From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Maciej W. Rozycki" Subject: [PATCH 5/5] defxx: Add missing DMA synchronisation calls Date: Sat, 5 Jul 2014 15:14:46 +0100 (BST) Message-ID: References: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Cc: Robert Coerver To: netdev@vger.kernel.org Return-path: Received: from eddie.linux-mips.org ([78.24.191.182]:59876 "EHLO cvs.linux-mips.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932098AbaGEOOr (ORCPT ); Sat, 5 Jul 2014 10:14:47 -0400 Received: from localhost.localdomain ([127.0.0.1]:48345 "EHLO localhost.localdomain" rhost-flags-OK-OK-OK-OK) by eddie.linux-mips.org with ESMTP id S6859971AbaGEOOqQNnoC (ORCPT ); Sat, 5 Jul 2014 16:14:46 +0200 In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: This adds DMA synchronisation calls needed in the receive path: 1. To retrieve the Receive Status word that is prepended by the PDQ DMA engine in the receive buffer, and provides information about the frame received, including its size and any errors. 2. To make data received available for copying in the small-frame case (size <= SKBUFF_RX_COPYBREAK) where the original DMA buffer will be returned to the receive descriptor ring and therefore its mapping retained. With DMA mapping error handling in place, added by the other patch, this may now also trigger where an attempt to map a newly allocated buffer for DMA has failed. In that case data from the original buffer will be copied out and the buffer returned to the DMA descriptor ring. These calls may do nothing when data is in the host DMA addressing range of the FDDI interface, such as always on 32-bit systems, however their absence makes frame reception stop functioning reliably on systems that have memory beyond the low 4GB of the address space. Reported-by: Robert Coerver Tested-by: Robert Coerver Signed-off-by: Maciej W. Rozycki --- Please apply, Maciej linux-defxx-dma-map-sync.patch Index: linux-20140623-swarm64-eb/drivers/net/fddi/defxx.c =================================================================== --- linux-20140623-swarm64-eb.orig/drivers/net/fddi/defxx.c +++ linux-20140623-swarm64-eb/drivers/net/fddi/defxx.c @@ -196,6 +196,7 @@ * 14 Jun 2005 macro Use irqreturn_t. * 23 Oct 2006 macro Big-endian host support. * 14 Dec 2006 macro TURBOchannel support. + * 01 Jul 2014 macro Fixes for DMA on 64-bit hosts. */ /* Include files */ @@ -224,8 +225,8 @@ /* Version information string should be updated prior to each new release! */ #define DRV_NAME "defxx" -#define DRV_VERSION "v1.10" -#define DRV_RELDATE "2006/12/14" +#define DRV_VERSION "v1.11" +#define DRV_RELDATE "2014/07/01" static char version[] = DRV_NAME ": " DRV_VERSION " " DRV_RELDATE @@ -3030,7 +3031,7 @@ static void dfx_rcv_queue_process( while (bp->rcv_xmt_reg.index.rcv_comp != p_type_2_cons->index.rcv_cons) { /* Process any errors */ - + dma_addr_t dma_addr; int entry; entry = bp->rcv_xmt_reg.index.rcv_comp; @@ -3039,6 +3040,11 @@ static void dfx_rcv_queue_process( #else p_buff = bp->p_rcv_buff_va[entry]; #endif + dma_addr = bp->descr_block_virt->rcv_data[entry].long_1; + dma_sync_single_for_cpu(bp->bus_dev, + dma_addr + RCV_BUFF_K_DESCR, + sizeof(u32), + DMA_FROM_DEVICE); memcpy(&descr, p_buff + RCV_BUFF_K_DESCR, sizeof(u32)); if (descr & PI_FMC_DESCR_M_RCC_FLUSH) @@ -3086,7 +3092,7 @@ static void dfx_rcv_queue_process( skb = (struct sk_buff *)bp->p_rcv_buff_va[entry]; dma_unmap_single(bp->bus_dev, - bp->descr_block_virt->rcv_data[entry].long_1, + dma_addr, PI_RCV_DATA_K_SIZE_MAX, DMA_FROM_DEVICE); skb_reserve(skb, RCV_BUFF_K_PADDING); @@ -3109,6 +3115,12 @@ static void dfx_rcv_queue_process( else { if (!rx_in_place) { /* Receive buffer allocated, pass receive packet up */ + dma_sync_single_for_cpu( + bp->bus_dev, + dma_addr + + RCV_BUFF_K_PADDING, + pkt_len + 3, + DMA_FROM_DEVICE); skb_copy_to_linear_data(skb, p_buff + RCV_BUFF_K_PADDING,