| 1 |
On 14/08/17 08:39, Zac Medico wrote: |
| 2 |
> From: Hidehiko Abe <hidehiko@××××××××.org> |
| 3 |
> |
| 4 |
> doins is written in bash. However, specifically in case that |
| 5 |
> too many files are installed, it is very slow. |
| 6 |
> This CL rewrites the script in python for performance. |
| 7 |
> |
| 8 |
> BUG=chromium:712659 |
| 9 |
> TEST=time (./setup_board --forace && \ |
| 10 |
> ./build_package --withdev && \ |
| 11 |
> ./build_image --noenable_rootfs_verification test) |
| 12 |
> ===Before=== |
| 13 |
> real 21m35.445s |
| 14 |
> user 93m40.588s |
| 15 |
> sys 21m31.224s |
| 16 |
> |
| 17 |
> ===After=== |
| 18 |
> real 17m30.106s |
| 19 |
> user 94m1.812s |
| 20 |
> sys 20m13.468s |
| 21 |
> |
| 22 |
I know I'm gonna get chewed out on this one, but here goes anyway ... |
| 23 |
|
| 24 |
Surely for a package like chromium, who's build time is already in the |
| 25 |
'hours' range anyway, surely a couple of minutes gain for the install |
| 26 |
phase is neither here nor there?! If there were some genuine filesystem |
| 27 |
iop gains/etc for this change, I think I'd likely support it further .. |
| 28 |
|
| 29 |
On this basis, what do the performance differences look like on an |
| 30 |
'average' package .. and are there any regressions in this regard?! |
| 31 |
|
| 32 |
I take issue with the copyright assignment, as I believe the legal .. |
| 33 |
err .. IANAL devs are campaigning for full rights to be owned and |
| 34 |
enforced by Gentoo Inc LLC in the US .. even if they have no idea what |
| 35 |
that means or does .. :] |
| 36 |
|
| 37 |
MJE |
Archives