Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On Sun, Aug 20, 2017 at 11:21:12AM +0200, Alan McKinnon wrote: > On 20/08/2017 05:47, Bill Kenworthy wrote: > > Hi all, > > I am getting random lockups on kernel 4.12.5. > > > > Three intel hosts (atom, early i7, i7 haswell) - its worse after a > > suspend resume session. Because these are in use Ive gone back to > > 4.9.34 which is stable. > > […] > > I'm seeing changes in 4.12 too and haven't bothered looking further as > 4.11 still works for me. My external monitors and USB keyboard & mouse > on laptops stopped working with 4.12 and other silliness which I forget. > > My techie spidey-sense is telling me it all smells a lot like someone > tidied up .config and things moved around, so make oldconfig got confused. I, too, did encounter problems with 4.12 (vanilla from kernel.org). I have a PC with radeon driver and a laptop with intel graphcis. I always do make oldconfig when building a new kernel version. But with 4.12, both those machines stopped working, they hung at boot, I think when switching graphics mode. Finally I found the time and reconfigured my kernel from scratch on the laptop, which eventually made it work again. The big PC is still waiting for its turn and running on 4.10 for the time being. -- Gruß | Greetings | Qapla’ Please do not share anything from, with or about me on any social network. I have seen things you lusers would not believe. I've seen Sun monitors on fire off the side of the multimedia lab. I've seen NTU lights glitter in the dark near the Mail Gate. All these things will be lost in time, like the root partition last week. Time to die. signature.asc Description: Digital signature
Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On Sun, 2017-08-20 at 11:21 +0200, Alan McKinnon wrote: > > I'm seeing changes in 4.12 too and haven't bothered looking further > as > 4.11 still works for me. My external monitors and USB keyboard & > mouse > on laptops stopped working with 4.12 and other silliness which I > forget. > > My techie spidey-sense is telling me it all smells a lot like someone > tidied up .config and things moved around, so make oldconfig got > confused. > Correct. Old Intel host here. No lock-ups but I had also USB-related problems with 4.12 till I set this (new) option in kernel config: CONFIG_USB_PCI: A lot of embeded system SOC (e.g. freescale T2080) have both PCI and USB modules. But USB module is controlled by registers directly, it have no relationship with PCI module. When you enable the above you get the next one: CONFIG_USB_UHCI_HCD: The Universal Host Controller Interface is a standard by Intel for accessing the USB hardware in the PC (which is also called the USB host controller). If your USB host controller conforms to this standard, you may want to say Y, but see below. All recent boards with Intel PCI chipsets (like intel 430TX, 440FX, 440LX, 440BX, i810, i820) conform to this standard. Also all VIA PCI chipsets (like VIA VP2, VP3, MVP3, Apollo Pro, Apollo Pro II or Apollo Pro 133) and LEON/GRLIB SoCs with the GRUSBHC controller. The first option's help line is a bit misleading, my system is not an embedded one but without UHCI I don't have any USB working. In 4.11 and earlier USB_UHCI_HCD was not tied to USB_PCI and that's why make oldconfig messed up in my case. bye, raffaele
Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On Sun, 20 Aug 2017 11:47:14 +0800 Bill Kenworthywrote: > Hi all, > I am getting random lockups on kernel 4.12.5. > > Three intel hosts (atom, early i7, i7 haswell) - its worse after a > suspend resume session. Because these are in use Ive gone back to > 4.9.34 which is stable. > > I also have some VM's and an MS surface pro 4 which are stable on > 4.12.5, 6 and 8) - confusing! > > Is anyone else seeing this? Yup, on one of my laptops I use, I have also experienced lockups, so I've been forced to downgrade to 4.9.x. On two other machines (one of it is my personal pc used at daily basis, the other is just a PC for storage purposes) I didn't noticed anything like that… -- -^- _ something is grinding the emptiness: _ /O)_\// Velvet Acid Christ - Caustic Disco (_(|__(_(_) grf.
Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On Sunday, 20 August 2017 10:39:51 BST Bill Kenworthy wrote: > On 20/08/17 17:21, Alan McKinnon wrote: > > On 20/08/2017 05:47, Bill Kenworthy wrote: > >> Hi all, > >> > >>I am getting random lockups on kernel 4.12.5. > >> > >> Three intel hosts (atom, early i7, i7 haswell) - its worse after a > >> suspend resume session. Because these are in use Ive gone back to > >> 4.9.34 which is stable. > >> > >> I also have some VM's and an MS surface pro 4 which are stable on > >> 4.12.5, 6 and 8) - confusing! > >> > >> Is anyone else seeing this? > >> > >> BillK > > > > Neil's reply elsewhere to Ralph may have the smoking gun clue. > > > > I'm seeing changes in 4.12 too and haven't bothered looking further as > > 4.11 still works for me. My external monitors and USB keyboard & mouse > > on laptops stopped working with 4.12 and other silliness which I forget. > > > > My techie spidey-sense is telling me it all smells a lot like someone > > tidied up .config and things moved around, so make oldconfig got confused. > > Not to mention changing the bcache device numbering - instead of > /dev/bacache0, /dev/bcache1 etc. its become /dev/bcache0, /dev/bcache16, > /dev/bcache32, ... (offset of 16) > > That bit me too ... > > BillK When running make oldconfig, which I prefer because it asks me to confirm settings for all new modules, Alan should make sure CONFIG_USB=y is set so. >From what I recall the oldconfig default setting is 'not set', so user input is required. Unfortunately, I have no idea why Bill's hard lockups occur. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On 20/08/17 17:21, Alan McKinnon wrote: > On 20/08/2017 05:47, Bill Kenworthy wrote: >> Hi all, >> I am getting random lockups on kernel 4.12.5. >> >> Three intel hosts (atom, early i7, i7 haswell) - its worse after a >> suspend resume session. Because these are in use Ive gone back to >> 4.9.34 which is stable. >> >> I also have some VM's and an MS surface pro 4 which are stable on >> 4.12.5, 6 and 8) - confusing! >> >> Is anyone else seeing this? >> >> BillK >> > > > Neil's reply elsewhere to Ralph may have the smoking gun clue. > > I'm seeing changes in 4.12 too and haven't bothered looking further as > 4.11 still works for me. My external monitors and USB keyboard & mouse > on laptops stopped working with 4.12 and other silliness which I forget. > > My techie spidey-sense is telling me it all smells a lot like someone > tidied up .config and things moved around, so make oldconfig got confused. > Not to mention changing the bcache device numbering - instead of /dev/bacache0, /dev/bcache1 etc. its become /dev/bcache0, /dev/bcache16, /dev/bcache32, ... (offset of 16) That bit me too ... BillK
Re: [gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
On 20/08/2017 05:47, Bill Kenworthy wrote: > Hi all, > I am getting random lockups on kernel 4.12.5. > > Three intel hosts (atom, early i7, i7 haswell) - its worse after a > suspend resume session. Because these are in use Ive gone back to > 4.9.34 which is stable. > > I also have some VM's and an MS surface pro 4 which are stable on > 4.12.5, 6 and 8) - confusing! > > Is anyone else seeing this? > > BillK > Neil's reply elsewhere to Ralph may have the smoking gun clue. I'm seeing changes in 4.12 too and haven't bothered looking further as 4.11 still works for me. My external monitors and USB keyboard & mouse on laptops stopped working with 4.12 and other silliness which I forget. My techie spidey-sense is telling me it all smells a lot like someone tidied up .config and things moved around, so make oldconfig got confused. -- Alan McKinnon alan.mckin...@gmail.com
[gentoo-user] Kernel 4.12.5 hard lockups, nothing in logs.
Hi all, I am getting random lockups on kernel 4.12.5. Three intel hosts (atom, early i7, i7 haswell) - its worse after a suspend resume session. Because these are in use Ive gone back to 4.9.34 which is stable. I also have some VM's and an MS surface pro 4 which are stable on 4.12.5, 6 and 8) - confusing! Is anyone else seeing this? BillK