[openib-general] Re: Re: Userspace testing results (manykernels, many svn trees)

[Michael S. Tsirkin]

Quoting r. Roland Dreier <rdreier at cisco.com>:
> Subject: Re: [openib-general] Re: Re: Userspace testing results (manykernels, many svn trees)
> Yes, 32 bit is fine.  That stuff in timex.h is there to fix up some
> sort of crazy PPC 601 stuff (and 601 is so old I don't think anyone
> cares or even has a box with both PPC601 and PCI).

I guessed that much.

>     Michael> Cant I just "mftb %0, 268"?
> 
> You can just do "mftb %0" to get the low 32 bits of the timebase, and
> "mftbu %0" to get the high 32 bits.  I'm not sure what the ", 268"
> would do.

Thats what trunk head does, but Nishanth Aravamudan here sees

1 sec = 5.37731e+14 usec

which seems to indicate something's still wrong.

Thanks anyway!

-- 
MST