Meltdown & Spectre list of cpus affected?

classic Classic list List threaded Threaded
22 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Meltdown & Spectre list of cpus affected?

Larry Stotler
I can't seem to pin down whether the Core 2 or Pentium 4/Ms are
affected by either of these issues.  Most reports say since 1995 or
Pentium Pro & Newer.    Intel's list starts with the Core ix series
and doesn't go back further.  Of course I can't find the list on their
site, but I found this one:

https://www.tweaktown.com/news/60411/heres-list-intel-cpus-affected-spectre-meltdown/index.html

Also what about AMD?  Their arch has differences.  Is there a list for
them?  Athlon, AthlonXP, Athlon64?

I've read that even the POWER chips have a spectre vuln(which seems to
be more mitigatable)..

While there seems to be a great deal of hype about all this, I have to
wonder how easy it is to compromise a system.  Most reports save
having to already have access to the machine.  If that's the case,
then I think you are already in trouble.  It's also hard to tell how
long the spy agencies have known about this and been (ab)using it.

I dunno.  I just can't see that much interest in mitigating this on
older chips/systems(even though a lot of us find them perfectly
usable).  Looks like a forced upgrade is going to be contemplated for
many even though it's really hard to tell how many other issues are
going to be found all things considering.  We could all rush out and
get the "fixed" hardware when it's released only to end up in the same
boat sooner or later.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Marcus Meissner
On Mon, Jan 08, 2018 at 02:50:09PM -0500, Larry Stotler wrote:

> I can't seem to pin down whether the Core 2 or Pentium 4/Ms are
> affected by either of these issues.  Most reports say since 1995 or
> Pentium Pro & Newer.    Intel's list starts with the Core ix series
> and doesn't go back further.  Of course I can't find the list on their
> site, but I found this one:
>
> https://www.tweaktown.com/news/60411/heres-list-intel-cpus-affected-spectre-meltdown/index.html
>
> Also what about AMD?  Their arch has differences.  Is there a list for
> them?  Athlon, AthlonXP, Athlon64?

Meltdown affects Intel chips. I think Core 2 for sure, Pentium 4 unclear.

FWIW, there is a Meltdown exploit available for testing.

        https://github.com/paboldin/meltdown-exploit

Spectre affects both Intel and AMD chips.
 
> I've read that even the POWER chips have a spectre vuln(which seems to
> be more mitigatable)..

This is also our current understanding.

> While there seems to be a great deal of hype about all this, I have to
> wonder how easy it is to compromise a system.  Most reports save
> having to already have access to the machine.  If that's the case,
> then I think you are already in trouble.  It's also hard to tell how
> long the spy agencies have known about this and been (ab)using it.
>
> I dunno.  I just can't see that much interest in mitigating this on
> older chips/systems(even though a lot of us find them perfectly
> usable).  Looks like a forced upgrade is going to be contemplated for
> many even though it's really hard to tell how many other issues are
> going to be found all things considering.  We could all rush out and
> get the "fixed" hardware when it's released only to end up in the same
> boat sooner or later.

One thing is that Javascript in Firefox and Chrome was/is able
to exploit Meltdown. This makes an attack over the web possible.

(Firefox and Chrome javascript engines had high precision timers
 available and some form of direct byte buffer access.
 They want to make the high precission timers less precise.)


Spectre is only beginning to be understood, it is actually more
tricky from a Mitigation point of view.

Ciao, Marcus

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
On Mon, Jan 8, 2018 at 2:59 PM, Marcus Meissner <[hidden email]> wrote:
> Meltdown affects Intel chips. I think Core 2 for sure, Pentium 4 unclear.
>
> FWIW, there is a Meltdown exploit available for testing.
>
>         https://github.com/paboldin/meltdown-exploit
>
> Spectre affects both Intel and AMD chips.

