Some Intel Skylake CPU models (SKL-U/Y, family 6, model 78, stepping 3)

have reports of system hangs when revision 0xdc of microcode, that is included

since microcode-20200609 update to address CVE-2020-0543, CVE-2020-0548,

and CVE-2020-0549, is applied[1].  In order to address this, microcode update

to the newer revision has been disabled by default on these systems,

and the previously published microcode revision 0xd6 is used by default

for the OS-driven microcode update.

[1] https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/issues/31

For the reference, SHA1 checksums of 06-4e-03 microcode files containing

microcode revisions in question are listed below:

 * 06-4e-03, revision 0xd6: 06432a25053c823b0e2a6b8e84e2e2023ee3d43e

 * 06-4e-03, revision 0xdc: cd1733458d187486999337ff8b51eeaa0cfbca6c

 * 06-4e-03, revision 0xe2: 41f4513cf563605bc85db38056ac430dec948366

Please contact your system vendor for a BIOS/firmware update that contains

the latest microcode version.  For the information regarding microcode versions

required for mitigating specific side-channel cache attacks, please refer

to the following knowledge base articles:

 * CVE-2017-5715 ("Spectre"):

   https://access.redhat.com/articles/3436091

 * CVE-2018-3639 ("Speculative Store Bypass"):

   https://access.redhat.com/articles/3540901

 * CVE-2018-3620, CVE-2018-3646 ("L1 Terminal Fault Attack"):

   https://access.redhat.com/articles/3562741

 * CVE-2018-12130, CVE-2018-12126, CVE-2018-12127, and CVE-2019-11091

   ("Microarchitectural Data Sampling"):

   https://access.redhat.com/articles/4138151

 * CVE-2019-0117 (Intel SGX Information Leak),

   CVE-2019-0123 (Intel SGX Privilege Escalation),

   CVE-2019-11135 (TSX Asynchronous Abort),

   CVE-2019-11139 (Voltage Setting Modulation):

   https://access.redhat.com/solutions/2019-microcode-nov

 * CVE-2020-0543 (Special Register Buffer Data Sampling),

   CVE-2020-0548 (Vector Register Data Sampling),

   CVE-2020-0549 (L1D Cache Eviction Sampling):

   https://access.redhat.com/solutions/5142751

 * CVE-2020-8695 (Information disclosure issue in Intel SGX via RAPL interface),

   CVE-2020-8696 (Vector Register Leakage-Active),

   CVE-2020-8698 (Fast Forward Store Predictor):

   https://access.redhat.com/articles/5569051

The information regarding enforcing microcode update is provided below.

To enforce usage of the latest 06-4e-03 microcode revision for a specific kernel

version, please create a file "force-intel-06-4e-03" inside

/lib/firmware/<kernel_version> directory, run

"/usr/libexec/microcode_ctl/update_ucode" to add it to firmware directory

where microcode will be available for late microcode update, and run

"dracut -f --kver <kernel_version>", so initramfs for this kernel version

is regenerated and the microcode can be loaded early, for example:

    touch /lib/firmware/3.10.0-862.9.1/force-intel-06-4e-03

    /usr/libexec/microcode_ctl/update_ucode

    dracut -f --kver 3.10.0-862.9.1

After that, it is possible to perform a late microcode update by executing

"/usr/libexec/microcode_ctl/reload_microcode" or by writing value "1" to

"/sys/devices/system/cpu/microcode/reload" directly.

To enforce addition of this microcode for all kernels, please create file

"/etc/microcode_ctl/ucode_with_caveats/force-intel-06-4e-03", run

"/usr/libexec/microcode_ctl/update_ucode" for enabling late microcode updates,

and "dracut -f --regenerate-all" for enabling early microcode updates:

    mkdir -p /etc/microcode_ctl/ucode_with_caveats

    touch /etc/microcode_ctl/ucode_with_caveats/force-intel-06-4e-03

    /usr/libexec/microcode_ctl/update_ucode

    dracut -f --regenerate-all

Please refer to /usr/share/doc/microcode_ctl/README.caveats for additional

information.