[bcain:bcain/hmx 105/114] arch/hexagon/kernel/process.c:76:2: error: call to undeclared function 'flush_smp_call_function_queue'; ISO C99 and later do not support implicit function declarations

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/bcain/linux.git bcain/hmx
head:   d18cfcbc89df32524d340991739f2a7695263a90
commit: a05fde3938d152d5f90a2f9a249590f592950a23 [105/114] hexagon: flush SMP call-function queue before entering idle
config: hexagon-randconfig-001-20260810 (https://download.01.org/0day-ci/archive/20260810/[email protected]/config)
compiler: clang version 24.0.0git (https://github.com/llvm/llvm-project 12df34b8469b8095359de8c249cb1b2753fadeea)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260810/[email protected]/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/

All errors (new ones prefixed by >>):

>> arch/hexagon/kernel/process.c:76:2: error: call to undeclared function 'flush_smp_call_function_queue'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
      76 |         flush_smp_call_function_queue();
         |         ^
   1 error generated.


vim +/flush_smp_call_function_queue +76 arch/hexagon/kernel/process.c

    55	
    56	/*
    57	 * Flush pending SMP call-function work before entering __vmwait().
    58	 *
    59	 * On platforms with high IPI delivery latency (e.g. Hexagon hardware
    60	 * threads behind a hypervisor under QEMU TCG emulation), the interrupt
    61	 * from __vmintop_post() can take an extremely long time to wake the
    62	 * target thread from __vmwait().  During this time, call_single_data_t
    63	 * objects sit locked on the call_single_queue.  Since v5.7 (commit
    64	 * 5a18ceca6350), smp_call_function_single_async() returns -EBUSY when
    65	 * it finds a locked CSD instead of spinning, and callers like the
    66	 * scheduler's NOHZ kick and hrtimer migration silently drop the work.
    67	 *
    68	 * By draining the queue here — with IRQs disabled but before the RCU
    69	 * idle context (ct_cpuidle_enter) — we unlock CSDs while the CPU is
    70	 * still responsive.  This shrinks the -EBUSY window to just the
    71	 * __vmwait() duration and prevents the cascading stalls that otherwise
    72	 * hang SMP boot on v5.7+ kernels.
    73	 */
    74	void arch_cpu_idle_enter(void)
    75	{
  > 76		flush_smp_call_function_queue();
    77	}
    78	

--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki