04:26 <yiy> Hi, has anyone had problems with NVMe not being visible from linux on the x13s? Not a single guide I've seen so far had any mention of an issue like this - they all assume that if you can boot then the NVMe is naturally available. No matter the image I try there's no device in /dev/ and no mention of NVMe in dmesg log. It's obviously there and working from the EUFI utility and windows. I don't see

04:26 <yiy> any relevant settings to change in UEFI to make it appear (changing linux mode on/off doesn't make a difference). Firmware version 1.57.

04:35 <HdkR> Works fine here, I have Linux installed on it even. Old kernel/devicetree or something?

04:35 <HdkR> Accidentally doing ACPI booting?

04:39 <yiy> Stupid question probably but how do I check?

04:41 <HdkR> I guess `ls /sys/firmware/devicetree/base` and see if there is anything there for device tree versus acpi boot

04:42 <HdkR> uname -a for old kernel

04:43 <yiy> OK I'll try it. For reference I tried that old debian linaro image, latest armbian and right now testing latest ironbin arch iso.

04:46 <yiy> /sys/firmware/devicetree/base exists and has many entries. My x86_64 desktop doesn't have that so I assume not ACPI boot. Dmesg log says "ACPI: Interpreter disabled"

04:47 <HdkR> Does lspci show the nvme drive?

04:48 <yiy> I don't know what it should look like but probably not. In order from top to bottom: pci bridge, pci bridge, wireless conroller, pci bridge, network controller.

04:49 <HdkR> Yea, there should be a "Non-Volatile memory controller" in there

04:52 <yiy> So strange. Online I see similar issues with x86 systems but problem is some intel/raid option in bios/uefi which does not exist on this device. Short of trying another drive, is there anything else I can try?

04:54 <HdkR> Might be able to check /var/log/kern.log for PCIe probing problems

04:56 <HdkR> or nvme

04:58 <yiy> log here: https://logpaste.com/EBIOUnx4 Search for 0002:00, is that "Timeout on hotplug command 0x03c0" the culprit you think?

05:00 <HdkR> Seems likely, I think that is the bridge that the NVMe is connected to

05:00 <yiy> Seems like it was fixed recently https://www.spinics.net/lists/linux-pci/msg139903.html

05:01 <yiy> Not sure how most people don't have this problem though, everyone has the same hardware yet no guides mention this.

05:03 <HdkR> Most guides are likely running downstream kernels that have been working around the issues

05:03 <HdkR> Seems like that's the fix though

05:05 <yiy> I'm running those same kernels and yet only I encounter the issue. Bewildering. I'll now go dig into those kernels to see if they have the patch, and if not, somehow crosscompile and make my own iso. Thanks for the help.

05:36 <yiy> Actually turns out armbian already had that patch, and yet it still doesn't work https://logpaste.com/EFGtsTgl Back to square 1.

05:39 <HdkR> quirky

05:53 <HdkR> Sadly I'm just a user, so I don't know how to debug any further

06:02 <yiy> Don't worry about it, I knew what I was getting into when I bought the device.

06:02 <yiy> But could I ask you to post your boot dmesg output? I'd like to compare and see if there are any other differences from a working system.

06:05 <HdkR> You definitely don't want to look at my dmesg output atm. I have the wwan port disabled in the bios and it spams errors constantly

06:06 <yiy> Ok then anyone else if they can please.

06:43 <yiy> thanks!

19:43 <altacus> yiy: Have you tried booting from the Ubuntu x13s concept? It allows you to boot from usb and gives you a live working version, which can be handy to try different things out with.

19:54 <altacus> I was wondering if we could ever expect battery life on the x13s to be near the reported 25+ hours that qualcomm claimed on their website.

21:40 <clover[m]> <steev> "clover: HdkR: pushed out 6.5.2..." <- cool, pushed it to my arch repo

22:07 <clover[m]> anyone else getting a bunch of these messages in their dmesg?... (full message at <https://matrix.org/_matrix/media/v3/download/matrix.org/CzCtGTKwUDgKHiORxwtmJbfO>)

22:08 <steev> nope

22:08 <steev> steev@wintermute:~/kernels/torvalds$ dmesg | grep glit

22:08 <steev> steev@wintermute:~/kernels/torvalds$ uptime

22:08 <steev> 17:08:13 up 7:42, 3 users, load average: 0.00, 0.00, 0.00

22:08 <steev> oh that's the wwan

22:09 <steev> drivers/net/wwan/mhi_wwan_mbim.c: net_err_ratelimited("sequence number glitch prev=%d curr=%d\n",

22:09 <steev> get rate limited noob

22:09 <steev> nah, i dunno what causes it but if you dig through that file, it should probably show

22:13 <clover[m]> lol

22:17 <steev> it's happening in verify_nth16 whatever nth is

22:35 <clover[m]> The screw holding down the wwan card is like fused in place

22:39 <clover[m]> I managed to coax it off with a tiny flat head piece. Thank you ifixit

22:56 <clover[m]> jhovold: just to verify, i removed the physical wwan card and SD card for good measure and still seeing my ath11k firmware issue

22:57 <clover[m]> I wonder if I can do a board transplant from the working x13s lol