net: ps3_gelic_net: handle skb allocation failures

Handle skb allocation failures in RX path, to avoid NULL pointer
dereference and RX stalls under memory pressure. If the refill fails
with -ENOMEM, complete napi polling and wake up later to retry via timer.
Also explicitly re-enable RX DMA after oom, so the dmac doesn't remain
stopped in this situation.

Previously, memory pressure could lead to skb allocation failures and
subsequent Oops like:

	Oops: Kernel access of bad area, sig: 11 [#2]
	Hardware name: SonyPS3 Cell Broadband Engine 0x701000 PS3
	NIP [c0003d0000065900] gelic_net_poll+0x6c/0x2d0 [ps3_gelic] (unreliable)
	LR [c0003d00000659c4] gelic_net_poll+0x130/0x2d0 [ps3_gelic]
	Call Trace:
	  gelic_net_poll+0x130/0x2d0 [ps3_gelic] (unreliable)
	  __napi_poll+0x44/0x168
	  net_rx_action+0x178/0x290

Steps to reproduce the issue:
	1. Start a continuous network traffic, like scp of a 20GB file
	2. Inject failslab errors using the kernel fault injection:
	    echo -1 > /sys/kernel/debug/failslab/times
	    echo 30 > /sys/kernel/debug/failslab/interval
	    echo 100 > /sys/kernel/debug/failslab/probability
	3. After some time, traces start to appear, kernel Oopses
	   and the system stops

Step 2 is not always necessary, as it is usually already triggered by
the transfer of a big enough file.

Fixes: 02c1889166b4 ("ps3: gigabit ethernet driver for PS3, take3")
Signed-off-by: Florian Fuchs <fuchsfl@gmail.com>
Link: https://patch.msgid.link/20251113181000.3914980-1-fuchsfl@gmail.com
Signed-off-by: Paolo Abeni <pabeni@redhat.com>

diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.c b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
index 5ee8e8980393c3491bf9cf91eb8e0dbb2df0f427..591866fc9055d635282e5f158b53ab30fd5ccaed 100644 (file)
--- a/drivers/net/ethernet/toshiba/ps3_gelic_net.c
+++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.c
@@ -260,6 +260,7 @@ void gelic_card_down(struct gelic_card *card)
        if (atomic_dec_if_positive(&card->users) == 0) {
                pr_debug("%s: real do\n", __func__);
                napi_disable(&card->napi);
+               timer_delete_sync(&card->rx_oom_timer);
                /*
                 * Disable irq. Wireless interrupts will
                 * be disabled later if any
@@ -970,7 +971,8 @@ static void gelic_net_pass_skb_up(struct gelic_descr *descr,
  * gelic_card_decode_one_descr - processes an rx descriptor
  * @card: card structure
  *
- * returns 1 if a packet has been sent to the stack, otherwise 0
+ * returns 1 if a packet has been sent to the stack, -ENOMEM on skb alloc
+ * failure, otherwise 0
  *
  * processes an rx descriptor by iommu-unmapping the data buffer and passing
  * the packet up to the stack
@@ -981,16 +983,18 @@ static int gelic_card_decode_one_descr(struct gelic_card *card)
        struct gelic_descr_chain *chain = &card->rx_chain;
        struct gelic_descr *descr = chain->head;
        struct net_device *netdev = NULL;
-       int dmac_chain_ended;
+       int dmac_chain_ended = 0;
+       int prepare_rx_ret;
 
        status = gelic_descr_get_status(descr);
 
        if (status == GELIC_DESCR_DMA_CARDOWNED)
                return 0;
 
-       if (status == GELIC_DESCR_DMA_NOT_IN_USE) {
+       if (status == GELIC_DESCR_DMA_NOT_IN_USE || !descr->skb) {
                dev_dbg(ctodev(card), "dormant descr? %p\n", descr);
-               return 0;
+               dmac_chain_ended = 1;
+               goto refill;
        }
 
        /* netdevice select */
@@ -1048,9 +1052,10 @@ static int gelic_card_decode_one_descr(struct gelic_card *card)
 refill:
 
        /* is the current descriptor terminated with next_descr == NULL? */
-       dmac_chain_ended =
-               be32_to_cpu(descr->hw_regs.dmac_cmd_status) &
-               GELIC_DESCR_RX_DMA_CHAIN_END;
+       if (!dmac_chain_ended)
+               dmac_chain_ended =
+                       be32_to_cpu(descr->hw_regs.dmac_cmd_status) &
+                       GELIC_DESCR_RX_DMA_CHAIN_END;
        /*
         * So that always DMAC can see the end
         * of the descriptor chain to avoid
@@ -1062,10 +1067,11 @@ refill:
        gelic_descr_set_status(descr, GELIC_DESCR_DMA_NOT_IN_USE);
 
        /*
-        * this call can fail, but for now, just leave this
-        * descriptor without skb
+        * this call can fail, propagate the error
         */
-       gelic_descr_prepare_rx(card, descr);
+       prepare_rx_ret = gelic_descr_prepare_rx(card, descr);
+       if (prepare_rx_ret)
+               return prepare_rx_ret;
 
        chain->tail = descr;
        chain->head = descr->next;
@@ -1087,6 +1093,13 @@ refill:
        return 1;
 }
 
+static void gelic_rx_oom_timer(struct timer_list *t)
+{
+       struct gelic_card *card = timer_container_of(card, t, rx_oom_timer);
+
+       napi_schedule(&card->napi);
+}
+
 /**
  * gelic_net_poll - NAPI poll function called by the stack to return packets
  * @napi: napi structure
@@ -1099,14 +1112,22 @@ static int gelic_net_poll(struct napi_struct *napi, int budget)
 {
        struct gelic_card *card = container_of(napi, struct gelic_card, napi);
        int packets_done = 0;
+       int work_result = 0;
 
        while (packets_done < budget) {
-               if (!gelic_card_decode_one_descr(card))
+               work_result = gelic_card_decode_one_descr(card);
+               if (work_result != 1)
                        break;
 
                packets_done++;
        }
 
+       if (work_result == -ENOMEM) {
+               napi_complete_done(napi, packets_done);
+               mod_timer(&card->rx_oom_timer, jiffies + 1);
+               return packets_done;
+       }
+
        if (packets_done < budget) {
                napi_complete_done(napi, packets_done);
                gelic_card_rx_irq_on(card);
@@ -1576,6 +1597,8 @@ static struct gelic_card *gelic_alloc_card_net(struct net_device **netdev)
        mutex_init(&card->updown_lock);
        atomic_set(&card->users, 0);
 
+       timer_setup(&card->rx_oom_timer, gelic_rx_oom_timer, 0);
+
        return card;
 }
 

diff --git a/drivers/net/ethernet/toshiba/ps3_gelic_net.h b/drivers/net/ethernet/toshiba/ps3_gelic_net.h
index f7d7931e51b77dd6620d69facf2424f1594f50cd..c10f1984a5a114cc40d160eec6195479959ab36c 100644 (file)
--- a/drivers/net/ethernet/toshiba/ps3_gelic_net.h
+++ b/drivers/net/ethernet/toshiba/ps3_gelic_net.h
@@ -268,6 +268,7 @@ struct gelic_vlan_id {
 struct gelic_card {
        struct napi_struct napi;
        struct net_device *netdev[GELIC_PORT_MAX];
+       struct timer_list rx_oom_timer;
        /*
         * hypervisor requires irq_status should be
         * 8 bytes aligned, but u64 member is