Ran it. Had to recompile it with suggested settings.  Got:

looking for linux_proc_banner in /proc/kallsyms
cached = 70, uncached = 463, threshold 180
read ffffffff81800060 = ff   (score=0/1000)
read ffffffff81800061 = ff   (score=0/1000)
read ffffffff81800062 = ff   (score=0/1000)
read ffffffff81800063 = ff   (score=0/1000)
read ffffffff81800064 = ff   (score=0/1000)
read ffffffff81800065 = ff   (score=0/1000)
read ffffffff81800066 = ff   (score=0/1000)
read ffffffff81800067 = ff   (score=0/1000)
read ffffffff81800068 = ff   (score=0/1000)
read ffffffff81800069 = ff   (score=0/1000)
read ffffffff8180006a = ff   (score=0/1000)
read ffffffff8180006b = ff   (score=0/1000)
read ffffffff8180006c = ff   (score=0/1000)
read ffffffff8180006d = ff   (score=0/1000)
read ffffffff8180006e = ff   (score=0/1000)
read ffffffff8180006f = ff   (score=0/1000)
NOT VULNERABLE
PLEASE POST THIS TO https://github.com/paboldin/meltdown-exploit/issues/22
NOT VULNERABLE ON
3.16.7-53-desktop #1 SMP PREEMPT Fri Dec 2 13:19:28 UTC 2016 (7b4a1f9) x86_64
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 15
model name      : Intel(R) Core(TM)2 CPU         T7200  @ 2.00GHz
stepping        : 6
microcode       : 0xd1
cpu MHz         : 2000.000
cache size      : 4096 KB
physical id     : 0

This laptop is still running 13.2/x64.  Been lazy.  However, it does
show not vulnerable, which is encouraging.  I do have a more current
setup on a Q9000 laptop I will try later.

> One thing is that Javascript in Firefox and Chrome was/is able
> to exploit Meltdown. This makes an attack over the web possible.
>
> (Firefox and Chrome javascript engines had high precision timers
>  available and some form of direct byte buffer access.
>  They want to make the high precission timers less precise.)

Good thing I run NoScript.  Maybe this will get sites to reconsider
relying on javascript so much(I'm not a fan).

> Spectre is only beginning to be understood, it is actually more
> tricky from a Mitigation point of view.

crap.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
On Mon, Jan 8, 2018 at 3:09 PM, Larry Stotler <[hidden email]> wrote:
> This laptop is still running 13.2/x64.  Been lazy.  However, it does
> show not vulnerable, which is encouraging.  I do have a more current
> setup on a Q9000 laptop I will try later.

The Q9000 passed as well.  However, I did not check it until I had ran
the update to 4.4.104-39-default.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

David C. Rankin
In reply to this post by Marcus Meissner
On 01/08/2018 01:59 PM, Marcus Meissner wrote:
> Meltdown affects Intel chips. I think Core 2 for sure, Pentium 4 unclear.
>
> FWIW, there is a Meltdown exploit available for testing.
>
> https://github.com/paboldin/meltdown-exploit

I'd take the test with a grain-of-salt. I've run in on both i7 and Core2 (both
of which are supposed to be affected) and it reports not vulnerable in both
cases. I'd be interested if anyone has had it report vulnerable?

--
David C. Rankin, J.D.,P.E.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Dave Plater lst


On 09/01/2018 11:29, David C. Rankin wrote:

> On 01/08/2018 01:59 PM, Marcus Meissner wrote:
>> Meltdown affects Intel chips. I think Core 2 for sure, Pentium 4 unclear.
>>
>> FWIW, there is a Meltdown exploit available for testing.
>>
>> https://github.com/paboldin/meltdown-exploit
>
> I'd take the test with a grain-of-salt. I've run in on both i7 and Core2 (both
> of which are supposed to be affected) and it reports not vulnerable in both
> cases. I'd be interested if anyone has had it report vulnerable?
>
Have you updated your kernel yet?
My i3-2120 was listed as vulnerable on intels site but it comes up as
not vulnerable. My system's already patched though.
Just a thought, imagine if meltdown-exploit actually installed an exploit.
Dave P

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
In reply to this post by David C. Rankin
On Tue, Jan 9, 2018 at 4:29 AM, David C. Rankin
<[hidden email]> wrote:
> I'd take the test with a grain-of-salt. I've run in on both i7 and Core2 (both
> of which are supposed to be affected) and it reports not vulnerable in both
> cases. I'd be interested if anyone has had it report vulnerable?

https://github.com/paboldin/meltdown-exploit/issues/19

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Marcus Meissner
In reply to this post by Larry Stotler
On Mon, Jan 08, 2018 at 08:45:21PM -0500, Larry Stotler wrote:
> On Mon, Jan 8, 2018 at 3:09 PM, Larry Stotler <[hidden email]> wrote:
> > This laptop is still running 13.2/x64.  Been lazy.  However, it does
> > show not vulnerable, which is encouraging.  I do have a more current
> > setup on a Q9000 laptop I will try later.
>
> The Q9000 passed as well.  However, I did not check it until I had ran
> the update to 4.4.104-39-default.

