Ok, so far CMake is working for what I need and I'll keep moving forward, I
welcome any feed back people might have and any other use cases. I need to
see how amenable CMake would be for the kernel itself as far as I can tell
it seems to be reasonable for user space. Although, it would be nice to see
if I can use CMake's gui to possibly replace Kconfig.
So I'll keep the list informed on my progress and again I welcome any feed
On Tue, Jul 28, 2015 at 7:20 PM Adrian Danis <adrian.danis(a)nicta.com.au>
There is always interest in a new build system as
there is little belief
that our current build system is actually good. The issues are typically
* Finding someone willing to work on it
* Finding someone with the time to work on it
* Covering all current use cases
* Having enough people understand it to provide maintenance
The problem with a new build system is it tends to be a bit all or
nothing. If it cannot be used for all use cases then it needs to at least
integrate with projects and libraries that are still on the old build
system, typically creating an annoying maintenance scenario.
Not trying to dissuade or discourage you, but it is hard to change a piece
of infrastructure like this. Especially as you cannot necessarily see use
cases that we haven't published or made public yet.
On Tue 28-Jul-2015 4:45 AM, Wink Saville wrote:
In my opinion the edit/compile/test loop with the standard build system
seeed slow, so I thought I'd try to see if CMake might be any better. Here
<https://github.com/winksaville/sel4-newlibs-manifest> is repo manifest
which uses CMake for building my sel4-newlibs helloworld application and
I'm happy to report there is some improvement.
For a "clean" build it is about 2x faster, 8 seconds as compared to 14
seconds. And for a small change to apps/helloworld/src/main.c it is 20x
faster, 0.3 seconds as compared to 7 seconds.
Is there any interest in using CMake or other build system to reduce the
builds times for seL4?
Devel mailing listDevel@sel4.systemshttps://sel4.systems/lists/listinfo/devel
The information in this e-mail may be confidential and subject to legal
professional privilege and/or copyright. National ICT Australia Limited
accepts no liability for any damage caused by this email or its attachments.