| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <1223572568-1298-5-git-send-email-mchan@broadcom.com>
Date: Thu, 9 Oct 2008 10:16:07 -0700
From: "Michael Chan" <mchan@...adcom.com>
To: davem@...emloft.net
cc: netdev@...r.kernel.org, "Michael Chan" <mchan@...adcom.com>,
"Benjamin Li" <benli@...adcom.com>
Subject: [PATCH net-next 5/6] bnx2: Handle DMA mapping errors.
Before, the driver would not care about the return codes from pci_map_*
functions. This could be potentially dangerous if a mapping failed.
Now, we will check all pci_map_* calls. On the transmit side, we switch
to use the new function skb_dma_map(). On the receive side, we add
pci_dma_mapping_error().
Signed-off-by: Benjamin Li <benli@...adcom.com>
Signed-off-by: Michael Chan <mchan@...adcom.com>
---
drivers/net/bnx2.c | 114 ++++++++++++++++++++++++++++------------------------
drivers/net/bnx2.h | 8 +++-
2 files changed, 67 insertions(+), 55 deletions(-)
diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index f147204..a95ca4f 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -2476,6 +2476,11 @@ bnx2_alloc_rx_page(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, u16 index)
return -ENOMEM;
mapping = pci_map_page(bp->pdev, page, 0, PAGE_SIZE,
PCI_DMA_FROMDEVICE);
+ if (pci_dma_mapping_error(bp->pdev, mapping)) {
+ __free_page(page);
+ return -EIO;
+ }
+
rx_pg->page = page;
pci_unmap_addr_set(rx_pg, mapping, mapping);
rxbd->rx_bd_haddr_hi = (u64) mapping >> 32;
@@ -2518,6 +2523,10 @@ bnx2_alloc_rx_skb(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, u16 index)
mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size,
PCI_DMA_FROMDEVICE);
+ if (pci_dma_mapping_error(bp->pdev, mapping)) {
+ dev_kfree_skb(skb);
+ return -EIO;
+ }
rx_buf->skb = skb;
pci_unmap_addr_set(rx_buf, mapping, mapping);
@@ -2592,7 +2601,7 @@ bnx2_tx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
sw_cons = txr->tx_cons;
while (sw_cons != hw_cons) {
- struct sw_bd *tx_buf;
+ struct sw_tx_bd *tx_buf;
struct sk_buff *skb;
int i, last;
@@ -2617,21 +2626,13 @@ bnx2_tx_int(struct bnx2 *bp, struct bnx2_napi *bnapi, int budget)
}
}
- pci_unmap_single(bp->pdev, pci_unmap_addr(tx_buf, mapping),
- skb_headlen(skb), PCI_DMA_TODEVICE);
+ skb_dma_unmap(&bp->pdev->dev, skb, DMA_TO_DEVICE);
tx_buf->skb = NULL;
last = skb_shinfo(skb)->nr_frags;
for (i = 0; i < last; i++) {
sw_cons = NEXT_TX_BD(sw_cons);
-
- pci_unmap_page(bp->pdev,
- pci_unmap_addr(
- &txr->tx_buf_ring[TX_RING_IDX(sw_cons)],
- mapping),
- skb_shinfo(skb)->frags[i].size,
- PCI_DMA_TODEVICE);
}
sw_cons = NEXT_TX_BD(sw_cons);
@@ -2672,11 +2673,31 @@ bnx2_reuse_rx_skb_pages(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr,
{
struct sw_pg *cons_rx_pg, *prod_rx_pg;
struct rx_bd *cons_bd, *prod_bd;
- dma_addr_t mapping;
int i;
- u16 hw_prod = rxr->rx_pg_prod, prod;
+ u16 hw_prod, prod;
u16 cons = rxr->rx_pg_cons;
+ cons_rx_pg = &rxr->rx_pg_ring[cons];
+
+ /* The caller was unable to allocate a new page to replace the
+ * last one in the frags array, so we need to recycle that page
+ * and then free the skb.
