Intel IFS Ready To Weed Out Faulty Silicon With Linux 6.2

[phoronix]

Phoronix: Intel IFS Ready To Weed Out Faulty Silicon With Linux 6.2

The x86/microcode changes that were merged this week into the Linux 6.2 kernel address prior shortcomings with the Intel In-Field Scan (IFS) driver so it's now deemed ready to help in spotting out faulty silicon across a fleet of systems in production or prior to commissioning new hardware...

https://www.phoronix.com/news/Linux-6.2-x86-Microcode

[coder]

This sounds good, but I have a nagging sense that it could've been at least partly motivated by SDDS (the new On-Demand capability that Intel added for unlocking CPU features by installing keys). You'd want to be able to test whether a given feature (still) works on your CPU, before buying a key to unlock it.

[lowflyer]

This article should have a disambiguing section to differentiate between IFS and IFS the Interplanetary File System. It may send unsuspecting readers down a rabbit hole...

[tildearrow]

This article should have a disambiguing section to differentiate between IFS and IFS the Interplanetary File System. It may send unsuspecting readers down a rabbit hole...

Wasn't the latter abbreviated as IPFS?

[oiaohm]

This sounds good, but I have a nagging sense that it could've been at least partly motivated by SDDS (the new On-Demand capability that Intel added for unlocking CPU features by installing keys). You'd want to be able to test whether a given feature (still) works on your CPU, before buying a key to unlock it.

Software Defined Silicon (SDSi) is what you were going with SDDS. If the presentation over SDSi by intel is to believed they started with Intel In-Field Scan (IFS). Of course if in-field scan found defect then comes the problem of how to turn the defective part off.

Intel was not wanting to release IFS and end up under a stack requests for replacement chips.

Most people think SDSi was first for selling updates to chips but the early presentations for it was here is refund for X chip being defective and here is downgrade that will make it function correctly. This being more targeted than microcode sledge hammer that alters complete batches.

I would say Intel In-Field Scan (IFS) and Software Defined Silicon (SDSi) are interlinked with mixed motivation for existence. Yes means to sell upgrades and means to reduce losses/court cases in case of defective chip. Intel presentations on these topics suggest both as motivation.

[coder]

This article should have a disambiguing section to differentiate between IFS and IFS the Interplanetary File System. It may send unsuspecting readers down a rabbit hole...

There's a more closely-related collision, which Intel Foundry Services -- what they call the semiconductor fabrication services that Intel is providing to other (fabless) semiconductor companies.

[oiaohm]

This article should have a disambiguing section to differentiate between IFS and IFS the Interplanetary File System. It may send unsuspecting readers down a rabbit hole...

InterPlanetary File System is IPFS that file system started with that acronym but at one point it mean "Internet Protocol File System" Yes "Internet Protocol" turns out to be trademarked and those making IPFS did not have approve to use it and that gets us to the name "InterPlanetary File System" just enough to change the name that trademark did not apply yet did not have to alter the codebase a lot.

Yes if InterPlanetary File System was the original name you would expect the acronym to be IFS. The IPFS is really a left over of it first name that the developers of IPFS could not use.

There's a more closely-related collision, which Intel Foundry Services -- what they call the semiconductor fabrication services that Intel is providing to other (fabless) semiconductor companies.

Intel "In-Field Scan (IFS)" I add that for a reason. IFS shorted form of the intel Internal hame Foundry Services is the internal name of Intel Foundry Services. So internally the two inside intel don't overlap at all. Yes if Intel Foundary Services are IFS then to keep everything constant "Intel In-Field Scan" should be IIFS.

Its very common in fact for acronyms to collide.

[lowflyer]

You guys tildearrow, coder and oiaohm point it out correctly:

Its very common in fact for acronyms to collide.

Why then is everybody trying to use them so excessively - usually while removing context? Trying to sound intelligent - they become unintelligible.

It's a simple but effective identification sign for vaporware and trolls. It works also for advertisements and product reviews.

I don't think that the black ink (or the black pixels on a screen) are so darn expensive that we must use acronyms to save some. Typing helpers and autocorrect will do the job for the keypresses.

[yump]

The hyperscalers are probably the most likely to notice and complain when their CPUs miscompute, but Intel ought to put this capability in all their products. Imagine if distros could include a 2-monthly on-reboot timer that would run the test and set a flag to warn the user that their CPU was going bad, and it was time to use the warranty or back off the overclock.

[oiaohm]

Why then is everybody trying to use them so excessively - usually while removing context? Trying to sound intelligent - they become unintelligible.

The most common answer to that is is annoying stupid. Its like why did the company Phillips change name to Philips yes this was cost cutting and time saving and space saving.

Formal writing rules tell you that the first usage of acronyms in a document should be acronym("what it is") or "what it is"(acronym) due to how often collisions happen. Please Note formal writing rules don't include the title.

The x86/microcode changes that were merged this week into the Linux 6.2 kernel address prior shortcomings with the Intel In-Field Scan (IFS) driver so it's now deemed ready to help in spotting out faulty silicon across a fleet of systems in production or prior to commissioning new hardware.

So the post this time was perfectly correct by formal writing rules. Now maybe we need to revise this on technical places like this to say "what it is"(acronym) need to be used in titles due to how often technical acronyms clash .

vaporware items you see not obeying formal writing rules with acronyms at all. I do try when I use acronyms to include the full non acronym form. I do human error and forget at times. Sometimes acronym without it long form is simple human error or belief that it that common.

GPL lot of people miss writing "GNU General Public License". Yes a horrible acronym creation rule I hate seeing applied. Yes IIFS due to it starting with 2 II you can drop a I just like "GNU General Public License" can drop a G. I had forgot that in my last post. So IIFS and IFS by acronym create rules can be exactly the same thing.

Constant rule of acronym in formal writing is a curse. Like it would be valid to do "GNU General Public License"(GGPL) then as long as you use GGPL all the way along your document its valid by formal writing. Example valid case by formal writing rules where you would be using GGPL as "GNU General Public License" acronym is if you were using "Game Programming Library"(GPL) acronym in your document to save space.

Acronyms are layers of cursed even that people like using them to save on typing.