Includes two test fixes (since chained mappings are no longer allowed)
and adds a new test for multiple overlays.
Uses other than CompilerInvocation.cpp are simple 1:1 mappings, but
without the need to read into a buffer first.
Depends on D121425
Details
Details
- Reviewers
keith dexonsmith JDevlieghere vsapsai
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
| llvm/include/llvm/Support/Error.h | ||
|---|---|---|
| 1284 | Should this be in a change all by itself? | |
Comment Actions
Includes two test fixes (since chained mappings are no longer allowed)
and adds a new test for multiple overlays.
Are we sure no one needs to support chained mappings? Has there been a clang-dev discourse discussion about it already? Just concerned that some vendor might rely on being able to support this.
| llvm/include/llvm/Support/Error.h | ||
|---|---|---|
| 1284 | Yes, but I also know you already split this out so I guess you just need to rebase :). | |
Comment Actions
I'm not *positive*, no, but I would be fairly surprised. You could just add the A -> C mapping if you really do want it. But I can start up that conversation if you think it needs having.
I actually didn't initially realise that there was a test for this case - vfsroot-with-overlay.c did test "indirection", but I completely missed it when I was looking through. I *thought* the only case was the one Nathan added in directory.c (and in that case what we really wanted was was what's now fallback).
@vsapsai do you know any clients of the chaining/nesting/indirection?
| llvm/include/llvm/Support/Error.h | ||
|---|---|---|
| 1284 | Heh, when you suggested I pull the rename out I basically thought "welll, that answer this question" So yep, it's out and I'll rebase this one :). Thanks! | |