[lkml]   [1996]   [Nov]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: What happened to the 4MB paging Pentium bug?

At 23:11 27.11.1996 +0000, Alan Cox wrote:
> Does anyone know how the B1,B3,B5 .. map to the version info
> in CPUID ??

At 15:22 27.11.1996 -0500, Andrew E. Mileski wrote:
>Version levels for the CPU.
>Note: My Pentium-166 has a stepping level of 12.

1. This should be fixed in /proc/cpuinfo:
stepping levels 10, 11, 12 should be written
as A, B and C (see Intel documentation).

2. There exist (I hope I've got them all looking through
the tables) steppings (CPUID) 0-7 and A-C.

3. Intel differentiates between steppings reported by CPUID
and "Manufacturing Steppings":

stepping man. stepping
0 mA4 (75+ MHz (model = 7 ?!?!?! - should be 2))
1 B1
2 B3
3 B1 (60/66 MHz only => model = 1)
4 B5
5 C1 (60/66 ...)
5 C2, mA1
6 E0
7 D1 (60/66 ...)

A tA0 (Overdrive 120/60, 133/66, model =1)
B cB1, mcB1
C cC0, mcC0

m - prefix means mobile (Voltage Reduction Technology, 3.3 V etc.)

Intel also says:
"The cB1 stepping is logically equivalent to the C2-step, but on a different
manufacturing process. The mcB1 step is logically equivalent to the cB1
step ..."

This means stepping B (=11) processors should have the same problems like
the 5th stepping CPU's. BUT I couldn't find anything about a 4MB paging bug
these steppings except for a "specification clarification" which states, that
after setting the CR4 PSE bit the TLB should be flushed by writing to CR3.


CONFUSED? -- read the full documentation's text at

 \ /
  Last update: 2005-03-22 13:38    [from the cache]
©2003-2017 Jasper Spaans. hosted at Digital OceanAdvertise on this site