[Intel] There is NO TLB bug in Nehalem.

[qurious63ss]

http://techreport.com/discussions.x/15979

TECH REPORT:

"While browsing news sites today, we came upon several reports saying Core i7 processors suffer from a translation lookaside buffer-related erratum. Those reports drew parallels between the TLB bug that plagued AMD's first quad-core CPUs and a note in Intel's Core i7 specification update, which states in part:

In rare instances, improper TLB invalidation may result in unpredictable system behavior, such as system hangs or incorrect data. Developers of operating systems should take this documentation into account when designing TLB invalidation algorithms. For the processors affected, Intel has provided a recommended update to system and BIOS vendors to incorporate into their BIOS to resolve this issue.

We asked Intel PR manager Dan Snyder for the chipmaker's official take on the issue, and he replied with the following:

This is simply a pointer to a previous document written in April 2007. This document is an application note (advises on programming techniques) that programmers have had since April of 2007. This item in the Nehalem spec sheet is a web pointer, under the heading "spec clarification". The reporter who wrote this did not contact us and we will try to clarify this with him.

In other words, Intel shareholders probably shouldn't be losing any sleep over this.

For reference, the AMD TLB erratum caused data corruption and system hangs in periods of high CPU utilization, and AMD halted shipments of quad-core Opterons for months because of it. On the desktop, the company released a BIOS fix that crippled the performance of early quad-core Phenoms in many apps. All Phenoms with model numbers ending in "50" and all shipping quad-core Opterons lack the erratum, though.

Update: Snyder has sent us a new statement that makes the situation even clearer:

The "AAJ1 Clarification of TRANSLATION LOOKASIDE BUFFERS" document is a SPEC CLARIFICATION, and is simply a pointer to a previous document written in April 2007.

SPEC CLARIFICATION AAJ1 was initially added due to an issue on the Intel® Core 2 Duo processor which was previously corrected with a BIOS update; this issue does not impact the Nehalem Family of CPUs. There are errata on the Intel® Core i7 processor that relate to the TLB. These all relate to improper translations or error reporting, and all of those that impact functionality have been fixed via BIOS updates prior to Core i7 launch.

(The emphasis is Intel's.)"

[George_o/c]

No way ... If FUD claims it's true, then it is ...
Can't understand the guys @ FUD ...

[qurious63ss]

For clarification purposes I was going to post this in the "[FUD TLB]" thread but Fugger closed it so I had to make a new thread and if Fugger wants to close this thread too then that's fine as long as people know the truth. FUDZILLA=FAIL

[Zucker2k]

Looks like FUD is about to eat some dirt from Intel.

[Shintai]

FUDzilla is the reason why I applaud goverment censorship.

[freeloader]

FUDzilla is the reason why I applaud goverment censorship.

Too bad the same government wouldn't censor you! Joking BTW.

The TLB is not an issue unless it causes data corruption and system crashes. No such issue here as it's been fixed. Good job by Intel catching it before releasing any CPU's to the public.

[Cooper]

There's at least one errata for 100k transistors in every complex IC.

[Hornet331]

FUDzilla is the reason why I applaud goverment censorship.

Thats the reason i just skip the thread when its from fud/inq.

As said in the other thread, i thing fuad needed some presents for christmas, that little "hoax" should have earned him a increase in hits for december.

[Rammsteiner]

Too bad the same government wouldn't censor you!

Fixed

Anyway, it was questionable in the first place Besides this clarification you could wonder whether it would have affected Intel in the same way as it did AMD, whether Intel was in the position to suffer from it or not like AMD.

Good to see this guy enlightened why this is no issue since saying 'There is no TLB bug' could have various meanings though

[foch3 -USA-]

FUDzilla is the reason why I applaud goverment censorship.

Wow are you serious... You can't be. A lot of things start off as "FUD" then turn out to be true.

Who get's to determine whats FUD or not? Thankfully FUGGER doesn't run this website that way.

[qurious63ss]

Too bad the same government wouldn't censor you! Joking BTW.

I gotta admit that was pretty funny.

[Shintai]

Wow are you serious... You can't be. A lot of things start off as "FUD" then turn out to be true.

Who get's to determine whats FUD or not? Thankfully FUGGER doesn't run this website that way.

I think you missed point...a few times

[ownage]

Wow are you serious... You can't be. A lot of things start off as "FUD" then turn out to be true.

Who gets to determine whats FUD or not? Thankfully FUGGER doesn't run this website that way.

