[PATCH 1/2] net/smc: Return directly after a failed kzalloc() in smc_fill_gid_list()

[PATCH 1/2] net/smc: Return directly after a failed kzalloc() in smc_fill_gid_list()

[Markus Elfring]

From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sun, 31 Dec 2023 15:15:19 +0100

The kfree() function was called in one case by
the smc_fill_gid_list() function during error handling
even if the passed variable contained a null pointer.
This issue was detected by using the Coccinelle software.

Thus return directly after a call of the function “kzalloc” failed
at the beginning.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 net/smc/af_smc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
index 7fc2f3c6d248..a396a9977ba9 100644
--- a/net/smc/af_smc.c
+++ b/net/smc/af_smc.c
@@ -1180,7 +1180,7 @@ void smc_fill_gid_list(struct smc_link_group *lgr,

 	alt_ini = kzalloc(sizeof(*alt_ini), GFP_KERNEL);
 	if (!alt_ini)
-		goto out;
+		return;

 	alt_ini->vlan_id = lgr->vlan_id;
 	alt_ini->check_smcrv2 = true;
--
2.43.0

Re: [PATCH 1/2] net/smc: Return directly after a failed kzalloc() in smc_fill_gid_list()

[Tony Lu]

On Sun, Dec 31, 2023 at 03:58:15PM +0100, Markus Elfring wrote:
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Sun, 31 Dec 2023 15:15:19 +0100
> 
> The kfree() function was called in one case by
> the smc_fill_gid_list() function during error handling
> even if the passed variable contained a null pointer.
> This issue was detected by using the Coccinelle software.
> 
> Thus return directly after a call of the function "kzalloc" failed
> at the beginning.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>

LGTM, thank you.

Also please add net tag in subject and fixes tag in commit body.

Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>

> ---
>  net/smc/af_smc.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c
> index 7fc2f3c6d248..a396a9977ba9 100644
> --- a/net/smc/af_smc.c
> +++ b/net/smc/af_smc.c
> @@ -1180,7 +1180,7 @@ void smc_fill_gid_list(struct smc_link_group *lgr,
> 
>  	alt_ini = kzalloc(sizeof(*alt_ini), GFP_KERNEL);
>  	if (!alt_ini)
> -		goto out;
> +		return;
> 
>  	alt_ini->vlan_id = lgr->vlan_id;
>  	alt_ini->check_smcrv2 = true;
> --
> 2.43.0

[PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()

[Markus Elfring]

From: Markus Elfring <elfring@users.sourceforge.net>
Date: Sun, 31 Dec 2023 15:42:07 +0100

The kfree() function was called in some cases by
the smc_llc_cli_add_link_invite() function during error handling
even if the passed variable contained a null pointer.
This issue was detected by using the Coccinelle software.

* Thus use another label.

* Merge two if statements.

* Omit an initialisation (for the variable “ini”)
  which became unnecessary with this refactoring.

Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
---
 net/smc/smc_llc.c | 15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/net/smc/smc_llc.c b/net/smc/smc_llc.c
index 018ce8133b02..2ff24a7feb26 100644
--- a/net/smc/smc_llc.c
+++ b/net/smc/smc_llc.c
@@ -1163,23 +1163,21 @@ static void smc_llc_cli_add_link_invite(struct smc_link *link,
 					struct smc_llc_qentry *qentry)
 {
 	struct smc_link_group *lgr = smc_get_lgr(link);
-	struct smc_init_info *ini = NULL;
+	struct smc_init_info *ini;

 	if (lgr->smc_version == SMC_V2) {
 		smc_llc_send_request_add_link(link);
-		goto out;
+		goto free_qentry;
 	}

 	if (lgr->type == SMC_LGR_SYMMETRIC ||
-	    lgr->type == SMC_LGR_ASYMMETRIC_PEER)
-		goto out;
-
-	if (lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
-		goto out;
+	    lgr->type == SMC_LGR_ASYMMETRIC_PEER ||
+	    lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
+		goto free_qentry;

 	ini = kzalloc(sizeof(*ini), GFP_KERNEL);
 	if (!ini)
-		goto out;
+		goto free_qentry;

 	ini->vlan_id = lgr->vlan_id;
 	smc_pnet_find_alt_roce(lgr, ini, link->smcibdev);
@@ -1190,6 +1188,7 @@ static void smc_llc_cli_add_link_invite(struct smc_link *link,
 			      ini->ib_gid, NULL, SMC_LLC_REQ);
 out:
 	kfree(ini);
+free_qentry:
 	kfree(qentry);
 }

