[paulmck-rcu:dev.2023.04.26a 35/51] arch/arm/kernel/smp.c:385:1: warning: function declared 'noreturn' should not return
From: kernel test robot
Date: Thu Apr 27 2023 - 06:26:30 EST
tree: https://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git dev.2023.04.26a
head: 56b38a90a12f76e463a7381d8cff854926fa2b59
commit: 2b927bfdda366c1719e021cd0e9688baf933c9cf [35/51] objtool: Fix for unreachable instruction warning
config: arm-milbeaut_m10v_defconfig (https://download.01.org/0day-ci/archive/20230427/202304271836.Rr0pBtSE-lkp@xxxxxxxxx/config)
compiler: clang version 17.0.0 (https://github.com/llvm/llvm-project 437b7602e4a998220871de78afcb020b9c14a661)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# install arm cross compiling tool for clang build
# apt-get install binutils-arm-linux-gnueabi
# https://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git/commit/?id=2b927bfdda366c1719e021cd0e9688baf933c9cf
git remote add paulmck-rcu https://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git
git fetch --no-tags paulmck-rcu dev.2023.04.26a
git checkout 2b927bfdda366c1719e021cd0e9688baf933c9cf
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=arm olddefconfig
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=arm SHELL=/bin/bash arch/arm/kernel/
If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <lkp@xxxxxxxxx>
| Link: https://lore.kernel.org/oe-kbuild-all/202304271836.Rr0pBtSE-lkp@xxxxxxxxx/
All warnings (new ones prefixed by >>):
>> arch/arm/kernel/smp.c:385:1: warning: function declared 'noreturn' should not return [-Winvalid-noreturn]
}
^
1 warning generated.
vim +/noreturn +385 arch/arm/kernel/smp.c
a054a811597a17 Russell King 2005-11-02 314
a054a811597a17 Russell King 2005-11-02 315 /*
a054a811597a17 Russell King 2005-11-02 316 * Called from the idle thread for the CPU which has been shutdown.
a054a811597a17 Russell King 2005-11-02 317 *
a054a811597a17 Russell King 2005-11-02 318 * Note that we disable IRQs here, but do not re-enable them
a054a811597a17 Russell King 2005-11-02 319 * before returning to the caller. This is also the behaviour
a054a811597a17 Russell King 2005-11-02 320 * of the other hotplug-cpu capable cores, so presumably coming
a054a811597a17 Russell King 2005-11-02 321 * out of idle fixes this.
a054a811597a17 Russell King 2005-11-02 322 */
9205b797dbe519 Stephen Boyd 2015-08-24 323 void arch_cpu_idle_dead(void)
a054a811597a17 Russell King 2005-11-02 324 {
a054a811597a17 Russell King 2005-11-02 325 unsigned int cpu = smp_processor_id();
a054a811597a17 Russell King 2005-11-02 326
a054a811597a17 Russell King 2005-11-02 327 idle_task_exit();
a054a811597a17 Russell King 2005-11-02 328
f36d340122ae87 Russell King 2010-11-30 329 local_irq_disable();
f36d340122ae87 Russell King 2010-11-30 330
51acdfd1fa38a2 Russell King 2013-04-18 331 /*
51acdfd1fa38a2 Russell King 2013-04-18 332 * Flush the data out of the L1 cache for this CPU. This must be
51acdfd1fa38a2 Russell King 2013-04-18 333 * before the completion to ensure that data is safely written out
51acdfd1fa38a2 Russell King 2013-04-18 334 * before platform_cpu_kill() gets called - which may disable
51acdfd1fa38a2 Russell King 2013-04-18 335 * *this* CPU and power down its cache.
51acdfd1fa38a2 Russell King 2013-04-18 336 */
51acdfd1fa38a2 Russell King 2013-04-18 337 flush_cache_louis();
51acdfd1fa38a2 Russell King 2013-04-18 338
51acdfd1fa38a2 Russell King 2013-04-18 339 /*
51acdfd1fa38a2 Russell King 2013-04-18 340 * Tell __cpu_die() that this CPU is now safe to dispose of. Once
51acdfd1fa38a2 Russell King 2013-04-18 341 * this returns, power and/or clocks can be removed at any point
51acdfd1fa38a2 Russell King 2013-04-18 342 * from this CPU and its cache by platform_cpu_kill().
