[dpdk-dev] [PATCH] eal/linuxapp: fix resource leak

[dpdk-dev] [PATCH] eal/linuxapp: fix resource leak

[Daniel Mrzyglod]

Fix issue reported by Coverity.
Coverity ID 97920

munmap structure of hugepage

leaked_storage: Variable hugepage going out of scope leaks the storage
it points to.

The system resource will not be reclaimed and reused, reducing the future
availability of the resource.

In rte_eal_hugepage_init: Leak of memory or pointers to system resources

Fixes: b6a468ad41d5 ("memory: add --socket-mem option")

Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 5b9132c..cd40cc9 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -1051,7 +1051,7 @@ int
 rte_eal_hugepage_init(void)
 {
 	struct rte_mem_config *mcfg;
-	struct hugepage_file *hugepage, *tmp_hp = NULL;
+	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
 	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
 
 	uint64_t memory[RTE_MAX_NUMA_NODES];
@@ -1374,6 +1374,8 @@ rte_eal_hugepage_init(void)
 
 fail:
 	free(tmp_hp);
+	if (hugepage != NULL)
+		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
 	return -1;
 }
 
-- 
2.5.5

Re: [dpdk-dev] [PATCH] eal/linuxapp: fix resource leak

[Sergio Gonzalez Monroy]

Hi,

On 11/05/2016 17:01, Daniel Mrzyglod wrote:
> Fix issue reported by Coverity.
> Coverity ID 97920
>
> munmap structure of hugepage
>
> leaked_storage: Variable hugepage going out of scope leaks the storage
> it points to.
>
> The system resource will not be reclaimed and reused, reducing the future
> availability of the resource.

I'm not really fond of this commit messages, but if no one minds them, 
so be it.

> In rte_eal_hugepage_init: Leak of memory or pointers to system resources
>
> Fixes: b6a468ad41d5 ("memory: add --socket-mem option")
>
> Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
> ---
>   lib/librte_eal/linuxapp/eal/eal_memory.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
> index 5b9132c..cd40cc9 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_memory.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
> @@ -1051,7 +1051,7 @@ int
>   rte_eal_hugepage_init(void)
>   {
>   	struct rte_mem_config *mcfg;
> -	struct hugepage_file *hugepage, *tmp_hp = NULL;
> +	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
>   	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
>   
>   	uint64_t memory[RTE_MAX_NUMA_NODES];
> @@ -1374,6 +1374,8 @@ rte_eal_hugepage_init(void)
>   
>   fail:
>   	free(tmp_hp);
> +	if (hugepage != NULL)
> +		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
>   	return -1;
>   }
>   

You missed the last conditional if(...0 of the function where it returns 
directly with error value.

Looking at the code a bit, we never check for anything other than < 0 
return value, so I'd just do a 'goto fail' instead.

Sergio

[dpdk-dev] [PATCH v2] eal/linuxapp: fix resource leak

[Daniel Mrzyglod]

Fix issue reported by Coverity.
Coverity ID 97920

munmap structure of hugepage

leaked_storage: Variable hugepage going out of scope leaks the storage
it points to.

The system resource will not be reclaimed and reused, reducing the future
availability of the resource.

In rte_eal_hugepage_init: Leak of memory or pointers to system resources

Fixes: b6a468ad41d5 ("memory: add --socket-mem option")

Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 5b9132c..9fd0d8d 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -1051,7 +1051,7 @@ int
 rte_eal_hugepage_init(void)
 {
 	struct rte_mem_config *mcfg;
-	struct hugepage_file *hugepage, *tmp_hp = NULL;
+	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
 	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
 
 	uint64_t memory[RTE_MAX_NUMA_NODES];
@@ -1367,13 +1367,15 @@ rte_eal_hugepage_init(void)
 			"of memory.\n",
 			i, nr_hugefiles, RTE_STR(CONFIG_RTE_MAX_MEMSEG),
 			RTE_MAX_MEMSEG);
-		return -ENOMEM;
+		goto fail;
 	}
 
 	return 0;
 
 fail:
 	free(tmp_hp);
+	if (hugepage != NULL)
+		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
 	return -1;
 }
 
-- 
2.5.5

Re: [dpdk-dev] [PATCH] eal/linuxapp: fix resource leak

[Thomas Monjalon]

2016-05-12 08:55, Sergio Gonzalez Monroy:
> On 11/05/2016 17:01, Daniel Mrzyglod wrote:
> > Fix issue reported by Coverity.
> > Coverity ID 97920
> >
> > munmap structure of hugepage
> >
> > leaked_storage: Variable hugepage going out of scope leaks the storage
> > it points to.
> >
> > The system resource will not be reclaimed and reused, reducing the future
> > availability of the resource.
> 
> I'm not really fond of this commit messages, but if no one minds them, 
> so be it.

Me too. It looks like a Machine2Machine message.
I'd prefer an explanation of what is wrong first, then an explanation of
the fix, and at the end, the Coverity ID (can be a tag like
Fixes-Coverity-ID: 97920).

> > In rte_eal_hugepage_init: Leak of memory or pointers to system resources
> >
> > Fixes: b6a468ad41d5 ("memory: add --socket-mem option")
> >
> > Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>

Re: [dpdk-dev] [PATCH v2] eal/linuxapp: fix resource leak

[Thomas Monjalon]

2016-05-12 11:12, Daniel Mrzyglod:
> Fix issue reported by Coverity.
> Coverity ID 97920
> 
> munmap structure of hugepage
> 
> leaked_storage: Variable hugepage going out of scope leaks the storage
> it points to.
> 
> The system resource will not be reclaimed and reused, reducing the future
> availability of the resource.
> 
> In rte_eal_hugepage_init: Leak of memory or pointers to system resources

Please reword.

[dpdk-dev] [PATCH v3] eal/linuxapp: fix resource leak

[Daniel Mrzyglod]

This patch fix all cases to do proper handle all munmap if pointer
of hugepage is not NULL which prohibits resource leak.

Coverity issue: 97920
Fixes: b6a468ad41d5 ("memory: add --socket-mem option")

Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 9251a5b..9b0d39a 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -1051,7 +1051,7 @@ int
 rte_eal_hugepage_init(void)
 {
 	struct rte_mem_config *mcfg;
-	struct hugepage_file *hugepage, *tmp_hp = NULL;
+	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
 	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
 
 	uint64_t memory[RTE_MAX_NUMA_NODES];
@@ -1367,13 +1367,15 @@ rte_eal_hugepage_init(void)
 			"of memory.\n",
 			i, nr_hugefiles, RTE_STR(CONFIG_RTE_MAX_MEMSEG),
 			RTE_MAX_MEMSEG);
-		return -ENOMEM;
+		goto fail;
 	}
 
 	return 0;
 
 fail:
 	free(tmp_hp);
+	if (hugepage != NULL)
+		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
 	return -1;
 }
 
-- 
2.7.4

Re: [dpdk-dev] [PATCH v3] eal/linuxapp: fix resource leak

[Sergio Gonzalez Monroy]

On 22/06/2016 17:50, Daniel Mrzyglod wrote:
> This patch fix all cases to do proper handle all munmap if pointer
> of hugepage is not NULL which prohibits resource leak.

I would reword the above commit to something like:

Current code does not munmap 'hugepage' mapping (hugepage info file) on 
function
exit, leaking resources.

