From: Paul Burton <paul.burton@mips.com>
To: <netdev@vger.kernel.org>
Cc: Hassan Naveed <hassan.naveed@mips.com>,
Matt Redfearn <matt.redfearn@mips.com>,
"David S . Miller" <davem@davemloft.net>,
<linux-mips@linux-mips.org>, Paul Burton <paul.burton@mips.com>
Subject: [PATCH v5 10/14] net: pch_gbe: Disable TX DMA whilst configuring descriptors
Date: Sat, 17 Feb 2018 12:10:33 -0800 [thread overview]
Message-ID: <20180217201037.3006-11-paul.burton@mips.com> (raw)
In-Reply-To: <20180217201037.3006-1-paul.burton@mips.com>
The pch_gbe driver enables TX DMA the first time we call
pch_gbe_configure_tx() and never disables it again, even if we
reconfigure the device & modify the transmit descriptor ring. This seems
unsafe, since the device may continue accessing descriptors whilst they
are in an unpredictable & possibly invalid state - especially on systems
where the CPUs writes to the descriptors is not coherent with DMA.
In the RX path pch_gbe_configure_rx() disables DMA before configuring
the descriptor pointers & before we set up the descriptors, then
pch_gbe_up() calls pch_gbe_enable_dma_rx() to enable DMA again after the
descriptors have been configured. Here we copy that same scheme for the
TX path - pch_gbe_configure_tx() calls pch_gbe_disable_dma_tx() to
disable DMA, and then after the descriptors have been configured
pch_gbe_up() calls pch_gbe_enable_dma_tx() to enable DMA. This should
ensure that the device doesn't begin reading descriptors before we have
configured them.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: linux-mips@linux-mips.org
Cc: netdev@vger.kernel.org
---
Changes in v5:
- New patch.
Changes in v4: None
Changes in v3: None
Changes in v2: None
.../net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c | 29 +++++++++++++++++-----
1 file changed, 23 insertions(+), 6 deletions(-)
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index b6cc4a34ed89..4354842b9b7e 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -851,6 +851,24 @@ static void pch_gbe_enable_dma_rx(struct pch_gbe_hw *hw)
iowrite32(rxdma, &hw->reg->DMA_CTRL);
}
+static void pch_gbe_disable_dma_tx(struct pch_gbe_hw *hw)
+{
+ u32 rxdma;
+
+ rxdma = ioread32(&hw->reg->DMA_CTRL);
+ rxdma &= ~PCH_GBE_TX_DMA_EN;
+ iowrite32(rxdma, &hw->reg->DMA_CTRL);
+}
+
+static void pch_gbe_enable_dma_tx(struct pch_gbe_hw *hw)
+{
+ u32 rxdma;
+
+ rxdma = ioread32(&hw->reg->DMA_CTRL);
+ rxdma |= PCH_GBE_TX_DMA_EN;
+ iowrite32(rxdma, &hw->reg->DMA_CTRL);
+}
+
/**
* pch_gbe_configure_tx - Configure Transmit Unit after Reset
* @adapter: Board private structure
@@ -858,7 +876,7 @@ static void pch_gbe_enable_dma_rx(struct pch_gbe_hw *hw)
static void pch_gbe_configure_tx(struct pch_gbe_adapter *adapter)
{
struct pch_gbe_hw *hw = &adapter->hw;
- u32 tdba, tdlen, dctrl, tx_mode, tcpip;
+ u32 tdba, tdlen, tx_mode, tcpip;
tx_mode = PCH_GBE_TM_LONG_PKT |
PCH_GBE_TM_ST_AND_FD |
@@ -876,17 +894,14 @@ static void pch_gbe_configure_tx(struct pch_gbe_adapter *adapter)
(unsigned long long)adapter->tx_ring->dma,
adapter->tx_ring->size);
+ pch_gbe_disable_dma_tx(hw);
+
/* Setup the HW Tx Head and Tail descriptor pointers */
tdba = adapter->tx_ring->dma;
tdlen = adapter->tx_ring->size - 0x10;
iowrite32(tdba, &hw->reg->TX_DSC_BASE);
iowrite32(tdlen, &hw->reg->TX_DSC_SIZE);
iowrite32(tdba, &hw->reg->TX_DSC_SW_P);
-
- /* Enables Transmission DMA */
- dctrl = ioread32(&hw->reg->DMA_CTRL);
- dctrl |= PCH_GBE_TX_DMA_EN;
- iowrite32(dctrl, &hw->reg->DMA_CTRL);
}
/**
@@ -1945,6 +1960,8 @@ int pch_gbe_up(struct pch_gbe_adapter *adapter)
pch_gbe_alloc_tx_buffers(adapter, tx_ring);
pch_gbe_alloc_rx_buffers(adapter, rx_ring, rx_ring->count);
adapter->tx_queue_len = netdev->tx_queue_len;
+
+ pch_gbe_enable_dma_tx(&adapter->hw);
pch_gbe_enable_dma_rx(&adapter->hw);
pch_gbe_enable_mac_rx(&adapter->hw);
--
2.16.1
next prev parent reply other threads:[~2018-02-17 20:14 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-17 20:10 [PATCH v5 00/14] net: pch_gbe: Fixes & MIPS support Paul Burton
2018-02-17 20:10 ` [PATCH v5 01/14] net: pch_gbe: Mark Minnow PHY reset GPIO active low Paul Burton
2018-02-17 22:14 ` Andrew Lunn
2018-02-17 20:10 ` [PATCH v5 02/14] net: pch_gbe: Pull PHY GPIO handling out of Minnow code Paul Burton
2018-02-17 22:29 ` Andrew Lunn
2018-02-17 22:50 ` Paul Burton
2018-02-17 23:34 ` Andrew Lunn
2018-02-18 15:51 ` Paul Burton
2018-02-18 16:14 ` Andrew Lunn
2018-02-17 20:10 ` [PATCH v5 03/14] dt-bindings: net: Document Intel pch_gbe binding Paul Burton
[not found] ` <20180217201037.3006-4-paul.burton-8NJIiSa5LzA@public.gmane.org>
2018-02-17 22:32 ` Andrew Lunn
2018-02-17 20:10 ` [PATCH v5 04/14] net: pch_gbe: Add device tree support Paul Burton
2018-02-17 20:10 ` [PATCH v5 05/14] net: pch_gbe: Always reset PHY along with MAC Paul Burton
2018-02-17 20:10 ` [PATCH v5 06/14] net: pch_gbe: Allow longer for resets Paul Burton
2018-02-18 15:29 ` kbuild test robot
2018-02-17 20:10 ` [PATCH v5 07/14] net: pch_gbe: Fix handling of TX padding Paul Burton
2018-02-19 14:01 ` David Laight
2018-02-19 16:42 ` Paul Burton
2018-02-19 16:41 ` David Miller
2018-02-17 20:10 ` [PATCH v5 08/14] net: pch_gbe: Fold pch_gbe_setup_[rt]ctl into pch_gbe_configure_[rt]x Paul Burton
2018-02-17 20:10 ` [PATCH v5 09/14] net: pch_gbe: Use pch_gbe_disable_dma_rx() in pch_gbe_configure_rx() Paul Burton
2018-02-17 20:10 ` Paul Burton [this message]
2018-02-17 20:10 ` [PATCH v5 11/14] net: pch_gbe: Ensure DMA is ordered with descriptor writes Paul Burton
2018-02-17 20:10 ` [PATCH v5 12/14] net: pch_gbe: Fix TX RX descriptor accesses for big endian systems Paul Burton
2018-02-18 16:01 ` kbuild test robot
2018-02-17 20:10 ` [PATCH v5 13/14] ptp: pch: Allow build on MIPS platforms Paul Burton
2018-02-17 20:10 ` [PATCH v5 14/14] net: pch_gbe: " Paul Burton
2018-02-18 15:31 ` [PATCH v5 00/14] net: pch_gbe: Fixes & MIPS support David Miller
2018-02-18 17:03 ` Paul Burton
2018-02-18 17:56 ` Andrew Lunn
2018-02-18 22:09 ` Paul Burton
2018-02-19 1:15 ` David Miller
2018-02-19 2:17 ` Florian Fainelli
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180217201037.3006-11-paul.burton@mips.com \
--to=paul.burton@mips.com \
--cc=davem@davemloft.net \
--cc=hassan.naveed@mips.com \
--cc=linux-mips@linux-mips.org \
--cc=matt.redfearn@mips.com \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).