2.6.16-rc1-mm3
in [Kernel]
|
Next: Grant
From: Reuben Farrelly on 27 Jan 2006 07:50 On 26/01/2006 6:39 p.m., Greg KH wrote: > On Wed, Jan 25, 2006 at 11:44:14PM +1300, Reuben Farrelly wrote: >> >> On 25/01/2006 8:24 p.m., Andrew Morton wrote: >>> http://www.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.16-rc1/2.6.16-rc1-mm3/ >>> >>> - Dropped the timekeeping patch series due to a complex timesource >>> selection >>> bug. >>> >>> - Various fixes and updates. >> Generally quite good again :) >> >> I'm seeing this USB "handoff" warning message logged when booting up: >> >> 0000:00:1d.7 EHCI: BIOS handoff failed (BIOS bug ?) >> >> This is not new to this -mm release, looking back over my bootlogs I note >> that 2.6-15-rc5-mm1 was OK, but 2.6.15-mm4 was showing this message. I'll >> narrow it down if it doesn't appear obvious what the problem is. > > When this happens, does your usb-ehci driver still work properly > (meaning do usb 2.0 devices connect and go at the proper high speeds)? It seems like every device I can lay my hands on either only supports 1.1 or doesn't specify what it is. I'll try find something which is known to be 2.0. Certainly my 1.1 devices seem to work ok. > And if you change the USB setting in your bios, does this error message > go away? If I change "USB 2.0 legacy support" from "High Speed 480Mbps" which it was originally to "Full Speed 12MBps" then yes it does go away. Although I'm not sure that's the best way to have it configured. There is a slight possibility that this setting was changed in the bios which is why I am now noticing it, and that it wasn't as a result of a change above. Reuben - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Greg KH on 27 Jan 2006 12:30 On Sat, Jan 28, 2006 at 01:46:20AM +1300, Reuben Farrelly wrote: > > > On 26/01/2006 6:39 p.m., Greg KH wrote: > >On Wed, Jan 25, 2006 at 11:44:14PM +1300, Reuben Farrelly wrote: > >> > >>On 25/01/2006 8:24 p.m., Andrew Morton wrote: > >>>http://www.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.16-rc1/2.6.16-rc1-mm3/ > >>> > >>>- Dropped the timekeeping patch series due to a complex timesource > >>>selection > >>> bug. > >>> > >>>- Various fixes and updates. > >>Generally quite good again :) > >> > >>I'm seeing this USB "handoff" warning message logged when booting up: > >> > >>0000:00:1d.7 EHCI: BIOS handoff failed (BIOS bug ?) > >> > >>This is not new to this -mm release, looking back over my bootlogs I note > >>that 2.6-15-rc5-mm1 was OK, but 2.6.15-mm4 was showing this message. > >>I'll narrow it down if it doesn't appear obvious what the problem is. > > > >When this happens, does your usb-ehci driver still work properly > >(meaning do usb 2.0 devices connect and go at the proper high speeds)? > > It seems like every device I can lay my hands on either only supports 1.1 > or doesn't specify what it is. I'll try find something which is known to > be 2.0. Certainly my 1.1 devices seem to work ok. > > >And if you change the USB setting in your bios, does this error message > >go away? > > If I change "USB 2.0 legacy support" from "High Speed 480Mbps" which it was > originally to "Full Speed 12MBps" then yes it does go away. Although I'm > not sure that's the best way to have it configured. How about just disabling USB legacy support in the bios completly? Unless you have a USB keyboard that you need for a bootloader screen or BIOS configuration, that's the recommended setting (due to all of the horrible BIOS USB bugs we have seen over the years.) If after you find a USB 2.0 device, and have any problems with it, please let us know. thanks, greg k-h - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Pete Zaitcev on 27 Jan 2006 13:00 On Fri, 27 Jan 2006 09:27:20 -0800, Greg KH <greg(a)kroah.com> wrote: > How about just disabling USB legacy support in the bios completly? > Unless you have a USB keyboard that you need for a bootloader screen or > BIOS configuration, that's the recommended setting (due to all of the > horrible BIOS USB bugs we have seen over the years.) I disagree with the "unless". Just disable it, period. Most of the time, disabling "USB Legacy Support" leaves the bootloader fully operational. I always recommend it as the first course of action in cases like this one. This setting was used to set two things: - INT 11h delivering characters to the bootloader - Emulation of port 0x60 However, most decent BIOS vendors figured that the first was bogus. They only use this setting to disable the port emulation. BTW, the BIOS often has its finger in the bus sharing between EHCI and its companion, there's never anything we can do about it. It is always a BIOS update that fixes it. But it sure produces lots of false regressions when our previous mode of operation worked by accident. -- Pete - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Reuben Farrelly on 27 Jan 2006 14:50 On 28/01/2006 6:49 a.m., Pete Zaitcev wrote: > On Fri, 27 Jan 2006 09:27:20 -0800, Greg KH <greg(a)kroah.com> wrote: > >> How about just disabling USB legacy support in the bios completly? >> Unless you have a USB keyboard that you need for a bootloader screen or >> BIOS configuration, that's the recommended setting (due to all of the >> horrible BIOS USB bugs we have seen over the years.) > > I disagree with the "unless". Just disable it, period. Most of the time, > disabling "USB Legacy Support" leaves the bootloader fully operational. > I always recommend it as the first course of action in cases like this one. That seemed to clear the error message, and my 1.1 devices at least, still work fine. Perhaps the error message could be changed to point users in the right direction to fixing this, perhaps something more like "0000:00:1d.7 EHCI: BIOS handoff failed (BIOS bug? Try disabling legacy USB support in BIOS if it is enabled)" Anyway, thanks for the hint (and sorry for the noise). reuben - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Pete Zaitcev on 27 Jan 2006 15:00
On Sat, 28 Jan 2006 08:40:18 +1300, Reuben Farrelly <reuben-lkml(a)reub.net> wrote: > direction to fixing this, perhaps something more like "0000:00:1d.7 EHCI: BIOS > handoff failed (BIOS bug? Try disabling legacy USB support in BIOS if it is > enabled)" I thought about this. However it seems that a blanket implication of BIOS vendors would not be ethical. Certainly our code is not infallible, and the handoff continues to improve. I expect BIOS developers to stop providing legacy support eventually. It's like the ISA bus, which seemed impossible to get rid of. -- Pete - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |