From: Thomas Falcon Date: Wed, 14 Aug 2019 19:57:05 +0000 (-0500) Subject: ibmvnic: Unmap DMA address of TX descriptor buffers after use X-Git-Tag: v4.19.72~24 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=ea78dc8b5e667d31c539192f54e214554c2a1b31;p=platform%2Fkernel%2Flinux-rpi.git ibmvnic: Unmap DMA address of TX descriptor buffers after use [ Upstream commit 80f0fe0934cd3daa13a5e4d48a103f469115b160 ] There's no need to wait until a completion is received to unmap TX descriptor buffers that have been passed to the hypervisor. Instead unmap it when the hypervisor call has completed. This patch avoids the possibility that a buffer will not be unmapped because a TX completion is lost or mishandled. Reported-by: Abdul Haleem Tested-by: Devesh K. Singh Signed-off-by: Thomas Falcon Signed-off-by: David S. Miller Signed-off-by: Sasha Levin --- diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c index 0ae43d2..255de7d 100644 --- a/drivers/net/ethernet/ibm/ibmvnic.c +++ b/drivers/net/ethernet/ibm/ibmvnic.c @@ -1586,6 +1586,8 @@ static int ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev) lpar_rc = send_subcrq_indirect(adapter, handle_array[queue_num], (u64)tx_buff->indir_dma, (u64)num_entries); + dma_unmap_single(dev, tx_buff->indir_dma, + sizeof(tx_buff->indir_arr), DMA_TO_DEVICE); } else { tx_buff->num_entries = num_entries; lpar_rc = send_subcrq(adapter, handle_array[queue_num], @@ -2747,7 +2749,6 @@ static int ibmvnic_complete_tx(struct ibmvnic_adapter *adapter, union sub_crq *next; int index; int i, j; - u8 *first; restart_loop: while (pending_scrq(adapter, scrq)) { @@ -2777,14 +2778,6 @@ restart_loop: txbuff->data_dma[j] = 0; } - /* if sub_crq was sent indirectly */ - first = &txbuff->indir_arr[0].generic.first; - if (*first == IBMVNIC_CRQ_CMD) { - dma_unmap_single(dev, txbuff->indir_dma, - sizeof(txbuff->indir_arr), - DMA_TO_DEVICE); - *first = 0; - } if (txbuff->last_frag) { dev_kfree_skb_any(txbuff->skb);