Hello,
sorry. Your mail got flagged by google mail as spam. :(
On 2020-08-02 15:29, Jia-Ju Bai wrote:
In p54p_tx(), skb->data is mapped to streaming DMA on line 337:
mapping = pci_map_single(..., skb->data, ...);
Then skb->data is accessed on line 349:
desc->device_addr = ((struct p54_hdr *)skb->data)->req_id;
This access may cause data inconsistency between CPU cache and hardware.
To fix this problem, ((struct p54_hdr *)skb->data)->req_id is stored in
a local variable before DMA mapping, and then the driver accesses this
local variable instead of skb->data.
Interesting. Please bear with me here. From my understanding, the
streaming direction is set to PCI_DMA_TODEVICE. So is it really possible
for the hardware to interfere with the data without the IOMMU catching this?
(That said, patch looks be fine. I'll need to dust off a old PCI PC to
check this with real hardware, if requested.)
Cheers,
Christian
Signed-off-by: Jia-Ju Bai <baijiaju@xxxxxxxxxxxxxxx>
---
drivers/net/wireless/intersil/p54/p54pci.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/wireless/intersil/p54/p54pci.c b/drivers/net/wireless/intersil/p54/p54pci.c
index 80ad0b7eaef4..f8c6027cab6b 100644
--- a/drivers/net/wireless/intersil/p54/p54pci.c
+++ b/drivers/net/wireless/intersil/p54/p54pci.c
@@ -329,10 +329,12 @@ static void p54p_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
struct p54p_desc *desc;
dma_addr_t mapping;
u32 idx, i;
+ __le32 device_addr;
spin_lock_irqsave(&priv->lock, flags);
idx = le32_to_cpu(ring_control->host_idx[1]);
i = idx % ARRAY_SIZE(ring_control->tx_data);
+ device_addr = ((struct p54_hdr *)skb->data)->req_id;
mapping = pci_map_single(priv->pdev, skb->data, skb->len,
PCI_DMA_TODEVICE);
@@ -346,7 +348,7 @@ static void p54p_tx(struct ieee80211_hw *dev, struct sk_buff *skb)
desc = &ring_control->tx_data[i];
desc->host_addr = cpu_to_le32(mapping);
- desc->device_addr = ((struct p54_hdr *)skb->data)->req_id;
+ desc->device_addr = device_addr;
desc->len = cpu_to_le16(skb->len);
desc->flags = 0;