[slow_vector_initialization]: clarify why Vec::new() + resize is worse

[y21]

#11198 extended this lint to also warn on Vec::new() + resize(0, len), but did not update the lint documentation, so it left some confused (#10938 (comment)).
This PR should make it a bit more clear. (cc @djc @vi what do you think about this?)

More details

Godbolt for Vec::new() + .resize(x, 0): https://godbolt.org/z/e7q9xc9rG

The resize call first does a normal allocation (__rust_alloc):

alloc::raw_vec::finish_grow:
  ...
  cmp     qword ptr [rcx + 8], 0
  je      .LBB1_7  ; if capacity == 0 -> LBB1_7

.LBB1_7:
  ...
  call    qword ptr [rip + __rust_alloc@GOTPCREL]

Then a memset for zero initialization:

example::f:
  ...
  xor     esi, esi  ; 0
  call    qword ptr [rip + memset@GOTPCREL]

---

Godbolt for vec![0; len]: https://godbolt.org/z/M3vr53vWY

Important bit:

example::f:
  ...
  call    qword ptr [rip + __rust_alloc_zeroed@GOTPCREL]

changelog: [slow_vector_initialization]: clarify why Vec::new() + resize is worse than vec![0; len]

[rustbot]

r? @xFrednet

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

[vi]

@vi what do you think about this?

It makes it clear that advantage is because of zero value. I though that zero values matter for .bss vs .data, but allocation optimisations based on zero values are maybe less typical yet.

Is the lint inhibited for Vec::new() + resize(len, 1), i.e. when it is statically known that the value is not zero, so vector initialisation needs to be an explicit step anyway? If not, the help message may be edited.

[djc]

This is great, thanks!

[y21]

Is the lint inhibited for Vec::new() + resize(len, 1), i.e. when it is statically known that the value is not zero, so vector initialisation needs to be an explicit step anyway? If not, the help message may be edited.

It only lints if the value argument in .resize(len, val) (or .extend(repeat(val).take(len))) is the integer literal "0", so yep it wouldn't lint for .resize(len, 1)

[djc]

Yup, looks great to me!

[xFrednet]

LGTM, thank you for the edit and review support @vi & @djc

bors r=xFrednet,djc

[xFrednet]

I forgot the @ ^^

@bors r=xFrednet,djc

[bors]

📌 Commit dd25cc3 has been approved by xFrednet,djc

It is now in the queue for this repository.

[bors]

⌛ Testing commit dd25cc3 with merge add2722...

[bors]

☀️ Test successful - checks-action_dev_test, checks-action_remark_test, checks-action_test
Approved by: xFrednet,djc
Pushing add2722 to master...