51acdfd1fa38a2 Russell King 2013-04-18 343 */
1d5087ab964d84 Paul E. McKenney 2015-05-12 344 (void)cpu_report_death();
3c030beabf937b Russell King 2010-11-30 345
a054a811597a17 Russell King 2005-11-02 346 /*
51acdfd1fa38a2 Russell King 2013-04-18 347 * Ensure that the cache lines associated with that completion are
51acdfd1fa38a2 Russell King 2013-04-18 348 * written out. This covers the case where _this_ CPU is doing the
51acdfd1fa38a2 Russell King 2013-04-18 349 * powering down, to ensure that the completion is visible to the
51acdfd1fa38a2 Russell King 2013-04-18 350 * CPU waiting for this one.
51acdfd1fa38a2 Russell King 2013-04-18 351 */
51acdfd1fa38a2 Russell King 2013-04-18 352 flush_cache_louis();
51acdfd1fa38a2 Russell King 2013-04-18 353
51acdfd1fa38a2 Russell King 2013-04-18 354 /*
51acdfd1fa38a2 Russell King 2013-04-18 355 * The actual CPU shutdown procedure is at least platform (if not
51acdfd1fa38a2 Russell King 2013-04-18 356 * CPU) specific. This may remove power, or it may simply spin.
51acdfd1fa38a2 Russell King 2013-04-18 357 *
51acdfd1fa38a2 Russell King 2013-04-18 358 * Platforms are generally expected *NOT* to return from this call,
51acdfd1fa38a2 Russell King 2013-04-18 359 * although there are some which do because they have no way to
51acdfd1fa38a2 Russell King 2013-04-18 360 * power down the CPU. These platforms are the _only_ reason we
51acdfd1fa38a2 Russell King 2013-04-18 361 * have a return path which uses the fragment of assembly below.
51acdfd1fa38a2 Russell King 2013-04-18 362 *
51acdfd1fa38a2 Russell King 2013-04-18 363 * The return path should not be used for platforms which can
51acdfd1fa38a2 Russell King 2013-04-18 364 * power off the CPU.
a054a811597a17 Russell King 2005-11-02 365 */
0a301110b7bd33 Russell King 2013-01-14 366 if (smp_ops.cpu_die)
0a301110b7bd33 Russell King 2013-01-14 367 smp_ops.cpu_die(cpu);
a054a811597a17 Russell King 2005-11-02 368
668bc38669f9a6 Russell King 2014-01-11 369 pr_warn("CPU%u: smp_ops.cpu_die() returned, trying to resuscitate\n",
668bc38669f9a6 Russell King 2014-01-11 370 cpu);
668bc38669f9a6 Russell King 2014-01-11 371
a054a811597a17 Russell King 2005-11-02 372 /*
a054a811597a17 Russell King 2005-11-02 373 * Do not return to the idle loop - jump back to the secondary
a054a811597a17 Russell King 2005-11-02 374 * cpu initialisation. There's some initialisation which needs
a054a811597a17 Russell King 2005-11-02 375 * to be repeated to undo the effects of taking the CPU offline.
a054a811597a17 Russell King 2005-11-02 376 */
a054a811597a17 Russell King 2005-11-02 377 __asm__("mov sp, %0\n"
faabfa0816916b Russell King 2010-12-20 378 " mov fp, #0\n"
19f29aebd929c3 Keith Packard 2021-09-18 379 " mov r0, %1\n"
a054a811597a17 Russell King 2005-11-02 380 " b secondary_start_kernel"
a054a811597a17 Russell King 2005-11-02 381 :
19f29aebd929c3 Keith Packard 2021-09-18 382 : "r" (task_stack_page(current) + THREAD_SIZE - 8),
19f29aebd929c3 Keith Packard 2021-09-18 383 "r" (current)
19f29aebd929c3 Keith Packard 2021-09-18 384 : "r0");
a054a811597a17 Russell King 2005-11-02 @385 }
a054a811597a17 Russell King 2005-11-02 386 #endif /* CONFIG_HOTPLUG_CPU */
a054a811597a17 Russell King 2005-11-02 387
:::::: The code at line 385 was first introduced by commit
:::::: a054a811597a17ffbe92bc4db04a4dc2f1b1ea55 [ARM SMP] Add hotplug CPU infrastructure
:::::: TO: Russell King <rmk@xxxxxxxxxxxxxxxxxxxxxxx>
:::::: CC: Russell King <rmk+kernel@xxxxxxxxxxxxxxxx>
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests