Good find, wrong solution. The chanspecs is a temporarily variable which should 
be freed when exiting the function. Not only when there is an error. I 
personally would have preferred just a  free at the end of the " if 
(request->n_channels) {". So something like this:

                }
                err = brcmf_p2p_escan(p2p, num_nodfs, chanspecs, search_state,
                                      action, P2PAPI_BSSCFG_DEVICE);
+               kfree(chanspecs);
        }

In this case the pointer doesn't have to be initialized to NULL.


Regards,
Hante

-----Original Message-----
From: Geyslan G. Bem [mailto:[email protected]] 
Sent: vrijdag 15 november 2013 12:54
To: [email protected]
Cc: Brett Rudley; Arend Van Spriel; Franky Lin; Hante Meuleman; John W. 
Linville; Pieter-Paul Giesberts; Piotr Haber; [email protected]; 
brcm80211-dev-list; [email protected]; [email protected]
Subject: [PATCH] brcmfmac: fix possible memory leak

In case of error free 'chanspecs'.

Signed-off-by: Geyslan G. Bem <[email protected]>
---
 drivers/net/wireless/brcm80211/brcmfmac/p2p.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/net/wireless/brcm80211/brcmfmac/p2p.c 
b/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
index d7a9745..aea2c2e 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/p2p.c
@@ -771,7 +771,7 @@ static s32 brcmf_p2p_run_escan(struct brcmf_cfg80211_info 
*cfg,
        struct brcmf_cfg80211_vif *vif;
        struct net_device *dev = NULL;
        int i, num_nodfs = 0;
-       u16 *chanspecs;
+       u16 *chanspecs = NULL;
 
        brcmf_dbg(TRACE, "enter\n");
 
@@ -825,8 +825,10 @@ static s32 brcmf_p2p_run_escan(struct brcmf_cfg80211_info 
*cfg,
                                      action, P2PAPI_BSSCFG_DEVICE);
        }
 exit:
-       if (err)
+       if (err) {
                brcmf_err("error (%d)\n", err);
+               kfree(chanspecs);
+       }
        return err;
 }
 
-- 
1.8.4.2



--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to