Nathan Ridge

> My suggestion is, could clangd keep using the symlink path, or at least give this symlink path higher priority, if so, all A, B, C headers would be located...

> > CompileFlags: Add: modifies the command from the CDB (compile_commands.json) entry used for a file. The entry (and thus the compile_commands.json file) still needs to exist. > > Why...

> Is there a way to point to the root of the project with some variable like `workspaceFolder`. For `-I` flags I can't just specify a _relative path_. Could be...

Right, usually in a case like this you will want to add `--query-driver=/home/cozy/Downloads/as4e-ide/plugins/com.atmel.avr.toolchains.linux.x86_64_3.0.0.201009140852/os/linux/x86_64/bin/avr32-gcc` to clangd's arguments.

> @i-ky https://logpaste.com/Of3j0b9O This log doesn't contain any "System include extraction: not allowed driver" error, so it sounds like you've resolved the problem in the issue description?

Since clangd is built on top of clang, a good rule of thumb is, if you can compile your code with clang as the compiler, you should be able to...

See also #1247 for configuring quoted vs. angled in particular.

@tom-anders FYI, Phabricator has been [taken down](https://discourse.llvm.org/t/llvm-phabricator-turndown/76137). Could you resubmit your patch at https://reviews.llvm.org/D141800 as a Github PR? (And sorry for being unresponsive on the Phabricator review thus far.)

@dmpolukhin Do you have any ideas about the performance concern discussed in earlier comments? To elaborate on it a bit: header insertion is implemented by placing a text edit describing...

> the main problem with performance happens due to not using `completionItem/resolve` requests I looked into this a bit, and it seems like an interesting approach, though I can think...