+ */
+ if (skb) {
+ struct page *page;
+ struct skb_shared_info *shinfo;
+
+ shinfo = skb_shinfo(skb);
+ shinfo->nr_frags--;
+ page = shinfo->frags[shinfo->nr_frags].page;
+ shinfo->frags[shinfo->nr_frags].page = NULL;
+
+ cons_rx_pg->page = page;
+ dev_kfree_skb(skb);
+ }
+
+ hw_prod = rxr->rx_pg_prod;
+
for (i = 0; i < count; i++) {
prod = RX_PG_RING_IDX(hw_prod);
@@ -2685,20 +2706,6 @@ bnx2_reuse_rx_skb_pages(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr,
cons_bd = &rxr->rx_pg_desc_ring[RX_RING(cons)][RX_IDX(cons)];
prod_bd = &rxr->rx_pg_desc_ring[RX_RING(prod)][RX_IDX(prod)];
- if (i == 0 && skb) {
- struct page *page;
- struct skb_shared_info *shinfo;
-
- shinfo = skb_shinfo(skb);
- shinfo->nr_frags--;
- page = shinfo->frags[shinfo->nr_frags].page;
- shinfo->frags[shinfo->nr_frags].page = NULL;
- mapping = pci_map_page(bp->pdev, page, 0, PAGE_SIZE,
- PCI_DMA_FROMDEVICE);
- cons_rx_pg->page = page;
- pci_unmap_addr_set(cons_rx_pg, mapping, mapping);
- dev_kfree_skb(skb);
- }
if (prod != cons) {
prod_rx_pg->page = cons_rx_pg->page;
cons_rx_pg->page = NULL;
@@ -2784,6 +2791,8 @@ bnx2_rx_skb(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, struct sk_buff *skb,
skb_put(skb, hdr_len);
for (i = 0; i < pages; i++) {
+ dma_addr_t mapping_old;
+
frag_len = min(frag_size, (unsigned int) PAGE_SIZE);
if (unlikely(frag_len <= 4)) {
unsigned int tail = 4 - frag_len;
@@ -2806,9 +2815,10 @@ bnx2_rx_skb(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, struct sk_buff *skb,
}
rx_pg = &rxr->rx_pg_ring[pg_cons];
- pci_unmap_page(bp->pdev, pci_unmap_addr(rx_pg, mapping),
- PAGE_SIZE, PCI_DMA_FROMDEVICE);
-
+ /* Don't unmap yet. If we're unable to allocate a new
+ * page, we need to recycle the page and the DMA addr.
+ */
+ mapping_old = pci_unmap_addr(rx_pg, mapping);
if (i == pages - 1)
frag_len -= 4;
@@ -2825,6 +2835,9 @@ bnx2_rx_skb(struct bnx2 *bp, struct bnx2_rx_ring_info *rxr, struct sk_buff *skb,
return err;
}
+ pci_unmap_page(bp->pdev, mapping_old,
+ PAGE_SIZE, PCI_DMA_FROMDEVICE);
+
frag_size -= frag_len;
skb->data_len += frag_len;
skb->truesize += frag_len;
@@ -4971,31 +4984,20 @@ bnx2_free_tx_skbs(struct bnx2 *bp)
continue;
for (j = 0; j < TX_DESC_CNT; ) {
- struct sw_bd *tx_buf = &txr->tx_buf_ring[j];
+ struct sw_tx_bd *tx_buf = &txr->tx_buf_ring[j];
struct sk_buff *skb = tx_buf->skb;
- int k, last;
if (skb == NULL) {
j++;
continue;
}
- pci_unmap_single(bp->pdev,
- pci_unmap_addr(tx_buf, mapping),
- skb_headlen(skb), PCI_DMA_TODEVICE);
+ skb_dma_unmap(&bp->pdev->dev, skb, DMA_TO_DEVICE);
tx_buf->skb = NULL;
- last = skb_shinfo(skb)->nr_frags;
- for (k = 0; k < last; k++) {
- tx_buf = &txr->tx_buf_ring[j + k + 1];
- pci_unmap_page(bp->pdev,
- pci_unmap_addr(tx_buf, mapping),
- skb_shinfo(skb)->frags[j].size,
- PCI_DMA_TODEVICE);
- }
+ j += skb_shinfo(skb)->nr_frags + 1;
dev_kfree_skb(skb);
- j += k + 1;
}
}
}
@@ -5373,8 +5375,11 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
