mbox series

[V4,0/3] patch set for flexcan

Message ID 20200929203041.29758-1-qiangqing.zhang@nxp.com
Headers show
Series patch set for flexcan | expand

Message

Joakim Zhang Sept. 29, 2020, 8:30 p.m. UTC
Add Flexcan ECC support.

Joakim Zhang (3):
  can: flexcan: initialize all flexcan memory for ECC function
  can: flexcan: add flexcan driver for i.MX8MP
  can: flexcan: disable runtime PM if register flexcandev failed

 drivers/net/can/flexcan.c | 62 +++++++++++++++++++++++++++++++++++++--
 1 file changed, 60 insertions(+), 2 deletions(-)

Comments

Marc Kleine-Budde Sept. 29, 2020, 12:44 p.m. UTC | #1
On 9/29/20 2:34 PM, Marc Kleine-Budde wrote:
> On 9/29/20 10:30 PM, Joakim Zhang wrote:
>> One issue was reported at a baremetal environment, which is used for
>> FPGA verification. "The first transfer will fail for extended ID
>> format(for both 2.0B and FD format), following frames can be transmitted
>> and received successfully for extended format, and standard format don't
>> have this issue. This issue occurred randomly with high possiblity, when
>> it occurs, the transmitter will detect a BIT1 error, the receiver a CRC
>> error. According to the spec, a non-correctable error may cause this
>> transfer failure."
>>
>> With FLEXCAN_QUIRK_DISABLE_MECR quirk, it supports correctable errors,
>> disable non-correctable errors interrupt and freeze mode. Platform has
>> ECC hardware support, but select this quirk, this issue may not come to
>> light. Initialize all FlexCAN memory before accessing them, at least it
>> can avoid non-correctable errors detected due to memory uninitialized.
>> The internal region can't be initialized when the hardware doesn't support
>> ECC.
>>
>> According to IMX8MPRM, Rev.C, 04/2020. There is a NOTE at the section
>> 11.8.3.13 Detection and correction of memory errors:
>> "All FlexCAN memory must be initialized before starting its operation in
>> order to have the parity bits in memory properly updated. CTRL2[WRMFRZ]
>> grants write access to all memory positions that require initialization,
>> ranging from 0x080 to 0xADF and from 0xF28 to 0xFFF when the CAN FD feature
>> is enabled. The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK registers need to
>> be initialized as well. MCR[RFEN] must not be set during memory initialization."
>>
>> Memory range from 0x080 to 0xADF, there are reserved memory (unimplemented
>> by hardware, e.g. only configure 64 MBs), these memory can be initialized or not.
>> In this patch, initialize all flexcan memory which includes reserved memory.
>>
>> In this patch, create FLEXCAN_QUIRK_SUPPORT_ECC for platforms which has ECC
>> feature. If you have a ECC platform in your hand, please select this
>> qurik to initialize all flexcan memory firstly, then you can select
>> FLEXCAN_QUIRK_DISABLE_MECR to only enable correctable errors.
>>
>> Signed-off-by: Joakim Zhang <qiangqing.zhang@nxp.com>
>> ---
>> ChangeLogs:
>> V1->V2:
>> 	* update commit messages, add a datasheet reference.
>> 	* initialize block memory instead of trivial memory.
>> 	* inilialize reserved memory.
>> V2->V3:
>> 	* add FLEXCAN_QUIRK_SUPPORT_ECC quirk.
>> 	* remove init_ram struct.
>> V3->V4:
>> 	* move register definition into flexcan_reg.
>> ---
>>  drivers/net/can/flexcan.c | 51 +++++++++++++++++++++++++++++++++++++--
>>  1 file changed, 49 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c
>> index e86925134009..ede25db42e87 100644
>> --- a/drivers/net/can/flexcan.c
>> +++ b/drivers/net/can/flexcan.c
>> @@ -239,6 +239,8 @@
>>  #define FLEXCAN_QUIRK_SETUP_STOP_MODE BIT(8)
>>  /* Support CAN-FD mode */
>>  #define FLEXCAN_QUIRK_SUPPORT_FD BIT(9)
>> +/* support memory detection and correction */
>> +#define FLEXCAN_QUIRK_SUPPORT_ECC BIT(10)
>>  
>>  /* Structure of the message buffer */
>>  struct flexcan_mb {
>> @@ -292,7 +294,16 @@ struct flexcan_regs {
>>  	u32 rximr[64];		/* 0x880 - Not affected by Soft Reset */
>>  	u32 _reserved5[24];	/* 0x980 */
>>  	u32 gfwr_mx6;		/* 0x9e0 - MX6 */
>> -	u32 _reserved6[63];	/* 0x9e4 */
>> +	u32 _reserved6[39];	/* 0x9e4 */
>> +	u32 _rxfir[6];		/* 0xa80 */
>> +	u32 _reserved8[2];	/* 0xa98 */
>> +	u32 _rxmgmask;		/* 0xaa0 */
>> +	u32 _rxfgmask;		/* 0xaa4 */
>> +	u32 _rx14mask;		/* 0xaa8 */
>> +	u32 _rx15mask;		/* 0xaac */
>> +	u32 tx_smb[4];		/* 0xab0 */
>> +	u32 rx_smb0[4];		/* 0xac0 */
>> +	u32 rx_smb1[4];		/* 0xad0 */
>>  	u32 mecr;		/* 0xae0 */
>>  	u32 erriar;		/* 0xae4 */
>>  	u32 erridpr;		/* 0xae8 */
>> @@ -305,9 +316,13 @@ struct flexcan_regs {
>>  	u32 fdctrl;		/* 0xc00 - Not affected by Soft Reset */
>>  	u32 fdcbt;		/* 0xc04 - Not affected by Soft Reset */
>>  	u32 fdcrc;		/* 0xc08 */
>> +	u32 _reserved9[199];	/* 0xc0c */
>> +	u32 tx_smb_fd[18];	/* 0xf28 */
>> +	u32 rx_smb0_fd[18];	/* 0xf70 */
>> +	u32 rx_smb1_fd[18];	/* 0xfb8 */
>>  };
>>  
>> -static_assert(sizeof(struct flexcan_regs) == 0x4 + 0xc08);
>> +static_assert(sizeof(struct flexcan_regs) ==  0x4 * 18 + 0xfb8);
>>  
>>  struct flexcan_devtype_data {
>>  	u32 quirks;		/* quirks needed for different IP cores */
>> @@ -1292,6 +1307,35 @@ static void flexcan_set_bittiming(struct net_device *dev)
>>  		return flexcan_set_bittiming_ctrl(dev);
>>  }
>>  
>> +static void flexcan_init_ram(struct net_device *dev)
>> +{
>> +	struct flexcan_priv *priv = netdev_priv(dev);
>> +	struct flexcan_regs __iomem *regs = priv->regs;
>> +	u32 reg_ctrl2;
>> +
>> +	/* 11.8.3.13 Detection and correction of memory errors:
>> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that
>> +	 * require initialization, ranging from 0x080 to 0xADF and
>> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.
>> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK registers need to
>> +	 * be initialized as well. MCR[RFEN] must not be set during memory
>> +	 * initialization.
>> +	 */
>> +	reg_ctrl2 = priv->read(&regs->ctrl2);
>> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;
>> +	priv->write(reg_ctrl2, &regs->ctrl2);
>> +
>> +	memset_io(&regs->mb[0][0], 0,
>> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);
> 
> why the cast?

Doh. Of cource you need to case, but you should better cast both. :) ... or use
offsetof()

Marc
Marc Kleine-Budde Sept. 29, 2020, 12:46 p.m. UTC | #2
On 9/29/20 2:38 PM, Joakim Zhang wrote:
>>>  		return flexcan_set_bittiming_ctrl(dev);  }
>>>
>>> +static void flexcan_init_ram(struct net_device *dev) {
>>> +	struct flexcan_priv *priv = netdev_priv(dev);
>>> +	struct flexcan_regs __iomem *regs = priv->regs;
>>> +	u32 reg_ctrl2;
>>> +
>>> +	/* 11.8.3.13 Detection and correction of memory errors:
>>> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that
>>> +	 * require initialization, ranging from 0x080 to 0xADF and
>>> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.
>>> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK registers
>> need to
>>> +	 * be initialized as well. MCR[RFEN] must not be set during memory
>>> +	 * initialization.
>>> +	 */
>>> +	reg_ctrl2 = priv->read(&regs->ctrl2);
>>> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;
>>> +	priv->write(reg_ctrl2, &regs->ctrl2);
>>> +
>>> +	memset_io(&regs->mb[0][0], 0,
>>> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);
>>
>> why the cast?
> 
> Due to mb is defined as a u8. And the count of memset_io is bytes.

right. this is why you don't need the 2nd cast.

