All of lore.kernel.org
 help / color / mirror / Atom feed
From: kernel test robot <lkp@intel.com>
To: Hongzhen Luo <hongzhen@linux.alibaba.com>
Cc: oe-kbuild-all@lists.linux.dev
Subject: Re: [RFC PATCH v5 4/4] erofs: introduce .fadvise for page cache share
Date: Mon, 6 Jan 2025 02:52:19 +0800	[thread overview]
Message-ID: <202501060221.LPQZaeA4-lkp@intel.com> (raw)
In-Reply-To: <20250105151208.3797385-5-hongzhen@linux.alibaba.com>

Hi Hongzhen,

[This is a private test report for your RFC patch.]
kernel test robot noticed the following build errors:

[auto build test ERROR on xiang-erofs/dev-test]
[also build test ERROR on xiang-erofs/dev xiang-erofs/fixes linus/master v6.13-rc5 next-20241220]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Hongzhen-Luo/erofs-move-struct-erofs_anon_fs_type-to-super-c/20250105-231438
base:   https://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs.git dev-test
patch link:    https://lore.kernel.org/r/20250105151208.3797385-5-hongzhen%40linux.alibaba.com
patch subject: [RFC PATCH v5 4/4] erofs: introduce .fadvise for page cache share
config: loongarch-randconfig-001-20250106 (https://download.01.org/0day-ci/archive/20250106/202501060221.LPQZaeA4-lkp@intel.com/config)
compiler: loongarch64-linux-gcc (GCC) 14.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250106/202501060221.LPQZaeA4-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/202501060221.LPQZaeA4-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from fs/erofs/pagecache_share.c:10:
   fs/erofs/pagecache_share.c: In function 'erofs_pcshr_init_mnt':
>> fs/erofs/pagecache_share.c:66:32: error: invalid application of 'sizeof' to incomplete type 'struct interval_tree_node'
      66 |                         sizeof(struct interval_tree_node), 0,
         |                                ^~~~~~
   include/linux/slab.h:430:55: note: in definition of macro 'kmem_cache_create'
     430 |                 default: __kmem_cache_create)(__name, __object_size, __args, __VA_ARGS__)
         |                                                       ^~~~~~~~~~~~~
   fs/erofs/pagecache_share.c: In function 'erofs_pcshr_free_inode':
>> fs/erofs/pagecache_share.c:163:15: error: implicit declaration of function 'interval_tree_iter_first'; did you mean 'vma_interval_tree_iter_first'? [-Wimplicit-function-declaration]
     163 |         seg = interval_tree_iter_first(&vi->segs, 0, LLONG_MAX);
         |               ^~~~~~~~~~~~~~~~~~~~~~~~
         |               vma_interval_tree_iter_first
>> fs/erofs/pagecache_share.c:163:13: error: assignment to 'struct interval_tree_node *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
     163 |         seg = interval_tree_iter_first(&vi->segs, 0, LLONG_MAX);
         |             ^
>> fs/erofs/pagecache_share.c:165:28: error: implicit declaration of function 'interval_tree_iter_next'; did you mean 'vma_interval_tree_iter_next'? [-Wimplicit-function-declaration]
     165 |                 next_seg = interval_tree_iter_next(seg, 0, LLONG_MAX);
         |                            ^~~~~~~~~~~~~~~~~~~~~~~
         |                            vma_interval_tree_iter_next
   fs/erofs/pagecache_share.c:165:26: error: assignment to 'struct interval_tree_node *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
     165 |                 next_seg = interval_tree_iter_next(seg, 0, LLONG_MAX);
         |                          ^
>> fs/erofs/pagecache_share.c:166:17: error: implicit declaration of function 'interval_tree_remove'; did you mean 'vma_interval_tree_remove'? [-Wimplicit-function-declaration]