> Coverity issue: 97920
> Fixes: b6a468ad41d5 ("memory: add --socket-mem option")
>
> Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
> ---
>   lib/librte_eal/linuxapp/eal/eal_memory.c | 6 ++++--
>   1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
> index 9251a5b..9b0d39a 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_memory.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
> @@ -1051,7 +1051,7 @@ int
>   rte_eal_hugepage_init(void)
>   {
>   	struct rte_mem_config *mcfg;
> -	struct hugepage_file *hugepage, *tmp_hp = NULL;
> +	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
>   	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
>   
>   	uint64_t memory[RTE_MAX_NUMA_NODES];
> @@ -1367,13 +1367,15 @@ rte_eal_hugepage_init(void)
>   			"of memory.\n",
>   			i, nr_hugefiles, RTE_STR(CONFIG_RTE_MAX_MEMSEG),
>   			RTE_MAX_MEMSEG);
> -		return -ENOMEM;
> +		goto fail;
>   	}
>   

I missed this but we should also munmap 'hugepage' on successful 
function exit.

Sergio

>   	return 0;
>   
>   fail:
>   	free(tmp_hp);
> +	if (hugepage != NULL)
> +		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
>   	return -1;
>   }
>   

[dpdk-dev] [PATCH v4] eal/linuxapp: fix resource leak

[Daniel Mrzyglod]

Current code does not munmap 'hugepage' mapping (hugepage info file) on
function exit, leaking resources.

Coverity issue: 97920
Fixes: b6a468ad41d5 ("memory: add --socket-mem option")

Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
---
 lib/librte_eal/linuxapp/eal/eal_memory.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/lib/librte_eal/linuxapp/eal/eal_memory.c b/lib/librte_eal/linuxapp/eal/eal_memory.c
index 5578c25..b663244 100644
--- a/lib/librte_eal/linuxapp/eal/eal_memory.c
+++ b/lib/librte_eal/linuxapp/eal/eal_memory.c
@@ -1136,7 +1136,7 @@ int
 rte_eal_hugepage_init(void)
 {
 	struct rte_mem_config *mcfg;
-	struct hugepage_file *hugepage, *tmp_hp = NULL;
+	struct hugepage_file *hugepage = NULL, *tmp_hp = NULL;
 	struct hugepage_info used_hp[MAX_HUGEPAGE_SIZES];
 
 	uint64_t memory[RTE_MAX_NUMA_NODES];
@@ -1479,14 +1479,19 @@ rte_eal_hugepage_init(void)
 			"of memory.\n",
 			i, nr_hugefiles, RTE_STR(CONFIG_RTE_MAX_MEMSEG),
 			RTE_MAX_MEMSEG);
-		return -ENOMEM;
+		goto fail;
 	}
 
+	munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
+
 	return 0;
 
 fail:
 	huge_recover_sigbus();
 	free(tmp_hp);
+	if (hugepage != NULL)
+		munmap(hugepage, nr_hugefiles * sizeof(struct hugepage_file));
+
 	return -1;
 }
 
-- 
2.7.4

Re: [dpdk-dev] [PATCH v4] eal/linuxapp: fix resource leak

[Sergio Gonzalez Monroy]

On 06/07/2016 11:44, Daniel Mrzyglod wrote:
> Current code does not munmap 'hugepage' mapping (hugepage info file) on
> function exit, leaking resources.
>
> Coverity issue: 97920
> Fixes: b6a468ad41d5 ("memory: add --socket-mem option")
>
> Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
> ---

Please next time remember to keep patch version history after the triple 
dashes.

Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>

Re: [dpdk-dev] [PATCH v4] eal/linuxapp: fix resource leak

[Thomas Monjalon]

2016-07-06 11:48, Sergio Gonzalez Monroy:
> On 06/07/2016 11:44, Daniel Mrzyglod wrote:
> > Current code does not munmap 'hugepage' mapping (hugepage info file) on
> > function exit, leaking resources.
> >
> > Coverity issue: 97920
> > Fixes: b6a468ad41d5 ("memory: add --socket-mem option")
> >
> > Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
> > ---
> 
> Please next time remember to keep patch version history after the triple 
> dashes.
> 
> Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>

Applied, thanks