2010-08-18 09:43:10

by Christian Samsel

[permalink] [raw]
Subject: runaway loop modprobe char-major-5-1 has returned?

Hi,
I wanted to build a minimal kernel as Xen DomU, but stumbled across

[ 0.029457] VFS: Disk quotas dquot_6.5.2
[ 0.029559] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
[ 0.029973] Installing knfsd (copyright (C) 1996 [email protected]).
[ 0.030662] msgmni has been set to 243
[ 0.099708] request_module: runaway loop modprobe char-major-5-1
[ 0.100482] request_module: runaway loop modprobe char-major-5-1
[ 0.101851] request_module: runaway loop modprobe char-major-5-1
[ 0.102534] request_module: runaway loop modprobe char-major-5-1
[ 0.106311] request_module: runaway loop modprobe char-major-5-1
(hangs)

after reading http://lwn.net/Articles/310472/ and comparing the output after
msgmni of a working kernel on a other machine i found the config option change
which breaks it:

1456c1456
< CONFIG_CRYPTO_MANAGER_TESTS=y
---
> # CONFIG_CRYPTO_MANAGER_TESTS is not set

(without crypto test the kernel is broken)

kernel version is 2.6.35.2 (current stable git)
cmdline: root=/dev/ram0 console=hvc0

non-working config attached

Christian


Attachments:
config-broken (37.35 kB)

2010-08-18 19:01:03

by Jiri Slaby

[permalink] [raw]
Subject: Re: runaway loop modprobe char-major-5-1 has returned?

On 08/18/2010 11:43 AM, Christian Samsel wrote:
> Hi,
> I wanted to build a minimal kernel as Xen DomU, but stumbled across
>
> [ 0.029457] VFS: Disk quotas dquot_6.5.2
> [ 0.029559] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
> [ 0.029973] Installing knfsd (copyright (C) 1996 [email protected]).
> [ 0.030662] msgmni has been set to 243
> [ 0.099708] request_module: runaway loop modprobe char-major-5-1
> [ 0.100482] request_module: runaway loop modprobe char-major-5-1
> [ 0.101851] request_module: runaway loop modprobe char-major-5-1
> [ 0.102534] request_module: runaway loop modprobe char-major-5-1
> [ 0.106311] request_module: runaway loop modprobe char-major-5-1
> (hangs)
>
> after reading http://lwn.net/Articles/310472/ and comparing the output after
> msgmni of a working kernel on a other machine i found the config option change
> which breaks it:

Hi, you may want this:
31d1d48e199e99077fb30f6fb9a793be7bec756f

Adding stable to consider it for inclusion in .35 stable too.

--
js

2010-08-23 22:47:12

by Greg KH

[permalink] [raw]
Subject: Re: [stable] runaway loop modprobe char-major-5-1 has returned?

On Wed, Aug 18, 2010 at 09:00:57PM +0200, Jiri Slaby wrote:
> On 08/18/2010 11:43 AM, Christian Samsel wrote:
> > Hi,
> > I wanted to build a minimal kernel as Xen DomU, but stumbled across
> >
> > [ 0.029457] VFS: Disk quotas dquot_6.5.2
> > [ 0.029559] Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
> > [ 0.029973] Installing knfsd (copyright (C) 1996 [email protected]).
> > [ 0.030662] msgmni has been set to 243
> > [ 0.099708] request_module: runaway loop modprobe char-major-5-1
> > [ 0.100482] request_module: runaway loop modprobe char-major-5-1
> > [ 0.101851] request_module: runaway loop modprobe char-major-5-1
> > [ 0.102534] request_module: runaway loop modprobe char-major-5-1
> > [ 0.106311] request_module: runaway loop modprobe char-major-5-1
> > (hangs)
> >
> > after reading http://lwn.net/Articles/310472/ and comparing the output after
> > msgmni of a working kernel on a other machine i found the config option change
> > which breaks it:
>
> Hi, you may want this:
> 31d1d48e199e99077fb30f6fb9a793be7bec756f
>
> Adding stable to consider it for inclusion in .35 stable too.

Now queued up for the next .35-stable kernel release.

thanks,

greg k-h