Re: Question about LAME in FreeBSD Ports

"Robert Hegemann" <[email protected]> Wed, 18 Oct 2017 10:07:06 +0200
Newsgroups gmane.comp.audio.mp3.lame
Message-ID <[email protected]>
Am .10.2017, 16:21 Uhr, schrieb Alexander Leidinger <[email protected]>:

> Quoting Daniel Engberg <[email protected]> (from Sun, 15
> Oct 2017 20:46:59 +0200):
>
>> Many thanks for keeping the LAME port up to date, a question
>> however. Is there a reason why the tmkk patch isn't added to the
>> port or perhaps this is something new to you?
>>
>> http://tmkk.undo.jp/lame/index_e.html
>
> CCing the lame-dev list.
>
> The FreeBSD lame port is providing the official version of lame.
> Additionally I was not aware of this website.
>
> I do not plan to change the FreeBSD lame port to track something else
> than the official lame version. It is easy to add your own patches,
> just created "files/patch-xxx" in the port and it will be applied.
>
> I may try to get some time to investigate if it feasible to integrate
> the patch in the official lame version (no promises, depends upon my
> free time and the conditions the patch is provided with (license)).
>
> Bye,
> Alexander.


I had seen this patch before. From my point of view, there are some
problems with it:
- it complicates maintaining the C-code
- it is compiler specific
- it requires rewrites with different CPUs
- it looks like some functions are in assembler, even though compilers
   don't have any problems vectorizing the code already

Saying that, the patch helps identifying hot spots, where it is worth
taking a look, to assist the C compiler to get faster code paths.

Ciao Robert

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot