Hi,
[no properly binding reference via In-Reply-To: available thus manually
re-creating, sorry]
> > So we can print warning in hibernation_die_notifier without
> > introducing a global variable?
> >
>
> Actually, I'd kill the machine right away.
>
> if (memcmp(result, buf, MD5_DIGEST_SIZE)) {
> pr_err("PM: e820 map conflict detected!\n");
> panic("BIOS is playing funny tricks with us.\n");
> }
> Best regards,
> Pavel
+1.
I would tend to think that
it's rather preferable to
kill an affected environment scope
(in this case: whole system), hard
(except for perhaps some emergency epilogue state saving),
in case of its state having become
suspicious/unpredictable/dangerous,
rather than having things carry on in a merry-go-round manner and thus
making improper state progress
which translates into
*continued activity* of *corrupting things*
(possibly even leading to *persistent* i.e. storage-recorded corruption!)
willy-nilly.
Plus, killing a machine hard in such a questionable case
would increase our influx of valuable reports
due to users being
hard-pressed to report this rather than
silently ignoring / not even knowing this issue
(in those cases where we already know that
no further development fixes will be determinable,
carrying on with a dire warning probably still is preferable to
making the machine completely unusable, eternally,
though).
Thus, think robustness.
Andreas Mohr
