Skip to contentSkip to conversation

CAN framework for U-Boot

RFC New Wolfgang Grandegger 4 participants 18 messages Nov 1, 2009 – Nov 5, 2009 Kbuild

lore
Share & tools
lore
#0 lore raw

cover CAN framework for U-Boot

From: Wolfgang Grandegger <[email protected]>

This patch series adds a simple CAN framework for U-Boot. The main purpose is to do simple RX/TX testing when the device boots up but the interface functions could also be used for more complex tasks. This is just a RFC and a few more features need to be added or issues to be fixed, like using a generic device interface, if it already exists, and board specific device configuration, e.g. via i82527_register(addr, cfg-params). The sources are based on GPL v2+ code to be compatible with future U-Boot licence requirement.

Please comment.

Wolfgang

Wolfgang Grandegger (5):
CAN interface library
CAN device test command
CAN device driver for the SJA1000
CAN device driver for the Intel 82527
CAN interface support for the TQM855L module

 Makefile                  |    1 +
 board/tqc/tqm8xx/tqm8xx.c |   17 ++
 common/Makefile           |    1 +
 common/cmd_can.c          |  119 +++++++++++++++
 drivers/can/Makefile      |   49 ++++++
 drivers/can/can.c         |   88 +++++++++++
 drivers/can/i82527.c      |  366 +++++++++++++++++++++++++++++++++++++++++++++
 drivers/can/sja1000.c     |  223 +++++++++++++++++++++++++++
 include/can.h             |   70 +++++++++
 include/configs/TQM855L.h |    8 +-
 include/i82527.h          |  201 +++++++++++++++++++++++++
 include/sja1000.h         |  159 ++++++++++++++++++++
 12 files changed, 1301 insertions(+), 1 deletions(-)
 create mode 100644 common/cmd_can.c
 create mode 100644 drivers/can/Makefile
 create mode 100644 drivers/can/can.c
 create mode 100644 drivers/can/i82527.c
 create mode 100644 drivers/can/sja1000.c
 create mode 100644 include/can.h
 create mode 100644 include/i82527.h
 create mode 100644 include/sja1000.h
#1 lore raw

RFC 1/5 CAN interface library

From: Wolfgang Grandegger <[email protected]>

---
 Makefile             |    1 +
 drivers/can/Makefile |   47 ++++++++++++++++++++++++++
 drivers/can/can.c    |   88 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/can.h        |   70 +++++++++++++++++++++++++++++++++++++++
 4 files changed, 206 insertions(+), 0 deletions(-)
 create mode 100644 drivers/can/Makefile
 create mode 100644 drivers/can/can.c
 create mode 100644 include/can.h
Makefile +1 -0
index bcb3fe9..dca15e0 100644
@@ -203,6 +203,7 @@ LIBS += net/libnet.a
 LIBS += disk/libdisk.a
 LIBS += drivers/bios_emulator/libatibiosemu.a
 LIBS += drivers/block/libblock.a
+LIBS += drivers/can/libcan.a
 LIBS += drivers/dma/libdma.a
 LIBS += drivers/fpga/libfpga.a
 LIBS += drivers/gpio/libgpio.a
drivers/can/Makefile new file +47 -0

47 lines not on this page · raw message

drivers/can/can.c new file +88 -0

88 lines not on this page · raw message

include/can.h new file +70 -0

70 lines not on this page · raw message

Signature · 1 line
1.6.2.5
#2 lore raw

RFC 2/5 CAN device test command

From: Wolfgang Grandegger <[email protected]>

---
 common/Makefile  |    1 +
 common/cmd_can.c |  119 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 120 insertions(+), 0 deletions(-)
 create mode 100644 common/cmd_can.c
common/Makefile +1 -0
index 3781738..b7f4c22 100644
@@ -72,6 +72,7 @@ COBJS-$(CONFIG_CMD_BEDBUG) += bedbug.o cmd_bedbug.o
 COBJS-$(CONFIG_CMD_BMP) += cmd_bmp.o
 COBJS-$(CONFIG_CMD_BOOTLDR) += cmd_bootldr.o
 COBJS-$(CONFIG_CMD_CACHE) += cmd_cache.o
+COBJS-$(CONFIG_CMD_CAN) += cmd_can.o
 COBJS-$(CONFIG_CMD_CONSOLE) += cmd_console.o
 COBJS-$(CONFIG_CMD_CPLBINFO) += cmd_cplbinfo.o
 COBJS-$(CONFIG_DATAFLASH_MMC_SELECT) += cmd_dataflash_mmc_mux.o
common/cmd_can.c new file +119 -0

119 lines not on this page · raw message

Signature · 1 line
1.6.2.5
#3 lore raw

RFC 3/5 CAN device driver for the SJA1000

From: Wolfgang Grandegger <[email protected]>

---
 drivers/can/Makefile  |    3 +-
 drivers/can/sja1000.c |  223 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/sja1000.h     |  159 +++++++++++++++++++++++++++++++++++
 3 files changed, 384 insertions(+), 1 deletions(-)
 create mode 100644 drivers/can/sja1000.c
 create mode 100644 include/sja1000.h
drivers/can/Makefile +2 -1
index 74d2ff5..e2b6bd6 100644
@@ -25,7 +25,8 @@ include $(TOPDIR)/config.mk
 
 LIB 	:= $(obj)libcan.a
 
-COBJS-$(CONFIG_CAN)	+= can.o
+COBJS-$(CONFIG_CAN)		+= can.o
+COBJS-$(CONFIG_CAN_SJA1000)	+= sja1000.o
 
 COBJS	:= $(COBJS-y)
 SRCS 	:= $(COBJS:.o=.c)
drivers/can/sja1000.c new file +223 -0

223 lines not on this page · raw message

include/sja1000.h new file +159 -0

159 lines not on this page · raw message

Signature · 1 line
1.6.2.5
#4 lore raw

RFC 4/5 CAN device driver for the Intel 82527

From: Wolfgang Grandegger <[email protected]>

---
 drivers/can/Makefile |    1 +
 drivers/can/i82527.c |  366 ++++++++++++++++++++++++++++++++++++++++++++++++++
 include/i82527.h     |  201 +++++++++++++++++++++++++++
 3 files changed, 568 insertions(+), 0 deletions(-)
 create mode 100644 drivers/can/i82527.c
 create mode 100644 include/i82527.h
drivers/can/Makefile +1 -0
index e2b6bd6..d550a45 100644
@@ -26,6 +26,7 @@ include $(TOPDIR)/config.mk
 LIB 	:= $(obj)libcan.a
 
 COBJS-$(CONFIG_CAN)		+= can.o
+COBJS-$(CONFIG_CAN_I82527)	+= i82527.o
 COBJS-$(CONFIG_CAN_SJA1000)	+= sja1000.o
 
 COBJS	:= $(COBJS-y)
drivers/can/i82527.c new file +366 -0

366 lines not on this page · raw message

include/i82527.h new file +201 -0

201 lines not on this page · raw message

Signature · 1 line
1.6.2.5
#5 lore raw

RFC 5/5 CAN interface support for the TQM855L module

From: Wolfgang Grandegger <[email protected]>

---
 board/tqc/tqm8xx/tqm8xx.c |   17 +++++++++++++++++
 include/configs/TQM855L.h |    8 +++++++-
 2 files changed, 24 insertions(+), 1 deletions(-)
board/tqc/tqm8xx/tqm8xx.c +17 -0
index f92c598..f1aab43 100644
@@ -26,6 +26,10 @@
 #ifdef CONFIG_PS2MULT
 #include <ps2mult.h>
 #endif
+#ifdef	CONFIG_CAN
+#include <can.h>
+#include <i82527.h>
+#endif
 
 extern flash_info_t flash_info[];	/* FLASH chips info */
 
@@ -447,6 +451,19 @@ int board_early_init_r (void)
 
 #endif /* CONFIG_PS2MULT */
 
+#if defined(CONFIG_CAN_DRIVER) && defined(CONFIG_CAN)
+static struct can_dev tqm8xx_can[2];
+
+#ifdef CONFIG_BOARD_EARLY_INIT_R
+int board_early_init_r (void)
+{
+	i82527_register (&tqm8xx_can[0], CONFIG_SYS_CAN_BASE);
+	i82527_register( &tqm8xx_can[1], CONFIG_SYS_CAN_BASE + 0x100);
+
+	return (0);
+}
+#endif
+#endif
 