Marc
Joakim Zhang Sept. 29, 2020, 12:50 p.m. UTC | #3
> -----Original Message-----

> From: Marc Kleine-Budde <mkl@pengutronix.de>

> Sent: 2020年9月29日 20:46

> To: Joakim Zhang <qiangqing.zhang@nxp.com>; linux-can@vger.kernel.org

> Cc: netdev@vger.kernel.org; dl-linux-imx <linux-imx@nxp.com>

> Subject: Re: [PATCH V4 1/3] can: flexcan: initialize all flexcan memory for ECC

> function

> 

> On 9/29/20 2:38 PM, Joakim Zhang wrote:

> >>>  		return flexcan_set_bittiming_ctrl(dev);  }

> >>>

> >>> +static void flexcan_init_ram(struct net_device *dev) {

> >>> +	struct flexcan_priv *priv = netdev_priv(dev);

> >>> +	struct flexcan_regs __iomem *regs = priv->regs;

> >>> +	u32 reg_ctrl2;

> >>> +

> >>> +	/* 11.8.3.13 Detection and correction of memory errors:

> >>> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that

> >>> +	 * require initialization, ranging from 0x080 to 0xADF and

> >>> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.

> >>> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK

> registers

> >> need to

> >>> +	 * be initialized as well. MCR[RFEN] must not be set during memory

> >>> +	 * initialization.

> >>> +	 */

> >>> +	reg_ctrl2 = priv->read(&regs->ctrl2);

> >>> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;

> >>> +	priv->write(reg_ctrl2, &regs->ctrl2);

> >>> +

> >>> +	memset_io(&regs->mb[0][0], 0,

> >>> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);

> >>

> >> why the cast?

> >

> > Due to mb is defined as a u8. And the count of memset_io is bytes.

> 

> right. this is why you don't need the 2nd cast.


Yes, do you need me send a patch to improve it with offsetof()?

Best Regards,
Joakim Zhang
> Marc

> 

> 

> --

> Pengutronix e.K.                 | Marc Kleine-Budde           |

> Embedded Linux                   | https://www.pengutronix.de  |

> Vertretung West/Dortmund         | Phone: +49-231-2826-924     |

> Amtsgericht Hildesheim, HRA 2686 | Fax:   +49-5121-206917-5555 |
Marc Kleine-Budde Sept. 29, 2020, 12:51 p.m. UTC | #4
On 9/29/20 2:50 PM, Joakim Zhang wrote:
> 
>> -----Original Message-----
>> From: Marc Kleine-Budde <mkl@pengutronix.de>
>> Sent: 2020年9月29日 20:46
>> To: Joakim Zhang <qiangqing.zhang@nxp.com>; linux-can@vger.kernel.org
>> Cc: netdev@vger.kernel.org; dl-linux-imx <linux-imx@nxp.com>
>> Subject: Re: [PATCH V4 1/3] can: flexcan: initialize all flexcan memory for ECC
>> function
>>
>> On 9/29/20 2:38 PM, Joakim Zhang wrote:
>>>>>  		return flexcan_set_bittiming_ctrl(dev);  }
>>>>>
>>>>> +static void flexcan_init_ram(struct net_device *dev) {
>>>>> +	struct flexcan_priv *priv = netdev_priv(dev);
>>>>> +	struct flexcan_regs __iomem *regs = priv->regs;
>>>>> +	u32 reg_ctrl2;
>>>>> +
>>>>> +	/* 11.8.3.13 Detection and correction of memory errors:
>>>>> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that
>>>>> +	 * require initialization, ranging from 0x080 to 0xADF and
>>>>> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.
>>>>> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK
>> registers
>>>> need to
>>>>> +	 * be initialized as well. MCR[RFEN] must not be set during memory
>>>>> +	 * initialization.
>>>>> +	 */
>>>>> +	reg_ctrl2 = priv->read(&regs->ctrl2);
>>>>> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;
>>>>> +	priv->write(reg_ctrl2, &regs->ctrl2);
>>>>> +
>>>>> +	memset_io(&regs->mb[0][0], 0,
>>>>> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);
>>>>
>>>> why the cast?
>>>
>>> Due to mb is defined as a u8. And the count of memset_io is bytes.
>>
>> right. this is why you don't need the 2nd cast.
> 
> Yes, do you need me send a patch to improve it with offsetof()?

If it looks better :), then yes :)

Marc
Jakub Kicinski Sept. 29, 2020, 2:48 p.m. UTC | #5
On Wed, 30 Sep 2020 04:30:39 +0800 Joakim Zhang wrote:
> @@ -1292,6 +1307,35 @@ static void flexcan_set_bittiming(struct net_device *dev)
>  		return flexcan_set_bittiming_ctrl(dev);
>  }
>  
> +static void flexcan_init_ram(struct net_device *dev)
> +{
> +	struct flexcan_priv *priv = netdev_priv(dev);
> +	struct flexcan_regs __iomem *regs = priv->regs;
> +	u32 reg_ctrl2;
> +
> +	/* 11.8.3.13 Detection and correction of memory errors:
> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that
> +	 * require initialization, ranging from 0x080 to 0xADF and
> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.
> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK registers need to
> +	 * be initialized as well. MCR[RFEN] must not be set during memory
> +	 * initialization.
> +	 */
> +	reg_ctrl2 = priv->read(&regs->ctrl2);
> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;
> +	priv->write(reg_ctrl2, &regs->ctrl2);
> +
> +	memset_io(&regs->mb[0][0], 0,
> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);
> +
> +	if (priv->can.ctrlmode & CAN_CTRLMODE_FD)
> +		memset_io(&regs->tx_smb_fd[0], 0,
> +			  (u8 *)&regs->rx_smb1_fd[17] - (u8 *)&regs->tx_smb_fd[0] + 0x4);
> +
> +	reg_ctrl2 &= ~FLEXCAN_CTRL2_WRMFRZ;
> +	priv->write(reg_ctrl2, &regs->ctrl2);
> +}
> +
>  /* flexcan_chip_start
>   *
>   * this functions is entered with clocks enabled

drivers/net/can/flexcan.c:1329:20: warning: cast removes address space '__iomem' of expression
drivers/net/can/flexcan.c:1329:43: error: subtraction of different types can't work (different address spaces)
Marc Kleine-Budde Sept. 29, 2020, 2:50 p.m. UTC | #6
On 9/29/20 4:48 PM, Jakub Kicinski wrote:
> On Wed, 30 Sep 2020 04:30:39 +0800 Joakim Zhang wrote:
>> @@ -1292,6 +1307,35 @@ static void flexcan_set_bittiming(struct net_device *dev)
>>  		return flexcan_set_bittiming_ctrl(dev);
>>  }
>>  
>> +static void flexcan_init_ram(struct net_device *dev)
>> +{
>> +	struct flexcan_priv *priv = netdev_priv(dev);
>> +	struct flexcan_regs __iomem *regs = priv->regs;
>> +	u32 reg_ctrl2;
>> +
>> +	/* 11.8.3.13 Detection and correction of memory errors:
>> +	 * CTRL2[WRMFRZ] grants write access to all memory positions that
>> +	 * require initialization, ranging from 0x080 to 0xADF and
>> +	 * from 0xF28 to 0xFFF when the CAN FD feature is enabled.
>> +	 * The RXMGMASK, RX14MASK, RX15MASK, and RXFGMASK registers need to
>> +	 * be initialized as well. MCR[RFEN] must not be set during memory
>> +	 * initialization.
>> +	 */
>> +	reg_ctrl2 = priv->read(&regs->ctrl2);
>> +	reg_ctrl2 |= FLEXCAN_CTRL2_WRMFRZ;
>> +	priv->write(reg_ctrl2, &regs->ctrl2);
>> +
>> +	memset_io(&regs->mb[0][0], 0,
>> +		  (u8 *)&regs->rx_smb1[3] - &regs->mb[0][0] + 0x4);
>> +
>> +	if (priv->can.ctrlmode & CAN_CTRLMODE_FD)
>> +		memset_io(&regs->tx_smb_fd[0], 0,
>> +			  (u8 *)&regs->rx_smb1_fd[17] - (u8 *)&regs->tx_smb_fd[0] + 0x4);
>> +
>> +	reg_ctrl2 &= ~FLEXCAN_CTRL2_WRMFRZ;
>> +	priv->write(reg_ctrl2, &regs->ctrl2);
>> +}
>> +
>>  /* flexcan_chip_start
>>   *
>>   * this functions is entered with clocks enabled

> drivers/net/can/flexcan.c:1329:20: warning: cast removes address space '__iomem' of expression
> drivers/net/can/flexcan.c:1329:43: error: subtraction of different types can't work (different address spaces)

Thanks, this patch is outdated, we use offset of now to calculate the range:

> 	memset_io(&regs->mb[0][0], 0,
> 		  offsetof(struct flexcan_regs, rx_smb1[3]) -
> 		  offsetof(struct flexcan_regs, mb[0][0]) + 0x4);

Marc