| 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
| ||
|
Message-ID: <b42b02d6-08f4-01b1-6ab3-72237ef58034@codeaurora.org>
Date: Tue, 18 May 2021 18:19:24 +0530
From: Faiyaz Mohammed <faiyazm@...eaurora.org>
To: Dan Carpenter <dan.carpenter@...cle.com>, kbuild@...ts.01.org,
cl@...ux.com, penberg@...nel.org, rientjes@...gle.com,
iamjoonsoo.kim@....com, akpm@...ux-foundation.org, vbabka@...e.cz,
linux-mm@...ck.org, linux-kernel@...r.kernel.org, glittao@...il.com
Cc: lkp@...el.com, kbuild-all@...ts.01.org, vinmenon@...eaurora.org
Subject: Re: [PATCH v5] mm: slub: move sysfs slab alloc/free interfaces to
debugfs
Hi,
Updated in the patch,
https://lore.kernel.org/linux-mm/1621341949-26762-1-git-send-email-faiyazm@codeaurora.org/T/#u
Thanks and regards,
Mohammed Faiyaz
On 5/7/2021 10:51 AM, Dan Carpenter wrote:
> Hi Faiyaz,
>
> url: https://github.com/0day-ci/linux/commits/Faiyaz-Mohammed/mm-slub-move-sysfs-slab-alloc-free-interfaces-to-debugfs/20210506-182420
> base: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 8404c9fbc84b741f66cff7d4934a25dd2c344452
> config: i386-randconfig-m021-20210506 (attached as .config)
> compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@...el.com>
> Reported-by: Dan Carpenter <dan.carpenter@...cle.com>
>
> smatch warnings:
> mm/slub.c:5891 slab_debugfs_start() warn: possible memory leak of 'spos'
>
> vim +/spos +5891 mm/slub.c
>
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5877 static void *slab_debugfs_start(struct seq_file *seq, loff_t *ppos)
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5878 {
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5879 struct kmem_cache_node *n;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5880 struct kmem_cache *s;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5881 enum track_item alloc;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5882 int node;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5883 loff_t *spos = kmalloc(sizeof(loff_t), GFP_KERNEL);
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> Generally avoid putting functions which can fail in the declaration
> block. Allocations inside the declaration block are a tiny percent of
> declarations over all but they are an outsize source of static checker
> bugs.
>
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5884
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5885 s = seq->file->f_inode->i_private;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5886
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5887 if (!spos)
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5888 return NULL;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5889
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5890 if (!(s->flags & SLAB_STORE_USER))
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 @5891 return ERR_PTR(-EOPNOTSUPP);
>
> "spos" is leaked.
>
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5892
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5893 if (*ppos == 0) {
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5894
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5895 t.count = 0;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5896 t.max = 0;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5897 t.loc = NULL;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5898 if (strcmp(seq->file->f_path.dentry->d_name.name, "alloc_traces") == 0)
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5899 alloc = TRACK_ALLOC;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5900 else
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5901 alloc = TRACK_FREE;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5902
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5903 if (!alloc_loc_track(&t, PAGE_SIZE / sizeof(struct location),
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5904 GFP_KERNEL)) {
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5905 seq_puts(seq, "Out of memory\n");
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5906 return ERR_PTR(-ENOMEM);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5907 }
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5908 /* Push back cpu slabs */
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5909 flush_all(s);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5910
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5911 for_each_kmem_cache_node(s, node, n) {
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5912 unsigned long flags;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5913 struct page *page;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5914
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5915 if (!atomic_long_read(&n->nr_slabs))
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5916 continue;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5917
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5918 spin_lock_irqsave(&n->list_lock, flags);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5919 list_for_each_entry(page, &n->partial, slab_list)
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5920 process_slab(&t, s, page, alloc);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5921 list_for_each_entry(page, &n->full, slab_list)
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5922 process_slab(&t, s, page, alloc);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5923 spin_unlock_irqrestore(&n->list_lock, flags);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5924 }
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5925 }
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5926
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5927 if (*ppos < t.count) {
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5928 *spos = *ppos;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5929 return spos;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5930 }
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5931
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5932 kfree(spos);
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5933 return NULL;
> 88e180b99466c1 Faiyaz Mohammed 2021-05-06 5934 }
>
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
>
Powered by blists - more mailing lists