 #ifdef CONFIG_MISC_INIT_R
 extern void load_sernum_ethaddr(void);
include/configs/TQM855L.h +7 -1
index 1255928..1603d30 100644
@@ -88,7 +88,13 @@
 
 #define	CONFIG_STATUS_LED	1	/* Status LED enabled		*/
 
-#undef	CONFIG_CAN_DRIVER		/* CAN Driver support disabled	*/
+#define	CONFIG_CAN_DRIVER			/* CAN Driver support enabled	*/
+#ifdef CONFIG_CAN_DRIVER
+#define CONFIG_CAN
+#define CONFIG_CAN_I82527
+#define CONFIG_CMD_CAN
+#define CONFIG_BOARD_EARLY_INIT_R
+#endif
 
 /*
  * BOOTP options
Signature · 1 line
1.6.2.5
#6 lore raw

RFC 3/5 CAN device driver for the SJA1000

In reply to Wolfgang Grandegger

Hi Wolfgang,

this patch conflicts with my simple SJA header posted some days ago

http://lists.denx.de/pipermail/u-boot/2009-October/063097.html

together with a fix for two of our boards - which has not much to do with CAN. WD asked me to use a C struct to access the SJA1000.

http://lists.denx.de/pipermail/u-boot/2009-October/062902.html

So where does this bring us? Either we want to use C structs for everything or decide it from patch to patch :-(

Matthias

Quoted 419 lines · Wolfgang Grandegger

From: Wolfgang Grandegger <[email protected]>

---
 drivers/can/Makefile  |    3 +-
 drivers/can/sja1000.c |  223 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/sja1000.h     |  159 +++++++++++++++++++++++++++++++++++
 3 files changed, 384 insertions(+), 1 deletions(-)
 create mode 100644 drivers/can/sja1000.c
 create mode 100644 include/sja1000.h
drivers/can/Makefile +2 -1
index 74d2ff5..e2b6bd6 100644
@@ -25,7 +25,8 @@ include $(TOPDIR)/config.mk
 
 LIB 	:= $(obj)libcan.a
 
-COBJS-$(CONFIG_CAN)	+= can.o
+COBJS-$(CONFIG_CAN)		+= can.o
+COBJS-$(CONFIG_CAN_SJA1000)	+= sja1000.o
 
