Home    |    View Topics    |    Search    |    Contact Us    |   



Category:   OS (Linux)  >   Linux Kernel Vendors:
(Oracle Issues Fix for Linux Kernel) Intel CPUs Let Local Users Obtain or Infer Portions of L1 Cache Memory on the Target System
SecurityTracker Alert ID:  1041653
SecurityTracker URL:
CVE Reference:   CVE-2018-3620   (Links to External Site)
Date:  Sep 14 2018
Impact:   Disclosure of system information, Disclosure of user information
Fix Available:  Yes  Vendor Confirmed:  Yes  
Version(s): Oracle Linux 5 ELS, Oracle Linux 6
Description:   A vulnerability was reported in Intel CPUs. A local user can obtain potentially sensitive information from system memory. Linux Kernel is affected.

A local user can conduct a speculative execution side-channel attack against the L1 cache to infer potentially sensitive information from L1 cache memory on the target system.

This method is referred to as the "L1 Terminal Fault (L1TF)" or "Foreshadow" attack.

Microprocessors that support Intel Software Guard Extensions (SGX) are affected [CVE-2018-3615].

Other Intel microprocessors that use speculative execution and address translations may be affected [CVE-2018-3620].

Other Intel microprocessors that use speculative execution and address translations and that host virtual systems may be affected [CVE-2018-3646].

The following models are affected:

Core i3 processor (45nm and 32nm)
Core i5 processor (45nm and 32nm)
Core i7 processor (45nm and 32nm)
Core M processor family (45nm and 32nm)
2nd generation Core processors
3rd generation Core processors
4th generation Core processors
5th generation Core processors
6th generation Core processors
7th generation Core processors
8th generation Core processors
Core X-series Processor Family for X99 platforms
Core X-series Processor Family for X299 platforms
Xeon processor 3400 series
Xeon processor 3600 series
Xeon processor 5500 series
Xeon processor 5600 series
Xeon processor 6500 series
Xeon processor 7500 series
Xeon Processor E3 Family
Xeon Processor E3 v2 Family
Xeon Processor E3 v3 Family
Xeon Processor E3 v4 Family
Xeon Processor E3 v5 Family
Xeon Processor E3 v6 Family
Xeon Processor E5 Family
Xeon Processor E5 v2 Family
Xeon Processor E5 v3 Family
Xeon Processor E5 v4 Family
Xeon Processor E7 Family
Xeon Processor E7 v2 Family
Xeon Processor E7 v3 Family
Xeon Processor E7 v4 Family
Xeon Processor Scalable Family
Xeon Processor D (1500, 2100)

The original advisory is available at:

Raoul Strackx, Jo Van Bulck, and Frank Piessens of imec-DistriNet, KU Leuven, Marina Minkin and Mark Silberstein of Technion, Ofir Weisse, Daniel Genkin, Baris Kasikci, and Thomas F. Wenisch of University of Michigan, and Yuval Yarom of University of Adelaide and Data61 reported this vulnerability.

Impact:   A local user can obtain potentially sensitive information from L1 cache memory on the target system.
Solution:   Oracle has issued a fix for CVE-2018-3620 for Linux Kernel.

The Oracle Linux advisory is available at:

Vendor URL: (Links to External Site)
Cause:   Access control error

Message History:   This archive entry is a follow-up to the message listed below.
Aug 14 2018 Intel CPUs Let Local Users Obtain or Infer Portions of L1 Cache Memory on the Target System

 Source Message Contents

Subject:  [El-errata] ELSA-2018-4214 Important: Oracle Linux 6 Unbreakable Enterprise kernel security update

Oracle Linux Security Advisory ELSA-2018-4214

The following updated rpms for Oracle Linux 6 have been uploaded to the 
Unbreakable Linux Network:




Description of changes:

