[PATCH v1] mm: slub: fix the leak of alloc/free traces debugfs interface

[PATCH v1] mm: slub: fix the leak of alloc/free traces debugfs interface

[Faiyaz Mohammed]

fix the leak of alloc/free traces debugfs interface, reported
by kmemleak like below,

unreferenced object 0xffff00091ae1b540 (size 64):
  comm "lsbug", pid 1607, jiffies 4294958291 (age 1476.340s)
  hex dump (first 32 bytes):
    02 00 00 00 00 00 00 00 6b 6b 6b 6b 6b 6b 6b 6b  ........kkkkkkkk
    6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b  kkkkkkkkkkkkkkkk
  backtrace:
    [<ffff8000106b06b8>] slab_post_alloc_hook+0xa0/0x418
    [<ffff8000106b5c7c>] kmem_cache_alloc_trace+0x1e4/0x378
    [<ffff8000106b5e40>] slab_debugfs_start+0x30/0x50
    slab_debugfs_start at mm/slub.c:5831
    [<ffff8000107b3dbc>] seq_read_iter+0x214/0xd50
    [<ffff8000107b4b84>] seq_read+0x28c/0x418
    [<ffff8000109560b4>] full_proxy_read+0xdc/0x148
    [<ffff800010738f24>] vfs_read+0x104/0x340
    [<ffff800010739ee0>] ksys_read+0xf8/0x1e0
    [<ffff80001073a03c>] __arm64_sys_read+0x74/0xa8
    [<ffff8000100358d4>] invoke_syscall.constprop.0+0xdc/0x1d8
    [<ffff800010035ab4>] do_el0_svc+0xe4/0x298
    [<ffff800011138528>] el0_svc+0x20/0x30
    [<ffff800011138b08>] el0t_64_sync_handler+0xb0/0xb8
    [<ffff80001001259c>] el0t_64_sync+0x178/0x17c

Fixes: 84a2bdb1b458fc968d6d9e07dab388dc679bd747 ("mm: slub: move sysfs slab alloc/free interfaces to debugfs")
Link: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/mm/slub.c?h=next-20210617&id=84a2bdb1b458fc968d6d9e07dab388dc679bd747

Signed-off-by: Faiyaz Mohammed <faiyazm@codeaurora.org>
---
 mm/slub.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/mm/slub.c b/mm/slub.c
index fcb0c50..f006766 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -5785,31 +5785,24 @@ static int slab_debugfs_show(struct seq_file *seq, void *v)
 
 static void slab_debugfs_stop(struct seq_file *seq, void *v)
 {
-	kfree(v);
 }
 
 static void *slab_debugfs_next(struct seq_file *seq, void *v, loff_t *ppos)
 {
-	loff_t *spos = v;
 	struct loc_track *t = seq->private;
 
+	v = ppos;
 	if (*ppos < t->count) {
-		*ppos = ++*spos;
-		return spos;
+		++*ppos;
+		return v;
 	}
-	*ppos = ++*spos;
+	++*ppos;
 	return NULL;
 }
 
 static void *slab_debugfs_start(struct seq_file *seq, loff_t *ppos)
 {
-	loff_t *spos = kmalloc(sizeof(loff_t), GFP_KERNEL);
-
-	if (!spos)
-		return NULL;
-
-	*spos = *ppos;
-	return spos;
+	return ppos;
 }
 
 static const struct seq_operations slab_debugfs_sops = {
-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a
member of the Code Aurora Forum, hosted by The Linux Foundation

Re: [PATCH v1] mm: slub: fix the leak of alloc/free traces debugfs interface

[Andy Shevchenko]

On Fri, Jun 18, 2021 at 3:38 PM Faiyaz Mohammed <faiyazm@codeaurora.org> wrote:
>
> fix the leak of alloc/free traces debugfs interface, reported

Fix

> by kmemleak like below,
>
> unreferenced object 0xffff00091ae1b540 (size 64):
>   comm "lsbug", pid 1607, jiffies 4294958291 (age 1476.340s)
>   hex dump (first 32 bytes):
>     02 00 00 00 00 00 00 00 6b 6b 6b 6b 6b 6b 6b 6b  ........kkkkkkkk
>     6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b  kkkkkkkkkkkkkkkk
>   backtrace:
>     [<ffff8000106b06b8>] slab_post_alloc_hook+0xa0/0x418
>     [<ffff8000106b5c7c>] kmem_cache_alloc_trace+0x1e4/0x378
>     [<ffff8000106b5e40>] slab_debugfs_start+0x30/0x50
>     slab_debugfs_start at mm/slub.c:5831
>     [<ffff8000107b3dbc>] seq_read_iter+0x214/0xd50
>     [<ffff8000107b4b84>] seq_read+0x28c/0x418
>     [<ffff8000109560b4>] full_proxy_read+0xdc/0x148
>     [<ffff800010738f24>] vfs_read+0x104/0x340
>     [<ffff800010739ee0>] ksys_read+0xf8/0x1e0
>     [<ffff80001073a03c>] __arm64_sys_read+0x74/0xa8
>     [<ffff8000100358d4>] invoke_syscall.constprop.0+0xdc/0x1d8
>     [<ffff800010035ab4>] do_el0_svc+0xe4/0x298
>     [<ffff800011138528>] el0_svc+0x20/0x30
>     [<ffff800011138b08>] el0t_64_sync_handler+0xb0/0xb8
>     [<ffff80001001259c>] el0t_64_sync+0x178/0x17c

Can you shrink this a bit?

