All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jianfeng Wang <jianfeng.w.wang@oracle.com>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: llvm@lists.linux.dev, oe-kbuild-all@lists.linux.dev, linux-mm@kvack.org
Subject: [vbabka-slab:slab/for-next 11/12] mm/slub.c:3256:14: error: call to undeclared function 'node_nr_objs'; ISO C99 and later do not support implicit function declarations
Date: Mon, 22 Apr 2024 22:06:13 -0700	[thread overview]
Message-ID: <f8629dfd-642c-4f1a-929d-ebe04d8f49ea@oracle.com> (raw)
In-Reply-To: <202404231035.P6HbfrHW-lkp@intel.com>


On 4/22/24 7:21 PM, kernel test robot wrote:
> tree:   git://git.kernel.org/pub/scm/linux/kernel/git/vbabka/slab.git slab/for-next
> head:   b6f00c4ef02065161c09fdf560e492cfbb1fec0a
> commit: 1c5610f451be71ca2f8b9c6b86ef4712aeed6437 [11/12] slub: introduce count_partial_free_approx()
> config: arm-mps2_defconfig (https://urldefense.com/v3/__https://download.01.org/0day-ci/archive/20240423/202404231035.P6HbfrHW-lkp@intel.com/config__;!!ACWV5N9M2RV99hQ!OME34mL_ADYQENhnXxNwsh7RcI01QQqhK9FJsBEMMQaQVDaJM3wclQX94a6zSFnvxan0Qx-_wzC34V4$ )
> compiler: clang version 19.0.0git (https://urldefense.com/v3/__https://github.com/llvm/llvm-project__;!!ACWV5N9M2RV99hQ!OME34mL_ADYQENhnXxNwsh7RcI01QQqhK9FJsBEMMQaQVDaJM3wclQX94a6zSFnvxan0Qx-_bxtcvFE$  5ef5eb66fb428aaf61fb51b709f065c069c11242)
> reproduce (this is a W=1 build): (https://urldefense.com/v3/__https://download.01.org/0day-ci/archive/20240423/202404231035.P6HbfrHW-lkp@intel.com/reproduce__;!!ACWV5N9M2RV99hQ!OME34mL_ADYQENhnXxNwsh7RcI01QQqhK9FJsBEMMQaQVDaJM3wclQX94a6zSFnvxan0Qx-_K2OZqIs$ )
> 
> 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://urldefense.com/v3/__https://lore.kernel.org/oe-kbuild-all/202404231035.P6HbfrHW-lkp@intel.com/__;!!ACWV5N9M2RV99hQ!OME34mL_ADYQENhnXxNwsh7RcI01QQqhK9FJsBEMMQaQVDaJM3wclQX94a6zSFnvxan0Qx-_gVX_avo$ 
> 
> All errors (new ones prefixed by >>):
> 
>    In file included from mm/slub.c:13:
>    In file included from include/linux/mm.h:2208:
>    include/linux/vmstat.h:522:36: warning: arithmetic between different enumeration types ('enum node_stat_item' and 'enum lru_list') [-Wenum-enum-conversion]
>      522 |         return node_stat_name(NR_LRU_BASE + lru) + 3; // skip "nr_"
>          |                               ~~~~~~~~~~~ ^ ~~~
>>> mm/slub.c:3256:14: error: call to undeclared function 'node_nr_objs'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
>     3256 |                 x = min(x, node_nr_objs(n));
>          |                            ^
>    mm/slub.c:3256:14: note: did you mean 'node_nr_slabs'?

I'm sorry that I didn't catch this when I submitted.
For the v3 patchset, I placed this new function under
"#if defined(CONFIG_SLUB_DEBUG) || defined(SLAB_SUPPORTS_SYSFS)".
However, this causes problems: (1) node_nr_objs() is defined under
CONFIG_SLUB_DEBUG; (2) count_partial_free_approx() is only used under
CONFIG_SLUB_DEBUG. So, if CONFIG_SLUB_DEBUG is not set and
SLAB_SUPPORTS_SYSFS is set, the compilation will fail.

I've submitted a v4 that moves count_partial_free_approx() under
CONFIG_SLUB_DEBUG only, as it is not used in sysfs. Only the first patch is modified.

Here is the link:
https://lore.kernel.org/linux-mm/20240423045554.15045-1-jianfeng.w.wang@oracle.com/T/#t

