On Sat, 23 Jan 2010 12:05:19 -0200 Mauro Carvalho Chehab <mchehab@xxxxxxxxxx> wrote: > > An associative memory is complex and its complexity grows > > exponentially with its size. So, usually, it is rather small. Then, > > the more the code is small and the less MMU interrupts occur... > > Linux doesn't use swap memory for the kernel. It will be using a > physical RAM memory for the entire kernel. So, I don't think MMU > applies here. Well, it is a long time since I did not program an embedded system! In the last one, the cpu was a PowerPC maybe 603 or similar. I retrieved the MMU sequence I was talking about in the linux kernel. The associative memory is called the TLB. In the file arch/powerpc/kernel/head_32.S you may see the treatment of the tlb interrupts (InstructionTLBMiss, DataLoadTLBMiss and DataStoreTLBMiss). Such interrupts occur at the lowest level, even when the cpu is running in the kernel state. These interrupts are different from page faults which occur when a memory page is not mapped, and, hopefully, the kernel code is always mapped... Cheers. -- Ken ar c'hentañ | ** Breizh ha Linux atav! ** Jef | http://moinejf.free.fr/ -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
