l> Everything slowed down. The easiest way to demontsrate this is by looking
at
> these figures:
>
> raid5: measuring checksumming speed
> - 8regs : 2343.600 MB/sec
> - 32regs : 1944.000 MB/sec
> - pIII_sse : 4163.600 MB/sec
> - pII_mmx : 3584.400 MB/sec
> - p5_mmx : 4600.800 MB/sec
> -raid5: using function: pIII_sse (4163.600 MB/sec)
> + 8regs : 228.400 MB/sec
> + 32regs : 199.200 MB/sec
> + pIII_sse : 352.000 MB/sec
> + pII_mmx : 316.800 MB/sec
> + p5_mmx : 432.800 MB/sec
> +raid5: using function: pIII_sse (352.000 MB/sec)
>
> Old motherboard above, new below. (Why it chose pIII_sse even when p5_mmx
> was faster is also an interesting question... :)
I have seen the same on a precompiled slackware 8.1 raid.s kernel I tried
for my promise controller. It's an AMD AthlonXP 2000+. Like you I found that
the PIII_SSE was slower than the P5_MMX and still got selected. I got higher
numers than you though, around your old mobo's speeds.... (KT333 chipset).
(MSI KT3 Ultra2-R). specs are in the KT333, IO-APIC, Promise Fasttak, Initrd
topic.
the 5 disks spanning ram image doesn't even load properly with me, maybe
it's caused by memory problems as well? Tho' I thought north bridges are for
memory access whilst I only get a message that my southbridge isn't
recognized...
-
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/
This archive was generated by hypermail 2b29 : Thu Oct 31 2002 - 22:00:32 EST