--
2.43.0

Re: [PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()

[kernel test robot]

Hi Markus,

kernel test robot noticed the following build warnings:

[auto build test WARNING on net/main]
[also build test WARNING on net-next/main linus/master v6.7-rc7 next-20231222]
[cannot apply to horms-ipvs/master]
[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/Markus-Elfring/net-smc-Return-directly-after-a-failed-kzalloc-in-smc_fill_gid_list/20231231-231406
base:   net/main
patch link:    https://lore.kernel.org/r/5253e660-6b66-4775-ae2f-06f5a1d40be5%40web.de
patch subject: [PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()
config: x86_64-randconfig-002-20231231 (https://download.01.org/0day-ci/archive/20240101/202401010536.XhzcWDPH-lkp@intel.com/config)
compiler: ClangBuiltLinux clang version 17.0.6 (https://github.com/llvm/llvm-project 6009708b4367171ccdbf4b5905cb6a803753fe18)
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240101/202401010536.XhzcWDPH-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/202401010536.XhzcWDPH-lkp@intel.com/

All warnings (new ones prefixed by >>):

>> net/smc/smc_llc.c:1175:34: warning: '&&' within '||' [-Wlogical-op-parentheses]
    1174 |             lgr->type == SMC_LGR_ASYMMETRIC_PEER ||
         |                                                  ~~
    1175 |             lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
         |             ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~
   net/smc/smc_llc.c:1175:34: note: place parentheses around the '&&' expression to silence this warning
    1175 |             lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
         |                                         ^                     
         |             (                                                 )
   1 warning generated.


vim +1175 net/smc/smc_llc.c

  1160	
  1161	/* as an SMC client, invite server to start the add_link processing */
  1162	static void smc_llc_cli_add_link_invite(struct smc_link *link,
  1163						struct smc_llc_qentry *qentry)
  1164	{
  1165		struct smc_link_group *lgr = smc_get_lgr(link);
  1166		struct smc_init_info *ini;
  1167	
  1168		if (lgr->smc_version == SMC_V2) {
  1169			smc_llc_send_request_add_link(link);
  1170			goto free_qentry;
  1171		}
  1172	
  1173		if (lgr->type == SMC_LGR_SYMMETRIC ||
  1174		    lgr->type == SMC_LGR_ASYMMETRIC_PEER ||
> 1175		    lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
  1176			goto free_qentry;
  1177	
  1178		ini = kzalloc(sizeof(*ini), GFP_KERNEL);
  1179		if (!ini)
  1180			goto free_qentry;
  1181	
  1182		ini->vlan_id = lgr->vlan_id;
  1183		smc_pnet_find_alt_roce(lgr, ini, link->smcibdev);
  1184		if (!ini->ib_dev)
  1185			goto out;
  1186	
  1187		smc_llc_send_add_link(link, ini->ib_dev->mac[ini->ib_port - 1],
  1188				      ini->ib_gid, NULL, SMC_LLC_REQ);
  1189	out:
  1190		kfree(ini);
  1191	free_qentry:
  1192		kfree(qentry);
  1193	}
  1194	

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

Re: [PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()

[kernel test robot]

Hi Markus,

kernel test robot noticed the following build warnings:

