On Mon, Jul 31, 2023 at 04:10:36PM +0200, Philippe Mathieu-Daudé wrote: > On 31/7/23 11:32, Daniel P. Berrangé wrote: > > > I was surprised to see that we're not using ccache in gitlab CI. It wouldn't > > help the from-clean compile time, but thereafter it ought to help. I'm doing > > some tests with that to see the impact. > > I tried that few years ago and this had very negative impact on custom > runners (maybe I wasn't doing it correctly). Hopefully that changed.
Our runner usage model has changed since then quite alot. What was previously mostly on shared runners, is now on Azure private runners. I can imagine it will vary tremendously on what you're using as a private runner. In the specific case of the windows jobs though, we're using the shared runners. Either way, if our jobs are all wired up for ccache correctly, it is then trivial to selectively turn off usage of ccache by just tweaking a few env vars. > See some previous comments: > https://lists.nongnu.org/archive/html/qemu-devel/2021-04/msg02220.html > > > Another option might be to try precompiled headers, which meson supports > > quite nicely / transparently. Might especially help on Windows where the > > entire world is declared in windows.h > > > > > > With regards, > > Daniel > With regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
