On Tue, 8 Oct 2019 22:04:45 +0200 Heiner Kallweit <hkallwe...@gmail.com> wrote:
> Thanks for the comprehensive analysis! Comparing the chip registers before > and after suspend your BIOS seems to change registers on resume from suspend, > and the driver doesn't configure jumbo when resuming. This may explain the > issue. The combination jumbo + suspend + BIOS bug seems to be quite rare, > else I think we should have seen such a report years ago already. I see. > Could you please check whether the following patch fixes the issue for you? I've tested your patch with minor change when applying on top of the current master tree. The result is: it is working great! No more problems, that I was observing before. Thank you very much, Heiner! You can add: Reported-by: Mariusz Bialonczyk <ma...@skyboo.net> Tested-by: Mariusz Bialonczyk <ma...@skyboo.net> regards, -- Mariusz Białończyk | xmpp/e-mail: ma...@skyboo.net http://manio.skyboo.net | https://github.com/manio