That kernel is fixed for Meltdown ;)

I ran it on my Tumbleweed before the update and it reported affecfted.

Ciao, Marcus

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Rodney Baker-2
In reply to this post by David C. Rankin
On Tuesday, 9 January 2018 19:59:10 ACDT David C. Rankin wrote:

> On 01/08/2018 01:59 PM, Marcus Meissner wrote:
> > Meltdown affects Intel chips. I think Core 2 for sure, Pentium 4 unclear.
> >
> > FWIW, there is a Meltdown exploit available for testing.
> >
> > https://github.com/paboldin/meltdown-exploit
>
> I'd take the test with a grain-of-salt. I've run in on both i7 and Core2
> (both of which are supposed to be affected) and it reports not vulnerable
> in both cases. I'd be interested if anyone has had it report vulnerable?

Yes. Same test, same CPU, two different kernels:

VULNERABLE ON
4.14.8-1.g674981b-default #1 SMP PREEMPT Wed Dec 20 10:40:41 UTC 2017
(674981b) x86_64
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 42
model name      : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping        : 7
microcode       : 0x29
cpu MHz         : 3099.882
cache size      : 6144 KB
physical id     : 0

NOT VULNERABLE ON
4.14.12-2.g7637ae2-default #1 SMP PREEMPT Sat Jan 6 09:10:30 UTC 2018
(7637ae2) x86_64
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 42
model name      : Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping        : 7
microcode       : 0x29
cpu MHz         : 3099.979
cache size      : 6144 KB
physical id     : 0





--
==============================================================
Rodney Baker VK5ZTV
[hidden email]
CCNA #CSCO12880208
==============================================================



--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
> Yes. Same test, same CPU, two different kernels:
> NOT VULNERABLE ON
> 4.14.12-2.g7637ae2-default #1 SMP PREEMPT Sat Jan 6 09:10:30 UTC 2018

The second kernel includes the fix.  The check looks for the
mitigation, not the flaw.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Carlos E. R.-2
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



On Tuesday, 2018-01-09 at 08:32 -0500, Larry Stotler wrote:

>> Yes. Same test, same CPU, two different kernels:
>> NOT VULNERABLE ON
>> 4.14.12-2.g7637ae2-default #1 SMP PREEMPT Sat Jan 6 09:10:30 UTC 2018
>
> The second kernel includes the fix.  The check looks for the
> mitigation, not the flaw.

Oh :-(

- --
Cheers,
        Carlos E. R.
        (from openSUSE 42.2 x86_64 "Malachite" at Telcontar)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iEYEARECAAYFAlpUyYwACgkQtTMYHG2NR9Vt/ACeJ63r/iTCzHXEMo0MX90eyXvx
6M0AoI7EviCCZ7lu9XhTlG5MV+QQL1P9
=hcuB
-----END PGP SIGNATURE-----

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
On Tue, Jan 9, 2018 at 8:54 AM, Carlos E. R.
<[hidden email]> wrote:
>> The second kernel includes the fix.  The check looks for the
>> mitigation, not the flaw.
>
>
> Oh :-(

Which begs the question:  Since my one laptop is still running
3.16.7-53, and it showed not vulnerable, is it?  Or is the check
expecting something that isn't there and not working properly?  The
check program doesn't say what the minimum kernel version is.
However, it does show that the specific exploit isn't working on that
system.

Even though everyone is saying anything from Pentium Pro forward, it
would seem some older chips may not be (as) affected.  That's why I
was trying to find a better list of affected cpus.  It's also possible
that while the flaw is there, it's more exploitable on the newer
chips.  Several people posted 45nm Core2s as being affected.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Lew Wolfgang
In reply to this post by Dave Plater lst
On 01/09/2018 02:03 AM, Dave Plater wrote:
> My i3-2120 was listed as vulnerable on intels site but it comes up as not
> vulnerable. My system's already patched though.
> Just a thought, imagine if meltdown-exploit actually installed an exploit.

That wouldn't be the first time that happened, would it?