Writing articles about every possible outcome creates lots of FUD and he has to be right about some things. I think its a good thing Shintai is always skeptical. Lots of guys here believe everything they want and create hundreds of BS post about new products.
A simple fact is that Fuad's articles are mostly BS (Nehalem TLB bug for example). Why is Fudzilla.com so popular while the quality is low? Because people love BS. The only thing Fuad wants is attention. Lots of page hits means more money for him. Don't you think he knew this supposed Nehalem TLB bug was BS? Off course he knew. He's laughing at all of you guys who think Fudzilla writes good articles from time to time.

Stop with the hatred towards Shintai and get a life ASAP!
Nobody in the business has respect for fudzilla, only nubs do. So yeah please go defend fudzilla lol.

[nn_step]

There's at least one errata for 100k transistors in every complex IC.

that point is debatable, but I do agree with the general principle given a complex enough device, there will be geometrically more errata as complexity grows.

[headman800]

Lol Fudzilla

[Ghostbuster]

As with the older thread.. The words "Told you so!" was pretty clear as I pointed out the same "bug" with the C2Ds...

[Zucker2k]

I think the point we're missing here is how such news could trigger people selling off stocks on the market; now I'm not blaming FUD for the huge loss on WS today, but I bet there are people who dumped shares on the news. In this fragile market, such bad journalism, especially involving a critical player in the technologies futures is very irresponsible and dangerous. I mean all FUD had to do was do a little bit of investigative journalism. I guess that's too much to ask. I'm beginning to think these guys do all their journalism from behind their PCs.

[flippin_waffles]

Well what else would Intel say. Of course they are going to deny anything negative in the press. That's standard procedure for crying out loud. lol Deny, deny, deny. Did people expect them to raise their hand and say, "guilty as charged". Maybe there is an issue, and maybe their isn't, but i'm certainly not going buy it hook, line and sinker because of a quick PR release.

[Ghostbuster]

Well what else would Intel say. Of course they are going to deny anything negative in the press. That's standard procedure for crying out loud. lol Deny, deny, deny. Did people expect them to raise their hand and say, "guilty as charged". Maybe there is an issue, and maybe their isn't, but i'm certainly not going buy it hook, line and sinker because of a quick PR release.

As with the other thread, I think you missed this...

Core 2 Duo E7000 and E8000 series http://download.intel.com/design/pro...pdt/318733.pdf
Same exact text written on page 47.

Core 2 Duo X6800, E6000 and E4000 series http://download.intel.com/design/pro...pdt/313279.pdf
Same exact text written on page 71.

Pentium Dual Core E2000 series http://download.intel.com/design/pro...pdt/316982.pdf
Same exact text written on page 59.

Intel Celeron 200 series http://download.intel.com/design/pro...t/31854702.pdf
Same exact text written on page 53

[flippin_waffles]

Actually no, I never missed it at all. But like I said, i'll give it a little more time to see if anything else turns up. Something doesn't feel right as far as i'm concerned and smells a little fishy. But, we'll see.

[Ghostbuster]

Actually no, I never missed it at all. But like I said, i'll give it a little more time to see if anything else turns up. Something doesn't feel right as far as i'm concerned and smells a little fishy. But, we'll see.

Believe what you want, as I said beforehand FUD is FUD. I've already pointed out the proof even before this verification was published..

Let me re-quote this...

The "AAJ1 Clarification of TRANSLATION LOOKASIDE BUFFERS" document is a SPEC CLARIFICATION, and is simply a pointer to a previous document written in April 2007.

SPEC CLARIFICATION AAJ1 was initially added due to an issue on the Intel® Core 2 Duo processor which was previously corrected with a BIOS update; this issue does not impact the Nehalem Family of CPUs. There are errata on the Intel® Core i7 processor that relate to the TLB. These all relate to improper translations or error reporting, and all of those that impact functionality have been fixed via BIOS updates prior to Core i7 launch.

[flippin_waffles]

Thanks for the permission, I believe I will! I'll wait for virtualization reviews. Maybe by that time we'll actually have some Shanghai VM articles.

[Spawne32]

there needs to be a warning label on anything that links back to fud or inq on this site

[Russian]

There is also a good explanation on Theo's Blog.
http://theovalich.wordpress.com/2008...-story-at-all/

[qurious63ss]

Actually no, I never missed it at all. But like I said, i'll give it a little more time to see if anything else turns up. Something doesn't feel right as far as i'm concerned and smells a little fishy. But, we'll see.

Sounds to me like it's wishful thinking on your part. Not a good way to cover up an issue by documenting it now is it.