Re: [PATCH bpf-next v9 2/6] mm, bpf: Introduce try_alloc_pages() for opportunistic page allocation

[kernel test robot <lkp@intel.com>]

Hi Alexei,

kernel test robot noticed the following build warnings:

[auto build test WARNING on bpf-next/master]

url:    https://github.com/intel-lab-lkp/linux/commits/Alexei-Starovoitov/locking-local_lock-Introduce-localtry_lock_t/20250222-104619
base:   https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next.git master
patch link:    https://lore.kernel.org/r/20250222024427.30294-3-alexei.starovoitov%40gmail.com
patch subject: [PATCH bpf-next v9 2/6] mm, bpf: Introduce try_alloc_pages() for opportunistic page allocation
config: x86_64-buildonly-randconfig-003-20250223 (https://download.01.org/0day-ci/archive/20250223/202502230935.fqsCRt0F-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250223/202502230935.fqsCRt0F-lkp@intel.com/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 <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202502230935.fqsCRt0F-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> mm/page_alloc.c:7103: warning: Function parameter or struct member 'nid' not described in 'try_alloc_pages'
>> mm/page_alloc.c:7103: warning: Function parameter or struct member 'order' not described in 'try_alloc_pages'


vim +7103 mm/page_alloc.c

  7087	
  7088	/**
  7089	 * try_alloc_pages - opportunistic reentrant allocation from any context
  7090	 * @nid - node to allocate from
  7091	 * @order - allocation order size
  7092	 *
  7093	 * Allocates pages of a given order from the given node. This is safe to
  7094	 * call from any context (from atomic, NMI, and also reentrant
  7095	 * allocator -> tracepoint -> try_alloc_pages_noprof).
  7096	 * Allocation is best effort and to be expected to fail easily so nobody should
  7097	 * rely on the success. Failures are not reported via warn_alloc().
  7098	 * See always fail conditions below.
  7099	 *
  7100	 * Return: allocated page or NULL on failure.
  7101	 */
  7102	struct page *try_alloc_pages_noprof(int nid, unsigned int order)
> 7103	{
  7104		/*
  7105		 * Do not specify __GFP_DIRECT_RECLAIM, since direct claim is not allowed.
  7106		 * Do not specify __GFP_KSWAPD_RECLAIM either, since wake up of kswapd
  7107		 * is not safe in arbitrary context.
  7108		 *
  7109		 * These two are the conditions for gfpflags_allow_spinning() being true.
  7110		 *
  7111		 * Specify __GFP_NOWARN since failing try_alloc_pages() is not a reason
  7112		 * to warn. Also warn would trigger printk() which is unsafe from
  7113		 * various contexts. We cannot use printk_deferred_enter() to mitigate,
  7114		 * since the running context is unknown.
  7115		 *
  7116		 * Specify __GFP_ZERO to make sure that call to kmsan_alloc_page() below
  7117		 * is safe in any context. Also zeroing the page is mandatory for
  7118		 * BPF use cases.
  7119		 *
  7120		 * Though __GFP_NOMEMALLOC is not checked in the code path below,
  7121		 * specify it here to highlight that try_alloc_pages()
  7122		 * doesn't want to deplete reserves.
  7123		 */
  7124		gfp_t alloc_gfp = __GFP_NOWARN | __GFP_ZERO | __GFP_NOMEMALLOC;
  7125		unsigned int alloc_flags = ALLOC_TRYLOCK;
  7126		struct alloc_context ac = { };
  7127		struct page *page;
  7128	
  7129		/*
  7130		 * In PREEMPT_RT spin_trylock() will call raw_spin_lock() which is
  7131		 * unsafe in NMI. If spin_trylock() is called from hard IRQ the current
  7132		 * task may be waiting for one rt_spin_lock, but rt_spin_trylock() will
  7133		 * mark the task as the owner of another rt_spin_lock which will
  7134		 * confuse PI logic, so return immediately if called form hard IRQ or
  7135		 * NMI.
  7136		 *
  7137		 * Note, irqs_disabled() case is ok. This function can be called
  7138		 * from raw_spin_lock_irqsave region.
  7139		 */
  7140		if (IS_ENABLED(CONFIG_PREEMPT_RT) && (in_nmi() || in_hardirq()))
  7141			return NULL;
  7142		if (!pcp_allowed_order(order))
  7143			return NULL;
  7144	

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