[auto build test WARNING on net/main]
[also build test WARNING on net-next/main linus/master v6.7-rc7 next-20231222]
[cannot apply to horms-ipvs/master]
[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/Markus-Elfring/net-smc-Return-directly-after-a-failed-kzalloc-in-smc_fill_gid_list/20231231-231406
base:   net/main
patch link:    https://lore.kernel.org/r/5253e660-6b66-4775-ae2f-06f5a1d40be5%40web.de
patch subject: [PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()
config: m68k-allyesconfig (https://download.01.org/0day-ci/archive/20240101/202401010657.eexbMD1F-lkp@intel.com/config)
compiler: m68k-linux-gcc (GCC) 13.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240101/202401010657.eexbMD1F-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/202401010657.eexbMD1F-lkp@intel.com/

All warnings (new ones prefixed by >>):

   net/smc/smc_llc.c: In function 'smc_llc_cli_add_link_invite':
>> net/smc/smc_llc.c:1175:41: warning: suggest parentheses around '&&' within '||' [-Wparentheses]
    1175 |             lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
         |             ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~


vim +1175 net/smc/smc_llc.c

  1160	
  1161	/* as an SMC client, invite server to start the add_link processing */
  1162	static void smc_llc_cli_add_link_invite(struct smc_link *link,
  1163						struct smc_llc_qentry *qentry)
  1164	{
  1165		struct smc_link_group *lgr = smc_get_lgr(link);
  1166		struct smc_init_info *ini;
  1167	
  1168		if (lgr->smc_version == SMC_V2) {
  1169			smc_llc_send_request_add_link(link);
  1170			goto free_qentry;
  1171		}
  1172	
  1173		if (lgr->type == SMC_LGR_SYMMETRIC ||
  1174		    lgr->type == SMC_LGR_ASYMMETRIC_PEER ||
> 1175		    lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
  1176			goto free_qentry;
  1177	
  1178		ini = kzalloc(sizeof(*ini), GFP_KERNEL);
  1179		if (!ini)
  1180			goto free_qentry;
  1181	
  1182		ini->vlan_id = lgr->vlan_id;
  1183		smc_pnet_find_alt_roce(lgr, ini, link->smcibdev);
  1184		if (!ini->ib_dev)
  1185			goto out;
  1186	
  1187		smc_llc_send_add_link(link, ini->ib_dev->mac[ini->ib_port - 1],
  1188				      ini->ib_gid, NULL, SMC_LLC_REQ);
  1189	out:
  1190		kfree(ini);
  1191	free_qentry:
  1192		kfree(qentry);
  1193	}
  1194	

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

Re: [PATCH 2/2] net/smc: Improve exception handling in smc_llc_cli_add_link_invite()

[Tony Lu]

On Sun, Dec 31, 2023 at 04:00:22PM +0100, Markus Elfring wrote:
> From: Markus Elfring <elfring@users.sourceforge.net>
> Date: Sun, 31 Dec 2023 15:42:07 +0100
> 
> The kfree() function was called in some cases by
> the smc_llc_cli_add_link_invite() function during error handling
> even if the passed variable contained a null pointer.
> This issue was detected by using the Coccinelle software.
> 
> * Thus use another label.
> 
> * Merge two if statements.
> 
> * Omit an initialisation (for the variable "ini")
>   which became unnecessary with this refactoring.
> 
> Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>

Thank you, LGTM. Also net and Fixes tags are needed.

