| 1 |
On Tue, 10 Jun 2008 09:56:18 -0400 |
| 2 |
Richard Freeman <rich0@g.o> wrote: |
| 3 |
> Joe Peterson wrote: |
| 4 |
> > No, I have not profiled PMs to try this, but you are saying that |
| 5 |
> > reading the first few lines of a file is not faster than sourcing |
| 6 |
> > the whole thing with bash? Remember that it could abort the minute |
| 7 |
> > it sees a non '#' or blank line, which would be after the first few. |
| 8 |
> > |
| 9 |
> |
| 10 |
> Actually, if you specified that the EAPI goes on the first line, then |
| 11 |
> you'd only need to read the first line. |
| 12 |
> |
| 13 |
> And to the extent that the rest got read into memory anyway by the |
| 14 |
> kernel, well then it is already in the cache when whatever mechanism |
| 15 |
> is invoked to parse the rest of the file. |
| 16 |
|
| 17 |
Except that currently, the ebuild file isn't opened for read. So it's |
| 18 |
not in memory at all. |
| 19 |
|
| 20 |
> I don't think that filename-vs-first-line is going to make a big |
| 21 |
> difference in practical performance. |
| 22 |
|
| 23 |
It's about a factor of five difference in cold-cache resolution |
| 24 |
performance for Paludis. |
| 25 |
|
| 26 |
-- |
| 27 |
Ciaran McCreesh |
Archives