Improve the error explanations for check_const #30932
Conversation
| const Y: usize = &X as *const u32 as usize; | ||
| println!("{}", Y); | ||
| static MY_STATIC_ADDR: &'static u32 = &MY_STATIC; | ||
| // ..and also |
There was a problem hiding this comment.
"// ..and also"? Isn't "// ... and also" better?
|
Looks like an overall improvement. |
|
improved |
| To fix this error, please do not assign this value to a constant expression. | ||
| Example: | ||
| To fix this, you should dereference your pointer at some point in | ||
| run-time. |
There was a problem hiding this comment.
This is wrong. This document already contains various cases of wrong advice.
There was a problem hiding this comment.
If the problem is that you need the value behind a pointer in a static variable, then not putting the value in a static variable does not fix the problem.
There was a problem hiding this comment.
It's not like you can solve the problem.
There was a problem hiding this comment.
I am not sure how important is the "how to fix this" advice, because each issue will have its own fix.
arielb1
force-pushed
the
raw-const-errors
branch
from
ba50fa0
to
5e16148
Compare
| Then MY_STATIC_ADDR would contain the address of MY_STATIC. However, | ||
| the address can change when the program is linked, as well as change | ||
| between different executions due to ASLR, and many linkers would | ||
| not be able to calculate the value of WHAT. |
There was a problem hiding this comment.
Nit: This paragraph should use inline code for the constant names.
|
r=me once the two nits are fixed. |
arielb1
force-pushed
the
raw-const-errors
branch
from
5e16148
to
cad3882
Compare
|
fixed ^ |
|
@bors r+ cad3882 |
|
@bors rollup |
|
☔ The latest upstream changes (presumably #31024) made this pull request unmergeable. Please resolve the merge conflicts. |
|
🔒 Merge conflict |
arielb1
force-pushed
the
raw-const-errors
branch
from
cad3882
to
47593da
Compare
|
@bors r=nagisa rollup |
|
📌 Commit 47593da has been approved by |
Fixes #30705
r? @nagisa