| Message ID | 20210916213336.1710044-1-f.fainelli@gmail.com |
|---|---|
| State | New |
| Headers | show |
| Series | [net] net: dsa: bcm_sf2: Fix array overrun in bcm_sf2_num_active_ports() | expand |
Hello: This patch was applied to netdev/net.git (refs/heads/master): On Thu, 16 Sep 2021 14:33:35 -0700 you wrote: > After d12e1c464988 ("net: dsa: b53: Set correct number of ports in the > DSA struct") we stopped setting dsa_switch::num_ports to DSA_MAX_PORTS, > which created an off by one error between the statically allocated > bcm_sf2_priv::port_sts array (of size DSA_MAX_PORTS). When > dsa_is_cpu_port() is used, we end-up accessing an out of bounds member > and causing a NPD. > > [...] Here is the summary with links: - [net] net: dsa: bcm_sf2: Fix array overrun in bcm_sf2_num_active_ports() https://git.kernel.org/netdev/net/c/02319bf15acf You are awesome, thank you! -- Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html
On 2021-09-16 23:33, Florian Fainelli wrote: > After d12e1c464988 ("net: dsa: b53: Set correct number of ports in the > DSA struct") we stopped setting dsa_switch::num_ports to DSA_MAX_PORTS, > which created an off by one error between the statically allocated > bcm_sf2_priv::port_sts array (of size DSA_MAX_PORTS). When > dsa_is_cpu_port() is used, we end-up accessing an out of bounds member > and causing a NPD. > > Fix this by iterating with the appropriate port count using > ds->num_ports. > > Fixes: d12e1c464988 ("net: dsa: b53: Set correct number of ports in > the DSA struct") > Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Tested-by: Rafał Miłecki <rafal@milecki.pl> This fixes: [ 0.515409] Unable to handle kernel read from unreadable memory at virtual address 0000000000000028 [ 0.524659] Mem abort info: [ 0.527522] ESR = 0x96000005 [ 0.530656] EC = 0x25: DABT (current EL), IL = 32 bits [ 0.536119] SET = 0, FnV = 0 [ 0.539262] EA = 0, S1PTW = 0 [ 0.542481] Data abort info: [ 0.545438] ISV = 0, ISS = 0x00000005 [ 0.549383] CM = 0, WnR = 0 [ 0.552427] [0000000000000028] user address but active_mm is swapper [ 0.558973] Internal error: Oops: 96000005 [#1] SMP [ 0.563986] Modules linked in: [ 0.567125] CPU: 1 PID: 24 Comm: kworker/1:1 Not tainted 5.10.64 #0 [ 0.573573] Hardware name: Netgear R8000P (DT) [ 0.578155] Workqueue: events deferred_probe_work_func [ 0.583431] pstate: 60400005 (nZCv daif +PAN -UAO -TCO BTYPE=--) [ 0.589617] pc : bcm_sf2_recalc_clock+0x58/0xe4 [ 0.594271] lr : bcm_sf2_port_setup+0xc0/0x2ac [ 0.598840] sp : ffffffc0109bb980 [ 0.602244] x29: ffffffc0109bb980 x28: ffffff801fef6f60 [ 0.607710] x27: ffffff8001242b30 x26: 0000000000039040 [ 0.613175] x25: 0000000000002380 x24: 0000000000000003 [ 0.618641] x23: ffffff800125f880 x22: 0000000000000003 [ 0.624107] x21: 0000000000000000 x20: 0000000000000000 [ 0.629572] x19: ffffff8001398280 x18: 0000002437b29c0a [ 0.635039] x17: 00008cad14430a3a x16: 0000000000000008 [ 0.640503] x15: 0000000000000000 x14: 6863746977732d74 [ 0.645969] x13: 656e72656874652e x12: 3030303038303038 [ 0.651435] x11: 0002001d00000000 x10: 6d726f6674616c70 [ 0.656900] x9 : ffffff800125f880 x8 : ffffff8001398800 [ 0.662366] x7 : ffffff80013989b8 x6 : 0000000000000001 [ 0.667832] x5 : ffffff800125f97c x4 : ffffff8001242b30 [ 0.673297] x3 : 0000000000000009 x2 : ffffff8001242b30 [ 0.678763] x1 : 0000000000000000 x0 : ffffff8001398280 [ 0.684230] Call trace: [ 0.686740] bcm_sf2_recalc_clock+0x58/0xe4
diff --git a/drivers/net/dsa/bcm_sf2.c b/drivers/net/dsa/bcm_sf2.c index 6ce9ec1283e0..b6c4b3adb171 100644 --- a/drivers/net/dsa/bcm_sf2.c +++ b/drivers/net/dsa/bcm_sf2.c @@ -68,7 +68,7 @@ static unsigned int bcm_sf2_num_active_ports(struct dsa_switch *ds) struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds); unsigned int port, count = 0; - for (port = 0; port < ARRAY_SIZE(priv->port_sts); port++) { + for (port = 0; port < ds->num_ports; port++) { if (dsa_is_cpu_port(ds, port)) continue; if (priv->port_sts[port].enabled)
After d12e1c464988 ("net: dsa: b53: Set correct number of ports in the DSA struct") we stopped setting dsa_switch::num_ports to DSA_MAX_PORTS, which created an off by one error between the statically allocated bcm_sf2_priv::port_sts array (of size DSA_MAX_PORTS). When dsa_is_cpu_port() is used, we end-up accessing an out of bounds member and causing a NPD. Fix this by iterating with the appropriate port count using ds->num_ports. Fixes: d12e1c464988 ("net: dsa: b53: Set correct number of ports in the DSA struct") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> --- drivers/net/dsa/bcm_sf2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)