[Freedos-devel] Wait don't fix you code bug!

[Jerome Shidel]

Hello all,

In case you haven't heard, there is a severe bug in Intel Skylake and Newer CPUs. I only know a little about it. From what I heard, it crops up with hyper-threading inside loops less than 64 bytes effecting several high byte registers (AH, BH, etc). They are working on a microcode patch and hopefully will have it out soon.

So, I guess it could effect DOS asm code running inside a VM.

I think it's been a while since there last major bug. Last one I recall was that Pentium 90 Floating Point problem.

[Rugxulo]

Hi,

Post by Jerome Shidel
In case you haven't heard, there is a severe bug in Intel Skylake and Newer CPUs.

Most of us aren't trendy. This Nehalem Westmere (Core gen #1) is my
newest machine (so far). ;-)

Post by Jerome Shidel
I only know a little about it. From what I heard, it crops up with hyper-threading
inside loops less than 64 bytes effecting several high byte registers (AH, BH, etc).
They are working on a microcode patch and hopefully will have it out soon.

Don't panic! Bugs are part of life ("par for the course").

http://www.os2museum.com/wp/vme-fixed-on-amd-ryzen/

Post by Jerome Shidel
So, I guess it could effect DOS asm code running inside a VM.
I think it's been a while since there last major bug. Last one I recall was that Pentium 90 Floating Point problem.

You mean the FDIV bug? That was fixed in 120 Mhz (and higher) models.

But there have always been other cpu bugs:

* https://en.wikipedia.org/wiki/Pentium_F00F_bug
* (P3 serializing bug)
http://web.archive.org/web/20070808143526/http://clio.rice.edu:80/djgpp/r5bug05.txt

So this is far from the last one we'll see.

[Jerome Shidel]

Post by Rugxulo
Hi,

Post by Jerome Shidel
In case you haven't heard, there is a severe bug in Intel Skylake and Newer CPUs.

Most of us aren't trendy. This Nehalem Westmere (Core gen #1) is my
newest machine (so far). ;-)

What? Are you telling me DOS users aren't trendy? ;-)

Post by Rugxulo

Post by Jerome Shidel
I only know a little about it. From what I heard, it crops up with hyper-threading
inside loops less than 64 bytes effecting several high byte registers (AH, BH, etc).
They are working on a microcode patch and hopefully will have it out soon.

Don't panic! Bugs are part of life ("par for the course").

At least they claim they can fix it. :-)

Post by Rugxulo
http://www.os2museum.com/wp/vme-fixed-on-amd-ryzen/

Post by Jerome Shidel
So, I guess it could effect DOS asm code running inside a VM.
I think it's been a while since there last major bug. Last one I recall was that Pentium 90 Floating Point problem.

You mean the FDIV bug? That was fixed in 120 Mhz (and higher) models.
* https://en.wikipedia.org/wiki/Pentium_F00F_bug
* (P3 serializing bug)
http://web.archive.org/web/20070808143526/http://clio.rice.edu:80/djgpp/r5bug05.txt

I was never bit by ether. So, my memory of them is foggy at best.

Post by Rugxulo
So this is far from the last one we'll see.

No doubt. CPUs are complex little monsters.

Post by Rugxulo
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
Freedos-devel mailing list
https://lists.sourceforge.net/lists/listinfo/freedos-devel