Reviewed-by: Tony Lu <tonylu@linux.alibaba.com>
> ---
>  net/smc/smc_llc.c | 15 +++++++--------
>  1 file changed, 7 insertions(+), 8 deletions(-)
> 
> diff --git a/net/smc/smc_llc.c b/net/smc/smc_llc.c
> index 018ce8133b02..2ff24a7feb26 100644
> --- a/net/smc/smc_llc.c
> +++ b/net/smc/smc_llc.c
> @@ -1163,23 +1163,21 @@ static void smc_llc_cli_add_link_invite(struct smc_link *link,
>  					struct smc_llc_qentry *qentry)
>  {
>  	struct smc_link_group *lgr = smc_get_lgr(link);
> -	struct smc_init_info *ini = NULL;
> +	struct smc_init_info *ini;
> 
>  	if (lgr->smc_version == SMC_V2) {
>  		smc_llc_send_request_add_link(link);
> -		goto out;
> +		goto free_qentry;
>  	}
> 
>  	if (lgr->type == SMC_LGR_SYMMETRIC ||
> -	    lgr->type == SMC_LGR_ASYMMETRIC_PEER)
> -		goto out;
> -
> -	if (lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
> -		goto out;
> +	    lgr->type == SMC_LGR_ASYMMETRIC_PEER ||
> +	    lgr->type == SMC_LGR_SINGLE && lgr->max_links <= 1)
> +		goto free_qentry;
> 
>  	ini = kzalloc(sizeof(*ini), GFP_KERNEL);
>  	if (!ini)
> -		goto out;
> +		goto free_qentry;
> 
>  	ini->vlan_id = lgr->vlan_id;
>  	smc_pnet_find_alt_roce(lgr, ini, link->smcibdev);
> @@ -1190,6 +1188,7 @@ static void smc_llc_cli_add_link_invite(struct smc_link *link,
>  			      ini->ib_gid, NULL, SMC_LLC_REQ);
>  out:
>  	kfree(ini);
> +free_qentry:
>  	kfree(qentry);
>  }
> 
> --
> 2.43.0

Re: [0/2] net/smc: Adjustments for two function implementations

[Markus Elfring]

…
>> A few update suggestions were taken into account
>> from static source code analysis.
…
>>    Return directly after a failed kzalloc() in smc_fill_gid_list()
>>    Improve exception handling in smc_llc_cli_add_link_invite()
>>
>>   net/smc/af_smc.c  |  2 +-
>>   net/smc/smc_llc.c | 15 +++++++--------
>>   2 files changed, 8 insertions(+), 9 deletions(-)
…
> I see you want to fix the kfree(NULL) issues in these two patches.

I propose to avoid redundant function calls at various source code places.


> But I am wondering if this is necessary, since kfree() can handle NULL correctly.

Would you prefer only required data processing in affected function implementations?

Regards,
Markus

Re: [0/2] net/smc: Adjustments for two function implementations

[Wen Gu]

On 2024/1/2 16:51, Markus Elfring wrote:
> …
>>> A few update suggestions were taken into account
>>> from static source code analysis.
> …
>>>     Return directly after a failed kzalloc() in smc_fill_gid_list()
>>>     Improve exception handling in smc_llc_cli_add_link_invite()
>>>
>>>    net/smc/af_smc.c  |  2 +-
>>>    net/smc/smc_llc.c | 15 +++++++--------
>>>    2 files changed, 8 insertions(+), 9 deletions(-)
> …
>> I see you want to fix the kfree(NULL) issues in these two patches.
> 
> I propose to avoid redundant function calls at various source code places.
> 
> 
>> But I am wondering if this is necessary, since kfree() can handle NULL correctly.
> 
> Would you prefer only required data processing in affected function implementations?
> 

Thank you Markus. I understood that you want to avoid redundant function calls.

But it is not very attractive to me since the calls occur on low-frequency paths
or unlikely condition, resulting in limited performance loss and the current
kfree() usage is fine and common. So what is the benfit?

I noticed that some other discussions are on-going. It seems like you are trying
to change other similiar places. Let's collect more opinions.

https://lore.kernel.org/netdev/828bb442-29d0-4bb8-b90d-f200bdd4faf6@web.de/
https://lore.kernel.org/netdev/90679f69-951c-47b3-b86f-75fd9fde3da3@web.de/
https://lore.kernel.org/netdev/dc0a1c9d-ceca-473d-9ad5-89b59e6af2e7@web.de/
https://lore.kernel.org/netdev/cde82080-c715-473c-97ac-6ef66bba6d64@web.de/

Thanks.

> Regards,
> Markus

Re: [0/2] net/smc: Adjustments for two function implementations

[Markus Elfring]

> But it is not very attractive to me since the calls occur on low-frequency paths
> or unlikely condition, resulting in limited performance loss and the current
> kfree() usage is fine and common.