 COBJS	:= $(COBJS-y)
 SRCS 	:= $(COBJS:.o=.c)
drivers/can/sja1000.c new file +223 -0
new file mode 100644
index 0000000..b75f01c
@@ -0,0 +1,223 @@
+/*
+ * (C) Copyright 2007-2009, Wolfgang Grandegger <[email protected]>
+ *
+ * Derived from Xenomai's RT-Socket-CAN driver for SJA1000:
+ *
+ * Copyright (C) 2005,2006 Sebastian Smolorz
+ *                        <[email protected]>
+ *
+ * Copyright (C) 2005, Sascha Hauer, Pengutronix
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+#include <common.h>
+#include <asm/io.h>
+
+#include <can.h>
+#include <sja1000.h>
+
+#define SJA1000_OCR	(SJA_OCR_MODE_NORMAL | SJA_OCR_TX0_PUSHPULL)
+#define SJA1000_CDR	SJA_CDR_CAN_MODE
+
+/*
+ * Basic functions to access registers
+ */
+#define sja1000_read_reg(dev, reg)			\
+	in_8 ((volatile u8 *)((dev)->base + (reg)))
+
+#define sja1000_write_reg(dev, reg, value)			\
+	out_8 ((volatile u8 *)((dev)->base + (reg)), value)
+
+/*
+ * Baudrate table
+ */
+
+static u16 sja1000_btr0btr1[] = {
+	0x031c,			/* 125K */
+	0x011c,			/* 250K */
+	0x001c,			/* 500K */
+};
+
+int sja1000_init (struct can_dev *dev, unsigned int ibaud)
+{
+	int i, wait = 1000;
+	u16 btr0btr1;
+
+	/* Disable the controller's interrupts */
+	sja1000_write_reg (dev, SJA_IER, 0x00);
+
+	/* Set reset mode bit */
+	sja1000_write_reg (dev, SJA_MOD, SJA_MOD_RM);
+
+	/* Read reset mode bit, multiple tests */
+	do {
+		udelay (100);
+		if (sja1000_read_reg (dev, SJA_MOD) & SJA_MOD_RM)
+			break;
+	} while (--wait);
+
+	sja1000_write_reg (dev, SJA_CDR, SJA1000_CDR);
+	sja1000_write_reg (dev, SJA_OCR, SJA1000_OCR);
+
+	sja1000_write_reg (dev, SJA_AMR0, 0xFF);
+	sja1000_write_reg (dev, SJA_AMR1, 0xFF);
+	sja1000_write_reg (dev, SJA_AMR2, 0xFF);
+	sja1000_write_reg (dev, SJA_AMR3, 0xFF);
+
+	sja1000_write_reg (dev, SJA_RXERR, 0);
+	sja1000_write_reg (dev, SJA_TXERR, 0);
+
+	i = sizeof (sja1000_btr0btr1) / sizeof (u16);
+	if (ibaud >= i)
+		ibaud = i - 1;
+	btr0btr1 = sja1000_btr0btr1[ibaud];
+	sja1000_write_reg (dev, SJA_BTR0, (btr0btr1 >> 8) & 0xff);
+	sja1000_write_reg (dev, SJA_BTR1, (btr0btr1 & 0xff));
+
+	/* Clear error code capture (i.e. read it) */
+	sja1000_read_reg (dev, SJA_ECC);
+
+	/* Clear reset mode bit in SJA1000 */
+	sja1000_write_reg (dev, SJA_MOD, 0);
+
+	return 0;
+}
+
+int sja1000_xmit (struct can_dev *dev, struct can_msg *msg)
+{
+	int i;
+	u8 fir;
+
+	if (msg->dlc > 8)
+		msg->dlc = 8;
+	fir = msg->dlc;
+
+	sja1000_write_reg (dev, SJA_ID1, msg->id >> 3);
+	sja1000_write_reg (dev, SJA_ID2, msg->id << 5);
+	for (i = 0; i < msg->dlc; i++)
+		sja1000_write_reg (dev, SJA_DATA_SFF (i), msg->data[i]);
+
+	/* Write frame information register */
+	sja1000_write_reg (dev, SJA_FIR, fir);
+
+	/* Push the 'send' button */
+	sja1000_write_reg (dev, SJA_CMR, SJA_CMR_TR);
+
+	/* Wait some time */
+	for (i = 0; i < CAN_XMIT_TIMEOUT_US; i += 10000) {
+		if (sja1000_read_reg (dev, SJA_SR) & SJA_SR_TCS)
+			return 0;
+		if (ctrlc ())
+			break;
+		udelay (10000);
+	}
+
+	return -1;
+}
+
+int sja1000_recv (struct can_dev *dev, struct can_msg *msg)
+{
+	int i;
+	u8 fir;
+
+	while (!(sja1000_read_reg (dev, SJA_SR) & SJA_SR_RBS)) {
+		if (ctrlc ())
+			return -1;
+	}
+
+	/* Read out frame information register */
+	fir = sja1000_read_reg (dev, SJA_FIR);
+
+	/* Extract data length code */
+	msg->dlc = fir & SJA_FIR_DLC_MASK;
+
+	/* If DLC exceeds 8 bytes adjust it to 8 (for the payload size) */
+	if (msg->dlc > 8)
+		msg->dlc = 8;
+
+	if (fir & SJA_FIR_EFF) {
+		printf ("Extended CAN messages not supported\n");
+		return -1;
+	} else {
+		msg->id = sja1000_read_reg (dev, SJA_ID1) << 3;
+		msg->id |= sja1000_read_reg (dev, SJA_ID2) >> 5;
+
+		if (!(fir & SJA_FIR_RTR)) {
+			for (i = 0; i < msg->dlc; i++)
+				msg->data[i] =
+				    sja1000_read_reg (dev, SJA_DATA_SFF (i));
+		}
+	}
+	if (fir & SJA_FIR_RTR)
+		msg->id |= CAN_RTR_FLAG;
+
+	/* Release Receive Buffer */
+	sja1000_write_reg (dev, SJA_CMR, SJA_CMR_RRB);
+
+	return 0;
+}
+
+int sja1000_status (struct can_dev *dev, int level)
+{
+	printf ("SJA1000 at %#x", dev->base);
+	if (level > 0) {
+		int stat = sja1000_read_reg (dev, SJA_SR) & 0xff;
+		printf (", status 0x%02x", stat);
+		if (stat & SJA_SR_BS)
+			puts (" busoff");
+		if (stat & SJA_SR_ES)
+			puts (" error");
+		if (stat & SJA_SR_TS)
+			puts (" txing");
+		if (stat & SJA_SR_RS)
+			puts (" rxing");
+		if (stat & SJA_SR_TCS)
+			puts (" txdone");
+		if (stat & SJA_SR_TBS)
+			puts (" txfree");
+		if (stat & SJA_SR_DOS)
+			puts (" overrun");
+		if (stat & SJA_SR_RBS)
+			puts (" rxfull");
+	}
+	puts ("\n");
+	if (level > 1) {
+		int i;
+		for (i = 0; i < SJA1000_SIZE; i++) {
+			if ((i % 0x10) == 0)
+				printf ("\n%02x:", i);
+			printf (" %02x", sja1000_read_reg (dev, i));
+		}
+		puts ("\n");
+	}
+	return 0;
+}
+
+void sja1000_register (struct can_dev *dev, unsigned long base)
+{
+	dev->name = "sja1000";
+	dev->base = base;
+	dev->init = sja1000_init;
+	dev->xmit = sja1000_xmit;
+	dev->recv = sja1000_recv;
+	dev->status = sja1000_status;
+
+	can_register (dev);
+}
include/sja1000.h new file +159 -0
new file mode 100644
index 0000000..56b43bf
@@ -0,0 +1,159 @@
+/*
+ * (C) Copyright 2007-2009, Wolfgang Grandegger <[email protected]>
+ *
+ * Derived from Xenomai's RT-Socket-CAN driver for SJA1000:
+ *
+ * Copyright (C) 2005,2006 Sebastian Smolorz
+ *                        <[email protected]>
+ *
+ * Copyright (C) 2005, Sascha Hauer, Pengutronix
+ *
+ * See file CREDITS for list of people who contributed to this
+ * project.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of
+ * the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston,
+ * MA 02111-1307 USA
+ */
+
+#ifndef __SJA1000_H__
+#define __SJA1000_H__
+
+#include <can.h>
+
+/* PeliCAN mode address map */
+
+/* reset and operating mode */
+#define SJA_MOD          0       /* Mode register                   */
+#define SJA_CMR          1       /* Command register                */
+#define SJA_SR           2       /* Status register                 */
+#define SJA_IR           3       /* Interrupt register              */
+#define SJA_IER          4       /* Interrupt enable register       */
+#define SJA_BTR0         6       /* Bus timing register 0           */
+#define SJA_BTR1         7       /* Bus timing register 1           */
+#define SJA_OCR          8       /* Output control register         */
+#define SJA_ALC         11       /* Arbitration lost capture        */
+#define SJA_ECC         12       /* Error code capture register     */
+#define SJA_RXERR       14       /* Receive error counter           */
+#define SJA_TXERR       15       /* Transmit error counter          */
+#define SJA_CDR         31       /* Clock divider register          */
+
+/* reset mode */
+#define SJA_ACR0        16       /* Acceptance code register 0      */
+#define SJA_ACR1        17       /* Acceptance code register 1      */
+#define SJA_ACR2        18       /* Acceptance code register 2      */
+#define SJA_ACR3        19       /* Acceptance code register 3      */
+#define SJA_AMR0        20       /* Acceptance mask register 0      */
+#define SJA_AMR1        21       /* Acceptance mask register 1      */
+#define SJA_AMR2        22       /* Acceptance mask register 2      */
+#define SJA_AMR3        23       /* Acceptance mask register 3      */
+
+/* operating mode */
+#define SJA_FIR         16       /* Frame information register      */
+#define SJA_ID1         17       /* Identifier 1                    */
+#define SJA_ID2         18       /* Identifier 2                    */
+#define SJA_ID3         19       /* Identifier 3 (EFF only)         */
+#define SJA_ID4         20       /* Identifier 4 (EFF only)         */
+
+#define SJA_DATA_SFF(x) (19 + (x)) /* Data registers in case of standard
+                                    * frame format; 0 <= x <= 7 */
+#define SJA_DATA_EFF(x) (21 + (x)) /* Data registers in case of extended
+                                    * frame format; 0 <= x <= 7 */
+/* Mode register */
+#define SJA_MOD_RM           (1<<0) /* Reset Mode                          */
+#define SJA_MOD_LOM          (1<<1) /* Listen Only Mode                    */
+#define SJA_MOD_STM          (1<<2) /* Self Test Mode                      */
+#define SJA_MOD_AFM          (1<<3) /* Acceptance Filter Mode              */
+#define SJA_MOD_SM           (1<<4) /* Sleep Mode                          */
