linux-kernel - Re: kernel/sched/isolation.c:143 housekeeping_setup() warn: always true condition '(first

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20240522131349.GA3582@redhat.com>
Date: Wed, 22 May 2024 15:13:49 +0200
From: Oleg Nesterov <oleg@...hat.com>
To: kernel test robot <lkp@...el.com>
Cc: oe-kbuild-all@...ts.linux.dev, linux-kernel@...r.kernel.org,
	Ingo Molnar <mingo@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>, Phil Auld <pauld@...hat.com>
Subject: Re: kernel/sched/isolation.c:143 housekeeping_setup() warn: always
 true condition '(first_cpu >= 0) => (0-u32max >= 0)'

On 05/02, Oleg Nesterov wrote:
>
> I am on PTO till May 9 without my working laptop, can't even read the source code.
> I'll return to this when I'm back. CONFIG_SMP=n I guess.

Sorry for late reply.

So smatch dislikes the

	first_cpu >= setup_max_cpus

check because setup_max_cpus is 0 without CONFIG_SMP.

I don't think it makes sense to try to avoid this warning, isolation.c should
not be compiled without CONFIG_SMP=y, but we have

	config CPU_ISOLATION
		bool "CPU isolation"
		depends on SMP || COMPILE_TEST

and this randconfig selects COMPILE_TEST.

Oleg.

> On 04/30, kernel test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head:   98369dccd2f8e16bf4c6621053af7aa4821dcf8e
> > commit: 257bf89d84121280904800acd25cc2c444c717ae sched/isolation: Fix boot crash when maxcpus < first housekeeping CPU
> > date:   2 days ago
> > config: xtensa-randconfig-r081-20240429 (https://download.01.org/0day-ci/archive/20240430/202404300915.WNvfwBz3-lkp@intel.com/config)
> > compiler: xtensa-linux-gcc (GCC) 13.2.0
> > 
> > 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 <lkp@...el.com>
> > | Closes: https://lore.kernel.org/oe-kbuild-all/202404300915.WNvfwBz3-lkp@intel.com/
> > 
> > New smatch warnings:
> > kernel/sched/isolation.c:143 housekeeping_setup() warn: always true condition '(first_cpu >= 0) => (0-u32max >= 0)'
> > 
> > Old smatch warnings:
> > arch/xtensa/include/asm/thread_info.h:97 current_thread_info() warn: inconsistent indenting
> > 
> > vim +143 kernel/sched/isolation.c
> > 
> >    117	
> >    118	static int __init housekeeping_setup(char *str, unsigned long flags)
> >    119	{
> >    120		cpumask_var_t non_housekeeping_mask, housekeeping_staging;
> >    121		unsigned int first_cpu;
> >    122		int err = 0;
> >    123	
> >    124		if ((flags & HK_FLAG_TICK) && !(housekeeping.flags & HK_FLAG_TICK)) {
> >    125			if (!IS_ENABLED(CONFIG_NO_HZ_FULL)) {
> >    126				pr_warn("Housekeeping: nohz unsupported."
> >    127					" Build with CONFIG_NO_HZ_FULL\n");
> >    128				return 0;
> >    129			}
> >    130		}
> >    131	
> >    132		alloc_bootmem_cpumask_var(&non_housekeeping_mask);
> >    133		if (cpulist_parse(str, non_housekeeping_mask) < 0) {
> >    134			pr_warn("Housekeeping: nohz_full= or isolcpus= incorrect CPU range\n");
> >    135			goto free_non_housekeeping_mask;
> >    136		}
> >    137	
> >    138		alloc_bootmem_cpumask_var(&housekeeping_staging);
> >    139		cpumask_andnot(housekeeping_staging,
> >    140			       cpu_possible_mask, non_housekeeping_mask);
> >    141	
> >    142		first_cpu = cpumask_first_and(cpu_present_mask, housekeeping_staging);
> >  > 143		if (first_cpu >= nr_cpu_ids || first_cpu >= setup_max_cpus) {
> >    144			__cpumask_set_cpu(smp_processor_id(), housekeeping_staging);
> >    145			__cpumask_clear_cpu(smp_processor_id(), non_housekeeping_mask);
> >    146			if (!housekeeping.flags) {
> >    147				pr_warn("Housekeeping: must include one present CPU, "
> >    148					"using boot CPU:%d\n", smp_processor_id());
> >    149			}
> >    150		}
> >    151	
> >    152		if (cpumask_empty(non_housekeeping_mask))
> >    153			goto free_housekeeping_staging;
> >    154	
> >    155		if (!housekeeping.flags) {
> >    156			/* First setup call ("nohz_full=" or "isolcpus=") */
> >    157			enum hk_type type;
> >    158	
> >    159			for_each_set_bit(type, &flags, HK_TYPE_MAX)
> >    160				housekeeping_setup_type(type, housekeeping_staging);
> >    161		} else {
> >    162			/* Second setup call ("nohz_full=" after "isolcpus=" or the reverse) */
> >    163			enum hk_type type;
> >    164			unsigned long iter_flags = flags & housekeeping.flags;
> >    165	
> >    166			for_each_set_bit(type, &iter_flags, HK_TYPE_MAX) {
> >    167				if (!cpumask_equal(housekeeping_staging,
> >    168						   housekeeping.cpumasks[type])) {
> >    169					pr_warn("Housekeeping: nohz_full= must match isolcpus=\n");
> >    170					goto free_housekeeping_staging;
> >    171				}
> >    172			}
> >    173	
> >    174			iter_flags = flags & ~housekeeping.flags;
> >    175	
> >    176			for_each_set_bit(type, &iter_flags, HK_TYPE_MAX)
> >    177				housekeeping_setup_type(type, housekeeping_staging);
> >    178		}
> >    179	
> >    180		if ((flags & HK_FLAG_TICK) && !(housekeeping.flags & HK_FLAG_TICK))
> >    181			tick_nohz_full_setup(non_housekeeping_mask);
> >    182	
> >    183		housekeeping.flags |= flags;
> >    184		err = 1;
> >    185	
> >    186	free_housekeeping_staging:
> >    187		free_bootmem_cpumask_var(housekeeping_staging);
> >    188	free_non_housekeeping_mask:
> >    189		free_bootmem_cpumask_var(non_housekeeping_mask);
> >    190	
> >    191		return err;
> >    192	}
> >    193	
> > 
> > -- 
> > 0-DAY CI Kernel Test Service
> > https://github.com/intel/lkp-tests/wiki
> > 
>