"
#asahi-dev · 2023-02-20 14:35
jannauintel nvme works: "Timing buffered disk reads: 4300 MB in 3.00 seconds = 1432.78 MB/sec"
#asahi-dev · 2023-02-20 14:22
jannauyes

plus ›