     166 |                 interval_tree_remove(seg, &vi->segs);
         |                 ^~~~~~~~~~~~~~~~~~~~
         |                 vma_interval_tree_remove
   fs/erofs/pagecache_share.c: In function 'erofs_pcshr_fadvise':
   fs/erofs/pagecache_share.c:276:13: error: assignment to 'struct interval_tree_node *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
     276 |         seg = interval_tree_iter_first(&vi->segs, start, end);
         |             ^
   fs/erofs/pagecache_share.c:278:26: error: assignment to 'struct interval_tree_node *' from 'int' makes pointer from integer without a cast [-Wint-conversion]
     278 |                 next_seg = interval_tree_iter_next(seg, start, end);
         |                          ^
   In file included from include/linux/kernel.h:28,
                    from include/linux/cpumask.h:11,
                    from arch/loongarch/include/asm/processor.h:9,
                    from arch/loongarch/include/asm/thread_info.h:15,
                    from include/linux/thread_info.h:60,
                    from include/asm-generic/current.h:6,
                    from ./arch/loongarch/include/generated/asm/current.h:1,
                    from include/linux/mutex.h:14,
                    from fs/erofs/pagecache_share.c:8:
>> fs/erofs/pagecache_share.c:283:35: error: invalid use of undefined type 'struct interval_tree_node'
     283 |                 l = max_t(u64, seg->start | 0ULL, start);
         |                                   ^~
   include/linux/minmax.h:93:23: note: in definition of macro '__cmp_once_unique'
      93 |         ({ type ux = (x); type uy = (y); __cmp(op, ux, uy); })
         |                       ^
   include/linux/minmax.h:221:27: note: in expansion of macro '__cmp_once'
     221 | #define max_t(type, x, y) __cmp_once(max, type, x, y)
         |                           ^~~~~~~~~~
   fs/erofs/pagecache_share.c:283:21: note: in expansion of macro 'max_t'
     283 |                 l = max_t(u64, seg->start | 0ULL, start);
         |                     ^~~~~
   fs/erofs/pagecache_share.c:284:35: error: invalid use of undefined type 'struct interval_tree_node'
     284 |                 r = min_t(u64, seg->last | 0ULL, end);
         |                                   ^~
   include/linux/minmax.h:93:23: note: in definition of macro '__cmp_once_unique'
      93 |         ({ type ux = (x); type uy = (y); __cmp(op, ux, uy); })
         |                       ^
   include/linux/minmax.h:213:27: note: in expansion of macro '__cmp_once'
     213 | #define min_t(type, x, y) __cmp_once(min, type, x, y)
         |                           ^~~~~~~~~~
   fs/erofs/pagecache_share.c:284:21: note: in expansion of macro 'min_t'
     284 |                 r = min_t(u64, seg->last | 0ULL, end);
         |                     ^~~~~
   fs/erofs/pagecache_share.c:289:24: error: invalid use of undefined type 'struct interval_tree_node'
     289 |                 if (seg->start < l) {
         |                        ^~
   fs/erofs/pagecache_share.c:291:32: error: invalid use of undefined type 'struct interval_tree_node'
     291 |                         new_seg->start = seg->start;
         |                                ^~
   fs/erofs/pagecache_share.c:291:45: error: invalid use of undefined type 'struct interval_tree_node'
     291 |                         new_seg->start = seg->start;
         |                                             ^~
   fs/erofs/pagecache_share.c:292:32: error: invalid use of undefined type 'struct interval_tree_node'
