| 1 |
El mar, 19-11-2019 a las 00:21 +0000, Sergei Trofimovich escribió: |
| 2 |
> repoman slows down ~linearly with amount of profiles being scanned. |
| 3 |
> In case of amd64 we have 28 stable profiles. |
| 4 |
> |
| 5 |
> To speed up processing and fit into time budged of various CIs we can |
| 6 |
> split the work across different processes that handle different profiles. |
| 7 |
> |
| 8 |
> Example benchmark on ::haskell overlay: |
| 9 |
> $ ./repoman full --include-arches=amd64 |
| 10 |
> ~65 minutes |
| 11 |
> $ ./repoman full --include-profiles=default/linux/amd64/17.0 |
| 12 |
> ~4 minutes |
| 13 |
> This allows for a crude sharding of work across processes and allows for |
| 14 |
> cheap tree-wide scans for early failures. |
| 15 |
> |
| 16 |
|
| 17 |
Just for knowing (as I guess there is a technical issue preventing that), why |
| 18 |
repoman is not trying to check one profile per core in parallel by default by |
| 19 |
itself? |
| 20 |
|
| 21 |
Thanks a lot for the info :) |
Archives