The prioritisation of development activities influences progress in related areas.


> So what is the benfit?

* Source code clarity

* Nicer run time characteristics


See also:
https://wiki.sei.cmu.edu/confluence/display/c/MEM12-C.+Consider+using+a+goto+chain+when+leaving+a+function+on+error+when+using+and+releasing+resources


> I noticed that some other discussions are on-going. It seems like you are trying
> to change other similiar places.

I would appreciate if improvements can be achieved also for similarly affected
software components.

Regards,
Markus

Re: [0/2] net/smc: Adjustments for two function implementations

[Simon Horman]

On Tue, Jan 02, 2024 at 07:33:18PM +0800, Wen Gu wrote:
> 
> 
> On 2024/1/2 16:51, Markus Elfring wrote:
> > …
> > > > A few update suggestions were taken into account
> > > > from static source code analysis.
> > …
> > > >     Return directly after a failed kzalloc() in smc_fill_gid_list()
> > > >     Improve exception handling in smc_llc_cli_add_link_invite()
> > > > 
> > > >    net/smc/af_smc.c  |  2 +-
> > > >    net/smc/smc_llc.c | 15 +++++++--------
> > > >    2 files changed, 8 insertions(+), 9 deletions(-)
> > …
> > > I see you want to fix the kfree(NULL) issues in these two patches.
> > 
> > I propose to avoid redundant function calls at various source code places.
> > 
> > 
> > > But I am wondering if this is necessary, since kfree() can handle NULL correctly.
> > 
> > Would you prefer only required data processing in affected function implementations?
> > 
> 
> Thank you Markus. I understood that you want to avoid redundant function calls.
> 
> But it is not very attractive to me since the calls occur on low-frequency paths
> or unlikely condition, resulting in limited performance loss and the current
> kfree() usage is fine and common. So what is the benfit?
> 
> I noticed that some other discussions are on-going. It seems like you are trying
> to change other similiar places. Let's collect more opinions.
> 
> https://lore.kernel.org/netdev/828bb442-29d0-4bb8-b90d-f200bdd4faf6@web.de/
> https://lore.kernel.org/netdev/90679f69-951c-47b3-b86f-75fd9fde3da3@web.de/
> https://lore.kernel.org/netdev/dc0a1c9d-ceca-473d-9ad5-89b59e6af2e7@web.de/
> https://lore.kernel.org/netdev/cde82080-c715-473c-97ac-6ef66bba6d64@web.de/

As as been explained to Markus many times recently,
calling kfree(NULL) is not only perfectly fine,
it is the preferred way of handling things.

Markus, please stop posting patches of this nature to Netdev.

-- 
pw-bot: rejected

Re: [0/2] net/smc: Adjustments for two function implementations

[Markus Elfring]

…
>>    Return directly after a failed kzalloc() in smc_fill_gid_list()
>>    Improve exception handling in smc_llc_cli_add_link_invite()
>>
>>   net/smc/af_smc.c  |  2 +-
>>   net/smc/smc_llc.c | 15 +++++++--------
>>   2 files changed, 8 insertions(+), 9 deletions(-)
…
> However, I'm on the same page with Wen Gu. I could not see the necessity of the patches.
> BTW, if you want to send fix patches,

I obviously propose to adjust specific implementation details.


> please provide the error messages you met,

This development concern does not apply here.


> the procedure of reproducing the issue and the correspoinding commit messages.

Would you like to extend the usage of source code analysis tools?


> If you want to send feature patches, I'd like to see a well thought-out patch or patch series.

I presented some thoughts for special transformation patterns
on several software components.


> E.g. In our component, the kfree(NULL) issue doesn't only occur in the positions where you mentioned in the patch series, also somewhere else.

Does your feedback indicate that you would support the avoidance of such a special function call
at more places?


> I would be grateful if all of them would be cleaned up, not just some pieces.

Do you find my patch series too small for the mentioned Linux module at the moment?

Regards,
Markus