     292 |                         new_seg->last = l;
         |                                ^~
>> fs/erofs/pagecache_share.c:293:25: error: implicit declaration of function 'interval_tree_insert'; did you mean 'vma_interval_tree_insert'? [-Wimplicit-function-declaration]
     293 |                         interval_tree_insert(new_seg, &vi->segs);
         |                         ^~~~~~~~~~~~~~~~~~~~
         |                         vma_interval_tree_insert
   fs/erofs/pagecache_share.c:297:28: error: invalid use of undefined type 'struct interval_tree_node'
     297 |                 if (r < seg->last) {
         |                            ^~
   fs/erofs/pagecache_share.c:299:32: error: invalid use of undefined type 'struct interval_tree_node'
     299 |                         new_seg->start = r;
         |                                ^~
   fs/erofs/pagecache_share.c:300:32: error: invalid use of undefined type 'struct interval_tree_node'
     300 |                         new_seg->last = seg->last;
         |                                ^~
   fs/erofs/pagecache_share.c:300:44: error: invalid use of undefined type 'struct interval_tree_node'
     300 |                         new_seg->last = seg->last;
         |                                            ^~
   fs/erofs/pagecache_share.c: In function 'erofs_pcshr_read_end':
   fs/erofs/pagecache_share.c:367:12: error: invalid use of undefined type 'struct interval_tree_node'
     367 |         seg->start = folio_index(folio);
         |            ^~
   fs/erofs/pagecache_share.c:368:12: error: invalid use of undefined type 'struct interval_tree_node'
     368 |         seg->last = seg->start + (folio_size(folio) >> PAGE_SHIFT);
         |            ^~
   fs/erofs/pagecache_share.c:368:24: error: invalid use of undefined type 'struct interval_tree_node'
     368 |         seg->last = seg->start + (folio_size(folio) >> PAGE_SHIFT);
         |                        ^~
   fs/erofs/pagecache_share.c:369:16: error: invalid use of undefined type 'struct interval_tree_node'
     369 |         if (seg->last > (vi->vfs_inode.i_size >> PAGE_SHIFT))
         |                ^~
   fs/erofs/pagecache_share.c:370:20: error: invalid use of undefined type 'struct interval_tree_node'
     370 |                 seg->last = vi->vfs_inode.i_size >> PAGE_SHIFT;
         |                    ^~
   In file included from fs/erofs/pagecache_share.c:13:
   fs/erofs/pagecache_share.c:371:22: error: invalid use of undefined type 'struct interval_tree_node'
     371 |         DBG_BUGON(seg->last < seg->start);
         |                      ^~
   fs/erofs/internal.h:32:41: note: in definition of macro 'DBG_BUGON'
      32 | #define DBG_BUGON(x)            ((void)(x))
         |                                         ^
   fs/erofs/pagecache_share.c:371:34: error: invalid use of undefined type 'struct interval_tree_node'
     371 |         DBG_BUGON(seg->last < seg->start);
         |                                  ^~
   fs/erofs/internal.h:32:41: note: in definition of macro 'DBG_BUGON'
      32 | #define DBG_BUGON(x)            ((void)(x))
         |                                         ^
   fs/erofs/pagecache_share.c: In function 'erofs_pcshr_readahead_end':
   fs/erofs/pagecache_share.c:422:12: error: invalid use of undefined type 'struct interval_tree_node'
     422 |         seg->start = start;
         |            ^~
   fs/erofs/pagecache_share.c:423:12: error: invalid use of undefined type 'struct interval_tree_node'
