[Info-vax] Intel junk...Kernel-memory-leaking Intel processor design flaw forces Linux, Windows redesign

[Jan-Erik Soderholm]

Den 2018-01-05 kl. 23:41, skrev DaveFroble:
> Bill Gunshannon wrote:
>> On 01/05/2018 04:04 PM, DaveFroble wrote:
>>> Bill Gunshannon wrote:
>>>> On 01/05/2018 08:50 AM, Alan Browne wrote:
>>>>> On 2018-01-04 15:43, DaveFroble wrote:
>>>>>> chrisv wrote:
>>>>>>> Designed By India H1B Engineers wrote:
>>>>>>>
>>>>>>>> Crucially, these updates to both Linux and Windows will incur a 
>>>>>>>> performance hit on Intel products. The effects are still being 
>>>>>>>> benchmarked, however we're looking at a ballpark figure of five to 
>>>>>>>> 30 per cent slow down, depending on the task and the processor model.
>>>>>>>
>>>>>>> This is ugly.  Think of the large computing centers, for example
>>>>>>> Google's data centers.  Suddenly, they will need significantly more
>>>>>>> CPU time, and thus electricity (and thus carbon), to get the job done?
>>>>>>>
>>>>>>
>>>>>> And once all the spanners are tossed into the works, which will slow 
>>>>>> things down, what happens when new CPUs without the issues are 
>>>>>> available?  Will computers forever be artificially slowed down?
>>>>>>
>>>>>> A whole bunch of someones has seriously dropped the ball on this. 
>>>>>> Protected memory should be just that, protected, with no way to avoid 
>>>>>> the protection.
>>>>>
>>>>> I presume it's an implementation flaw, not a principle-of-design flaw. 
>>>>> So once addressed, it should result in both proper memory protection 
>>>>> and increased performance in future cores.  Alas (per the article) 
>>>>> this can't be addressed with a microcode patch.
>>>>>
>>>>
>>>> Sounds more like a "principle-of-design" flaw to me.  Hard to
>>>> believe all those different companies all made the same mistake
>>>> building on a sound design.
>>>>
>>>> bill
>>>>
>>>
>>> I wonder whether VAX would have these problems?
>>>
>>> :-)
>>>
>>>
>>
>> VAX didn't have the capabilities that lead to this problem.
>> I think Alpha does, however.
>>
>> bill
>>
> 
> :-)
> 
> Yeah, I know that.  No predictive speculation in VAX.
> 
> Now, as for Alpha, yes, OoO and such, but, the question would be, does it 
> allow "illegal" access to memory?...

The problem with the current issue seems to be that the speculative
pre-fetch is done on a lower level then the page protection checks.
When the page-protection kicks in, the fetch has already been done.

I have no idea how that is designed on the Alpha.


> If Alpha does not allow loading memory 
> it should not into cache, then perhaps not a problem.
> 
>