On Fri, 26 Jun 2020 at 08:21, <fedora@eyal.emu.id.au> wrote:
On 2020-06-26 20:40, George N. White III wrote:
> On Fri, 26 Jun 2020 at 04:56, <fedora@eyal.emu.id.au <mailto:fedora@eyal.emu.id.au>> wrote:
>
>     On 2020-06-26 16:22, Samuel Sieb wrote:
>      > On 6/24/20 11:36 PM, fedora@eyal.emu.id.au <mailto:fedora@eyal.emu.id.au> wrote:
>      >> Jun 23 18:06:17 e7 kernel: usb 2-7: Disable of device-initiated U1 failed.
>      >> Jun 23 18:06:17 e7 kernel: usb 2-7: Disable of device-initiated U2 failed.
>      >
>      > There is some issue with handling the SD card modes.
>
>     Testing on my other machine (already on f32) did not show the above messages.
>     However, being a 2009 vintage, makes the hware less sophisticated, and maybe the newer hware
>     is too smart for its own good?
>
>
> 10+ years is enough to find and fix all the bugs, especially if you were lucky enough to catch
> the first use of a chip that became popular.   OTOH, if you happen to catch the last machine
> that used a problematic chipset, new kernels may not even have the drivers.   Also, if you
> have a problem running linux on older hardware, there will be discussions of the issue.  A
> search for the problem with the make and model of the PC and USB ID of the prime suspect
> often provides useful information and may tell you which kernel version has the fix.

Maybe I am misreading this response, but let me clarify:
The problem happened on a new machine (a Gigabyte Z390 board) running f30(*1).
The old (2009 vintage) machine runs f32 and is where I tested, not showing the problem.

I meant that lack of problems on older hardware is because the problems have been found
and fixed over the years.   New machines bring new problems.   Gigabyte Z390 board does
have a good reputation and lots of Phoronix benchmark results for various linux distros, 
including Fedora 31.   This points to the external hardware, cables, or a previously
unexplored "use case".

USB3 does have RF interference issues that can cause USB2 problems if the USB2 plug is 
close to a low-quality (poorly shielded) USB3 cable.   Older USB2 devices like card readers
may not be shielded to work in a USB3 environment.

It would be worth testing with all USB devices but the card reader disconnected.  Ideally
you would use a USB2 port with the card reader (via header on the system board so 
could be on a front panel port).
 

>     [...]
>
>      > I suggest upgrading the system because there is likely a kernel fix for the initial trigger that caused the whole problem.  And possibly a dvb fix as well to avoid the oops.  I second the suggestion of testing with a live boot, but if you're going to upgrade anyway, might as well just do that.
>
>     Agreed, my sentiment as well.
>
>     Still, as this is repeatable, I wonder if running a debug kernel (kernel-debug) will reveal more?
>
>
> Google may tell you someone else has already done the heavy lifting, and which kernel has the
> fix.

Naturally, and I searched and did not find a relevant report.

Google search results vary depending on past searches.  I found
many complaints about the BIOS and with "Gigabyte Z390 +USB +linux":
 


--
George N. White III