![]() | |||||||||||||
Messages in this thread |
On Llu, 2003-06-02 at 03:34, chas williams wrote: > smp machines either. i dont believe the i386 will reorder > read/writes from multiple cpus so in theory it would be safe to > do away with this lock on smp i386's. the only arch that > can reorder (and actually has posted read/writes) is the ia64/sn2. The only i386 platforms that will re-order writes I know of are the Natsemi Geode (which will do magic so that the write order seen by PCI is always ordered), Pentium Pro (fence bug) and IDT Winchip (but only to main memory) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2005-03-22 12:35 [W:1.700 / U:0.000 seconds] ©2003-2008 Jasper Spaans | |||||||||||||