Intel is the worlds largest chip-marker. They primarily make computer CPUs with high-clockspeed and low core-count. This makes their chips very attractive for applications that require high single-core performance.
e-mail addresses from @intel.com have been a very frequent occurrence on the Linux Kernel Mailing List since the early 2000s. They are, and has long been, one of the largest corporate contributors to the Linux kernel. Intel is also making regular contributions to the Mesa graphics stack.
Intel did have some screw-ups with their i915 graphics driver for their GPUs between Linux kernel 5.3 and 5.5 which makes their iGPUs essentially unusable in 5.3.x and 5.4.x series kernels. Most of these issuers seem to be resolved as of 5.5rc4.
Intel is all-in-all one of the overall most free software friendly hardware manufacturers.
Intel is a large manufacturer of networking equipment. Their cheaper networking chips are found on higher-end consumer motherboards and their high-end network chips are frequently found on enterprise-level network cards. Older enterprise-level networking cards with Intel network chips, such as the quad gigabit HP NC364T, are readily available in the used market. Such cards are typically very well supported by the Linux kernel.
Wireless consumer network cards from Intel will typically work as they should right out of the box on GNU/Linux systems. Intel has long had the best free software kernel driver for wireless networking.
Bugs in the system
Intel CPUs have been plagued with a very high number of serious CPU bugs which require a lot of software workarounds. It is possible to disable many of the Linux kernel workarounds for Intel CPU bugs. Doing so is not advisable but quite possible for those who value pure performance over security.
Intel's the #1 chip-marker
Intel was the number one chip-marker for a decade before being knocked down by Samsung in 2017. Intel regained the #1 spot in 2019.
First And Second-class "Employees"
Intel is notable for its extensive use of contractors. More than half the labour-force are contractors at many of Intels larger sites.
Employees get blue badges. Contractors, who are in practice employees and have, in many cases, worked at Intel for 10+ years, get green badges.
"Green badges are treated as a sub-life form even though there are more green badges than blue badges on this site and Intel could not get the job done with out green badges. very poor collaboration between green badges and blue badges."
Blue badge employees get a lot of benefits the second-class green badge employees do not get. Blue badges pay less for lunch and they get free fruit. Green badges risk getting fired if they touch the, to them, forbidden fruit.
The clear distinction between blue/green badge employees is noteworthy because of the large percentage of in practice employees who are contractors. It is not like contractors/green badges are used for temporary positions lasting a month or two when a regular employee is sick. The green badges are a permanent part of the workforce who, at most of Intels locations, make up more than half the workforce. They are, in practice, regular employees who are not given regular jobs at Intel in order to save the cost of hiring employees directly.
Linus Torvalds on Intel CPUs
" A *competent* CPU engineer would fix this by making sure speculation doesn't happen across protection domains. Maybe even a L1 I$ that is keyed by CPL.
I think somebody inside of Intel needs to really take a long hard look at their CPU's, and actually admit that they have issues instead of writing PR blurbs that say that everything works as designed.
.. and that really means that all these mitigation patches should be written with "not all CPU's are crap" in mind.
Or is Intel basically saying "we are committed to selling you shit forever and ever, and never fixing anything"?
Because if that's the case, maybe we should start looking towards the ARM64 people more.
Please talk to management. Because I really see exactly two possibibilities:
- Intel never intends to fix anything
- these workarounds should have a way to disable them.
Which of the two is it?"