We plan to ship to clang 8 where a clang is needed in Firefox 68. We should create a separate project for it and retire the llvm one we have later on as the consensus seems to be to use "clang" instead of "llvm". This ticket is motivated as well by comment:43:ticket:28716.
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information
Child items ...
Show closed items
Linked items 0
Link issues together to show that they're related.
Learn more.
with https://gitweb.torproject.org/user/boklm/tor-browser-build.git/commit/?h=bug_30585&id=f0c2046ecf2e9da1574ddd6d67395fa67399b00e we move all the patch into a win-patches directory, so we don't need to list all patches individually. We also use git apply instead of patch so we don't need to handle the binary file bigobj.o.gz separately. This required fixing paths in some of the patches as git apply doesn't like them. Looking at tor-browser/build/build-clang/build-clang.py I also see that mozilla is using patch to apply the patches, so I'm wondering how it works for them.
Looks good. Merged to master as 7b6444dd34d7a9e32d619c194e078464529fd5c1. I was a bit worried about creating yet another container-image by just using git for applying patch in the clang project. However, I think that's fine as it is a temporary measure only anyway as all the LLVM patches landed upstream (dunno either why patch is working for Mozilla but not for us, weird).
Trac: Status: needs_review to closed Resolution: N/Ato fixed