Hi Mario,
On Thu, Feb 23, 2023 at 03:07:40PM -0600, Mario Limonciello wrote:
> TBT3 devices when connected to an AMD USB4 router occasionally fail to
> properly respond to requests for the DROM via bit banging.
>
> Depending upon which part of the request failed will impact the severity.
> A number of workarounds have been put in place to let the driver handle
> the failed requests:
>
> commit e87491a9fd4e3 ("thunderbolt: Retry DROM reads for more failure scenarios")
> commit a283de3ec646f ("thunderbolt: Do not resume routers if UID is not set")
> commit 6915812bbd109 ("thunderbolt: Do not make DROM read success compulsory")
> commit f022ff7bf377 ("thunderbolt: Retry DROM read once if parsing fails")
>
> Still even with these changes the failures do make it through. In comparing
> other CM implementations utilized on AMD systems, they all access the
> DROM directly from the NVM.
>
> To avoid triggering this issue, try to get the DROM directly from the NVM
> in Linux as well when devices have an LC.
>
> v4:
> * Style fixups
> * Fixup for wrong path for USB4 devices
>
> Mario Limonciello (3):
> thunderbolt: Adjust how NVM reading works
> thunderbolt: use `tb_eeprom_get_drom_offset` to discover DROM offset
> thunderbolt: Refactor DROM reading
I split the device side into a separate function too, renamed root
switch to host router (as that's the correct USB4 term), and fixed a
couple style issues and applied to thunderbolt.git/next, thanks!
Please check that I did not mess up anything :)
