ACK: [SRU Focal, Jammy, OEM-5.17, OEM-6.0, OEM-6.1, Kinetic, Lunar 0/1] LP: #2023220
Cengiz Can
cengiz.can at canonical.com
Thu Jun 8 12:53:24 UTC 2023
- Previous message (by thread): [SRU Jammy, OEM-5.17, OEM-6.0, OEM-6.1, Kinetic/Lunar] x86/mm: Avoid incomplete Global INVLPG flushes
- Next message (by thread): ACK: [SRU Focal, Jammy, OEM-5.17, OEM-6.0, OEM-6.1, Kinetic, Lunar 0/1] LP: #2023220
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
On Wed, 2023-06-07 at 17:33 -0300, Thadeu Lima de Souza Cascardo wrote:
> [Impact]
> When PCIDs are enabled on Alder Lake and Raptor Lake, INVLPG will not
> flush
> the global TLB entries. This can lead to info leak or undefined
> behavior.
>
> [Mitigation]
> Boot with nopcid on affected systems.
>
> [Backports]
> Clean cherry-pick, except for 5.4, where backport from linux-5.4.y
> was picked.
>
> [Test case]
> There is no particular test case.
>
> [Potential regressions]
> This affects performance on the affected systems. TLB behavior could
> also be
> affected.
>
> Dave Hansen (1):
> x86/mm: Avoid incomplete Global INVLPG flushes
Acked-by: Cengiz Can <cengiz.can at canonical.com>
>
> arch/x86/include/asm/intel-family.h | 5 +++++
> arch/x86/mm/init.c | 25 +++++++++++++++++++++++++
> 2 files changed, 30 insertions(+)
>
> --
> 2.34.1
>
>
- Previous message (by thread): [SRU Jammy, OEM-5.17, OEM-6.0, OEM-6.1, Kinetic/Lunar] x86/mm: Avoid incomplete Global INVLPG flushes
- Next message (by thread): ACK: [SRU Focal, Jammy, OEM-5.17, OEM-6.0, OEM-6.1, Kinetic, Lunar 0/1] LP: #2023220
- Messages sorted by:
[ date ]
[ thread ]
[ subject ]
[ author ]
More information about the kernel-team
mailing list