[Discuss] I need a recommendation for a Linux-compatible high-end PC builder

Alan W. Irwin irwin at beluga.phys.uvic.ca
Thu Jan 25 18:34:59 PST 2018


On 2018-01-25 13:04-0800 Murray Strome wrote:

> This article might be of interest to you, and others following Meltdown and 
> Spectre issues:
> https://insights.ubuntu.com/2018/01/24/meltdown-spectre-and-ubuntu-what-you-need-to-know/?utm_source=marketo&utm_medium=email&mkt_tok=eyJpIjoiWW1FeFkyVm1OekZpTmpZNSIsInQiOiJ2YStZckp3RW42eEprWVlITFdyNU1vMmdsV2dvRzd5MHd5TjVZMVVHWjBqSlB3ZDZGQk5LNHU0MjNVSW9RUU9MUzc4OVBKbzJRV0tVc3NpTFlQWWN0emMrOGh0TmVUMDAzdGR4NVNseFQxYmk5dmtua2VNWGhKQnpNWEdXR0xuNyJ9

Thanks for that reference which is one of the better summaries I have
seen.  It is interesting that he states "All modern Intel processors
and IBM POWER7, POWER8, and POWER9" are affected by Meltdown without
specifically stating that AMD claims they are not subject to it (which
so far seems to be accepted wisdom).  Also, the article makes clear
that the firmware updates from Intel (and presumably the rest) are not
real fixes that make it impossible to speculatively fetch memory that
a programme does not have correct privileges to read.  We are going to
have to wait a long time for such definitive fixes (which will be
restricted to new hardware) and what is going on now with the firmware
upgrades are workarounds for the faults in existing hardware.  And my
own attitude is if you have a choice between a workaround written in
proprietary firmware versus one written in open source code for the
Linux kernel under intense public scrutiny, the latter approach will
typically always generate a much better result.

For others here, a better URL for this article is
<https://insights.ubuntu.com/2018/01/24/meltdown-spectre-and-ubuntu-what-you-need-to-know/>.
Note that
<https://blog.barkly.com/meltdown-spectre-patches-list-windows-update-help>
is also a excellent summary and continues to be updated (e.g., today).

With regard to the original question of who will build my dream computer, I am
virtually positive I will go with ZaReason in mid February.  They have been
extraordinarily responsive to my many questions, with the usual answer
being, yes, "we can adjust the build that way." For example, I needed
an esata port which their MB did not have by default and they were
willing to throw in a card to provide esata, gratis!

I am going to reluctantly delay (since I am looking forward to using a
computer that is roughly an order of magnitude faster than what I have
now) ordering from ZaReason until mid February, in the hope that the
remaining debates about *how* to mitigate the speculative execution
exploits (especially the ongoing debate about whether to use the open
software retpoline approach versus proprietary firmware upgrade to
mitigate Spectre Variant 2) will be more resolved by then. Also, even
though responsible articles emphasize that this mess caused by
hardware faults will be with us a long time with mitigations evolving
over a long time, it is also clear we are still in the middle of the
first burst of mitigations.  Thus, I am hoping delaying the buy until
mid February will somewhat reduce the large intensity of the Linux
kernel changes (and mistakes that will typically be the result of such
a large development intensity) that are going on now in an initial
response to these issues.

Alan
__________________________
Alan W. Irwin

Astronomical research affiliation with Department of Physics and Astronomy,
University of Victoria (astrowww.phys.uvic.ca).

Programming affiliations with the FreeEOS equation-of-state
implementation for stellar interiors (freeeos.sf.net); the Time
Ephemerides project (timeephem.sf.net); PLplot scientific plotting
software package (plplot.sf.net); the libLASi project
(unifont.org/lasi); the Loads of Linux Links project (loll.sf.net);
and the Linux Brochure Project (lbproject.sf.net).
__________________________

Linux-powered Science
__________________________



More information about the Discuss mailing list