Re: Average number of instructions per line of kernel code

["Mohamed Bamakhrama" <bamakhrama at gmail dot com>]

On 8/30/07, Tom St Denis <tstdenis@ellipticsemi.com> wrote:
>
>  Mohamed Bamakhrama wrote:
>  Hi,
> I understand that it doesn't sound interesting at all for programmers.
> Nevertheless, this metric is interesting for hardware architects since
> it gives you a hint about the instruction cache behavior of your code.
> If somehow you know how many instructions can result from one
> statement *on average*, then it can help you somehow in calculating
> optimal parameters for the cache (assuming that you can find the
> length of the basic blocks within your C code)
>  I have no idea what you're trying to do.  When I worked at
> $insert_name_of_x86_processor_vendor_here$ we would use
> things like the I-CACHE miss/hit rates, code size, block size, etc.  Not the
> number of opcodes per line.
>
>  Really you have to profile your code, find the hotspots, see how large they
> are and target that.  The answer could be more ways instead of more cache,
> for instance.

Hi,
We have gone through all this cycle already :-)
Anyhow, thanks a lot for the remarks.

Greetings,

-- 
Mohamed