Bug 1537625

Summary: PIII reported to carry "bugs: cpu_meltdown spectre_v1 spectre_v2"
Product: [Fedora] Fedora Reporter: Ralf Corsepius <rc040203>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED WONTFIX QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 27CC: airlied, ajax, bskeggs, dhgutteridge, ewk, fweimer, hdegoede, ichavero, itamar, jarodwilson, jforbes, jglisse, john.j5live, jonathan, josef, kernel-maint, labbott, linville, mchehab, mjg59, steved
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-01-23 18:40:51 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---
Bug Depends On:    
Bug Blocks: 1489998    

Description Ralf Corsepius 2018-01-23 16:27:05 UTC
Description of problem:

# cat /proc/cpuinfo  | grep bugs
bugs		: cpu_meltdown spectre_v1 spectre_v2

# cat /proc/cpuinfo  | grep model
model		: 8
model name	: Pentium III (Coppermine)


Version-Release number of selected component (if applicable):
kernel-4.14.13-300.fc27.i686

Additional info:
It would be news to me, if old Pentium IIIs are were subject to meltdown and spectre_v1|v2.

Seems to me as if the logic to set these flags doesn't work at all.

Comment 1 Ralf Corsepius 2018-01-23 16:42:23 UTC
Same on an Atom N270:

bugs		: cpu_meltdown spectre_v1 spectre_v2

model		: 28
model name	: Intel(R) Atom(TM) CPU N270   @ 1.60GHz

Comment 2 Justin M. Forbes 2018-01-23 18:40:51 UTC
Your PIII is impacted by Meltdown and potentially spectre, everything back to the P6 architecture (1995, P Pro and P III).  The requirement is out of order execution. You Atom is old enough that it is not, they didn't add that feature to Atom until 2013.  Realistically though, Pre 2013 atom CPUs still in use is such a small number of systems that it is not worth putting in rules to determine which ones are vulnerable and which are not. Patches are floating around for mitigation on 32bit Intel, but they are going to hurt performance on already seriously old systems.  When they are ready, we will carry them.

Closing this as won't fix because we aren't going to change the reporting, and this bug is against the reporting.   We do intend to include mitigation for 32bit systems as it becomes available, so that part should be fixed.

Comment 3 Ralf Corsepius 2018-01-24 07:03:09 UTC
With all due respect, Justin, what do you expect people to think about your reply?

I think, RHAT and RHAT's kernel devs need to change their attitude. RHAT needs tp comprehend Fedora is not a RHAT play ball, colony or egotrip.

That said, I consider your response a manifestation of disrespect towards the community.

Comment 4 Josh Boyer 2018-01-24 07:31:17 UTC
(In reply to Ralf Corsepius from comment #3)
> With all due respect, Justin, what do you expect people to think about your
> reply?
> 
> I think, RHAT and RHAT's kernel devs need to change their attitude. RHAT
> needs tp comprehend Fedora is not a RHAT play ball, colony or egotrip.
> 
> That said, I consider your response a manifestation of disrespect towards
> the community.

I honestly have no idea why you responded this way.  Justin's comments were factual and he said the issue would eventually be fixed.

Comment 5 David H. Gutteridge 2018-02-18 02:01:12 UTC
FWIW, this reporting was corrected beginning with kernel 4.14.18. E.g., the /proc/cpuinfo "bugs" identification changed so an Atom N270 no longer reports "cpu_meltdown spectre_v1 spectre_v2".

commit d9b47a41160c71b217582b9876129e3fbf587225
Author: David Woodhouse <dwmw@amazon.co.uk>
Date:   Thu Jan 25 16:14:13 2018 +0000

    x86/pti: Do not enable PTI on CPUs which are not vulnerable to Meltdown
    
    commit fec9434a12f38d3aeafeb75711b71d8a1fdef621
    
    Also, for CPUs which don't speculate at all, don't report that they're
    vulnerable to the Spectre variants either.
    
    Leave the cpu_no_meltdown[] match table with just X86_VENDOR_AMD in it
    for now, even though that could be done with a simple comparison, on the
    assumption that we'll have more to add.
    
    Based on suggestions from Dave Hansen and Alan Cox.