     423 |         seg->last = readahead_pos(rac) >> PAGE_SHIFT;
         |            ^~
   fs/erofs/pagecache_share.c:424:16: error: invalid use of undefined type 'struct interval_tree_node'
     424 |         if (seg->last > (vi->vfs_inode.i_size >> PAGE_SHIFT))
         |                ^~
   fs/erofs/pagecache_share.c:425:20: error: invalid use of undefined type 'struct interval_tree_node'
     425 |                 seg->last = vi->vfs_inode.i_size >> PAGE_SHIFT;
         |                    ^~


vim +66 fs/erofs/pagecache_share.c

   > 8	#include <linux/mutex.h>
     9	#include <uapi/linux/fadvise.h>
  > 10	#include <linux/slab.h>
    11	#include <linux/pagemap.h>
    12	#include "pagecache_share.h"
    13	#include "internal.h"
    14	#include "xattr.h"
    15	
    16	#define PCSHR_FPRT_IDX	4
    17	#define PCSHR_FPRT_NAME	"erofs.fingerprint"
    18	#define PCSHR_FPRT_MAXLEN (sizeof(size_t) + 1024)
    19	
    20	struct erofs_pcshr_counter {
    21		struct mutex mutex;
    22		struct kref ref;
    23		struct vfsmount *mnt;
    24		/* kmem cache for each inode's first-read segments */
    25		struct kmem_cache *segsp;
    26	};
    27	
    28	struct erofs_pcshr_private {
    29		char fprt[PCSHR_FPRT_MAXLEN];
    30		struct mutex mutex;
    31	};
    32	
    33	static struct erofs_pcshr_counter mnt_counter = {
    34		.mutex = __MUTEX_INITIALIZER(mnt_counter.mutex),
    35		.mnt = NULL,
    36	};
    37	
    38	static void erofs_pcshr_counter_release(struct kref *ref)
    39	{
    40		struct erofs_pcshr_counter *counter = container_of(ref,
    41				struct erofs_pcshr_counter, ref);
    42	
    43		DBG_BUGON(!counter->mnt);
    44		kern_unmount(counter->mnt);
    45		counter->mnt = NULL;
    46		kmem_cache_destroy(counter->segsp);
    47		counter->segsp = NULL;
    48	}
    49	
    50	int erofs_pcshr_init_mnt(void)
    51	{
    52		int ret;
    53		struct vfsmount *tmp;
    54	
    55		mutex_lock(&mnt_counter.mutex);
    56		if (!mnt_counter.mnt) {
    57			tmp = kern_mount(&erofs_anon_fs_type);
    58			if (IS_ERR(tmp)) {
    59				ret = PTR_ERR(tmp);
    60				goto out;
    61			}
    62			mnt_counter.mnt = tmp;
    63			kref_init(&mnt_counter.ref);
    64	
    65			mnt_counter.segsp = kmem_cache_create("erofs_segs",
  > 66				sizeof(struct interval_tree_node), 0,
    67				SLAB_RECLAIM_ACCOUNT | SLAB_ACCOUNT, NULL);
    68			if (!mnt_counter.segsp) {
    69				ret = -ENOMEM;
    70				goto out;
    71			}
    72		} else
    73			kref_get(&mnt_counter.ref);
    74		ret = 0;
    75	out:
    76		mutex_unlock(&mnt_counter.mutex);
    77		return ret;
    78	}
    79	
    80	void erofs_pcshr_free_mnt(void)
    81	{
    82		mutex_lock(&mnt_counter.mutex);
    83		kref_put(&mnt_counter.ref, erofs_pcshr_counter_release);
    84		mutex_unlock(&mnt_counter.mutex);
    85	}
    86	
    87	static struct interval_tree_node *erofs_pcshr_alloc_seg(void)
    88	{
    89		return kmem_cache_alloc(mnt_counter.segsp, GFP_KERNEL);
    90	}
    91	
    92	static void erofs_pcshr_free_seg(struct interval_tree_node *seg)
    93	{
    94		kmem_cache_free(mnt_counter.segsp, seg);
    95	}
    96	
    97	static int erofs_fprt_eq(struct inode *inode, void *data)
    98	{
    99		struct erofs_pcshr_private *ano_private = inode->i_private;
   100	
   101		return ano_private && memcmp(ano_private->fprt, data,
   102				sizeof(size_t) + *(size_t *)data) == 0 ? 1 : 0;
   103	}
   104	
   105	static int erofs_fprt_set(struct inode *inode, void *data)
   106	{
   107		struct erofs_pcshr_private *ano_private;
   108	
   109		ano_private = kmalloc(sizeof(struct erofs_pcshr_private), GFP_KERNEL);
