Friday, August 24th 2018

Intel Updates Microcode License Deleting "No-Benchmarks" Clause

A huge controversy erupted earlier this week as the license governing Intel's latest CPU microcode updates redistribution inserted a legally-binding clause that gagged its customers from publishing benchmarks or comparative testing that showed the performance impact of microcode updates that mitigate security vulnerabilities in Intel processors. Intel has since started reaching out to media sites. "We are updating the license now to address this and will have a new version available soon. As an active member of the open source community, we continue to welcome all feedback," the opening remarks from the Intel spokesperson read. Not long after, Intel updated the license terms to have just three conditions:
Redistribution and use in binary form, without modification, are permitted, provided that the following conditions are met:
  • Redistributions must reproduce the above copyright notice and the following disclaimer in the documentation and/or other materials provided with the distribution.
  • Neither the name of Intel Corporation nor the names of its suppliers may be used to endorse or promote products derived from this software without specific prior written permission.
  • No reverse engineering, decompilation, or disassembly of this software is permitted.
"Binary form" includes any format that is commonly used for electronic conveyance that is a reversible, bit-exact translation of binary representation to ASCII or ISO text, for example "uuencode."


Sounds good so far, however, the language in the opening remarks got us thinking, whether Intel has two different licenses targeted at two different groups:
1. Big cloud-computing providers such as Amazon Web Services (AWS), Microsoft Azure, Alibaba, Google, etc; and
2. The "open source community" that downloads the microcode update from Intel website and posts performance numbers in news blogs.

TechPowerUp doesn't just cater to the open source community. We are equally interested to know whether cloud-computing providers are gagged from disclosing performance impact of microcode updates, because a lot more money and jobs are lost as a result of lowered performance/$ or stunted performance/$ growth from cloud-providers.

We are awaiting a specific affirmative/negative from the Intel spokesperson on whether a cloud-computing firm like Amazon Web Services (AWS), for example, is free to disclose performance impact of the latest microcode update to its downstream customers without violating the applicable license governing the microcode update distribution.

Watch this space.
Add your own comment

24 Comments on Intel Updates Microcode License Deleting "No-Benchmarks" Clause

#1
Vayra86
Drunk driving at its finest
Posted on Reply
#2
AlwaysHope
Chipzilla upset about self inflicted ignorance with Hardware based security on their products...
SUCK IT UP & cop the blow back!
Posted on Reply
#3
FordGT90Concept
"I go fast!1!11!1!"
They kind of had to. There's no way they could enforce that clause in their EULA.
Posted on Reply
#4
techy1
obviously only a public outcry and shaming works these days - do not hesitate to click, comment and share BS like that.
Posted on Reply
#5
R0H1T
techy1
obviously only a public outcry and shaming works these days - do not hesitate to click, comment and share BS like that.
Too bad politicains aren't affected, & no I'm not targeting anyone specific - they're all like that, shameless!
Posted on Reply
#6
GlacierNine
FordGT90Concept
They kind of had to. There's no way they could enforce that clause in their EULA.
Even if they could, and even if they *won* the case, how many big server companies would immediately abandon Xeon platforms if Intel started threatening to sue them for talking about performance degradation like this?

Not only was it not enforceable, it would also have been suicide to enforce it even if it were possible.
Posted on Reply
#7
Tsukiyomi91
I say both parties are partially at fault. Why enforce such controversial rules when you know people are gonna bite back? ==
Posted on Reply
#8
R-T-B
techy1
obviously only a public outcry and shaming works these days - do not hesitate to click, comment and share BS like that.
That's a double edged sword, I'm tempted to point out. There are plenty of instances where "public outcry" has led to something bad happening.

In this case though, yes. Good outcome.
Posted on Reply
#9
VulkanBros
Tsukiyomi91
I say both parties are partially at fault. Why enforce such controversial rules when you know people are gonna bite back? ==
They assume no one reads the EULA´s - this is sick
Posted on Reply
#11
techy1
R-T-B
That's a double edged sword, I'm tempted to point out. There are plenty of instances where "public outcry" has led to something bad happening.