>    mm/slub.c:1862:29: note: 'node_nr_slabs' declared here
>     1862 | static inline unsigned long node_nr_slabs(struct kmem_cache_node *n)
>          |                             ^
>>> mm/slub.c:3256:7: error: static assertion failed due to requirement '__builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((unsigned long)(-1)) < (unsigned long)1)) * 0L)) : (int *)8))), (((unsigned long)(-1)) < (unsigned long)1), 0) == __builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((int)(-1)) < (int)1)) * 0L)) : (int *)8))), (((int)(-1)) < (int)1), 0) || __builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((unsigned long)(-1)) < (unsigned long)1)) * 0L)) : (int *)8))), (((unsigned long)(-1)) < (unsigned long)1), 0) == __builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((int)(-1)) < (int)1)) * 0L)) : (int *)8))), (((int)(-1)) < (int)1), 0) || (__builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)(x) * 0L)) : (int *)8))) && __builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((unsigned long)(-1)) < (unsigned long)1)) * 0L)) : (int *)8))), (((unsigned long)(-1)) < (unsigned long)1), 0), x, -1) >= 0) || (__builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)(node_nr_objs(n)) * 0L)) : (int *)8))) && __builtin_choose_expr((sizeof(int) == sizeof (*(8 ? ((void *)((long)((((int)(-1)) < (int)1)) * 0L)) : (int *)8))), (((int)(-1)) < (int)1), 0), node_nr_objs(n), -1) >= 0)': min(x, node_nr_objs(n)) signedness error, fix types or consider umin() before min_t()
>     3256 |                 x = min(x, node_nr_objs(n));
>          |                     ^~~~~~~~~~~~~~~~~~~~~~~
>    include/linux/minmax.h:85:19: note: expanded from macro 'min'
>       85 | #define min(x, y)       __careful_cmp(min, x, y)
>          |                         ^~~~~~~~~~~~~~~~~~~~~~~~
>    include/linux/minmax.h:58:3: note: expanded from macro '__careful_cmp'
>       58 |                 __cmp_once(op, x, y, __UNIQUE_ID(__x), __UNIQUE_ID(__y)))
>          |                 ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>    include/linux/minmax.h:51:16: note: expanded from macro '__cmp_once'
>       51 |         static_assert(__types_ok(x, y),                 \
>          |         ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>       52 |                 #op "(" #x ", " #y ") signedness error, fix types or consider u" #op "() before " #op "_t()"); \
>          |                 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>    note: (skipping 1 expansions in backtrace; use -fmacro-backtrace-limit=0 to see all)
>    include/linux/minmax.h:31:2: note: expanded from macro '__is_signed'
>       31 |         __builtin_choose_expr(__is_constexpr(is_signed_type(typeof(x))),        \
>          |         ^
>    include/linux/build_bug.h:77:50: note: expanded from macro 'static_assert'
>       77 | #define static_assert(expr, ...) __static_assert(expr, ##__VA_ARGS__, #expr)
>          |                                  ~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
>    include/linux/build_bug.h:78:56: note: expanded from macro '__static_assert'
>       78 | #define __static_assert(expr, msg, ...) _Static_assert(expr, msg)
>          |                                                        ^~~~
>    1 warning and 2 errors generated.
> 
> 
> vim +/node_nr_objs +3256 mm/slub.c
> 
>   3226	
>   3227	static unsigned long count_partial_free_approx(struct kmem_cache_node *n)
>   3228	{
>   3229		unsigned long flags;
>   3230		unsigned long x = 0;
>   3231		struct slab *slab;
>   3232	
>   3233		spin_lock_irqsave(&n->list_lock, flags);
>   3234		if (n->nr_partial <= MAX_PARTIAL_TO_SCAN) {
>   3235			list_for_each_entry(slab, &n->partial, slab_list)
>   3236				x += slab->objects - slab->inuse;
>   3237		} else {
>   3238			/*
>   3239			 * For a long list, approximate the total count of objects in
>   3240			 * it to meet the limit on the number of slabs to scan.
>   3241			 * Scan from both the list's head and tail for better accuracy.
>   3242			 */
>   3243			unsigned long scanned = 0;
>   3244	
>   3245			list_for_each_entry(slab, &n->partial, slab_list) {
>   3246				x += slab->objects - slab->inuse;
>   3247				if (++scanned == MAX_PARTIAL_TO_SCAN / 2)
>   3248					break;
>   3249			}
>   3250			list_for_each_entry_reverse(slab, &n->partial, slab_list) {
>   3251				x += slab->objects - slab->inuse;
>   3252				if (++scanned == MAX_PARTIAL_TO_SCAN)
>   3253					break;
>   3254			}
>   3255			x = mult_frac(x, n->nr_partial, scanned);
>> 3256			x = min(x, node_nr_objs(n));
>   3257		}
>   3258		spin_unlock_irqrestore(&n->list_lock, flags);
>   3259		return x;
>   3260	}
>   3261	
> 

      reply	other threads:[~2024-04-23  5:06 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-23  2:21 [vbabka-slab:slab/for-next 11/12] mm/slub.c:3256:14: error: call to undeclared function 'node_nr_objs'; ISO C99 and later do not support implicit function declarations kernel test robot
2024-04-23  5:06 ` Jianfeng Wang [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=f8629dfd-642c-4f1a-929d-ebe04d8f49ea@oracle.com \
    --to=jianfeng.w.wang@oracle.com \
    --cc=linux-mm@kvack.org \
    --cc=llvm@lists.linux.dev \
    --cc=oe-kbuild-all@lists.linux.dev \
    --cc=vbabka@suse.cz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.