Skip to contentSkip to conversation

mm: check the function kmalloc_slab return value

PATCH New Ren Yu 3 participants 7 messages Jun 13, 2022 – Jun 14, 2022 Memory management

lore
Share & tools
lore

Ren Yu

author
#0 lore raw

PATCH 1/1 mm: check the function kmalloc_slab return value

As the possible failure of the kmalloc_slab,
it should be better to check it.

---
 mm/slab.c | 2 ++
 1 file changed, 2 insertions(+)
mm/slab.c +2 -0
index f8cd00f4ba13..72135e555827 100644
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags)
 	if (OFF_SLAB(cachep)) {
 		cachep->freelist_cache =
 			kmalloc_slab(cachep->freelist_size, 0u);
+		if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))
+			return cachep->freelist_cache;
 	}
 
 	err = setup_cpu_cache(cachep, gfp);
Signature · 1 line
2.11.0
#1 lore raw
Thank you for the patch! Perhaps something to improve:

Hi Ren,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on akpm-mm/mm-everything]

url:    https://github.com/intel-lab-lkp/linux/commits/Ren-Yu/mm-check-the-function-kmalloc_slab-return-value/20220613-182849
base:   https://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm.git mm-everything
config: um-i386_defconfig (https://download.01.org/0day-ci/archive/20220613/[email protected]/config)
compiler: gcc-11 (Debian 11.3.0-3) 11.3.0
reproduce (this is a W=1 build):
        # https://github.com/intel-lab-lkp/linux/commit/5ba2024be9e85177c986e9078e903798cac72f74
        git remote add linux-review https://github.com/intel-lab-lkp/linux
        git fetch --no-tags linux-review Ren-Yu/mm-check-the-function-kmalloc_slab-return-value/20220613-182849
        git checkout 5ba2024be9e85177c986e9078e903798cac72f74
        # save the config file
        mkdir build_dir && cp config build_dir/.config
        make W=1 O=build_dir ARCH=um SUBARCH=i386 SHELL=/bin/bash

If you fix the issue, kindly add following tag where applicable

All warnings (new ones prefixed by >>):

mm/slab.c: In function '__kmem_cache_create':

>> mm/slab.c:2068:38: warning: returning 'struct kmem_cache *' from a function with return type 'int' makes integer from pointer without a cast [-Wint-conversion]
    2068 |                         return cachep->freelist_cache;
         |                                ~~~~~~^~~~~~~~~~~~~~~~

vim +2068 mm/slab.c

2063
2064 if (OFF_SLAB(cachep)) {
2065 cachep->freelist_cache =
2066 kmalloc_slab(cachep->freelist_size, 0u);
2067 if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))

> 2068				return cachep->freelist_cache;

2069 }
2070
2071 err = setup_cpu_cache(cachep, gfp);
2072 if (err) {
2073 __kmem_cache_release(cachep);
2074 return err;
2075 }
2076
2077 return 0;
2078 }
2079

Signature · 2 lines
0-DAY CI Kernel Test Service
https://01.org/lkp
author
#2 lore raw

PATCH 1/1 mm: check the function kmalloc_slab return value

In reply to Ren Yu

As the possible failure of the kmalloc_slab,
it should be better to check it.

---
v2:
- fix build waring integer from pointer without a cast
---

---
 mm/slab.c | 2 ++
 1 file changed, 2 insertions(+)
mm/slab.c +2 -0
index f8cd00f4ba13..72135e555827 100644
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags)
 	if (OFF_SLAB(cachep)) {
 		cachep->freelist_cache =
 			kmalloc_slab(cachep->freelist_size, 0u);
+		if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))
+			return cachep->freelist_cache;
 	}
 
 	err = setup_cpu_cache(cachep, gfp);
Signature · 1 line
2.11.0
#3 lore raw
Quoted 2 lines · Ren Yu

As the possible failure of the kmalloc_slab,
it should be better to check it.

AFAIK failure is not possible, kmalloc_slab() is not an allocation function, it just returns a member of kmalloc_caches array, which is initialized elsewhere and shouldn't contain NULLs. So the patch seems unnecessary to me.

Quoted 15 lines

---
v2:
- fix build waring integer from pointer without a cast
---

---
 mm/slab.c | 2 ++
 1 file changed, 2 insertions(+)
mm/slab.c +0 -0
index f8cd00f4ba13..72135e555827 100644

mm/slab.c int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags) · line 2067

Quoted 4 lines
mm/slab.c +1 -0
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags)
 	if (OFF_SLAB(cachep)) {
 		cachep->freelist_cache =
 			kmalloc_slab(cachep->freelist_size, 0u);
+		if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))

The usual way is "if (!cachep->freelist_cache)". Not sure why check for ZERO.

mm/slab.c int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags) · line 2068

Quoted 1 line
mm/slab.c +1 -0
+			return cachep->freelist_cache;

So in case of NULL this would return NULL, thus 0, but __kmem_cache_create() return 0 on success, so it's wrong. You would have to return e.g. -ENOMEM.

