Thread (8 messages) 8 messages, 3 authors, 2025-12-29

Re: [PATCH net] net: stmmac: fix the crash issue for zero copy XDP_TX action

From: Hariprasad Kelam <hkelam@marvell.com>
Date: 2025-12-17 11:09:14
Also in: bpf, imx, linux-arm-kernel, lkml

On 2025-12-04 at 12:43:32, Wei Fang (wei.fang@nxp.com) wrote:
quoted hunk ↗ jump to hunk
There is a crash issue when running zero copy XDP_TX action, the crash
log is shown below.

[  216.122464] Unable to handle kernel paging request at virtual address fffeffff80000000
[  216.187524] Internal error: Oops: 0000000096000144 [#1]  SMP
[  216.301694] Call trace:
[  216.304130]  dcache_clean_poc+0x20/0x38 (P)
[  216.308308]  __dma_sync_single_for_device+0x1bc/0x1e0
[  216.313351]  stmmac_xdp_xmit_xdpf+0x354/0x400
[  216.317701]  __stmmac_xdp_run_prog+0x164/0x368
[  216.322139]  stmmac_napi_poll_rxtx+0xba8/0xf00
[  216.326576]  __napi_poll+0x40/0x218
[  216.408054] Kernel panic - not syncing: Oops: Fatal exception in interrupt

For XDP_TX action, the xdp_buff is converted to xdp_frame by
xdp_convert_buff_to_frame(). The memory type of the resulting xdp_frame
depends on the memory type of the xdp_buff. For page pool based xdp_buff
it produces xdp_frame with memory type MEM_TYPE_PAGE_POOL. For zero copy
XSK pool based xdp_buff it produces xdp_frame with memory type
MEM_TYPE_PAGE_ORDER0. However, stmmac_xdp_xmit_back() does not check the
memory type and always uses the page pool type, this leads to invalid
mappings and causes the crash. Therefore, check the xdp_buff memory type
in stmmac_xdp_xmit_back() to fix this issue.

Fixes: bba2556efad6 ("net: stmmac: Enable RX via AF_XDP zero-copy")
Signed-off-by: Wei Fang <wei.fang@nxp.com>
---
 .../net/ethernet/stmicro/stmmac/stmmac_main.c   | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index 7b90ecd3a55e..a6664f300e4a 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -88,6 +88,7 @@ MODULE_PARM_DESC(phyaddr, "Physical device address");
 #define STMMAC_XDP_CONSUMED	BIT(0)
 #define STMMAC_XDP_TX		BIT(1)
 #define STMMAC_XDP_REDIRECT	BIT(2)
+#define STMMAC_XSK_CONSUMED	BIT(3)
 
 static int flow_ctrl = 0xdead;
 module_param(flow_ctrl, int, 0644);
@@ -4988,6 +4989,7 @@ static int stmmac_xdp_get_tx_queue(struct stmmac_priv *priv,
 static int stmmac_xdp_xmit_back(struct stmmac_priv *priv,
 				struct xdp_buff *xdp)
 {
+	bool zc = !!(xdp->rxq->mem.type == MEM_TYPE_XSK_BUFF_POOL);
 	struct xdp_frame *xdpf = xdp_convert_buff_to_frame(xdp);
 	int cpu = smp_processor_id();
 	struct netdev_queue *nq;
@@ -5004,9 +5006,18 @@ static int stmmac_xdp_xmit_back(struct stmmac_priv *priv,
 	/* Avoids TX time-out as we are sharing with slow path */
 	txq_trans_cond_update(nq);
 
-	res = stmmac_xdp_xmit_xdpf(priv, queue, xdpf, false);
-	if (res == STMMAC_XDP_TX)
+	/* For zero copy XDP_TX action, dma_map is true */
+	res = stmmac_xdp_xmit_xdpf(priv, queue, xdpf, zc);
	Seems stmmac_xdp_xmit_xdpf is using dma_map_single if we pass zc is true.
        Ideally in case of zc, driver can use page_pool_get_dma_addr, may be you
        need pass zc param as false. Please check

Thanks,
Hariprasad k
	
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help