Critical update for Intel Core CPUs is out

[CalvinHobbes]

http://www.theinquirer.net/default.aspx?article=40567

A COUPLE OF WEEKS ago, we heard that Dell was dealing with a certain situation considering Intel dual-core MCW and quad-core KC marchitecture, and that the company was releasing urgent BIOS and microcode versions for its line up.
We learned that the affected CPUs are the Core 2 Duo E4000/E6000, Core 2 Quad Q6600, Core 2 Xtreme X6800, XC6700 and XC6800. In the mobile world, the Core 2 Duo T5000 and T7000 need to visit Microsoft's site, while server guys will want to use motherboard BIOSes if they do not rely on Microsoft Windows operating systems. The affected servers are Xeon 3000, 3200, 5100 and 5300s - or just about every model from second generation of Core marchitecture, Oddly enough, Yonah - 32-bit Core Duo processor - isn't among the affected cores.

We are assured that no product recall will happen, and that La Intella took all appropriate steps in order to minimise public image, because if a product recall happened, Intel's credibility would be ruined for good.

Anyway, if you have a Core CPU based machine, go to the link below to download the update. AMD processors are not affected at all, in case you were wondering.

http://support.microsoft.com/?kbid=936357

A microcode reliability update is available that improves the reliability of systems that use Intel processors

 

[A554SS1N]

Is it really that critical, or just FUD?

 

[imported_Dekard]

its from the inq... prolly fud (fear, uncertainty and doubt).

 

[Phynaz]

Update

Definition: An update is a broadly released fix for a specific problem. An update addresses a non-critical, non-security-related bug.

Not critical, and also a Windows fix, not an Intel fix.

 

[Duvie]

what was the fix for? What is the errata? or is there one a normal user would ever see......

Microsuck is a virus that needs to be eradicated once and for all...

 

[adairusmc]

Originally posted by: Duvie


Microsuck is a virus that needs to be eradicated once and for all...

:roll:

 

[JeffreyLebowski]

Originally posted by: Duvie
what was the fix for? What is the errata? or is there one a normal user would ever see......

Microsuck is a virus that needs to be eradicated once and for all...

You do realize if it weren't for Microsoft we'd all be wearing birkenstoks, driving VW's and have un-upgradeable PCs

 

[aka1nas]

Originally posted by: Phynaz
Update

Definition: An update is a broadly released fix for a specific problem. An update addresses a non-critical, non-security-related bug.

Not critical, and also a Windows fix, not an Intel fix.

If it is actually a micro-code update, then it's not a windows patch. More like flashing the firmware on your CPU.

 

[Vee]

There is no technical info anywhere from what I've seen sofar.

I assume the following:

1: There is a bug in the Core 2 processor that is more serious than these bugs normally are.
2: Intel is keeping very quiet and will not do production recall because that would not only be disaster for their killing market squeeze on AMD but a disaster for Intel as company as well. Windows will be blamed for crashes anyway, not Intel.
3: Intel have finally come up with this fix instead.
4: The fix will actually not be able to solve the problem. Only reduce the frequency at which it will come into play.
5: The fix will lower performance. But extremely little.

That is, as I said, what I assume.

 

[Fallen Kell]

Originally posted by: JeffreyLebowski
<div class="FTQUOTE"><begin quote>Originally posted by: Duvie
what was the fix for? What is the errata? or is there one a normal user would ever see......

Microsuck is a virus that needs to be eradicated once and for all...</end quote></div>

You do realize if it weren't for Microsoft we'd all be wearing birkenstoks, driving VW's and have un-upgradeable PCs

You do realize that upgradeable PCs existed before MS release DOS don't you?

 

[coolpurplefan]

Uh, this is weird. We had gaming problems with X2s and now this? What's this? A secret problem that Intel doesn't want to reveal to the public?

 

[kotrtim]

http://www-304.ibm.com/jct01004c/sy...display?brandind=5000008&lndocid=MIGR-5070941

# A marginal design issue in the L1 cache circuit that could lead to a system hang, blue screen, or kernel panic in Linux.

It just affects Linux?

It is already stated in the warranty card that intel is not responsible for errata, AMD is the same too, so there will no recall as long it works correctly.. Intel Pentium D has the standby resume errata too IIRC

 

[VirtualLarry]

Originally posted by: Vee
There is no technical info anywhere from what I've seen sofar.

I assume the following:

1: There is a bug in the Core 2 processor that is more serious than these bugs normally are.
2: Intel is keeping very quiet and will not do production recall because that would not only be disaster for their killing market squeeze on AMD but a disaster for Intel as company as well. Windows will be blamed for crashes anyway, not Intel.
3: Intel have finally come up with this fix instead.
4: The fix will actually not be able to solve the problem. Only reduce the frequency at which it will come into play.
5: The fix will lower performance. But extremely little.

That is, as I said, what I assume.

That sounds about right.

I'm wondering, if this is a microcode fix, why isn't it being distributed to mainboard BIOS companies instead. This issue must be more serious than just a microcode update, if it requires a Windows driver fix. I wonder if it is resident (runtime), or just a static, boot-time fix?

 

[gramboh]

Originally posted by: kotrtim
http://www-304.ibm.com/jct01004c/sy...display?brandind=5000008&lndocid=MIGR-5070941

# A marginal design issue in the L1 cache circuit that could lead to a system hang, blue screen, or kernel panic in Linux.

It just affects Linux?

It is already stated in the warranty card that intel is not responsible for errata, AMD is the same too, so there will no recall as long it works correctly.. Intel Pentium D has the standby resume errata too IIRC

No, a kernal panic is a symptom in Linux. The blue screen is obviously a windows only symptom, so it is not platform related.

