Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

clarify what is UB #149

Merged
merged 21 commits into from
Aug 16, 2019
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
44 changes: 35 additions & 9 deletions src/what-unsafe-does.md
Original file line number Diff line number Diff line change
Expand Up @@ -16,18 +16,43 @@ to your program. You definitely *should not* invoke Undefined Behavior.
Unlike C, Undefined Behavior is pretty limited in scope in Rust. All the core
language cares about is preventing the following things:

* Dereferencing null, dangling, or unaligned pointers
* Reading [uninitialized memory][]
* Dereferencing (using the `*` operator on) dangling, or unaligned pointers, or
wide pointers with invalid metadata (see below)
RalfJung marked this conversation as resolved.
Show resolved Hide resolved
* Breaking the [pointer aliasing rules][]
* Producing invalid primitive values:
* dangling/null references
* null `fn` pointers
* a `bool` that isn't 0 or 1
* an undefined `enum` discriminant
* a `char` outside the ranges [0x0, 0xD7FF] and [0xE000, 0x10FFFF]
* A non-utf8 `str`
* Unwinding into another language
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is it really strictly defined to unwind between two rust libraries that are directly calling eachother via extern "C"? Otherwise we should have a more precise definition.

Specifically it might be that we just need to enumerate that you can only "exit a function via unwinding" if it has one of these 3 rust-specific calling conventions:

  • extern "rust" (the default of all fns)
  • extern "rust-call" (rust closures)
  • extern "rust-intrinsic" (things in core::intrinsics)

I think "rust-intrinsic" isn't supposed to unwind, but I might be wrong on that (perhaps the builtin operator impls have this ABI, somewhere?). But also "rust-intrinsic" kinda doesn't matter since those implementations are only provided by rustc. Only slightly matters if we care about helping devs assume raw re-exported intrinsics never unwind. But making an intrinsic a "normal" function would hardly be considered a breaking change!

Also I'm guessing incompatible compilation flags like mixing panic=abort is subsumed by the "target features" bullet below?

extern "platform-intrinsic" is also a thing but I think not relevant. I'm guessing it has something to do with libc functions that llvm is allowed to make implementation assumptions about, like malloc?)

Also I thought we had automatic guards against unwinding from an extern fn. Is that not the case? (I only work on panic=abort software so I never worry about this issue...)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is currently de facto undefined to unwind out of a Rust function defined as extern "C" no matter the caller, because our LLVM IR generation adds nounwind to such functions which makes it UB to unwind out of. IIUC lang team is saying this is deliberately and explicitly UB until explicitly stated to be supported in some cases, for which there's an ongoing RFC, rust-lang/rfcs#2699. The safeguard that aborts when you try to unwind out of an extern "C" Rust function did exist but was reverted twice so far because of fallout for software that relied on it despite it being UB, see rust-lang/rust#58794 if you want the messy history and part of the discussion that led to the aforementioned RFC.

* Causing a [data race][race]
RalfJung marked this conversation as resolved.
Show resolved Hide resolved
* Executing code compiled with [target features][] that the current thread of execution does
not support
* Producing invalid values (either alone or as a field of a compound type such
as `enum`/`struct`/array/tuple):
* a `bool` that isn't 0 or 1
* an `enum` with an invalid discriminant
* a null `fn` pointer
* a `char` outside the ranges [0x0, 0xD7FF] and [0xE000, 0x10FFFF]
* a `!` (all values are invalid for this type)
* a reference that is dangling, unaligned, points to an invalid value, or
that has invalid metadata (if wide)
* slice metadata is invalid if the slice has a total size larger than
`isize::MAX` bytes in memory
* `dyn Trait` metadata is invalid if it is not a pointer to a vtable for
`Trait` that matches the actual dynamic trait the reference points to
* a `str` that isn't valid UTF-8
* an integer (`i*`/`u*`), floating point value (`f*`), or raw pointer read from
[uninitialized memory][]
* a type with custom invalid values that is one of those values, such as a
`NonNull` that is null. (Requesting custom invalid values is an unstable
feature, but some stable libstd types, like `NonNull`, make use of it.)

"Producing" a value happens any time a value is assigned, passed to a
function/primitive operation or returned from a function/primitive operation.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

suggested reword and massive clarification:

Many have trouble accepting the consequences of invalid values, so they merit some extra discussion. The claim being made here is a very strong one, so read carefully.

A value is produced whenever it is assigned, passed to something, or returned from something. Keep in mind references get to assume their referents are valid, so you can't even create a reference to an invalid value. Additionally, uninitialized memory is always invalid, so you can't assign it to anything, pass it to anything, return it from anything, or take a reference to it. (Padding bytes are not technically part of a value's memory, and so may be left uninitialized.)

In simple and blunt terms: you cannot ever even suggest the existence of an invalid value. No, it's not ok if you "don't use" or "don't read" the value. Invalid values are instant Undefined Behaviour. The only correct way to manipulate memory that could be invalid is with raw pointers using methods like write and copy. If you want to leave a local variable or struct field uninitialized (or otherwise invalid), you must use a union or enum which clearly indicates at the type level that this memory may contain no values (see MaybeUninit for details).

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I applied most of your suggestions but this one is big enough that it is probably easier to hand the PR off to you. ;) I'd love to do a pass over what you got when you are done, if you don't mind.

I like this new text, as usual in you very pointed style! One comment though:

Additionally, uninitialized memory is always invalid, so you can't assign it to anything

That's not true for MaybeUninit.


A reference/pointer is "dangling" if it is null or not all of the bytes it
points to are part of the same allocation (so in particular they all have to be
part of *some* allocation). The span of bytes it points to is determined by the
pointer value and the size of the pointee type. As a consequence, if the span is
empty, "dangling" is the same as "non-null". Note that slices point to their
entire range, so it's very important that the length metadata is never too
large. If for some reason this is too cumbersome, consider using raw pointers.

That's it. That's all the causes of Undefined Behavior baked into Rust. Of
course, unsafe functions and traits are free to declare arbitrary other
Expand Down Expand Up @@ -58,3 +83,4 @@ these problems are considered impractical to categorically prevent.
[pointer aliasing rules]: references.html
[uninitialized memory]: uninitialized.html
[race]: races.html
[target features]: ../reference/attributes/codegen.html#the-target_feature-attribute