From: Dan W. (danw_at_[hidden])
Date: 2003-12-12 12:38:28
>Is this some kind of joke? What did you expect when you posted this
>mail? Please refrain from going wild on a public-mailing list.
Did I put smiley's on it?
>Most processors have an opcode to round a floating-point value to an
>integer. For example, as I said, this instruction is called frndint on
>x86 (how do you think floor, rint and ceil are coded?). Don't try to
>emulate a floating-point unit by using a handful of integer operations.
Sorry, when you mentioned frndint I thought it was a function name.
I've never used floating point instructions. I learned assembler just
to code in MMX and 3DNow. I know that when coding for floating point
using SIMD instructions, one tries as much as possible to use integer
instructions instead, as a way to parallelize execution, rather than
overburden the floating point pipe while the integer pipe sits idle.
>First it will be completely wrong (it's not as easy as you seem to
>believe), and second it will be slow.
I believe you are wrong. I coded a lot of linear algebra using a mix
of floating point 3DNow! and integer MMX, and I don't discount the
possibility I may have made a mistake in the code, but to accuse me
of joking or posting for any reason other than try to offer a solution
to a problem...
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk