News for hackers. Stuff that matters.
Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by jononor on Friday April 21, 2017 @06:14AM (#14164288)
For such inlining optimization it is the compiler that 'introduces' the UB in its intermediate step. That is not a fault of the input code, so no warning should be emitted?

If the compiler can know that the pointer is non-null, why would it need to rely on UB to optimize it? Problem I guess is the ever-present possibility of pointer aliasing and such nearly allowing anything to change anything else at any time... Not easy making things more sound on such a foundation.

Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by derf_ on Friday April 21, 2017 @07:08AM (#14164499)
If the compiler can know that the pointer is non-null, why would it need to rely on UB to optimize it?

You've got this backwards. The compilers "knows" the pointer is non-null because you dereferenced it, and dereferencing null would have been undefined behavior. That's what allows the compiler to assume that didn't happen.

There are many operations which will produce undefined behavior for some inputs (e.g., every pointer dereference, every signed integer arithmetic operation). Figuring out if it is possible for a program to encounter such inputs at runtime is equivalent to the halting problem.

This is why any attempt to craft a -Wundefined-behavior that works on non-trivial programs is doomed to failure. It may be possible to prove a program will definitely invoke undefined behavior in some useful subset of cases. Compilers already do this in some of them, even. But those cases (and many more, besides) would also be caught with ubsan and a unit test.

Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by jononor on Friday April 21, 2017 @07:41AM (#14164629)
You're right, it is an intractable problem in general. I guess just use Rust if one cares about this..
Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by masklinn on Friday April 21, 2017 @07:29AM (#14164583)
> For such inlining optimization it is the compiler that 'introduces' the UB in its intermediate step.

Technically the UB is in the original code, it's the dereferencing of a pointer without checking it. The compiler can merely use that to assert dead code (checking for a null pointer after having already deref'd the pointer is nonsensical) and optimise it away.

Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by jononor on Friday April 21, 2017 @07:47AM (#14164650)
I don't support this unsafe compiler behavior. There should be a check on the pointer in sight, or a manual declaration that "this is safe, trust me". Otherwise, emit a UB warning...
Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by adrianN on Friday April 21, 2017 @09:20AM (#14165283)
Then you want to use a different language. Java for example throws NullPointerException instead of having UB for null pointer accesses.
Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by benibela on Friday April 21, 2017 @01:25PM (#14167473)
Object Pascal just calls the function with this == null (well self = nil)
Re:Cap'n'Proto remote vuln: pointer overflow check optimized... by adrianN on Friday April 21, 2017 @01:45PM (#14167636)
ON ERROR RESUME NEXT