- x86/speculation/l1tf: Fix overflow in l1tf_pfn_limit() on 32bit 
(Vlastimil Babka)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Exempt zeroed PTEs from inversion (Sean 
Christopherson)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Protect PAE swap entries against L1TF (Vlastimil 
Babka)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Extend 64bit swap file size limit (Vlastimil 
Babka)  [Orabug: 28505519]  {CVE-2018-3620}
- mm, fremap: mitigate L1TF in remap_file_pages (Daniel Jordan) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation: Don't mark cpu_no_l1tf __initconst (Boris Ostrovsky) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/mm/kmmio: Make the tracer robust against L1TF (Andi Kleen) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Make pmd/pud_mknotpresent() invert (Andi Kleen) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Invert all not present mappings (Andi Kleen) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/bugs: Move the l1tf function and define pr_fmt properly (Konrad 
Rzeszutek Wilk)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Limit swap file size to MAX_PA/2 (Andi Kleen) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Disallow non privileged high MMIO PROT_NONE 
mappings (Andi Kleen)  [Orabug: 28505519]  {CVE-2018-3620}
- mm/pagewalk.c: prevent positive return value of walk_page_test() from 
being passed to callers (Naoya Horiguchi)  [Orabug: 28505519] 
- mm: pagewalk: fix misbehavior of walk_page_range for vma(VM_PFNMAP) 
(Naoya Horiguchi)  [Orabug: 28505519]  {CVE-2018-3620}
- pagewalk: improve vma handling (Naoya Horiguchi)  [Orabug: 28505519] 
- mm/pagewalk: remove pgd_entry() and pud_entry() (Naoya Horiguchi) 
[Orabug: 28505519]  {CVE-2018-3620}
- mm/pagewalk.c: fix walk_page_range() access of wrong PTEs (Chen LinX) 
[Orabug: 28505519]  {CVE-2018-3620}
- mm/pagewalk.c: walk_page_range should avoid VM_PFNMAP areas (Cliff 
Wickman)  [Orabug: 28505519]  {CVE-2018-3620}
- pagewalk: don't look up vma if walk->hugetlb_entry is unused (KOSAKI 
Motohiro)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Add sysfs reporting for l1tf (Andi Kleen) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/bugs: Concentrate bug reporting into a separate function (Konrad 
Rzeszutek Wilk)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/bugs: Concentrate bug detection into a separate function (Konrad 
Rzeszutek Wilk)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/cpufeature: Add X86_FEATURE_IA32_ARCH_CAPS and 
X86_FEATURE_IBRS_ATT (Konrad Rzeszutek Wilk)  [Orabug: 28505519] 
- x86/speculation/l1tf: Make sure the first page is always reserved 
(Andi Kleen)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Fix up pte->pfn conversion for PAE (Michal 
Hocko)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Protect PROT_NONE PTEs against speculation (Andi 
Kleen)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Protect swap entries against L1TF (Linus 
Torvalds)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Change order of offset/type in swap entry (Linus 
Torvalds)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/mm: Fix swap entry comment and macro (Dave Hansen)  [Orabug: 
28505519]  {CVE-2018-3620}
- x86/mm: Move swap offset/type up in PTE to work around erratum (Dave 
Hansen)  [Orabug: 28505519]  {CVE-2018-3620}
- mm: x86 pgtable: drop unneeded preprocessor ifdef (Cyrill Gorcunov) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/speculation/l1tf: Increase 32bit PAE __PHYSICAL_PAGE_SHIFT (Andi 
Kleen)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/mm: Fix regression with huge pages on PAE (Kirill A. Shutemov) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/asm: Fix pud/pmd interfaces to handle large PAT bit (Toshi Kani) 
[Orabug: 28505519]  {CVE-2018-3620}
- x86/asm: Add pud/pmd mask interfaces to handle large PAT bit (Toshi 
Kani)  [Orabug: 28505519]  {CVE-2018-3620}
- x86/asm: Move PUD_PAGE macros to page_types.h (Toshi Kani)  [Orabug: 
28505519]  {CVE-2018-3620}
- x86/cpufeature: uniquely define *IA32_ARCH_CAPS and *IBRS_ATT (Daniel 
Jordan)  [Orabug: 28505519]  {CVE-2018-3620}
- Disable kaiser if the cpu is not vulnerable to X86_BUG_CPU_MELTDOWN 
(Kanth Ghatraju)  [Orabug: 28001909]
- x86/pti: Do not enable PTI on CPUs which are not vulnerable to 
Meltdown (David Woodhouse)  [Orabug: 28001909]
- x86/msr: Add definitions for IA32_ARCH_CAPABILITIES MSR (David 
Woodhouse)  [Orabug: 28001909]
- Add driver auto probing for x86 features v4 (Andi Kleen)  [Orabug: 
- mm/mempolicy: fix use after free when calling get_mempolicy (zhong 
jiang)  [Orabug: 28022110]  {CVE-2018-10675}
- xen-netback: do not requeue skb if xenvif is already disconnected 
(Dongli Zhang)  [Orabug: 28247698]
- posix-timer: Properly check sigevent->sigev_notify (Thomas Gleixner) 
[Orabug: 28481397]  {CVE-2017-18344}

El-errata mailing list

Go to the Top of This SecurityTracker Archive Page

Home   |    View Topics   |    Search   |    Contact Us

This web site uses cookies for web analytics. Learn More

Copyright 2020, LLC