+
+/* Command register */
+#define SJA_CMR_TR           (1<<0) /* Transmission request                */
+#define SJA_CMR_AT           (1<<1) /* Abort Transmission                  */
+#define SJA_CMR_RRB          (1<<2) /* Release Receive Buffer              */
+#define SJA_CMR_CDO          (1<<3) /* Clear Data Overrun                  */
+#define SJA_CMR_SRR          (1<<4) /* Self reception request              */
+
+/* Status register */
+#define SJA_SR_RBS           (1<<0) /* Receive Buffer Status               */
+#define SJA_SR_DOS           (1<<1) /* Data Overrun Status                 */
+#define SJA_SR_TBS           (1<<2) /* Transmit Buffer Status              */
+#define SJA_SR_TCS           (1<<3) /* Transmission Complete Status        */
+#define SJA_SR_RS            (1<<4) /* Receive Status                      */
+#define SJA_SR_TS            (1<<5) /* Transmit Status                     */
+#define SJA_SR_ES            (1<<6) /* Error Status                        */
+#define SJA_SR_BS            (1<<7) /* Bus Status                          */
+
+/* Interrupt register */
+#define SJA_IR_RI            (1<<0) /* Receive Interrupt                   */
+#define SJA_IR_TI            (1<<1) /* Transmit Interrupt                  */
+#define SJA_IR_EI            (1<<2) /* Error Warning Interrupt             */
+#define SJA_IR_DOI           (1<<3) /* Data Overrun Interrupt              */
+#define SJA_IR_WUI           (1<<4) /* Wake-Up Interrupt                   */
+#define SJA_IR_EPI           (1<<5) /* Error Passive Interrupt             */
+#define SJA_IR_ALI           (1<<6) /* Arbitration Lost Interrupt          */
+#define SJA_IR_BEI           (1<<7) /* Bus Error Interrupt                 */
+
+/* Interrupt enable register */
+#define SJA_IER_RIE          (1<<0) /* Receive Interrupt Enable            */
+#define SJA_IER_TIE          (1<<1) /* Transmit Interrupt Enable           */
+#define SJA_IER_EIE          (1<<2) /* Error Warning Interrupt Enable      */
+#define SJA_IER_DOIE         (1<<3) /* Data Overrun Interrupt Enable       */
+#define SJA_IER_WUIE         (1<<4) /* Wake-Up Interrupt Enable            */
+#define SJA_IER_EPIE         (1<<5) /* Error Passive Interrupt Enable      */
+#define SJA_IER_ALIE         (1<<6) /* Arbitration Lost Interrupt Enable   */
+#define SJA_IER_BEIE         (1<<7) /* Bus Error Interrupt Enable          */
+
+/* Output control register */
+#define SJA_OCR_MODE_BIPHASE 0
+#define SJA_OCR_MODE_TEST    1
+#define SJA_OCR_MODE_NORMAL  2
+#define SJA_OCR_MODE_CLOCK   3
+#define SJA_OCR_TX0_INVERT   (1<<2)
+#define SJA_OCR_TX0_PULLDOWN (1<<3)
+#define SJA_OCR_TX0_PULLUP   (2<<3)
+#define SJA_OCR_TX0_PUSHPULL (3<<3)
+#define SJA_OCR_TX1_INVERT   (1<<5)
+#define SJA_OCR_TX1_PULLDOWN (1<<6)
+#define SJA_OCR_TX1_PULLUP   (2<<6)
+#define SJA_OCR_TX1_PUSHPULL (3<<6)
+
+/* Error code capture register */
+
+/*
+ * The segmentation field gives information about the location of
+ * errors on the bus
+ */
+#define SJA_ECC_SEG_MASK     31     /* Segmentation field mask             */
+#define SJA_ECC_DIR          (1<<5) /* Transfer direction                  */
+#define SJA_ECC_ERR_BIT      (0<<6)
+#define SJA_ECC_ERR_FORM     (1<<6)
+#define SJA_ECC_ERR_STUFF    (2<<6)
+#define SJA_ECC_ERR_MASK     (3<<6) /* Error code mask                     */
+
+/* Frame information register */
+#define SJA_FIR_DLC_MASK     15     /* Data length code mask               */
+#define SJA_FIR_RTR          (1<<6) /* Remote transmission request         */
+#define SJA_FIR_EFF          (1<<7) /* Extended frame format               */
+
+/* Clock divider register */
+#define SJA_CDR_CLK_OFF      (1<<3) /* Clock off (CLKOUT pin)              */
+#define SJA_CDR_CBP          (1<<6) /* CAN input comparator bypass         */
+#define SJA_CDR_CAN_MODE     (1<<7) /* CAN mode: 1 = PeliCAN               */
+
+#define SJA1000_SIZE	     0x80
+
+void sja1000_register (struct can_dev *dev, unsigned long base);
+
+#endif  /* __SJA1000_H__ */
+
Signature · 13 lines
-------------------------------------------------------------------------
Dipl.-Ing. Matthias Fuchs
Head of System Design

esd electronic system design gmbh
Vahrenwalder Str. 207 - 30165 Hannover - GERMANY
Phone: +49-511-37298-0 - Fax: +49-511-37298-68
Please visit our homepage http://www.esd.eu
Quality Products - Made in Germany
-------------------------------------------------------------------------
Gesch?ftsf?hrer: Klaus Detering, Dr. Werner Schulze
Amtsgericht Hannover HRB 51373 - VAT-ID DE 115672832
-------------------------------------------------------------------------
#7 lore raw

RFC 3/5 CAN device driver for the SJA1000

Quoted 6 lines · Matthias Fuchs

Hi Wolfgang,

this patch conflicts with my simple SJA header posted some days ago

http://lists.denx.de/pipermail/u-boot/2009-October/063097.html

Quoted 7 lines

together with a fix for two of our boards - which has not much to do with CAN. WD asked me to use a C struct to access the SJA1000.

http://lists.denx.de/pipermail/u-boot/2009-October/062902.html

So where does this bring us? Either we want to use C structs for everything or decide it from patch to patch :-(

Then it should be changed, of course. This patch is far from being accepted and for the moment it's an implementation detail. I'm especially interested to hear if such a generic CAN interface would serve your purposes as well, as you require access to the SJA1000 somehow.

Wolfgang.

#8 lore raw

RFC 3/5 CAN device driver for the SJA1000

Hi Wolfgang,

of course I can think of situations where some simple CAN mechanism might be helpful (e.g. simple hardware testing).

But do we really need this inside a bootloader? Surely not for a production build. But please keep on hacking!

Quoted 12 lines · Wolfgang Grandegger
Quoted 11 lines · Matthias Fuchs

Hi Wolfgang,

this patch conflicts with my simple SJA header posted some days ago

http://lists.denx.de/pipermail/u-boot/2009-October/063097.html

together with a fix for two of our boards - which has not much to do with CAN. WD asked me to use a C struct to access the SJA1000.

http://lists.denx.de/pipermail/u-boot/2009-October/062902.html

Quoted 7 lines
Quoted 2 lines

So where does this bring us? Either we want to use C structs for everything or decide it from patch to patch :-(

Then it should be changed, of course. This patch is far from being accepted and for the moment it's an implementation detail. I'm especially interested to hear if such a generic CAN interface would serve your purposes as well, as you require access to the SJA1000 somehow.