Quoted 3 lines
mm/slab.c +0 -0
 	}
 
 	err = setup_cpu_cache(cachep, gfp);
author
#4 lore raw
Quoted 28 lines · Vlastimil Babka
Quoted 2 lines · Ren Yu

As the possible failure of the kmalloc_slab,
it should be better to check it.

AFAIK failure is not possible, kmalloc_slab() is not an allocation function, it just returns a member of kmalloc_caches array, which is initialized elsewhere and shouldn't contain NULLs. So the patch seems unnecessary to me.

Quoted 15 lines

---
v2:
- fix build waring integer from pointer without a cast
---

---
  mm/slab.c | 2 ++
  1 file changed, 2 insertions(+)
mm/slab.c +0 -0
index f8cd00f4ba13..72135e555827 100644

mm/slab.c int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags) · line 2067

Quoted 4 lines
mm/slab.c +1 -0
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags)
  	if (OFF_SLAB(cachep)) {
  		cachep->freelist_cache =
  			kmalloc_slab(cachep->freelist_size, 0u);
+		if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))

The usual way is "if (!cachep->freelist_cache)". Not sure why check for ZERO.

Quoted 3 lines

mm/slab.c int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags) · line 2068

Quoted 1 line
mm/slab.c +1 -0
+			return cachep->freelist_cache;

So in case of NULL this would return NULL, thus 0, but __kmem_cache_create() return 0 on success, so it's wrong. You would have to return e.g. -ENOMEM.

Thanks for the advice ,I'll be re-patching

Quoted 5 lines
Quoted 4 lines
mm/slab.c +0 -0
  	}
 
  	err = setup_cpu_cache(cachep, gfp);
#5 lore raw
Quoted 32 lines · tury
Quoted 29 lines · Vlastimil Babka
Quoted 2 lines · Ren Yu

As the possible failure of the kmalloc_slab,
it should be better to check it.

AFAIK failure is not possible, kmalloc_slab() is not an allocation function, it just returns a member of kmalloc_caches array, which is initialized elsewhere and shouldn't contain NULLs. So the patch seems unnecessary to me.

Quoted 17 lines

---
v2:
- fix build waring integer from pointer without a cast
---
---
  mm/slab.c | 2 ++

  1 file changed, 2 insertions(+)
mm/slab.c +0 -0
index f8cd00f4ba13..72135e555827 100644
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep,
 slab_flags_t flags)
       if (OFF_SLAB(cachep)) {
Quoted 3 lines

          cachep->freelist_cache = kmalloc_slab(cachep->freelist_size, 0u); +        if (unlikely(ZERO_OR_NULL_PTR(cachep->freelist_cache)))

The usual way is "if (!cachep->freelist_cache)". Not sure why check for ZERO.

Quoted 4 lines
Quoted 3 lines
Quoted 1 line

+            return cachep->freelist_cache;

So in case of NULL this would return NULL, thus 0, but __kmem_cache_create() return 0 on success, so it's wrong. You would have to return e.g. -ENOMEM.

Thanks for the advice ,I'll be re-patching

However that was meant just for your information/learning, the patch is still unecessary as I wrote above, so I will not merge it so we don't complicate the code needlessly.

Quoted 5 lines
Quoted 4 lines
Quoted 3 lines

      }
        err = setup_cpu_cache(cachep, gfp);

author
#6 lore raw

PATCH v2 1/1 mm, slab: check the function kmalloc_slab return value

In reply to Ren Yu

As the possible failure of the kmalloc_slab,
it should be better to check it.

---
v2:
- change return value of error path to '-ENOMEM'
- not check for zero

---
 mm/slab.c | 2 ++
 1 file changed, 2 insertions(+)
mm/slab.c +2 -0
index f8cd00f4ba13..eb3fb042f4f4 100644
@@ -2064,6 +2064,8 @@ int __kmem_cache_create(struct kmem_cache *cachep, slab_flags_t flags)
 	if (OFF_SLAB(cachep)) {
 		cachep->freelist_cache =
 			kmalloc_slab(cachep->freelist_size, 0u);
+		if (!cachep->freelist_cache)
+			return -ENOMEM;
 	}
 
 	err = setup_cpu_cache(cachep, gfp);
Signature · 1 line
2.11.0