In this case though, yes. Good outcome.
I should have specified that I am talking about Tech industry only. I can not remember an instance where public outcry and shaming in Tech industry was the "other edge" of the sword.. oh yea - we did nuke the nvidias GPP... how could we - it was for our own sake and would beffit us (as nvida after cancelation stated :D)
GinoLatino
What where they thinking?
they were thinking they will get away with it.
Posted on Reply
#12
DarkStalker
GinoLatino
What where they thinking?
my dear friends... this clearly shows that not all people at Intel have IQ higher then 85
Posted on Reply
#13
R-T-B
techy1
I should have specified that I am talking about Tech industry only. I can not remember an instance where public outcry and shaming in Tech industry was the "other edge" of the sword..
I can but they mostly relate to software development. That is "tech" but OT for here.
Posted on Reply
#14
GlacierNine
Tsukiyomi91
I say both parties are partially at fault. Why enforce such controversial rules when you know people are gonna bite back? ==
What are you talking about? The only party responsible here is Intel.

It's Intel's issue that their CPUs are vulnerable.
It's Intel's issue that their fixes hurt performance.
It's Intel's issue that the license had this clause added to it.

What exactly did anyone do wrong here except for Intel? It certainly wasn't datacenters publishing benchmarks. It certainly wasn't consumers taking issue with a gag clause, and once you've run through those things, what else is there?
Posted on Reply
#15
eidairaman1
The Exiled Airman
GlacierNine
What are you talking about? The only party responsible here is Intel.

It's Intel's issue that their CPUs are vulnerable.
It's Intel's issue that their fixes hurt performance.
It's Intel's issue that the license had this clause added to it.

What exactly did anyone do wrong here except for Intel? It certainly wasn't datacenters publishing benchmarks. It certainly wasn't consumers taking issue with a gag clause, and once you've run through those things, what else is there?
Just means they want people in a bed of lies with such clauses. So nope, bad business practice.
Posted on Reply
#16
GlacierNine
eidairaman1
Just means they want people in a bed of lies with such clauses. So nope, bad business practice.
Again, what are you talking about? Who is "they"?

Tsukiyomi's post says "both parties" - the only parties involved with *any* of this are:

1 - Intel.
2 - Datacenter customers.
3 - Regular consumers.

Of those 3, only Intel have done anything wrong.
Posted on Reply
#17
lexluthermiester
FordGT90Concept
They kind of had to. There's no way they could enforce that clause in their EULA.
Another big company trying to enforce something that isn't right. Shocking.
Posted on Reply
#18
GorbazTheDragon
I like how it was worded over on wccf, one way to "make sure everyone benches the hell out of the microcode updates"
Posted on Reply
#19
lexluthermiester
GorbazTheDragon
I like how it was worded over on wccf, one way to "make sure everyone benches the hell out of the microcode updates"
Right? It's not like everyone hasn't been doing the benchmarks anyway.
Posted on Reply
#20
Ahhzz
Tsukiyomi91
I say "both parties" are partially at fault. Why enforce such controversial rules when you know people are gonna bite back? ==
what? what possible fault lies in a company complaining that a provider introduced code to their product substantially reducing that products efficiency?
Posted on Reply
#21
HD64G
lexluthermiester
Another big company trying to enforce something that isn't right. Shocking.
Getting used to malicious business-mass market practices to the point where one doesn't react at all shows much about human kind's getting apathetic and refusing to react against its "rapists" who do that by brute force just to get more money.
Posted on Reply
#22
GlacierNine
HD64G
Getting used to malicious business-mass market practices to the point where one doesn't react at all shows much about human kind's getting apathetic and refusing to react against its "rapists" who do that by brute force just to get more money.
Agreed. Do these people genuinely not realise how sad and pathetic it is, to defend abuse of power by pretending it's an acceptable way for things to be?

Like, what grinds a person down to that degree, that they can believe something so clearly, completely nonsensical?
Posted on Reply
#23
Hossein Almet
Well, after updating the BIOS, the sequential read and the random read of my 960 EVO drop by some 12%, the sequential write and random write stay the same.
Posted on Reply
#24
lemonadesoda
Benchmarkgate. I only just caught this news. Shame on Intel.
Posted on Reply
Add your own comment