Regards,
Lew


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Knurpht-openSUSE
Op dinsdag 9 januari 2018 15:56:57 CET schreef Lew Wolfgang:
> On 01/09/2018 02:03 AM, Dave Plater wrote:
> > My i3-2120 was listed as vulnerable on intels site but it comes up as not
> > vulnerable. My system's already patched though.
> > Just a thought, imagine if meltdown-exploit actually installed an exploit.
>
> That wouldn't be the first time that happened, would it?
>
> Regards,
> Lew

If you clone the github source, you can read the meltdown-exploit.c file and
the Makefile. Or, if you still don't trust it, the included stuff. I did the
first and decided it was safe to compile and run.

--
Gertjan Lettink, a.k.a. Knurpht

openSUSE Board Member
openSUSE Forums Team



--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Richmond
In reply to this post by Larry Stotler
Larry Stotler wrote:

> I can't seem to pin down whether the Core 2 or Pentium 4/Ms are
> affected by either of these issues.  Most reports say since 1995 or
> Pentium Pro & Newer.    Intel's list starts with the Core ix series
> and doesn't go back further.  Of course I can't find the list on their
> site, but I found this one:
>
> https://www.tweaktown.com/news/60411/heres-list-intel-cpus-affected-spectre-meltdown/index.html
>
> Also what about AMD?  Their arch has differences.  Is there a list for
> them?  Athlon, AthlonXP, Athlon64?
>
> I've read that even the POWER chips have a spectre vuln(which seems to
> be more mitigatable)..
>
> While there seems to be a great deal of hype about all this, I have to
> wonder how easy it is to compromise a system.  Most reports save
> having to already have access to the machine.  If that's the case,
> then I think you are already in trouble.  It's also hard to tell how
> long the spy agencies have known about this and been (ab)using it.
>
> I dunno.  I just can't see that much interest in mitigating this on
> older chips/systems(even though a lot of us find them perfectly
> usable).  Looks like a forced upgrade is going to be contemplated for
> many even though it's really hard to tell how many other issues are
> going to be found all things considering.  We could all rush out and
> get the "fixed" hardware when it's released only to end up in the same
> boat sooner or later.
>
It sounds like it is pretty much all of them.

https://www.kb.cert.org/vuls/id/584653

Maybe some old ones would be OK.


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

medwinz
In reply to this post by Larry Stotler
On Tue, Jan 9, 2018 at 2:50 AM, Larry Stotler <[hidden email]> wrote:

> I can't seem to pin down whether the Core 2 or Pentium 4/Ms are
> affected by either of these issues.  Most reports say since 1995 or
> Pentium Pro & Newer.    Intel's list starts with the Core ix series
> and doesn't go back further.  Of course I can't find the list on their
> site, but I found this one:
>
> https://www.tweaktown.com/news/60411/heres-list-intel-cpus-affected-spectre-meltdown/index.html
>
> Also what about AMD?  Their arch has differences.  Is there a list for
> them?  Athlon, AthlonXP, Athlon64?
>
> I've read that even the POWER chips have a spectre vuln(which seems to
> be more mitigatable)..
>
> While there seems to be a great deal of hype about all this, I have to
> wonder how easy it is to compromise a system.  Most reports save
> having to already have access to the machine.  If that's the case,
> then I think you are already in trouble.  It's also hard to tell how
> long the spy agencies have known about this and been (ab)using it.
>
> I dunno.  I just can't see that much interest in mitigating this on
> older chips/systems(even though a lot of us find them perfectly
> usable).  Looks like a forced upgrade is going to be contemplated for
> many even though it's really hard to tell how many other issues are
> going to be found all things considering.  We could all rush out and
> get the "fixed" hardware when it's released only to end up in the same
> boat sooner or later.


HI,

I use https://github.com/speed47/spectre-meltdown-checker
My laptop use Intel i5. This is Tumbleweed 20180107 with kernel 4.14.12-1.
The result is:

CVE-2017-5753 [bounds check bypass] aka 'Spectre Variant 1'
* Checking count of LFENCE opcodes in kernel:  YES  (77 opcodes found,
which is >= 70)
> STATUS:  NOT VULNERABLE  (heuristic to be improved when official patches become available)

CVE-2017-5715 [branch target injection] aka 'Spectre Variant 2'
* Mitigation 1
*   Hardware (CPU microcode) support for mitigation:  NO
*   Kernel support for IBRS:  YES
*   IBRS enabled for Kernel space:  UNKNOWN
*   IBRS enabled for User space:  UNKNOWN
* Mitigation 2
*   Kernel compiled with retpoline option:  NO
*   Kernel compiled with a retpoline-aware compiler:  NO
> STATUS:  VULNERABLE  (IBRS hardware + kernel support OR kernel with retpoline are needed to mitigate the vulnerability)

CVE-2017-5754 [rogue data cache load] aka 'Meltdown' aka 'Variant 3'
* Kernel supports Page Table Isolation (PTI):  YES
* PTI enabled and active:  YES
> STATUS:  NOT VULNERABLE  (PTI mitigates the vulnerability)

It is still vulnerable for Spectre Variant 2 with mitigation 2.


