runtime: potential for self-deadlock in runtime write barriers #28993
In the runtime there's an implicit invariant that's being maintained: write barriers may not be called into while the mheap.lock is held or the runtime could deadlock. This invariant is not documented anywhere nor is it enforced via
Consider the following situation:
After some manual inspection, I'm fairly convinced that this isn't really a problem today because everywhere the heap lock is held, only
However, this invariant adds an additional mental burden when making changes to the runtime, and one day we may want to allow such write barriers (see https://go-review.googlesource.com/c/go/+/46751/12/src/runtime/mgc.go#266 for example), so perhaps the invariant doesn't need to exist at all.
traceBufs for example are managed via sysAlloc and sysFree directly, whereas GC mark work buffers live in heap-allocated spans. If we were to move GC mark work buffers to use the same model, this would effectively remove this invariant. Making this change will require some work, so I'm kicking this issue to 1.13.
The text was updated successfully, but these errors were encountered: