On Friday 16 November 2007 06:09 pm, David Wolfskill wrote:
> On Fri, Nov 16, 2007 at 04:26:43PM -0500, Jung-uk Kim wrote:
> > ...
> > 'Undefined symbol "__mb_sb_limit" is totally unrelated to ACPI,
> > i.e., it is caused by recent API breakage.
>
> That's what I thought, but I also thought it would be useful for me
> to report what I saw, rather than what I thought I should have
> seen. :-}
>
> > I think 'make buildworld; make
> > installworld' should fix it. Anyway, thanks for testing!
>
> Well, I re-booted again -- in a slightly different environment (one
> where I have a DHCP server on the net), and didn't see a recurrence
> of that whine about __mb_sb_limit.
>
> (I always do a "make buildworld && make kernel && make
> installworld" (with appropriate invocations of assorted other
> things, such as mergemaster).)
>
> I tried the Fn+Esc keyboard chord to suspend the machine (from
> vty0, rather than X); the power LED started blinking, but screen
> didn't even blank. I then tried to resume the machine (by pressing
> & releasing the power button quickly); that caused the display to
> blank, but I couldn't get it "un-blanked." So I flipped to vty2
> and logged in (blind) and issued "sudo reboot" (which worked).
>
> On reboot, I tried using "sudo zzz" to suspend the machine. That
> did not cause the power LED to blink, but it did cause the display
> to blank. I tried pressing & releasing the power button quickly; no
> noticable effect. I tried using the Fn+F8 keyboard chord to cause
> the display to show something, and it showed what looked like a
> strange "test pattern" (sort like a bunch of vertical lines made of
> hyphens). Pressing the Fn+F8 keybaord chord again powered the
> machine off again, rather quickly. Apparently the disk buffers
> were flaushed though, as no file systems were dirty.
>
> FWIW, I'm pretty sure this is the same (fairly broken) behavior as
> I see now.
>
> The machine is a Dell Inspiron 8200.
You mean it did not improve nor regress anything, right?
Jung-uk Kim