Ping for this patch https://gcc.gnu.org/pipermail/gcc-patches/2025-November/699826.html
On 4/7/26 2:02 PM, Federico Terraneo wrote:
Ping for this patch https://gcc.gnu.org/pipermail/gcc-patches/2025-November/699826.html On 2/11/26 10:19 PM, Federico Terraneo wrote:Ping for this patch On 1/24/26 9:16 AM, Federico Terraneo wrote:Ping for this patch On 12/18/25 9:29 PM, Federico Terraneo wrote:Ping for this patch On 12/3/25 10:33 AM, Federico Terraneo wrote:Ping for this patch On 11/19/25 11:59 AM, Federico Terraneo wrote:Ping for this patch Best regards, Federico On 11/6/25 6:40 PM, Federico Terraneo wrote:As discussed in https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122078GCC defines the __atomic_test_and_set builtin in gcc/sync- builtins.def On architectures without atomic operations this builtin gets implementedas a call to a function with the same name. Libatomic does not provide a symbol named __atomic_test_and_set resulting in an "undefined reference to `__atomic_test_and_set'" link-time error.This bug affects different targets such as m68k-rtems and microblaze whofiled separate bug reports https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118977 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118280 It also affects armv6-m (cortex-m0), as we discovered testing our target, arm-miosix-eabi which is for the Miosix RTOS for which we maintain patches outside of the GCC repository. This patch modifies libatomic to provide the __atomic_test_and_setsymbol as an alias to __atomic_test_and_set_1, a compatible symbol already present in libatomic.We tested against our target (arm-miosix-eabi). The patch was also tested on rtems for both ARM and m68k:https://gcc.gnu.org/bugzilla/show_bug.cgi?id=118977#c9 PS: This is the first patch I propose to GCC, comments welcome. Best regards, Federico
OpenPGP_signature.asc
Description: OpenPGP digital signature
