| 1 |
For those unaware, dev-util/ninja is a make-replacement created by one |
| 2 |
of the Chromium guys at Google. Its focus is on making incremental |
| 3 |
builds of large software faster. |
| 4 |
|
| 5 |
In the latest chromium ebuild (chromium-29.0.1516.3), we are using ninja |
| 6 |
instead of make. ninja auto-detects the number of jobs to run by |
| 7 |
default; in the chromium ebuild, we override this by passing -j |
| 8 |
$(makeopts_jobs) instead. |
| 9 |
|
| 10 |
Another place ninja is used cmake-utils.eclass, where it may be used |
| 11 |
optionally in place of make depending on the CMAKE_MAKEFILE_GENERATOR |
| 12 |
variable. The ninja_src_make function in cmake-utils.eclass calls ninja |
| 13 |
with ${MAKEOPTS} or not depending on the value of CMAKE_VERBOSE. This is |
| 14 |
a bit inconsistent and could cause problems because ninja only accepts a |
| 15 |
limited subset of GNU make's command line options. |
| 16 |
|
| 17 |
I'm wondering if we should create a more global function for calling |
| 18 |
ninja in a consistent way. Maybe we should introduce a NINJAOPTS |
| 19 |
variable for users to set in make.conf. Should we create a new eclass |
| 20 |
for this, or just stick it in eutils? |
| 21 |
|
| 22 |
Thoughts? |
Archives