All of lore.kernel.org
 help / color / mirror / Atom feed
* Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case?
@ 2018-05-19 19:53 Christian Krüger
  2018-05-23 10:53 ` Pavel Machek
  0 siblings, 1 reply; 4+ messages in thread
From: Christian Krüger @ 2018-05-19 19:53 UTC (permalink / raw)
  To: linux-kernel

Hi,

Since the old "in-order-execution" Intel CPUs like the Intel Atom N270 
(known for being installed in many Netbooks and Nettops) are not 
sensitive for "Meltdown" & "Spectre" , wouldn't it be a good idea to 
exclude these anyway "weak" CPUs from the costly patches by default?

Browsing the web, I can "feel the difference" if the matching kernel 
options are applied on such a device.

BR
C.Krüger

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case?
  2018-05-19 19:53 Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case? Christian Krüger
@ 2018-05-23 10:53 ` Pavel Machek
  2018-05-23 17:07   ` Johannes Hirte
  0 siblings, 1 reply; 4+ messages in thread
From: Pavel Machek @ 2018-05-23 10:53 UTC (permalink / raw)
  To: Christian Krüger; +Cc: linux-kernel

[-- Attachment #1: Type: text/plain, Size: 766 bytes --]

On Sat 2018-05-19 21:53:02, Christian Krüger wrote:
> Hi,
> 
> Since the old "in-order-execution" Intel CPUs like the Intel Atom N270
> (known for being installed in many Netbooks and Nettops) are not sensitive
> for "Meltdown" & "Spectre" , wouldn't it be a good idea to exclude these
> anyway "weak" CPUs from the costly patches by default?
> 
> Browsing the web, I can "feel the difference" if the matching kernel options
> are applied on such a device.

Can you also measure the difference? Placebo effect is hard to avoid.

But yes, we do not need to do workarounds on non-buggy machines...

								Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case?
  2018-05-23 10:53 ` Pavel Machek
@ 2018-05-23 17:07   ` Johannes Hirte
  2018-05-23 18:40     ` Christian Krüger
  0 siblings, 1 reply; 4+ messages in thread
From: Johannes Hirte @ 2018-05-23 17:07 UTC (permalink / raw)
  To: Pavel Machek; +Cc: Christian Krüger, linux-kernel

On 2018 Mai 23, Pavel Machek wrote:
> On Sat 2018-05-19 21:53:02, Christian Krüger wrote:
> > Hi,
> > 
> > Since the old "in-order-execution" Intel CPUs like the Intel Atom N270
> > (known for being installed in many Netbooks and Nettops) are not sensitive
> > for "Meltdown" & "Spectre" , wouldn't it be a good idea to exclude these
> > anyway "weak" CPUs from the costly patches by default?
> > 
> > Browsing the web, I can "feel the difference" if the matching kernel options
> > are applied on such a device.
> 
> Can you also measure the difference? Placebo effect is hard to avoid.
> 
> But yes, we do not need to do workarounds on non-buggy machines...
> 
> 								Pavel

On my Atom N270 there doesn't seem to be any workaround active with
kernel 4.14.42:

localhost ~ # cat /sys/devices/system/cpu/vulnerabilities/*
Not affected
Not affected
Not affected

Christian, did you verified the mitigations are active on your system?
What kernels are affected? 

-- 
Regards,
  Johannes

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case?
  2018-05-23 17:07   ` Johannes Hirte
@ 2018-05-23 18:40     ` Christian Krüger
  0 siblings, 0 replies; 4+ messages in thread
From: Christian Krüger @ 2018-05-23 18:40 UTC (permalink / raw)
  To: linux-kernel

Am 23.05.2018 um 19:07 schrieb Johannes Hirte:
> On 2018 Mai 23, Pavel Machek wrote:
>> On Sat 2018-05-19 21:53:02, Christian Krüger wrote:
>>> Hi,
>>>
>>> Since the old "in-order-execution" Intel CPUs like the Intel Atom N270
>>> (known for being installed in many Netbooks and Nettops) are not sensitive
>>> for "Meltdown" & "Spectre" , wouldn't it be a good idea to exclude these
>>> anyway "weak" CPUs from the costly patches by default?
>>>
>>> Browsing the web, I can "feel the difference" if the matching kernel options
>>> are applied on such a device.
>>
>> Can you also measure the difference? Placebo effect is hard to avoid.
>>
>> But yes, we do not need to do workarounds on non-buggy machines...
>>
>> 								Pavel
> 
> On my Atom N270 there doesn't seem to be any workaround active with
> kernel 4.14.42:
> 
> localhost ~ # cat /sys/devices/system/cpu/vulnerabilities/*
> Not affected
> Not affected
> Not affected
> 
> Christian, did you verified the mitigations are active on your system?
> What kernels are affected?
> 


$ uname -a
Linux user-device 3.13.0-143-generic #192-Ubuntu SMP Tue Feb 27 10:46:44 
UTC 2018 i686 i686 i686 GNU/Linux

$ cat /sys/devices/system/cpu/vulnerabilities/*
Vulnerable
Mittigation: OSB (observable speculation barrier, Intel v6)
Mittigation: Full generic repoline

OK, so I've updated the OS on my (usually off-line) Netbook:

$ uname -a
Linux user-device 3.13.0-149-generic #199-Ubuntu SMP Thu May 17 10:12:57 
UTC 2018 i686 i686 i686 GNU/Linux

$ cat /sys/devices/system/cpu/vulnerabilities/*
Not affected
Not affected
Not affected
Not affected

Conclusion:
Don't bother others before not checking the latest version. (Actually 
this is the first time it made a difference... ).

Sorry for the noise and thanks for listening and all the great work!
Christian

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2018-05-23 18:40 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-05-19 19:53 Suggestion: „spectre_v2=off“ and „nopti“ per default in "Intel Atom N270" case? Christian Krüger
2018-05-23 10:53 ` Pavel Machek
2018-05-23 17:07   ` Johannes Hirte
2018-05-23 18:40     ` Christian Krüger

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.