Stephanie Wilde-Hobbs
Stephanie Wilde-Hobbs
I think we should remove or rename `exit_status`. It's nothing more than a platform-specific encoding of the exit code + signal number.
I'd like to keep it one type too.
@asterite the request is for a special case to allow `sizeof` for primitive types only, like Pointer, and Int/Float types in this situation.
I saw this on 6.14 after an online resize, also with a need_discard key.
It just happened again after a resize: ``` [1693291.408531] bcachefs (f7fa14ed-5a8e-4b14-b39a-8b5c21f8bc25): reference to invalid bucket u64s 13 type alloc_v4 9:2099818:0 len 0 ver 0: gen 1 oldest_gen 0 data_type need_discard...
Oh, and `/dev/vdc` is the device index 9 from the journal log. Seems the bucket index from the error is very likely to be greater than the old nbuckets. The...
Most applications that add XDG support don't *change* the config file location, they support both. So they check if `~/.yarnrc` exists, if not use the XDG dirs. This means new...
Big +1 to vectored IO from me. `read()` and `write()` should support vectored IO overloads. Unfortunately the `to_s(io)` pattern doesn't support buffering the whole line, unless we added a `print_atomic`...
@carlhoerberg vectored IO isn't relevant to this issue, there will have to be buffering unless the design of `to_s` is radically changed to add something like IO lists from elixir....