[llvmlinux] Quick build against stable versions of toolchain (llvm and clang) and Linux-kernel?

Sedat Dilek sedat.dilek at gmail.com
Mon Jan 21 14:58:34 UTC 2013


On Mon, Jan 21, 2013 at 3:37 PM, Behan Webster
<behanw at converseincode.com> wrote:
> On 01/21/13 07:39, Sedat Dilek wrote:
>>
>> I think people are interested in "stable" work against stable releases.
>
> We don't have "releases". The output of this project isn't a compiler and a
> kernel; the output is test results, and upstreamed patches. The LLVMLinux
> project is a meta-project.
>
> The point of the project is to figure out the what changes are needed in
> both LLVM and the Linux kernel and how to get those changes upstream. Since
> a really big part of that decision is made by the upstream maintainers (they
> don't always want to accept our changes as is, or in the way we've
> implemented them) we have absolutely no way of providing a "stable release"
> without forking each of the upstream projects; forking code is definitely
> not the point here.
>
> The LLVMLinux project's goal is to contribute to a stable upstream LLVM
> toolchain release which has the fixes in it in order to be able to compile a
> Linux kernel, and to contribute to a stable upstream kernel code release
> which works with that LLVM toolchain.
>
> The checkpoint mechanism that Tinti is talking about is a configuration file
> which allows you to use the build environment we have to checkout, patch and
> configure LLVM, and the Linux kernel for a particular build target such that
> you can test against that one target. But it does allow people to build and
> test against a particular version of the code cross the multiple git repos
> involved in order to try a known-good version of the patches against
> known-good versions of the involved code.
>
> However, it you aren't using the LLVMLinux build environment, the settings
> file won't help too much.
>
> Is there any way we could convince you to try out the build system? If you
> work within it your efforts are a lot more useful to the rest of us.
>

Hmmm, I don't like this approach as I am concerned this will lead to a
"development loop and never-ending cycle".
Working against mainline (upstream) is hard enough.
Even harder you do against toolchain (LLVM plus CLANG) and the Linux-kernel!

See for example the development of enlightenment E17:
You could always build-from-svn (which was quite stable IMHO), but it
took 12 years to throw out a "stable" release.
And the same concerns I have with your approach.
This might be more a psychological thingie.

Is there something speaking against to backport LLVMLinux against the
latest ***stable*** toolchain and Linux-kernel release-versions?
And integrate others work into the build-system?
What is speaking against offering snapshots or even pre-build
toolchains for downloading?
The LLVM project itself offers prebuild-toolchains for Ubuntu as an example.
BTW, it is not ***your*** time :-).
Think a bit of a more user and/or developer friendly way.

Personally, I do not want to waste my time with building toolchains
the whole day.
I want to spend my time squashing CLANG bugs in mainline kernels
(preferred -rcX).
Switching the toolchain can cause extra work, I am not willing to
parallelly fix my toolchain.
I want to use my toolchain.

These are my personal thought, hope I am not hurting your feelings!

Shall I sent or sent not my patches to the LLVMLinux ML?
If there is no interest, please let me know.

- Sedat -

> Thanks,
>
>
> Behan
>
> --
> Behan Webster
> behanw at converseincode.com
>


More information about the LLVMLinux mailing list