Skip to content
Do you have adblock enabled?
 
If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(The Register)   "If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce"   (theregister.com) divider line
    More: Cool, X86, Advanced Micro Devices, Linus Torvalds, Advanced Configuration and Power Interface, power management issues, Central processing unit, hasty patch, kernel developers  
•       •       •

1828 clicks; posted to STEM » on 27 Sep 2022 at 10:17 PM (8 weeks ago)   |   Favorite    |   share:  Share on Twitter share via Email Share on Facebook



14 Comments     (+0 »)
View Voting Results: Smartest and Funniest
 
2022-09-27 11:04:05 PM  
Prolly too late.
 
2022-09-27 11:06:20 PM  
What about Windows 95 era Plug 'n Pray,...er Play?
 
2022-09-27 11:18:50 PM  
Fark user imageView Full Size
 
2022-09-28 12:03:20 AM  
Intel or Incel?
 
2022-09-28 12:53:19 AM  
ACPI implies that you've got your x86 cores fully powered. That is actually rather arcane in the age of race to idle, but then again Linux and freeBSD do often try to out-arcane each other.

/not really a fair fight once Stallman starts flicking boogers
//and Torvalds commences his litany of swearing
 
2022-09-28 4:13:19 AM  
So does this mean bitcoin mining will be more efficient? Because I was thinking about buying the NFT of bitcoin server output.
 
2022-09-28 5:29:56 AM  
His patch links to some benchmark numbers, but the bottom line is that the minimum throughput increases by about 14 times, and the mean throughput by just over a half.

Putin isn't the only one who needs to hang.
 
2022-09-28 6:08:04 AM  

PoweredByIrony: So does this mean bitcoin mining will be more efficient? Because I was thinking about buying the NFT of bitcoin server output.


"Your lava tubes are perfect."
"You are wise, Starfleet. We will hear you out."
 
2022-09-28 6:45:36 AM  
"If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce"

Fark user imageView Full Size
 
2022-09-28 2:51:17 PM  
The result has been a hasty patch by Intel's Dave Hansen, which doesn't remove the workaround but limits it to Intel chips, where it shouldn't cause any problem as they use a different method to switch to idle.

Oh come the fark on. Just remove it. Don't put in a check for a different processor type. This type of shiat makes no sense as a fix. And could have unintended side effects.

Dave Hansen, you suck.
 
2022-09-28 7:49:15 PM  

Quantumbunny: The result has been a hasty patch by Intel's Dave Hansen, which doesn't remove the workaround but limits it to Intel chips, where it shouldn't cause any problem as they use a different method to switch to idle.

Oh come the fark on. Just remove it. Don't put in a check for a different processor type. This type of shiat makes no sense as a fix. And could have unintended side effects.

Dave Hansen, you suck.


hacking the hack is a perfectly cromulent method according to most professional developers
 
2022-09-28 8:17:42 PM  

moos: Quantumbunny: The result has been a hasty patch by Intel's Dave Hansen, which doesn't remove the workaround but limits it to Intel chips, where it shouldn't cause any problem as they use a different method to switch to idle.

Oh come the fark on. Just remove it. Don't put in a check for a different processor type. This type of shiat makes no sense as a fix. And could have unintended side effects.

Dave Hansen, you suck.

hacking the hack is a perfectly cromulent method according to most professional developers


Also, since a lot of work for a long time has been based how things are based on that being in place, just flat out removing it might well do something unexpected, even if it's fine with it just being disabled.  I have seen dumber shiat happen.  When you hear "Well that's ancient useless code, and it's farking stuff up, why is this even still in here?  I'll just take it out." those words are often followed by "Shiat!  What..?  Why is this...?  HOW?"
 
2022-09-28 8:21:13 PM  

moos: Quantumbunny: The result has been a hasty patch by Intel's Dave Hansen, which doesn't remove the workaround but limits it to Intel chips, where it shouldn't cause any problem as they use a different method to switch to idle.

Oh come the fark on. Just remove it. Don't put in a check for a different processor type. This type of shiat makes no sense as a fix. And could have unintended side effects.

Dave Hansen, you suck.

hacking the hack is a perfectly cromulent method according to most professional developers


It's the Linux kernel, it's literally open source. Why would you just switch the check to a different check? That's objectively stupid because it still wastes a cycle or two, and sends what are hopefully ignored calls to another processor type.

Cleaning out the check entirely makes sense. Putting in a more specific check to the K2s for the theory that someone, somewhere may still run it, makes sense. Switching it to Intel CPUs is the kind of thing I'd never approve as a solution from my developers.
 
2022-09-28 9:09:48 PM  

Quantumbunny: moos: Quantumbunny: The result has been a hasty patch by Intel's Dave Hansen, which doesn't remove the workaround but limits it to Intel chips, where it shouldn't cause any problem as they use a different method to switch to idle.

Oh come the fark on. Just remove it. Don't put in a check for a different processor type. This type of shiat makes no sense as a fix. And could have unintended side effects.

Dave Hansen, you suck.

hacking the hack is a perfectly cromulent method according to most professional developers

It's the Linux kernel, it's literally open source. Why would you just switch the check to a different check? That's objectively stupid because it still wastes a cycle or two, and sends what are hopefully ignored calls to another processor type.

Cleaning out the check entirely makes sense. Putting in a more specific check to the K2s for the theory that someone, somewhere may still run it, makes sense. Switching it to Intel CPUs is the kind of thing I'd never approve as a solution from my developers.


I'm not disagreeing with you.. I was using the topic to biatch about what I consider poor behavior by people I know while posting what I consider to be an idiotic line of thinking.

FOTU is the most common reason supporting such approaches, and the thing @Some_Junkie_Cosmonaut mentions is real, but "Hey, try it now" should be limited to test environments.
 
Displayed 14 of 14 comments

View Voting Results: Smartest and Funniest

This thread is closed to new comments.

Continue Farking




On Twitter


  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.