Am Montag, den 05.03.2018, 16:27 +0200 schrieb Mathias Nyman: > On 26.02.2018 16:05, Oliver Neukum wrote: > > > > Am Dienstag, den 13.02.2018, 13:56 -0800 schrieb The Real Bev: > > > > > it looks like a relatively old issue with D3 on some instances of XHCI. > > > > Yep, The "Host halt failed, -19" (ENODEV) is seen when xhci driver tries to > halt the host during setup, polling a mmio register but reading only 0xffffffff. > > Which I guess is expected if controller is still on D3. > > No idea why this controller fails to change from D3 to D0. > Any PCI changes betewen 4.8 and 4.9? commit 4132a577a0a7e75b938d2ae49c7a16b358f60661 Author: Lukas Wunner <lukas@xxxxxxxxx> Date: Sun Sep 18 05:39:20 2016 +0200 PCI: Afford direct-complete to devices with non-standard PM commit cc7cc02bada84f0d707aa5b6d2ef8728a2e1f911 Author: Lukas Wunner <lukas@xxxxxxxxx> Date: Sun Sep 18 05:39:20 2016 +0200 PCI: Query platform firmware for device power state commit a6a64026c0cd1a76a0c8ab1c05a421aa4821887b Author: Lukas Wunner <lukas@xxxxxxxxx> Date: Sun Sep 18 05:39:20 2016 +0200 PCI: Recognize D3cold in pci_update_current_state() commit a0d2a959d3da343554523d26902de1d40a9e5c28 Author: Lukas Wunner <lukas@xxxxxxxxx> Date: Sun Sep 18 05:39:20 2016 +0200 PCI: Avoid unnecessary resume after direct-complete Could you test with a revert of a0d2a959d3da343554523d26902de1d40a9e5c28 ? Regards Oliver -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
