Re: Pentium DEATH in user-mode

Mike Jagdis (mike@roan.co.uk)
Mon, 10 Nov 1997 11:16:09 +0000 (GMT/BST)


On Sat, 8 Nov 1997, Andre Derrick Balsa wrote:

> Setting the NO_LOCK bit in CCR1 will prevent the deadlock caused by the
> above code sequence. Here is a short call to set6x86 that does this:
>
> set6x86 -p 0xc1 -s 0x10.
>
> Page table accesses and interrupt acknowledge cycles will still be
> executed in locked cycles, but the xchgl instruction will *not* generate
> locked cycles anymore.
>
> I don't know if setting the NO_LOCK bit will cause problems when running
> Linux. I don't think so...

It doesn't. I don't think explicit locks can be relied on to be
visible to anything that isn't sitting directly on the memory/CPU
bus. The only time Linux cares whether explicit locks are honoured
or not is when they are used to synchronize SMP systems. As far as
I know there isn't yet a Cyrix based SMP design available.

Mike

-- 
.----------------------------------------------------------------------.
|  Mike Jagdis                  |  Internet:  mailto:mike@roan.co.uk   |
|  Roan Technology Ltd.         |                                      |
|  54A Peach Street, Wokingham  |  Telephone:  +44 118 989 0403        |
|  RG40 1XG, ENGLAND            |  Fax:        +44 118 989 1195        |
`----------------------------------------------------------------------'