Reducing the size of the ports tree (brainstorm v2)

Adam Vande More amvandemore at gmail.com
Wed Nov 5 01:59:33 UTC 2014


On Tue, Nov 4, 2014 at 5:39 PM, Matthias Andree <matthias.andree at gmx.de>
wrote:

> Am 03.11.2014 um 21:24 schrieb Tijl Coosemans:
>
> > Other tools won't change anything.  It's the file system that would
> > have to change which is not going to happen.  When the ports tree was
> > created disks were much smaller and file systems were better (still not
> > good) at storing small files.  Today disks are much bigger and file
> > systems have adapted to that.  Now it's time for the ports tree to adapt.
>
> So you're saying the only answer we've had to growing storage capacities
> was growing block sizes, without adding support for "many small files"
> back in.


What is this 'support for "many small files"' you are referring to?


> That's still the fault of the tool (here: tool == file system)
> and not of the ports tree.


It's a problem with disk's themselves, not an svn or fs problem.  Tail
packing is already a part of UFS, and disks for the purposes here basically
only read one block at a time.  If you have a method of overcoming the
inherent seek slowness of reading many small files scattered across the
spinning media then please share it so progress can be made.

This is definitely a problem with the ports system considering the
underlying hw and fs has changed characteristics and ports hasn't accounted
for them.

-- 
Adam


More information about the freebsd-ports mailing list