On Thu, May 17, 2018 at 03:26:25PM -0700, Adrian Chadd wrote:

On Thu, 17 May 2018 at 16:16, Niklas Cassel [off-list ref]
wrote:

quoted

diff --git a/drivers/net/wireless/ath/ath10k/txrx.c

b/drivers/net/wireless/ath/ath10k/txrx.c

quoted

index cda164f6e9f6..1d3b2d2c3fee 100644

--- a/drivers/net/wireless/ath/ath10k/txrx.c
+++ b/drivers/net/wireless/ath/ath10k/txrx.c

@@ -95,6 +95,9 @@ int ath10k_txrx_tx_unref(struct ath10k_htt *htt,
                 wake_up(&htt->empty_tx_wq);
         spin_unlock_bh(&htt->tx_lock);

quoted

+       if (htt->num_pending_tx <= 3 && !list_empty(&ar->txqs))
+               ath10k_mac_tx_push_pending(ar);
+

Just sanity checking - what's protecting htt->num_pending_tx? or is it
serialised some other way?

Hello Adrian,

I did not take the htt->tx_lock lock for this since it should not be
needed.

There are however some compiler optimizations that you have to look out
for:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/memory-barriers.txt?h=v4.17-rc6#n1547

I can't see that any of the examples applies, but let's add READ_ONCE(),
to make sure that the compiler doesn't try to optimize this.

Will send a V2 for this.

Regards,
Niklas

quoted

         dma_unmap_single(dev, skb_cb->paddr, msdu->len, DMA_TO_DEVICE);

quoted

         ath10k_report_offchan_tx(htt->ar, msdu);
--
2.17.0