Re: [3/3] 2.6.23-rc2: known regressions v2

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Saturday, 11 August 2007 12:53, Meelis Roos wrote:
> > > > 3. Can you please check if this patch changes anything:
> > > > http://marc.info/?l=linux-mm-commits&m=118669680811669&w=2
> > > 
> > > Tried it on current git, kernel/power/Kconfig gave 2 rejects.
> > > Tried on 2.6.23-rc2, still 2 rejects from the same file, .rej below:
> > 
> > Sorry, but:
> > 
> > rafael@albercik:~/src/linux-2.6> git-pull git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
> > Already up-to-date.
> > rafael@albercik:~/src/linux-2.6> cat ~/pm-fix-dependencies-of-config_suspend-and-config_hibernation-updated-3x.patch | patch -p1
> > patching file arch/x86_64/defconfig
> > patching file include/linux/cpu.h
> > patching file kernel/cpu.c
> > patching file kernel/power/Kconfig
> > 
> > How come?
> 
> Hmm, must have been mangled on the web (downloaded via wget so should be 
> byte-by-byte exact). I now downloaded it from akpm-s broken-out tarball 
> and this one applied fine.
> 
> The patch changes the warning to a different warning, it's now
> kernel/power/Kconfig:79:warning: 'select' used by config symbol 'PM_SLEEP_SMP' refers to undefined symbol 'HOTPLUG_CPU'
> 
> The warning also appears when running just "make ARCH=ppc" after
> "make ARCH=ppc oldconfig".

Yes, I have reproduced it now.

Well, it's clearly a false positive, since none of the options PM_SLEEP_SMP
depends on is set in your .config.  Also, it seems to be ppc-specific (for
example, if I change the architecture to powerpc, the warning doesn't show up).

The only problem I see here is that the warning in fact should not be printed,
but I don't know how to make the build system stop printing it.

Greetings,
Rafael


-- 
"Premature optimization is the root of all evil." - Donald Knuth