[PATCH] nfp: abm: fix a memory leak bug

[PATCH] nfp: abm: fix a memory leak bug

[wu000273]

From: Qiushi Wu <wu000273@umn.edu>

In function nfp_abm_vnic_set_mac, pointer nsp is allocated by nfp_nsp_open.
But when nfp_nsp_has_hwinfo_lookup fail, the pointer is not released,
which can lead to a memory leak bug. Fix this issue by adding
nfp_nsp_close(nsp) in the error path.

Signed-off-by: Qiushi Wu <wu000273@umn.edu>
---
 drivers/net/ethernet/netronome/nfp/abm/main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/netronome/nfp/abm/main.c b/drivers/net/ethernet/netronome/nfp/abm/main.c
index 9183b3e85d21..354efffac0f9 100644
--- a/drivers/net/ethernet/netronome/nfp/abm/main.c
+++ b/drivers/net/ethernet/netronome/nfp/abm/main.c
@@ -283,6 +283,7 @@ nfp_abm_vnic_set_mac(struct nfp_pf *pf, struct nfp_abm *abm, struct nfp_net *nn,
 	if (!nfp_nsp_has_hwinfo_lookup(nsp)) {
 		nfp_warn(pf->cpp, "NSP doesn't support PF MAC generation\n");
 		eth_hw_addr_random(nn->dp.netdev);
+		nfp_nsp_close(nsp);
 		return;
 	}
 
-- 
2.17.1

Re: [PATCH] nfp: abm: fix a memory leak bug

[Jakub Kicinski]

On Sat,  2 May 2020 17:42:59 -0500 wu000273@umn.edu wrote:
> From: Qiushi Wu <wu000273@umn.edu>
> 
> In function nfp_abm_vnic_set_mac, pointer nsp is allocated by nfp_nsp_open.
> But when nfp_nsp_has_hwinfo_lookup fail, the pointer is not released,
> which can lead to a memory leak bug. Fix this issue by adding
> nfp_nsp_close(nsp) in the error path.
> 
> Signed-off-by: Qiushi Wu <wu000273@umn.edu>

Fixes: f6e71efdf9fb1 ("nfp: abm: look up MAC addresses via management FW")
Acked-by: Jakub Kicinski <kuba@kernel.org>

Re: [PATCH] nfp: abm: fix a memory leak bug

[David Miller]

From: wu000273@umn.edu
Date: Sat,  2 May 2020 17:42:59 -0500

> From: Qiushi Wu <wu000273@umn.edu>
> 
> In function nfp_abm_vnic_set_mac, pointer nsp is allocated by nfp_nsp_open.
> But when nfp_nsp_has_hwinfo_lookup fail, the pointer is not released,
> which can lead to a memory leak bug. Fix this issue by adding
> nfp_nsp_close(nsp) in the error path.
> 
> Signed-off-by: Qiushi Wu <wu000273@umn.edu>

Applied and queued up for -stable, thanks.

Re: [PATCH] nfp: abm: fix a memory leak bug

[Markus Elfring]

…
> Fix this issue by adding nfp_nsp_close(nsp) in the error path.

How do you think about a wording variant like the following?

   Subject:
   [PATCH v2] nfp: abm: Fix incomplete release of system resources in nfp_abm_vnic_set_mac()

   Change description:
   …
   Thus add a call of the function “nfp_nsp_close” for the completion
   of the exception handling.


…
> +++ b/drivers/net/ethernet/netronome/nfp/abm/main.c
> @@ -283,6 +283,7 @@  nfp_abm_vnic_set_mac(struct nfp_pf *pf, struct nfp_abm *abm, struct nfp_net *nn,
>  	if (!nfp_nsp_has_hwinfo_lookup(nsp)) {
>  		nfp_warn(pf->cpp, "NSP doesn't support PF MAC generation\n");
>  		eth_hw_addr_random(nn->dp.netdev);
> +		nfp_nsp_close(nsp);
>  		return;
>  	}

I suggest to reconsider the order for resource clean-up function calls
a bit more.

+		nfp_nsp_close(nsp);
-		eth_hw_addr_random(nn->dp.netdev);
-		return;
+		goto generate_random_address;


Should such a jump target be added in another update step?

Would you like to add the tag “Fixes”?

Regards,
Markus