| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Thu, 14 Mar 2019 15:38:23 +0100
From: Wolfgang Grandegger <wg@...ndegger.com>
To: Dan Murphy <dmurphy@...com>, mkl@...gutronix.de,
davem@...emloft.net
Cc: linux-can@...r.kernel.org, netdev@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v9 1/4] can: m_can: Create a m_can platform framework
Hello,
Am 14.03.19 um 15:04 schrieb Dan Murphy:
> Hello
>
> On 3/14/19 2:26 AM, Wolfgang Grandegger wrote:
>> Hello Dan,
>>
>> we are close...
>>
>> Am 13.03.19 um 17:26 schrieb Dan Murphy:
>>> Create a m_can platform framework that peripherial
>>> devices can register to and use common code and register sets.
>>> The peripherial devices may provide read/write and configuration
>>> support of the IP.
>>>
>>> Signed-off-by: Dan Murphy <dmurphy@...com>
>>> ---
>>>
>>> v9 - Added back the CSR clearing as this is needed for TCAN device, fixed clean
>>> function to clean index for version > 30, removed extra skb free and fixed work
>>> handler to handle FIFO full for perpheral devices - https://lore.kernel.org/patchwork/patch/1050120/
>>>
>>> v8 - Added a clean function for the tx_skb, cleaned up skb on BUS_OFF and on xmit
>>> BUSY - https://lore.kernel.org/patchwork/patch/1047980/
>>> v7 - Fixed remaining new checkpatch issues, removed CSR setting, fixed tx hard
>>> start function to return tx_busy, and renamed device callbacks - https://lore.kernel.org/patchwork/patch/1047220/
>>> v6 - Squashed platform patch to this patch for bissectablity, fixed coding style
>>> issues, updated Kconfig help, placed mcan reg offsets back into c file, renamed
>>> priv->skb to priv->tx_skb and cleared perp interrupts at ISR start -
>>> Patch 1 comments - https://lore.kernel.org/patchwork/patch/1042446/
>>> Patch 2 comments - https://lore.kernel.org/patchwork/patch/1042442/
>>>
>>> drivers/net/can/m_can/Kconfig | 13 +-
>>> drivers/net/can/m_can/Makefile | 1 +
>>> drivers/net/can/m_can/m_can.c | 733 +++++++++++++------------
>>> drivers/net/can/m_can/m_can.h | 110 ++++
>>> drivers/net/can/m_can/m_can_platform.c | 202 +++++++
>>> 5 files changed, 714 insertions(+), 345 deletions(-)
>>> create mode 100644 drivers/net/can/m_can/m_can.h
>>> create mode 100644 drivers/net/can/m_can/m_can_platform.c
>>>
>>> diff --git a/drivers/net/can/m_can/Kconfig b/drivers/net/can/m_can/Kconfig
>>> index 04f20dd39007..f7119fd72df4 100644
>>> --- a/drivers/net/can/m_can/Kconfig
>>> +++ b/drivers/net/can/m_can/Kconfig
>>> @@ -1,5 +1,14 @@
>>> config CAN_M_CAN
>>> + tristate "Bosch M_CAN support"
>>> + ---help---
>>> + Say Y here if you want support for Bosch M_CAN controller framework.
>>> + This is common support for devices that embed the Bosch M_CAN IP.
>>> +
>>> +config CAN_M_CAN_PLATFORM
>>> + tristate "Bosch M_CAN support for io-mapped devices"
>>> depends on HAS_IOMEM
>>> - tristate "Bosch M_CAN devices"
>>> + depends on CAN_M_CAN
>>> ---help---
>>> - Say Y here if you want to support for Bosch M_CAN controller.
>>> + Say Y here if you want support for IO Mapped Bosch M_CAN controller.
>>> + This support is for devices that have the Bosch M_CAN controller
>>> + IP embedded into the device and the IP is IO Mapped to the processor.
>>> diff --git a/drivers/net/can/m_can/Makefile b/drivers/net/can/m_can/Makefile
>>> index 8bbd7f24f5be..057bbcdb3c74 100644
>>> --- a/drivers/net/can/m_can/Makefile
>>> +++ b/drivers/net/can/m_can/Makefile
>>> @@ -3,3 +3,4 @@
>>> #
>>>
>>> obj-$(CONFIG_CAN_M_CAN) += m_can.o
>>> +obj-$(CONFIG_CAN_M_CAN_PLATFORM) += m_can_platform.o
>>> diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
>>> index 9b449400376b..56336aefc567 100644
>>> --- a/drivers/net/can/m_can/m_can.c
>>> +++ b/drivers/net/can/m_can/m_can.c
... snip ...
>>> +static netdev_tx_t m_can_start_xmit(struct sk_buff *skb,
>>> + struct net_device *dev)
>>> +{
>>> + struct m_can_priv *priv = netdev_priv(dev);
>>> +
>>> + if (can_dropped_invalid_skb(dev, skb))
>>> + return NETDEV_TX_OK;
>>> +
>>> + if (priv->is_peripherial) {
>>> + if (priv->tx_skb) {
>>> + netdev_err(dev, "hard_xmit called while tx busy\n");
>>> + return NETDEV_TX_BUSY;
>>> + }
>>> +
>>> + if (priv->can.state == CAN_STATE_BUS_OFF) {
>>> + m_can_clean(dev);
>>> + } else {
>>> + priv->tx_skb = skb;
>>> + queue_work(priv->tx_wq, &priv->tx_work);
>>
>> You removed "netif_stop_queue(cdev->net)" here. Does that work with just
>> *one* "tx_skb"? I think you need an array then!? I'm afraid, this will
>> require quite some changes and you should also measure the benefit of
>> queuing outgoing messages first. For the time being, it would be fine
>> for me to stop the queue here and add an appropriate comment.
>>
>
> Yes I removed this in favor of allowing the work handler to stop the queue.
>
> If the version is < 30 then the queue is stopped immediately. If the version > 30 then the queue is
> only stopped if FIFO is full.
Yes, and the next call to m_can_start_xmit() will probably return with
NETDEV_TX_BUSY. You should at least add a "netif_stop_queue(cdev->net)"
before returning NETDEV_TX_BUSY, otherwice the upper layer will retry
immediately resulting in high CPU load. Anyway, the purpose of this
return code is to signal errors/bugs in the start/stop flow control:
https://elixir.bootlin.com/linux/latest/source/Documentation/networking/netdevices.txt#L81
>
> I did see the message "tcan4x5x spi1.0 can0: hard_xmit called while tx busy"
> more with no gap and no polling in the cangen request.
You usually should never see that message.
>
>> BTW, how big is the FIFO size on your system:
>>
>
> This value is set to 1.
>
>> tx_fifo_size = mram_config_vals[7];
Then I'm surprised that you see "hard_xmit called while tx busy" at all.
Are you sure that "cdev->version" is 32 on the tcan4x5?
Wolfgang
Powered by blists - more mailing lists