[Bug c++/99858] New: Wrong throw-expression behaviour with reference to pointer

[Bug c++/99858] New: Wrong throw-expression behaviour with reference to pointer

[ibrbulat at yandex dot ru]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99858

            Bug ID: 99858
           Summary: Wrong throw-expression behaviour with reference to
                    pointer
           Product: gcc
           Version: 10.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: ibrbulat at yandex dot ru
  Target Milestone: ---

If you catch an exception of pointer type by reference to pointer, change the
value of the pointer inside this catch block and then rethrow it via *throw*
with no operand, compiler will create a copy of initial exception(!) and no
changes will be seen in the next handler.
Minimal example is here https://godbolt.org/z/T11939EYM

This behavior is contrary to the C ++ language standard (e.g. C++17):
[expr.throw]: "... 3. A throw-expression with no operand rethrows the currently
handled exception (18.3). The exception is reactivated with the existing
exception object; no new exception object is created. ..."
[except.throw]: "... If a handler exits by rethrowing, control is passed to
another handler for the same exception object.  ..."

However it works well with other (non-pointer) types - no additional copy is
created, all handlers work with the same object and see each other changes
(example: https://godbolt.org/z/Ea6r1z7rE)

[Bug c++/99858] Wrong throw-expression behaviour with reference to pointer

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99858

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2021-04-01
     Ever confirmed|0                           |1
           Keywords|                            |wrong-code

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
clang behaves as you expect, ICC behaves the same as GCC so it's unclear (it
might be that the ABI makes GCC/ICC behavior more natural to follow).

[Bug c++/99858] Wrong throw-expression behaviour with reference to pointer

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99858

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
As requested by https://gcc.gnu.org/bugs/ please provide code, not just a URL.

Reduced:

extern "C" int printf(const char*, ...);

int main() try {
    try {
        throw (void*)1;
    } catch (void*& ptr) {
        printf("%p %p\n", ptr, &ptr);
        ptr = nullptr;
        throw;
    }
} catch (void*& ptr) {
    printf("%p %p\n", ptr, &ptr);
}

Not a regression. I'll take a look at the EH code.

[Bug c++/99858] Wrong throw-expression behaviour with reference to pointer

[foom at fuhm dot net]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99858

James Y Knight <foom at fuhm dot net> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |foom at fuhm dot net

--- Comment #3 from James Y Knight <foom at fuhm dot net> ---
This is a mess.

Clang gets it "right" only through a really awful hack, which breaks
std::rethrow_exception, and which I'm now proposing to remove.
See https://github.com/llvm/llvm-project/issues/55340#issuecomment-1152755112

I don't believe there's a way to actually get this correct in the Itanium ABI
as it stands.

[Bug c++/99858] Wrong throw-expression behaviour with reference to pointer

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99858

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fchelnokov at gmail dot com

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
*** Bug 112620 has been marked as a duplicate of this bug. ***