Re: Issue 37 - new work

[Han-Wen Nienhuys]

On Fri, Jan 28, 2011 at 6:01 PM, Graham Percival
<address@hidden> wrote:
> On Fri, Jan 28, 2011 at 05:42:56PM -0200, Han-Wen Nienhuys wrote:
>> On Fri, Jan 28, 2011 at 4:43 PM, Graham Percival
>> <address@hidden> wrote:
>> > That's actually precisely why I'm suggesting a
>> >  -p X
>> > option.  People (generally) aren't going to look into the depths
>>
>> that's even worse, because people that will want to use this option
>> won't even understand what it does.  It's the same time of idiocy of
>> Windows' and IE settings: do you want low, medium or high for hardware
>> acceleration, internet privacy.
>
> address@hidden:~$ gunzip --help
> Usage: gzip [OPTION]... [FILE]...
> Compress or uncompress FILEs (by default, compress FILES
> in-place).
>
> [...]
>
>  -V, --version     display version number
>  -1, --fast        compress faster
>  -9, --best        compress better
>
>
> oh dear.  I hope that nobody uses these terrible command-line
> switches to gzip!  At least bzip2 doesn't make this same
> mistake... oops, no wait, it also has -1 .. -9 options!
>
> This is terribly disheartening.  I'd better go off and download
> some pirated tv shows... oh no!  Most popular video encoding
> software has a "quality" setting, too!
>
> ok, I guess I'll go drown myself in C programming.  Surely gcc
> wouldn't be stupid enough to provide multiple optimization
> levels... oh noes!  -O -O0 -O1 -O2 -O3 -Os !
>
> I see nothing wrong with providing "easy-to-use" optimization
> levels, as long as it's possible to "dig down" and find out what
> they all do.  I've happily used -O2 for years without once looking
> at it.  A few months ago, I tried doing some heavy optimizations,
> so I went and looked at the difference between -O2 and -O3 and
> what --funsafe-math-optimizations did (I was especially curious
> because the latter produced incorrect answers for my program!).
> But I don't see anything dangerous about having -Ox switches in
> general.

Quick:  in what case should you use gcc option -ftree-vrp ?


The -O2 option to GCC demonstrates what I mean.  Nobody knows what the
levels do, so people settle on either -O2 which is the maximum that is
not a tradeoff, or no optimization at all.

In the case of lilypond, how would someone know which level to choose,
if it werent either maximum or minimum level?

-- 
Han-Wen Nienhuys - address@hidden - http://www.xs4all.nl/~hanwen