   110		if (!ano_private)
   111			return -ENOMEM;
   112		memcpy(ano_private, data, sizeof(size_t) + *(size_t *)data);
   113		mutex_init(&ano_private->mutex);
   114		inode->i_private = ano_private;
   115		return 0;
   116	}
   117	
   118	int erofs_pcshr_fill_inode(struct inode *inode)
   119	{
   120		struct erofs_inode *vi = EROFS_I(inode);
   121		/* | fingerprint length | fingerprint content | */
   122		char fprt[PCSHR_FPRT_MAXLEN];
   123		struct inode *ano_inode;
   124		unsigned long fprt_hash;
   125		size_t fprt_len;
   126		int ret = -1;
   127	
   128		vi->ano_inode = NULL;
   129		memset(fprt, 0, sizeof(fprt));
   130		fprt_len = erofs_getxattr(inode, PCSHR_FPRT_IDX, PCSHR_FPRT_NAME,
   131				fprt + sizeof(size_t), PCSHR_FPRT_MAXLEN);
   132		if (fprt_len > 0 && fprt_len <= PCSHR_FPRT_MAXLEN) {
   133			*(size_t *)fprt = fprt_len;
   134			fprt_hash = xxh32(fprt + sizeof(size_t), fprt_len, 0);
   135			ano_inode = iget5_locked(mnt_counter.mnt->mnt_sb, fprt_hash,
   136						 erofs_fprt_eq, erofs_fprt_set, fprt);
   137			DBG_BUGON(!ano_inode);
   138			vi->ano_inode = ano_inode;
   139			vi->segs = RB_ROOT_CACHED;
   140			mutex_init(&vi->segs_mutex);
   141			if (ano_inode->i_state & I_NEW) {
   142				if (erofs_inode_is_data_compressed(vi->datalayout))
   143					ano_inode->i_mapping->a_ops = &z_erofs_aops;
   144				else
   145					ano_inode->i_mapping->a_ops = &erofs_aops;
   146				ano_inode->i_size = inode->i_size;
   147				unlock_new_inode(ano_inode);
   148			}
   149			ret = 0;
   150		}
   151		return ret;
   152	}
   153	
   154	void erofs_pcshr_free_inode(struct inode *inode)
   155	{
   156		struct interval_tree_node *seg, *next_seg;
   157		struct erofs_inode *vi = EROFS_I(inode);
   158	
   159		if (S_ISREG(inode->i_mode) &&  vi->ano_inode) {
   160			iput(vi->ano_inode);
   161			vi->ano_inode = NULL;
   162		}
 > 163		seg = interval_tree_iter_first(&vi->segs, 0, LLONG_MAX);
   164		while (seg) {
 > 165			next_seg = interval_tree_iter_next(seg, 0, LLONG_MAX);
 > 166			interval_tree_remove(seg, &vi->segs);
   167			erofs_pcshr_free_seg(seg);
   168			seg = next_seg;
   169		}
   170	}
   171	

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

  reply	other threads:[~2025-01-05 18:53 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-01-05 15:12 [RFC PATCH v5 0/4] erofs: page cache share feature Hongzhen Luo
2025-01-05 15:12 ` Hongzhen Luo
2025-01-05 15:12 ` [RFC PATCH v5 1/4] erofs: move `struct erofs_anon_fs_type` to super.c Hongzhen Luo
2025-01-05 15:12   ` Hongzhen Luo
2025-01-05 15:12 ` [RFC PATCH v5 2/4] erofs: introduce the page cache share feature Hongzhen Luo
2025-01-05 15:12   ` Hongzhen Luo
2025-01-06  2:27   ` Gao Xiang
2025-01-06  3:03     ` Hongzhen Luo
2025-01-05 15:12 ` [RFC PATCH v5 3/4] erofs: apply " Hongzhen Luo
2025-01-05 15:12   ` Hongzhen Luo
2025-01-06  2:15   ` Gao Xiang
2025-01-21 11:59   ` Hongbo Li via Linux-erofs
2025-01-21 11:59     ` Hongbo Li
2025-01-21 12:16     ` Gao Xiang
2025-01-21 14:48     ` Hongzhen Luo
2025-01-05 15:12 ` [RFC PATCH v5 4/4] erofs: introduce .fadvise for page cache share Hongzhen Luo
2025-01-05 15:12   ` Hongzhen Luo
2025-01-05 18:52   ` kernel test robot [this message]
2025-01-06  3:40   ` kernel test robot
2025-01-06 11:52   ` kernel test robot

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=202501060221.LPQZaeA4-lkp@intel.com \
    --to=lkp@intel.com \
    --cc=hongzhen@linux.alibaba.com \
    --cc=oe-kbuild-all@lists.linux.dev \
    /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.