[Pm-utils] [patch] disable suspend when kernel image changed
victor.lowther at gmail.com
Sat Apr 10 09:21:16 PDT 2010
On Sat, Apr 10, 2010 at 9:53 AM, Dan Nicholson <dbn.lists at gmail.com> wrote:
> On Fri, Apr 9, 2010 at 3:37 PM, Richard Zidlicky <rz at linux-m68k.org> wrote:
>> On Fri, Apr 09, 2010 at 06:24:59AM -0700, Dan Nicholson wrote:
>>> I'm not sure I think this is really a big issue for pm-utils. 99% of
>>> people will be using their distro's kernel that has the full version
>>> in the filename, so they'll never have their current kernel image
>> if it is so simple to for pm-utils to fix it why not do it? Crashing
>> a system just because it may affect only less than 1% of Linux users
>> is inexcusable.
It is only "simple" in the specific cases the distros create. There
is no general way for pm-utils to know what kernel the user will boot
into on resume from hibernate, so we don't even try -- it is up to the
user or the distro to get things right.
If the user is hand-rolling and installing their own kernels, any
breaks are strictly their own problem.
>>> .... Overwriting the image of your currently running kernel
>>> and then intentionally hibernating seems to be a pretty risky
>>> maneuver, and I don't think pm-utils should be trying to safeguard the
>>> very few people who try to do it. There's a simpler solution to this
>> afaics it is not documented anywhere and fixing it is easier than documenting
> I disagree. Fixing it without causing regressions is much harder than
> documenting it.
True -- every distro does it their own way, so the solution is one for
the distros, not for pm-utils.
I encourage distros to roll their own hooks to handle this situation,
and (in the long term) I think that being able to resume from
hibernate from a different kernel than you suspended from is the best
> The only people who ever would get into this situation are people
> installing their own kernel and intentionally overwriting the running
> kernel. Why would they logout and walk away without rebooting into the
> new kernel on a multiuser system?
> No distro I've ever heard of is overwriting the current kernel, which
> means that this feature could only serve to cause regressions for at
> least 99% of people.
Arch Linux does, but it is very much a "you asked for it, you got it"
sort of distro.
> Pm-utils mailing list
> Pm-utils at lists.freedesktop.org
More information about the Pm-utils