Regards,
--
Edwin

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Rodney Baker-2
In reply to this post by Larry Stotler
On Wednesday, 10 January 2018 0:02:52 ACDT Larry Stotler wrote:
> > Yes. Same test, same CPU, two different kernels:
> > NOT VULNERABLE ON
> > 4.14.12-2.g7637ae2-default #1 SMP PREEMPT Sat Jan 6 09:10:30 UTC 2018
>
> The second kernel includes the fix.  The check looks for the
> mitigation, not the flaw.

No, the test attempts to exploit the flaw - if successful, it reports as
vulnerable (the combination of processor and kernel); if unsuccessful, then
either the processor is not vulnerable (as with my Intel Atom tablet) or the
kernel has been patched to successfully mitigate the flaw on an otherwise
vulnerable processor (hence why I ran 2 tests).

The first test shows that the processor is vulnerable - the second that the
specific vulnerability being tested has been mitigated by the fix in the later
kernel.

There are other tests (flagged by others on this list) that test both Spectre
and Meltdown vulnerabilities - this specific test is only for Meltdown.

The current 4.14.12-2.g7637ae2-default kernel does not mitigate Spectre
Variant 2.

Regards,
Rodney.

--
==============================================================
Rodney Baker VK5ZTV
[hidden email]
CCNA #CSCO12880208
==============================================================



--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Andrei Borzenkov
In reply to this post by Larry Stotler
On Tue, Jan 9, 2018 at 4:32 PM, Larry Stotler <[hidden email]> wrote:
>> Yes. Same test, same CPU, two different kernels:
>> NOT VULNERABLE ON
>> 4.14.12-2.g7637ae2-default #1 SMP PREEMPT Sat Jan 6 09:10:30 UTC 2018
>
> The second kernel includes the fix.  The check looks for the
> mitigation, not the flaw.
>

Boot with "nopti" kernel option.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
On Wed, Jan 10, 2018 at 8:41 AM, Andrei Borzenkov <[hidden email]> wrote:
> Boot with "nopti" kernel option.

Is there any way to create like a live CD that could run the checks
regardless of the installed OS?  On my machine with the older 3.16
kernel, it showed not vulnerable, but is that the case or is it
because the kernel was too old for the check to work properly?
Further, something that can do some simple benchmarks to see how much
the fixes will slow them down.

I have some older Pentium 4 systems that I will have to find time to
put up and install a newer install on and see if they show vulnerable.

It would be interesting to see if the older machines would need a
different type of exploit that the newer chips.  If so, then this
could create a new second class of older machines that will never see
a proper fix since so few of them are still in use.  One more way to
try to push people to buy newer hardware when what they have is
working perfectly fine.  Or, it could be that the fix doesn't slow
down the older chips as much and they may not have more value since
the newest chips are getting the slowdown.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Meltdown & Spectre list of cpus affected?

Larry Stotler
On Wed, Jan 10, 2018 at 6:34 PM, Larry Stotler <[hidden email]> wrote:
> Or, it could be that the fix doesn't slow
> down the older chips as much and they may not have more value since
> the newest chips are getting the slowdown.

From Phoronix( https://www.phoronix.com/scan.php?page=article&item=linux-more-x86pti&num=1)
:

- On that old Clarksfield-era ThinkPad I wasn't going to be surprised
if the performance was disastrous, but it wound up being better than I
had anticipated given all the ongoing drama... In general purpose
workloads there was no reportable performance difference in our
frequent benchmark test cases. Under I/O, the PTI-using kernel did
yield some slower results but not by the margins seen on the newer
systems with faster storage. The laptop consumer-grade HDD in this
laptop appeared to be the main bottleneck and kernel inefficiencies
weren't causing as dramatic slowdowns.

- To some surprise, when carrying out network benchmarks with
netperf/iperf3, in at least those contexts PTI didn't have a
noticeable impact on the network throughput performance.

--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

12