Bugzilla – Bug 1205938
D-Installer - Slowness initialization on real hardware
Last modified: 2023-03-21 14:37:14 UTC
Created attachment 863250 [details] Hwinfo and usbmon logs We have observed that, sometimes, the installation on bare-metal (using an USB stick) is rather slow. Even the storage probing (running `hwinfo --disk`) can take up to 1 minute. Unfortunately, the problem does not happen 100% of the time. Some observations: * We have observed the same behavior on TW Live images. The installer images used can be found at: - https://build.opensuse.org/package/show/YaST:Head:D-Installer/d-installer-live We have reproduced the problem using a Dell Latitude E7470 with a Toshiba USB stick (I need to try with a different stick) and in a Dell Precision Tower 5810 even in a running system using just `hwinfo --usb` or `hwinfo --disk` But the same USB Stick works fine in a Lenovo ThinkPad T14s and int other machines. I collected usbmon and hwinfo logs from two calls where hwinfo was fine and where get stuck.
Additionally, we have reproduced the same problem (at least in the Latitude E7470) on an already installed system, so the problem triggered by hwinfo might not be specific to D-Installer.
The proximate case is here in the log: ffff8fd30d480180 674142717 C Bi:1:010:1 0 13 = 55534253 58010000 00000000 00 ffff8fd30d480180 674142832 S Bo:1:010:2 -115 31 = 55534243 59010000 24000000 80000612 01800024 00000000 00000000 000000 ffff8fd30d480180 674142849 C Bo:1:010:2 0 31 > ffff8fd5a59dcc00 674142856 S Bi:1:010:1 -115 36 < ffff8fd5a59dcc00 674143236 C Bi:1:010:1 -75 0 ffff8fd30d480180 674143279 S Bi:1:010:1 -115 13 < But that does not give us the root cause. Please also provide dmesg of the error case.
(In reply to Oliver Neukum from comment #2) > The proximate case is here in the log: > > ffff8fd30d480180 674142717 C Bi:1:010:1 0 13 = 55534253 58010000 00000000 00 > ffff8fd30d480180 674142832 S Bo:1:010:2 -115 31 = 55534243 59010000 24000000 > 80000612 01800024 00000000 00000000 000000 > ffff8fd30d480180 674142849 C Bo:1:010:2 0 31 > > ffff8fd5a59dcc00 674142856 S Bi:1:010:1 -115 36 < > ffff8fd5a59dcc00 674143236 C Bi:1:010:1 -75 0 > ffff8fd30d480180 674143279 S Bi:1:010:1 -115 13 < > > But that does not give us the root cause. Please also provide dmesg of the > error case. The only thing reported is something like: [21121.998831] usb 1-5: reset high-speed USB device number 10 using xhci_hcd [21214.145941] usb 1-5: reset high-speed USB device number 10 using xhci_hcd
(In reply to Knut Alejandro Anderssen González from comment #3) > [21121.998831] usb 1-5: reset high-speed USB device number 10 using xhci_hcd > [21214.145941] usb 1-5: reset high-speed USB device number 10 using xhci_hcd Yes, but I need to know when that happens. Please just attach dmesg.
(In reply to Oliver Neukum from comment #4) > (In reply to Knut Alejandro Anderssen González from comment #3) > > > [21121.998831] usb 1-5: reset high-speed USB device number 10 using xhci_hcd > > [21214.145941] usb 1-5: reset high-speed USB device number 10 using xhci_hcd > > Yes, but I need to know when that happens. Please just attach dmesg. Ok, I will reproduce it once more time providing requested logs, do you need something else?
(In reply to Knut Alejandro Anderssen González from comment #5) > Ok, I will reproduce it once more time providing requested logs, do you need > something else? Right now I cannot tell. I need to evaluate the full logs first.