Bug 28917

Mga 7 rpms:

SRPM:
kernel-linus-5.10.37-1.mga7.src.rpm


i586:
kernel-linus-5.10.37-1.mga7-1-1.mga7.i586.rpm
kernel-linus-devel-5.10.37-1.mga7-1-1.mga7.i586.rpm
kernel-linus-devel-latest-5.10.37-1.mga7.i586.rpm
kernel-linus-doc-5.10.37-1.mga7.noarch.rpm
kernel-linus-latest-5.10.37-1.mga7.i586.rpm
kernel-linus-source-5.10.37-1.mga7-1-1.mga7.noarch.rpm
kernel-linus-source-latest-5.10.37-1.mga7.noarch.rpm


x86_64:
kernel-linus-5.10.37-1.mga7-1-1.mga7.x86_64.rpm
kernel-linus-devel-5.10.37-1.mga7-1-1.mga7.x86_64.rpm
kernel-linus-devel-latest-5.10.37-1.mga7.x86_64.rpm
kernel-linus-doc-5.10.37-1.mga7.noarch.rpm
kernel-linus-latest-5.10.37-1.mga7.x86_64.rpm
kernel-linus-source-5.10.37-1.mga7-1-1.mga7.noarch.rpm
kernel-linus-source-latest-5.10.37-1.mga7.noarch.rpm

Whiteboard: (none) => MGA7TOO
Summary: Update request: kernel-linus-5.10.37-1.mga8 => Update request: kernel-linus-5.10.37-1.mga8/7

Moved to another system on the machine referenced in #28908 comment 9.
Kernel: 5.10.37-1.mga8 x86_64
Quad Core Intel Core i7-4790 [MT MCP]
NVIDIA GM204 [GeForce GTX 970] driver: nvidia v: 460.73.01

Installed the linus kernel alongside desktop-latest.
Booted linus.  nvidia driver built during boot.
Stress tests OK.
Sound and video and various desktop applications all working.

CC: (none) => tarazed25

(In reply to Len Lawrence from comment #2)
> Moved to another system on the machine referenced in #28908 comment 9.
> Kernel: 5.10.37-1.mga8 x86_64
> Quad Core Intel Core i7-4790 [MT MCP]
> NVIDIA GM204 [GeForce GTX 970] driver: nvidia v: 460.73.01
> 
> Installed the linus kernel alongside desktop-latest.
> Booted linus.  nvidia driver built during boot.
> Stress tests OK.
> Sound and video and various desktop applications all working.

Works also ok on MGA7 and MGA8. Both with current nvidia and 460.80 from Bug 28906

CC: (none) => ouaurelien

Sigh,

turns out there is some iommu regressions in 5.10.37 causing some systems to hang on boot... so more fixes are needed...

Keywords: (none) => feedback

as the iommu change does not affect everyone, and this is a reference kernel I will release it anyway for the security fixes it contains

Keywords: feedback => (none)

Advisory, added to svn:

type: security
subject: Updated kernel-linus packages fix security vulnerabilities
CVE:
 - CVE-2021-3491
 - CVE-2021-3506
 - CVE-2021-23133
 - CVE-2021-31440
 - CVE-2021-31829
 - CVE-2021-32399
 - CVE-2021-33034
src:
  8:
   core:
     - kernel-linus-5.10.37-1.mga8
  7:
   core:
     - kernel-linus-5.10.37-1.mga7
description: |
  This kernel-linus update is based on upstream 5.10.37 and fixes atleast the
  following security issues:

  It was discovered that the io_uring implementation of the Linux kernel did
  not properly enforce the MAX_RW_COUNT limit in some situations. A local
  attacker could use this to cause a denial of service (system crash) or
  execute arbitrary code (CVE-2021-3491).

  An out-of-bounds (OOB) memory access flaw was found in fs/f2fs/node.c in
  the f2fs module in the Linux kernel in versions before 5.12.0-rc4. A bounds
  check failure allows a local attacker to gain access to out-of-bounds
  memory leading to a system crash or a leak of internal kernel information
  (CVE-2021-3506).

  A race condition in Linux kernel SCTP sockets (net/sctp/socket.c) before
  5.12-rc8 can lead to kernel privilege escalation from the context of a
  network service or an unprivileged process. If sctp_destroy_sock is called
  without sock_net(sk)->sctp.addr_wq_lock then an element is removed from
  the auto_asconf_splist list without any proper locking. This can be
  exploited by an attacker with network service privileges to escalate to
  root or from the context of an unprivileged user directly if a
  BPF_CGROUP_INET_SOCK_CREATE is attached which denies creation of some
  SCTP socket.
  NOTE! This already had a fix in kernel-5.10.33, but that fix caused some
  systems to deadlock, so this is now fixed in a better way (CVE-2021-23133).

  bpf: Fix propagation of 32 bit unsigned bounds from 64 bit bounds
  (CVE-2021-31440).

  kernel/bpf/verifier.c in the Linux kernel through 5.12.1 performs undesirable
  speculative loads, leading to disclosure of stack content via side-channel
  attacks. The specific concern is not protecting the BPF stack area against
  speculative loads. Also, the BPF stack can contain uninitialized data that
  might represent sensitive information previously operated on by the kernel
  (CVE-2021-31829).

  net/bluetooth/hci_request.c in the Linux kernel through 5.12.2 has a race
  condition for removal of the HCI controller (CVE-2021-32399).

  In the Linux kernel before 5.12.4, net/bluetooth/hci_event.c has a
  use-after-free when destroying an hci_chan. This leads to writing an
  arbitrary value. (CVE-2021-33034).

  For other upstream fixes, see the referenced changelogs.
references:
 - https://bugs.mageia.org/show_bug.cgi?id=28917
 - https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.10.34
 - https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.10.35
 - https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.10.36
 - https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.10.37

Keywords: (none) => advisory

Phys Hardware - MGA8 - Gnome - Laptop AMD A6, R4 graphics

- cpupower-5.10.37-2.mga8.x86_64
- kernel-desktop-5.10.37-2.mga8-1-1.mga8.x86_64
- kernel-desktop-latest-5.10.37-2.mga8.x86_64

-- rebooted and confirmed 37-2

sleep works.  wifi working, etc.

works for me on AMD graphics platform.

CC: (none) => brtians1

An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGASA-2021-0215.html

Status: NEW => RESOLVED
Resolution: (none) => FIXED