[PATCH v2] drivers: net: cpsw: fix kmemleak false-positive reports for sk buffers
From: Grygorii Strashko
Date: Wed Aug 10 2016 - 17:23:09 EST
Kmemleak reports following false positive memory leaks for each sk
buffers allocated by CPSW (__netdev_alloc_skb_ip_align()) in
cpsw_ndo_open() and cpsw_rx_handler():
unreferenced object 0xea915000 (size 2048):
comm "systemd-network", pid 713, jiffies 4294938323 (age 102.180s)
hex dump (first 32 bytes):
00 58 91 ea ff ff ff ff ff ff ff ff ff ff ff ff .X..............
ff ff ff ff ff ff fd 0f 00 00 00 00 00 00 00 00 ................
backtrace:
[<c0108680>] __kmalloc_track_caller+0x1a4/0x230
[<c0529eb4>] __alloc_skb+0x68/0x16c
[<c052c884>] __netdev_alloc_skb+0x40/0x104
[<bf1ad29c>] cpsw_ndo_open+0x374/0x670 [ti_cpsw]
[<c053c3d4>] __dev_open+0xb0/0x114
[<c053c690>] __dev_change_flags+0x9c/0x14c
[<c053c760>] dev_change_flags+0x20/0x50
[<c054bdcc>] do_setlink+0x2cc/0x78c
[<c054c358>] rtnl_setlink+0xcc/0x100
[<c054b34c>] rtnetlink_rcv_msg+0x184/0x224
[<c056467c>] netlink_rcv_skb+0xa8/0xc4
[<c054b1c0>] rtnetlink_rcv+0x2c/0x34
[<c0564018>] netlink_unicast+0x16c/0x1f8
[<c0564498>] netlink_sendmsg+0x334/0x348
[<c052015c>] sock_sendmsg+0x1c/0x2c
[<c05213e0>] SyS_sendto+0xc0/0xe8
unreferenced object 0xec861780 (size 192):
comm "softirq", pid 0, jiffies 4294938759 (age 109.540s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 b0 5a ed 00 00 00 00 00 00 00 00 ......Z.........
backtrace:
[<c0107830>] kmem_cache_alloc+0x190/0x208
[<c052c768>] __build_skb+0x30/0x98
[<c052c8fc>] __netdev_alloc_skb+0xb8/0x104
[<bf1abc54>] cpsw_rx_handler+0x68/0x1e4 [ti_cpsw]
[<bf11aa30>] __cpdma_chan_free+0xa8/0xc4 [davinci_cpdma]
[<bf11ab98>] __cpdma_chan_process+0x14c/0x16c [davinci_cpdma]
[<bf11abfc>] cpdma_chan_process+0x44/0x5c [davinci_cpdma]
[<bf1adc78>] cpsw_rx_poll+0x1c/0x9c [ti_cpsw]
[<c0539180>] net_rx_action+0x1f0/0x2ec
[<c003881c>] __do_softirq+0x134/0x258
[<c0038a00>] do_softirq+0x68/0x70
[<c0038adc>] __local_bh_enable_ip+0xd4/0xe8
[<c0640994>] _raw_spin_unlock_bh+0x30/0x34
[<c05f4e9c>] igmp6_group_added+0x4c/0x1bc
[<c05f6600>] ipv6_dev_mc_inc+0x398/0x434
[<c05dba74>] addrconf_dad_work+0x224/0x39c
This happens because CPSW allocates SK buffers and then passes
pointers on them in CPDMA where they stored in internal CPPI RAM
(SRAM) which belongs to DEV MMIO space. Kmemleak does not scan IO
memory and so reports memory leaks.
Hence, mark allocated sk buffers as false positive explicitly.
Acked-by: Catalin Marinas <catalin.marinas@xxxxxxx>
Signed-off-by: Grygorii Strashko <grygorii.strashko@xxxxxx>
---
changes in v2:
- comments added
drivers/net/ethernet/ti/cpsw.c | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 0805855..5caef77 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -732,6 +732,11 @@ static void cpsw_rx_handler(void *token, int len, int status)
netif_receive_skb(skb);
ndev->stats.rx_bytes += len;
ndev->stats.rx_packets++;
+ /* SKB pointer will be stored in CPPI RAM (SRAM) which belongs
+ * to MMIO space, as result false positive memory leak report
+ * will be generated.
+ */
+ kmemleak_not_leak(new_skb);
} else {
ndev->stats.rx_dropped++;
new_skb = skb;
@@ -1323,6 +1328,11 @@ static int cpsw_ndo_open(struct net_device *ndev)
kfree_skb(skb);
goto err_cleanup;
}
+ /* SKB pointer will be stored in CPPI RAM (SRAM) which
+ * belongs to MMIO space, as result false positive
+ * memory leak report will be generated.
+ */
+ kmemleak_not_leak(skb);
}
/* continue even if we didn't manage to submit all
* receive descs
--
2.9.2