ci: Fix CMakeLists.txt for llvm-jit to allow user define the llvm path

[Zheaoli]

Description

Fixes #187

In this PR, I make some changes following below:

1. Use the LLVM managed by the system package manager
2. Add LLVM_FIND_PATH to let people customize the LLVM path themself
3. Add ENABLE_LLVM_SHARED for some special distributions which are make the LLVM all dynamic lib (like Arch Linux, FYI https://gitlab.archlinux.org/archlinux/packaging/packages/llvm/-/commit/3e452a15b68d156a9ef5608e8f6e1870d192d954)

Type of change

• [x] Bug fix (non-breaking change which fixes an issue)
• [ ] New feature (non-breaking change which adds functionality)
• [ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)
• [ ] This change requires a documentation update

How Has This Been Tested?

• [ ] Test A
• [ ] Test B

Test Configuration:

• Firmware version:
• Hardware:
• Toolchain:
• SDK:

Checklist

• [x] My code follows the style guidelines of this project
• [x] I have performed a self-review of my own code
• [x] I have commented my code, particularly in hard-to-understand areas
• [x] I have made corresponding changes to the documentation
• [x] My changes generate no new warnings
• [x] I have added tests that prove my fix is effective or that my feature works
• [x] New and existing unit tests pass locally with my changes
• [x] Any dependent changes have been merged and published in downstream modules
• [x] I have checked my code and corrected any misspellings

[Zheaoli]

@Officeyutong Hi, thanks for the comment, I can't quote your previous comment(would you mind to keep the discussion next time?)

After the discussion, I think it would be great keep only find_package(LLVM REQUIRED CONFIG) here. The developer could use the LLVM_ROOT or -DLLVM_ROOT to configure it. I think it would be better than hard code.

Meanwhile, the PR have following thing to do:

1. Forbid LLVM-14
2. Make CI Happy

WDYT?

[Officeyutong]

@Officeyutong Hi, thanks for the comment, I can't quote your previous comment(would you mind to keep the discussion next time?)

After the discussion, I think it would be great keep only find_package(LLVM REQUIRED CONFIG) here. The developer could use the LLVM_ROOT or -DLLVM_ROOT to configure it. I think it would be better than hard code.

Meanwhile, the PR have following thing to do:

1. Forbid LLVM-14
2. Make CI Happy

WDYT?

In fact I don't know which is better, may be you should consult @yunwei37

[Officeyutong]

@Officeyutong Hi, thanks for the comment, I can't quote your previous comment(would you mind to keep the discussion next time?) After the discussion, I think it would be great keep only find_package(LLVM REQUIRED CONFIG) here. The developer could use the LLVM_ROOT or -DLLVM_ROOT to configure it. I think it would be better than hard code. Meanwhile, the PR have following thing to do:

1. Forbid LLVM-14
2. Make CI Happy

WDYT?

In fact I don't know which is better, may be you should consult @yunwei37

Note that the following workflows should be updated:

• test-aot-cli.yml
• test-examples.yml
• test-llvm-jit.yml
• test-vm.yml

[yunwei37]

I think using find_package is ok.

Maybe we can update the bpftime build-and-test documents in eunomia.dev, to tell others the best way to compile it and set the cmake variables.

Thanks for your help!

[yunwei37]

And also, is it better to delete the makefile in project root? There are too many options for the makefile to handle, maintaining the makefile and cmake together will also make user confused. What do you think?

If it's better, we can update the ci and documents in

https://eunomia.dev/bpftime/documents/build-and-test/

It can also help with #173

[Zheaoli]

I think it could be a split PR. I will simplify the build process and the doc

[yunwei37]

Yes. It should be another issue and pr. Thanks a lot!

[Zheaoli]

Also fix #203

[yunwei37]

LGTM, Thanks!