Slashdot Mirror
FreeBSD Boots on x86-64
craig2787 writes "FreeBSD developer Peter Wemm has successfully booted FreeBSD on a real AMD ClawHammer CPU, in both 64- and 32-bit modes. Original posting to the -current mailing list is here."
← Back to Stories (view on slashdot.org)
Seriously, though, I'd eventually like to see some real world performance specs of Hammer running in 32 bit mode [...]
Benchmarks will come in time. Right now everyone that has a Hammer system is under NDA from AMD. Think about it, if everyone was posting performance numbers months before Hammer was ready for introduction, that would give Intel plenty of time to come up with some sort of response. When April 23 comes and the chip is officially released, I suspect that a ton of performance numbers will be released within a few seconds.
As for the 32-bit-on-64-bit problem, remember that the amd64 architecture is just an extension on ia32, much like ia32 was an extension of the 16-bit stuff. Code either uses the wider registers or it doesn't. The real fear is that 64-bit code won't perform as fast as 32-bit code, since 64-bit pointers/integers/etc means less efficient cache usage.
Cat, the other, tastier white meat.
According to reports I've read, the new Opteron actually outperforms the current Athlon clock-for-clock in 32-bit Legacy mode (32-bit OS running 32-bit code) because at base, the decode paths and functional unit pipelines are similar to the present generation processor. There are some additional tricks they have pulled to get some more speed, like a "smart" TLB that only flushes its cache of page table entries when truly necessary (not at every context switch).
In 64-bit Mode (where a 64-bit OS runs 64-bit code), average instruction length has increased because of the addition of a preface byte to every 64-bit instruction, but overall code size has DECREASED because 8 additional general purpose registers have been added (reducing compiler generated load/store code). This decrease in code size compensates for the larger average instruction length and enables performance to remain on-par with 32-bit.
However, I will feel much better about these claims once I have seen some true performance comparisons at an independent reviewer's site! :)
It was a joke, I changed the clock speed strings to 'pi'. This is a 10 month old pre-pre-pre-production A0-step silicon machine. I'm not allowed to talk about the speed of this particular machine. But I can say that I'm rather impressed given this particular machine's early production state. It does run very nicely in 32 bit mode. It is faster than my home desktop machine, and is faster than my work desktop (in single processor mode).
Also, note this is a Clawhammer cpu, not the Sledgehammer/Opteron that is coming up for release on April 23.
Who let the Iraqi Information Minister out of Baghdad?
Everyone sing!!
Ok, now we now go back to your regularly scheduled BSD is dying trolls