fix(ptr_as_ptr): handle std::ptr::null{_mut}

[KisaragiEffective]

close #11066
close #11665
close #11911

Please write a short comment explaining your change (or "none" for internal only changes)

changelog: [ptr_as_ptr]: handle std::ptr::null and std::ptr::null_mut

[rustbot]

r? @llogiq

(rustbot has picked a reviewer for you, use r? to override)

[KisaragiEffective]

I've rebased onto latest master.

[KisaragiEffective]

Note for future improvement:
This should be replaced with check of:

1. if calling function is <T>(...args) -> *{const,mut} T
2. one of:
   • args does not contain or refer T -> we can give arbitrary T
   • check if it still pass type-checking if we change T=Src to T=Dst

[y21]

Does this also work for stuff like let p = ptr::null(); p as *const u8;? I imagine that runs into the same kinds of inference issues? Though not sure if it's possible to make it more "general" (avoiding FPs through other kinds of expressions) since we'd need to know if the source pointer type had inference variables before they were unified with the cast target, or something like that (or maybe it's stored somewhere?) 🤔

[llogiq]

This looks like a solid improvement to me. Any further work can be done in a follow-up PR.

@bors r+

[bors]

📌 Commit 8eea8b1 has been approved by llogiq

It is now in the queue for this repository.

[bors]

⌛ Testing commit 8eea8b1 with merge 018cf8b...

[bors]

💔 Test failed - checks-action_test

[llogiq]

That's a weird one; doesn't seem like it has to do with this PR. Let's see if other PRs also fail.

[llogiq]

@bors retry

[bors]

⌛ Testing commit 8eea8b1 with merge 2793e8d...

[bors]

☀️ Test successful - checks-action_dev_test, checks-action_remark_test, checks-action_test
Approved by: llogiq
Pushing 2793e8d to master...