 

[jhtrico1850]

If it ain't broke, don't fix it. People on XS are reporting problems so I'd wait for more info before messing with this.

 

[coolpurplefan]

http://forums.anandtech.com/me...=2064923&enterthread=y

Well, here's a dude talking about Windows Vista issues before the patch. Maybe he was guessing wrong, I don't know.

 

[covert24]

sounds like a load of garbage being that its from the inquirer...

 

[Cookie Monster]

Link

Around April 10th, 2007, Intel released a microcode update (BIOS flash to the rest of us) so resolve several major issues with how Intel Core 2 series processors operate.

Intel has provided a microcode update as a critical software fix. This Intel microcode update addresses an improper Translation Lookaside Buffer (TLB) invalidation that may result in unpredictable system behavior such as system hangs or incorrect data. OEM's are making available a system BIOS upgrade that integrates a new version of this Intel microcode to address this potential but extremely rare issue. While the occurrence of this issue is extremely rare, OEM's strongly recommends upgrading the system BIOS to an updated version. Intel has indicated that the microcode update has no known measurable performance impact and provides a complete and comprehensive resolution to this issue.

The issues can affect any workstation that uses Intel Core2 Duo desktop processor E4000 and E6000 sequences, Intel Core2 Quad desktop processor Q6600 and Intel Core2 Extreme processors X6800 and QX6700 processors.

While the implications of the issue are difficult to quantify, any of the following symptoms can occur:

The system may stop responding to keyboard or mouse input.
A system operating in a Microsoft Windows environment may generate a blue screen.
A system operating in a Linux environment may generate a kernel panic.

 

[apoppin]

Are the MB manufacturers gonna offer this fix with a BIOS flash?

do they *already* offer it?

i got my gigabyte Xfire MB last week and flashed it Friday night to their very latest BIOS ... since this has been a "known issue" since April, wouldn't the fix already be included?

 

[Roguestar]

Originally posted by: Vee
There is no technical info anywhere from what I've seen sofar.

I assume the following:

1: There is a bug in the Core 2 processor that is more serious than these bugs normally are.
2: Intel is keeping very quiet and will not do production recall because that would not only be disaster for their killing market squeeze on AMD but a disaster for Intel as company as well. Windows will be blamed for crashes anyway, not Intel.
3: Intel have finally come up with this fix instead.
4: The fix will actually not be able to solve the problem. Only reduce the frequency at which it will come into play.
5: The fix will lower performance. But extremely little.

That is, as I said, what I assume.

Based on ...?

 

[crzychikn]

so is there anything to this? i had a load of blue screen random crash errors that culminated in registry and memory errors making me format and reinstall until i updated vista and my bios.

i dont know if that bios incorporated any fix for this tho

 

[CTho9305]

Originally posted by: Roguestar
<div class="FTQUOTE"><begin quote>Originally posted by: Vee
There is no technical info anywhere from what I've seen sofar.

I assume the following:

1: There is a bug in the Core 2 processor that is more serious than these bugs normally are.
2: Intel is keeping very quiet and will not do production recall because that would not only be disaster for their killing market squeeze on AMD but a disaster for Intel as company as well. Windows will be blamed for crashes anyway, not Intel.
3: Intel have finally come up with this fix instead.
4: The fix will actually not be able to solve the problem. Only reduce the frequency at which it will come into play.
5: The fix will lower performance. But extremely little.

That is, as I said, what I assume.</end quote></div>

Based on ...?

Ignorance, probably. OCing fanboys aren't the only people who use computers - some people actually do real work using CPUs, and CPU vendors have to spend tremendous resources ensuring that CPUs work correctly for the people who don't consider "passes prime95 for 24 hrs, but occasionally programs crash" stable. When bugs are found, those resources are put to use to provide fixes that don't introduce new bugs. It's not like Intel is the only company that makes mistakes either - this post covers a situation where AMD did something similar.

 

[orion23]

Yeah, Intel should really find a way to lock the FSB, just like the multiplier on their CPU's.

I can only imagine the thousands of CPU's that have been returned to Intel because they didn't overclock to 3.6ghz like most people would like them to. I have seen people crying because their E6400's @ 2.133ghz won't go past 3.4ghz or so.

I mean, WTF.

Also, running orthos and prime95 for days became a sport for so many idiots around these boards lately. They no longer use their computers because Orthos comes firts and if it crashes while overclocked @ 3.6ghz after 36 hours, then it is somehow UNSTABLE.

I think those morons will ruin the overclocking game for all of us!

 

[VirtualLarry]

From thread

Originally posted by: n7
Memtest is okay for testing RAM, but that's it.
It's rather useless for OCing i find, as running it doesn't test stability in Windows, & i think most people run their systems with their OS running

I used to use S&M as an alternative, but with the issue of their program reporting failed L1 cache on C2Ds, i haven't bothered with it lately.

To get to my point, i find that overall, SP20004's Orthos is by far the best overall stability indicator (excluding graphics of course) out there right now.

Could this L1 cache bug have already been detected back in 2006 with the right diagnostic software?

 

[VirtualLarry]

Originally posted by: orion23
Also, running orthos and prime95 for days became a sport for so many idiots around these boards lately. They no longer use their computers because Orthos comes firts and if it crashes while overclocked @ 3.6ghz after 36 hours, then it is somehow UNSTABLE.
I think those morons will ruin the overclocking game for all of us!

I don't get it. Running Orthos for 24hour and Memtest86+ for 24hour is essential to determining if your computer is stable. For longer periods, not so sure, but I recommend a minimum of 24hours of each.