On 2/7/24 11:22, Torbjorn SVENSSON wrote:
Hi,
Is it okay to backport 71804526d3a71a8c0f189a89ce3aa615784bfd8b to
releases/gcc-13?
OK.
Without this backport, I see these failures on arm-none-eabi:
FAIL: g++.dg/contracts/contracts-ctor-dtor2.C (test for errors, line 23)
FAIL: g++.dg/contracts/contracts-ctor-dtor2.C (test for errors, line 27
Kind regards,
Torbjörn
On 2023-11-20 16:40, Jason Merrill wrote:
On 11/19/23 02:28, Alexandre Oliva wrote:
When targetm.cxx.cdtor_return_this() holds, cdtors have a
non-VOID_TYPE_P result, but IMHO this ABI implementation detail
shouldn't leak to the abstract language conceptual framework, in which
cdtors don't have return values. For contracts, specifically those
that establish postconditions on results, such a leakage is present,
and the present patch puts an end to it: with it, cdtors get an error
for result postconditions regardless of the ABI. This fixes
g++.dg/contracts/contracts-ctor-dtor2.C on arm-eabi.
Regstrapped on x86_64-linux-gnu, also tested on arm-eabi with default
cpu on trunk, and with tms570 on gcc-13. Ok to install?
OK.
for gcc/cp/ChangeLog
* contracts.cc (check_postcondition_result): Cope with
cdtor_return_this.
---
gcc/cp/contracts.cc | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/gcc/cp/contracts.cc b/gcc/cp/contracts.cc
index 66d2298a9bfac..035ca4827e758 100644
--- a/gcc/cp/contracts.cc
+++ b/gcc/cp/contracts.cc
@@ -636,7 +636,11 @@ make_postcondition_variable (cp_expr id)
bool
check_postcondition_result (tree decl, tree type, location_t loc)
{
- if (VOID_TYPE_P (type))
+ /* Do not be confused by targetm.cxx.cdtor_return_this ();
+ conceptually, cdtors have no return value. */
+ if (VOID_TYPE_P (type)
+ || DECL_CONSTRUCTOR_P (decl)
+ || DECL_DESTRUCTOR_P (decl))
{
error_at (loc,
DECL_CONSTRUCTOR_P (decl)