Re: [PATCH net-next 4/4] net: add dedicated kmem_cache for typical/small skb->head

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

Hi Eric,

I love your patch! Yet something to improve:

[auto build test ERROR on net-next/master]

url:    https://github.com/intel-lab-lkp/linux/commits/Eric-Dumazet/net-add-SKB_HEAD_ALIGN-helper/20230203-031126
patch link:    https://lore.kernel.org/r/20230202185801.4179599-5-edumazet%40google.com
patch subject: [PATCH net-next 4/4] net: add dedicated kmem_cache for typical/small skb->head
config: arc-randconfig-r014-20230204 (https://download.01.org/0day-ci/archive/20230204/202302040329.E10xZHbY-lkp@intel.com/config)
compiler: arceb-elf-gcc (GCC) 12.1.0
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
        # https://github.com/intel-lab-lkp/linux/commit/002bb9238e98f3fbeb0402c97f711420c3321b93
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Eric-Dumazet/net-add-SKB_HEAD_ALIGN-helper/20230203-031126
        git checkout 002bb9238e98f3fbeb0402c97f711420c3321b93
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=arc olddefconfig
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=arc SHELL=/bin/bash net/core/

If you fix the issue, kindly add following tag where applicable
| Reported-by: kernel test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

   net/core/skbuff.c: In function 'kmalloc_reserve':
>> net/core/skbuff.c:503:23: error: 'KMALLOC_NOT_NORMAL_BITS' undeclared (first use in this function)
     503 |             !(flags & KMALLOC_NOT_NORMAL_BITS)) {
         |                       ^~~~~~~~~~~~~~~~~~~~~~~
   net/core/skbuff.c:503:23: note: each undeclared identifier is reported only once for each function it appears in


vim +/KMALLOC_NOT_NORMAL_BITS +503 net/core/skbuff.c

   486	
   487	/*
   488	 * kmalloc_reserve is a wrapper around kmalloc_node_track_caller that tells
   489	 * the caller if emergency pfmemalloc reserves are being used. If it is and
   490	 * the socket is later found to be SOCK_MEMALLOC then PFMEMALLOC reserves
   491	 * may be used. Otherwise, the packet data may be discarded until enough
   492	 * memory is free
   493	 */
   494	static void *kmalloc_reserve(unsigned int *size, gfp_t flags, int node,
   495				     bool *pfmemalloc)
   496	{
   497		bool ret_pfmemalloc = false;
   498		unsigned int obj_size;
   499		void *obj;
   500	
   501		obj_size = SKB_HEAD_ALIGN(*size);
   502		if (obj_size <= SKB_SMALL_HEAD_CACHE_SIZE &&
 > 503		    !(flags & KMALLOC_NOT_NORMAL_BITS)) {
   504	
   505			/* skb_small_head_cache has non power of two size,
   506			 * likely forcing SLUB to use order-3 pages.
   507			 * We deliberately attempt a NOMEMALLOC allocation only.
   508			 */
   509			obj = kmem_cache_alloc_node(skb_small_head_cache,
   510					flags | __GFP_NOMEMALLOC | __GFP_NOWARN,
   511					node);
   512			if (obj) {
   513				*size = SKB_SMALL_HEAD_CACHE_SIZE;
   514				goto out;
   515			}
   516		}
   517		*size = obj_size = kmalloc_size_roundup(obj_size);
   518		/*
   519		 * Try a regular allocation, when that fails and we're not entitled
   520		 * to the reserves, fail.
   521		 */
   522		obj = kmalloc_node_track_caller(obj_size,
   523						flags | __GFP_NOMEMALLOC | __GFP_NOWARN,
   524						node);
   525		if (obj || !(gfp_pfmemalloc_allowed(flags)))
   526			goto out;
   527	
   528		/* Try again but now we are using pfmemalloc reserves */
   529		ret_pfmemalloc = true;
   530		obj = kmalloc_node_track_caller(obj_size, flags, node);
   531	
   532	out:
   533		if (pfmemalloc)
   534			*pfmemalloc = ret_pfmemalloc;
   535	
   536		return obj;
   537	}
   538	

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