[Linux-Desktops] Dell & latest CPU microcode to address CVE-2017-5715

Peter Reilly peter_reilly at wycliffe.org
Fri Jan 12 13:51:36 CST 2018


I have a Dell Optiplex 9010. 

Dell lists new BIOS firmware available to address CVE-2017-5715 aka
'Spectre Variant 2'.


Here is the BIOS update:


Which states:

- Update to the latest CPU microcode to address CVE-2017-5715.

Intel released new microcode here:

As documented on this site:

According to the release notes for the Intel download only these
processor families have new microcode

IVT C0        (06-3e-04:ed) 428->42a
SKL-U/Y D0    (06-4e-03:c0) ba->c2
BDW-U/Y E/F    (06-3d-04:c0) 25->28
HSW-ULT Cx/Dx    (06-45-01:72) 20->21
Crystalwell Cx    (06-46-01:32) 17->18
BDW-H E/G    (06-47-01:22) 17->1b
HSX-EX E0    (06-3f-04:80) 0f->10
SKL-H/S R0    (06-5e-03:36) ba->c2
HSW Cx/Dx    (06-3c-03:32) 22->23
HSX C0        (06-3f-02:6f) 3a->3b
BDX-DE V0/V1    (06-56-02:10) 0f->14
BDX-DE V2    (06-56-03:10) 700000d->7000011
KBL-U/Y H0    (06-8e-09:c0) 62->80
KBL Y0 / CFL D0    (06-8e-0a:c0) 70->80
KBL-H/S B0    (06-9e-09:2a) 5e->80
CFL U0        (06-9e-0a:22) 70->80
CFL B0        (06-9e-0b:02) 72->80
SKX H0        (06-55-04:b7) 2000035->200003c
GLK B0        (06-7a-01:01) 1e->22

So my processor is this:
Intel(R) Core(TM) i5-3550 CPU @ 3.30GHz

Which CPU World http://www.cpu-world.com/cgi-bin/CPUID.pl  lists as:
Family:    6 (06h)
Model:    58 (03Ah)
Stepping:    9 (09h)

So would match the microcode for 06-3a-09 which is indeed in the Intel
microcode available in microcode-20180108.tgz.

But it's not in the list that the release notes has:
cat releasenote | grep -i 06-3a-09

So it looks like Dell is releasing new BIOS updates that supply the
latest microcode, and possibly labeling the download as fixing
CVE-2017-5715 just because it's in the bundle, and not because it's been

Can anyone validate if this is actually correct?  Or was this possibly a
mixup just for this Dell desktop?

When I test with spectre-meltdown-checker from:
It shows:
*   Hardware (CPU microcode) support for mitigation:  NO

The part of the script that does that check fails here:
dd if=/dev/cpu/0/msr of=/dev/null bs=8 count=1 skip=9
dd: error reading '/dev/cpu/0/msr': Input/output error

Which seems (to me) to indicate that the microcode to mitigate Spectre
really isn't present.



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.us.dell.com/pipermail/linux-desktops/attachments/20180112/947c5e7e/attachment.html>

More information about the Linux-Desktops mailing list