APPLIED: [SRU][Trusty][PATCH 0/7] Follow-up fixes for CVE-2018-3620/CVE-2018-3646
Kleber Souza
kleber.souza at canonical.com
Fri Aug 24 10:38:16 UTC 2018
On 08/22/18 08:40, Juerg Haefliger wrote:
> This patch series reenables NUMA balancing by disentangling _PAGE_NUMA from
> _PAGE_PROTNONE. It also fixes the remap_file_pages syscall to invert the PTE
> offset bits. This was an oversight since that code is no longer present in the
> Xenial kernel. Note that *not* inverting the offset bits doesn't cause any
> functional problems but simply doesn't protect swapped out entries against
> L1TF.
>
> I'm still trying to get access to a NUMA system to do some testing to
> verify that NUMA is working as expected (whatever that means).
>
> Compile-tested all supported architectures and verified that tomcat7
> (java) starts succesfully.
>
> Signed-off-by: Juerg Haefliger <juergh at canonical.com>
>
>
> Aneesh Kumar K.V (1):
> mm: Move change_prot_numa outside CONFIG_ARCH_USES_NUMA_PROT_NONE
>
> Cyrill Gorcunov (1):
> mm: x86 pgtable: drop unneeded preprocessor ifdef
>
> Hugh Dickins (1):
> x86,mm: fix pte_special versus pte_numa
>
> Juerg Haefliger (2):
> Revert "UBUNTU: [Config] disable NUMA_BALANCING"
> UBUNTU: SAUCE: x86/fremap: Invert the offset when converting to/from a
> PTE
>
> Mel Gorman (2):
> x86: require x86-64 for automatic NUMA balancing
> x86: define _PAGE_NUMA by reusing software bits on the PMD and PTE
> levels
>
> arch/x86/Kconfig | 2 +-
> arch/x86/include/asm/pgtable-2level.h | 10 ----
> arch/x86/include/asm/pgtable.h | 20 +++++--
> arch/x86/include/asm/pgtable_64.h | 46 ++++++++++++----
> arch/x86/include/asm/pgtable_types.h | 66 ++++++++++++-----------
> arch/x86/mm/pageattr-test.c | 2 +-
> debian.master/config/config.common.ubuntu | 4 +-
> include/asm-generic/pgtable.h | 8 ++-
> include/linux/mm.h | 2 +-
> include/linux/swapops.h | 2 +-
> mm/fremap.c | 6 +++
> mm/memory.c | 10 ++--
> mm/mempolicy.c | 5 +-
> 13 files changed, 112 insertions(+), 71 deletions(-)
>
Applied to trusty/master-next branch.
Thanks,
Kleber
More information about the kernel-team
mailing list