Messages in this thread | | | Date | Tue, 27 Sep 2005 13:36:06 -0700 | From | George Anzinger <> | Subject | Re: [ANNOUNCE] ktimers subsystem |
| |
Tim Bird wrote: > Roman Zippel wrote: > >>On Sun, 25 Sep 2005, Thomas Gleixner wrote: >> >>>Can you please point out which architectures do not have a 32x32->64 >>>instruction ? > > > <snip> > >>For the rest you might want to check <asm/div64.h>, if div64 has to be >>emulated, there are good chances this instruction has to be emulated as >>well (especially in smaller embedded archs). > > > Hmmm. In my experience, there are several embedded platforms > with a 32x32->64 instruction, which are lacking a div64 instruction. > I don't think checking for div64 is a very good metric here.
Also, even having a div64 instruction does not eliminate the asm/div64.h as it checks for results that are >32-bits and does the right thing. For example, letting the x86 do this divide with such a result, results in a trap.
This is why we need to be very careful where we use the div_ll_l_rem() which accesses just this instruction.
-- George Anzinger george@mvista.com HRT (High-res-timers): http://sourceforge.net/projects/high-res-timers/ - 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/
| |