for (i = 14; i < pkt_size; i++)
packet[i] = (unsigned char) (i & 0xff);
- map = pci_map_single(bp->pdev, skb->data, pkt_size,
- PCI_DMA_TODEVICE);
+ if (skb_dma_map(&bp->pdev->dev, skb, DMA_TO_DEVICE)) {
+ dev_kfree_skb(skb);
+ return -EIO;
+ }
+ map = skb_shinfo(skb)->dma_maps[0];
REG_WR(bp, BNX2_HC_COMMAND,
bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW_WO_INT);
@@ -5409,7 +5414,7 @@ bnx2_run_loopback(struct bnx2 *bp, int loopback_mode)
udelay(5);
- pci_unmap_single(bp->pdev, map, pkt_size, PCI_DMA_TODEVICE);
+ skb_dma_unmap(&bp->pdev->dev, skb, DMA_TO_DEVICE);
dev_kfree_skb(skb);
if (bnx2_get_hw_tx_cons(tx_napi) != txr->tx_prod)
@@ -5970,13 +5975,14 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
struct bnx2 *bp = netdev_priv(dev);
dma_addr_t mapping;
struct tx_bd *txbd;
- struct sw_bd *tx_buf;
+ struct sw_tx_bd *tx_buf;
u32 len, vlan_tag_flags, last_frag, mss;
u16 prod, ring_prod;
int i;
struct bnx2_napi *bnapi;
struct bnx2_tx_ring_info *txr;
struct netdev_queue *txq;
+ struct skb_shared_info *sp;
/* Determine which tx ring we will be placed on */
i = skb_get_queue_mapping(skb);
@@ -6041,11 +6047,16 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
} else
mss = 0;
- mapping = pci_map_single(bp->pdev, skb->data, len, PCI_DMA_TODEVICE);
+ if (skb_dma_map(&bp->pdev->dev, skb, DMA_TO_DEVICE)) {
+ dev_kfree_skb(skb);
+ return NETDEV_TX_OK;
+ }
+
+ sp = skb_shinfo(skb);
+ mapping = sp->dma_maps[0];
tx_buf = &txr->tx_buf_ring[ring_prod];
tx_buf->skb = skb;
- pci_unmap_addr_set(tx_buf, mapping, mapping);
txbd = &txr->tx_desc_ring[ring_prod];
@@ -6064,10 +6075,7 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev)
txbd = &txr->tx_desc_ring[ring_prod];
len = frag->size;
- mapping = pci_map_page(bp->pdev, frag->page, frag->page_offset,
- len, PCI_DMA_TODEVICE);
- pci_unmap_addr_set(&txr->tx_buf_ring[ring_prod],
- mapping, mapping);
+ mapping = sp->dma_maps[i + 1];
txbd->tx_bd_haddr_hi = (u64) mapping >> 32;
txbd->tx_bd_haddr_lo = (u64) mapping & 0xffffffff;
diff --git a/drivers/net/bnx2.h b/drivers/net/bnx2.h
index edc7774..617d953 100644
--- a/drivers/net/bnx2.h
+++ b/drivers/net/bnx2.h
@@ -6526,10 +6526,14 @@ struct sw_pg {
DECLARE_PCI_UNMAP_ADDR(mapping)
};
+struct sw_tx_bd {
+ struct sk_buff *skb;
+};
+
#define SW_RXBD_RING_SIZE (sizeof(struct sw_bd) * RX_DESC_CNT)
#define SW_RXPG_RING_SIZE (sizeof(struct sw_pg) * RX_DESC_CNT)
#define RXBD_RING_SIZE (sizeof(struct rx_bd) * RX_DESC_CNT)
-#define SW_TXBD_RING_SIZE (sizeof(struct sw_bd) * TX_DESC_CNT)
+#define SW_TXBD_RING_SIZE (sizeof(struct sw_tx_bd) * TX_DESC_CNT)
#define TXBD_RING_SIZE (sizeof(struct tx_bd) * TX_DESC_CNT)
/* Buffered flash (Atmel: AT45DB011B) specific information */
@@ -6609,7 +6613,7 @@ struct bnx2_tx_ring_info {
u32 tx_bseq_addr;
struct tx_bd *tx_desc_ring;
- struct sw_bd *tx_buf_ring;
+ struct sw_tx_bd *tx_buf_ring;
u16 tx_cons;
u16 hw_tx_cons;
--
1.5.6.GIT
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists