http://bugzilla.suse.com/show_bug.cgi?id=1039737 http://bugzilla.suse.com/show_bug.cgi?id=1039737#c18 --- Comment #18 from Jochen Hansper <hansper@t-online.de> --- (In reply to Vlastimil Babka from comment #17)

(In reply to Jochen Hansper from comment #16)

...

I get lots of these now:

Weird, this part of the second patch should be preventing these errors:

-#if !defined(CONFIG_DEBUG_PAGEALLOC) && !defined(CONFIG_KMEMCHECK) +#if !defined(CONFIG_DEBUG_PAGEALLOC) && !defined(CONFIG_KMEMCHECK) && 0

Does your dmesg after boot contain the string "Using GB pages for direct mapping" ?

Yes, it does: Jun 01 19:56:35 mframe kernel: found SMP MP-table at [mem 0x000fcc60-0x000fcc6f] mapped at [ffff8800000fcc60] Jun 01 19:56:35 mframe kernel: Scanning 1 areas for low memory corruption Jun 01 19:56:35 mframe kernel: Base memory trampoline at [ffff880000097000] 97000 size 24576 Jun 01 19:56:35 mframe kernel: Using GB pages for direct mapping Jun 01 19:56:35 mframe kernel: BRK [0x023e0000, 0x023e0fff] PGTABLE Jun 01 19:56:35 mframe kernel: BRK [0x023e1000, 0x023e1fff] PGTABLE Jun 01 19:56:35 mframe kernel: BRK [0x023e2000, 0x023e2fff] PGTABLE Jun 01 19:56:35 mframe kernel: BRK [0x023e3000, 0x023e3fff] PGTABLE Jun 01 19:56:35 mframe kernel: BRK [0x023e4000, 0x023e4fff] PGTABLE Jun 01 19:56:35 mframe kernel: BRK [0x023e5000, 0x023e5fff] PGTABLE Jun 01 19:56:35 mframe kernel: RAMDISK: [mem 0x36b84000-0x375b9fff] -- I've built the patched kernel in a kvm like this: As root: Installed default kernel and its sources via Yast2 # mkdir -p /build/src # cp -Rp /usr/src/linux-4.4.62-18.6/* /build/src/ # chmod -R 775 /build/src # chgrp -R users /build/src As non-root user dummy: Applied patches (I just checked that arch/x86/mm/init.c was patched) #cd /build/src #cp /boot/config-4.4.62-18.6-default .config #make rpm I then copied the rpms in /home/dummy/rpmbuild/RPMS/x86_64 to the real machine and installed them there. -- You are receiving this mail because: You are on the CC list for the bug.