> Fixes: 84a2bdb1b458fc968d6d9e07dab388dc679bd747 ("mm: slub: move sysfs slab alloc/free interfaces to debugfs")

We use 12, which is shorter.

> Link: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/mm/slub.c?h=next-20210617&id=84a2bdb1b458fc968d6d9e07dab388dc679bd747

>

Must be no blank lines in the tag block.

> Signed-off-by: Faiyaz Mohammed <faiyazm@codeaurora.org>

...

>  static void *slab_debugfs_next(struct seq_file *seq, void *v, loff_t *ppos)
>  {
> -       loff_t *spos = v;
>         struct loc_track *t = seq->private;
>
> +       v = ppos;
>         if (*ppos < t->count) {
> -               *ppos = ++*spos;
> -               return spos;
> +               ++*ppos;
> +               return v;
>         }
> -       *ppos = ++*spos;
> +       ++*ppos;
>         return NULL;

Can it be

       v = ppos;
       ++*ppos;
       if (*ppos <= t->count)
              return v;
       return NULL;

?  (basically the question is, is the comparison equivalent in this case or not)

>  }

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v1] mm: slub: fix the leak of alloc/free traces debugfs interface

[Faiyaz Mohammed]

On 6/18/2021 6:45 PM, Andy Shevchenko wrote:
> On Fri, Jun 18, 2021 at 3:38 PM Faiyaz Mohammed <faiyazm@codeaurora.org> wrote:
>>
>> fix the leak of alloc/free traces debugfs interface, reported
> 
> Fix
> 
Okay, I will update in next patch version.

>> by kmemleak like below,
>>
>> unreferenced object 0xffff00091ae1b540 (size 64):
>>   comm "lsbug", pid 1607, jiffies 4294958291 (age 1476.340s)
>>   hex dump (first 32 bytes):
>>     02 00 00 00 00 00 00 00 6b 6b 6b 6b 6b 6b 6b 6b  ........kkkkkkkk
>>     6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b  kkkkkkkkkkkkkkkk
>>   backtrace:
>>     [<ffff8000106b06b8>] slab_post_alloc_hook+0xa0/0x418
>>     [<ffff8000106b5c7c>] kmem_cache_alloc_trace+0x1e4/0x378
>>     [<ffff8000106b5e40>] slab_debugfs_start+0x30/0x50
>>     slab_debugfs_start at mm/slub.c:5831
>>     [<ffff8000107b3dbc>] seq_read_iter+0x214/0xd50
>>     [<ffff8000107b4b84>] seq_read+0x28c/0x418
>>     [<ffff8000109560b4>] full_proxy_read+0xdc/0x148
>>     [<ffff800010738f24>] vfs_read+0x104/0x340
>>     [<ffff800010739ee0>] ksys_read+0xf8/0x1e0
>>     [<ffff80001073a03c>] __arm64_sys_read+0x74/0xa8
>>     [<ffff8000100358d4>] invoke_syscall.constprop.0+0xdc/0x1d8
>>     [<ffff800010035ab4>] do_el0_svc+0xe4/0x298
>>     [<ffff800011138528>] el0_svc+0x20/0x30
>>     [<ffff800011138b08>] el0t_64_sync_handler+0xb0/0xb8
>>     [<ffff80001001259c>] el0t_64_sync+0x178/0x17c
> 
> Can you shrink this a bit?
>
Okay

>> Fixes: 84a2bdb1b458fc968d6d9e07dab388dc679bd747 ("mm: slub: move sysfs slab alloc/free interfaces to debugfs")
> 
> We use 12, which is shorter.
> 
>> Link: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/mm/slub.c?h=next-20210617&id=84a2bdb1b458fc968d6d9e07dab388dc679bd747
> 
>>
> 
> Must be no blank lines in the tag block.
> >> Signed-off-by: Faiyaz Mohammed <faiyazm@codeaurora.org>
> 
Okay
> ...
> 
>>  static void *slab_debugfs_next(struct seq_file *seq, void *v, loff_t *ppos)
>>  {
>> -       loff_t *spos = v;
>>         struct loc_track *t = seq->private;
>>
>> +       v = ppos;
>>         if (*ppos < t->count) {
>> -               *ppos = ++*spos;
>> -               return spos;
>> +               ++*ppos;
>> +               return v;
>>         }
>> -       *ppos = ++*spos;
>> +       ++*ppos;
>>         return NULL;
> 
> Can it be
> 
>        v = ppos;
>        ++*ppos;
>        if (*ppos <= t->count>               return v;
>        return NULL;
> 
> ?  (basically the question is, is the comparison equivalent in this case or not)
> 
>>  }
>Yes, we can update it and slab_debugfs_show has the index check as well.
I will update in next patch version.

Thanks and regards,
Mohammed Faiyaz