I just need to bit bang around in the OCR register. So no need for a full blown and flash consuming CAN implementation. Of course I could life with your register access style. Especially because it makes the code more common with Socket-CAN files which prevents us from rewriting fully functional code ;-)

Matthias

#9 lore raw

RFC 3/5 CAN device driver for the SJA1000

Quoted 28 lines · Matthias Fuchs

Hi Wolfgang,

of course I can think of situations where some simple CAN mechanism might be helpful (e.g. simple hardware testing).

But do we really need this inside a bootloader? Surely not for a production build. But please keep on hacking!

Quoted 10 lines · Wolfgang Grandegger
Quoted 9 lines · Matthias Fuchs

Hi Wolfgang,

this patch conflicts with my simple SJA header posted some days ago

http://lists.denx.de/pipermail/u-boot/2009-October/063097.html

together with a fix for two of our boards - which has not much to do with CAN. WD asked me to use a C struct to access the SJA1000.

Quoted 8 lines
Quoted 4 lines

http://lists.denx.de/pipermail/u-boot/2009-October/062902.html

So where does this bring us? Either we want to use C structs for everything or decide it from patch to patch :-(

Then it should be changed, of course. This patch is far from being accepted and for the moment it's an implementation detail. I'm especially interested to hear if such a generic CAN interface would serve your purposes as well, as you require access to the SJA1000 somehow.

Quoted 4 lines

I just need to bit bang around in the OCR register. So no need for a full blown and flash consuming CAN implementation. Of course I could life with your register access style. Especially because it makes the code more common with Socket-CAN files which prevents us from rewriting fully functional code ;-)

Well, I think Wolfgang will tell me to use structs sooner than later.

Wolfgang.

#10 lore raw

RFC 3/5 CAN device driver for the SJA1000

Dear Wolfgang Grandegger,

Quoted 2 lines · In message <[email protected]> you wrote:

Well, I think Wolfgang will tell me to use structs sooner than later.

Why should I? You already know it!

Best regards,

Wolfgang Denk

Signature · 5 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"There is no statute of limitations on stupidity."
- Randomly produced by a computer program called Markov3.
#11 lore raw

RFC 2/5 CAN device test command

In reply to Wolfgang Grandegger

common/cmd_can.c line 104

Quoted 5 lines · On Sunday 01 November 2009 06:33:34 Wolfgang Grandegger wrote:
common/cmd_can.c new file +5 -0
+	if (op == 's') {
+	else if (op == 'i') {
+	else if (op == 'r') {
+	} else if (op == 'x') {
+	} else {

your if style here is inconsistent, but ignoring that, shouldnt this really be a switch() ? although, by only checking the first char, you allow people to encode typos into their commands and not realize it until some point in the future where things get stricter. i.e. people can do can ilovecandy ...

Quoted 4 lines
common/cmd_can.c new file +4 -0
+		unsigned int dev_num = 0, ibaud = 0;
+		struct can_dev *dev;
+
+		if (argc > 2)

common/cmd_can.c line 72

Quoted 10 lines
common/cmd_can.c new file +10 -0
+			dev_num = simple_strtoul (argv[2], NULL, 10);
+		if (argc > 3) {
+			ibaud = simple_strtoul (argv[3], NULL, 10);
+			if (ibaud > 2)
+				ibaud = 2;
+		}
+		dev = can_init (dev_num, ibaud);
+		if (!dev)
+			return 1;
+		can_dev = dev;

if i told CAN to init an unknown device, i would expect to get an error and the command state to remain in said error state until i selected a proper CAN device. otherwise, a script that didnt check the can init status would incorrectly operate on the previously selected can device.

how do other commands work ? am i complaining about common convention here ?

common/cmd_can.c line 105

Quoted 1 line
common/cmd_can.c new file +1 -0
+		printf ("Usage:\n%s\n", cmdtp->usage);

cmd_usage() ?

common/cmd_can.c line 118

Quoted 6 lines
common/cmd_can.c new file +6 -0
+	can, 3,	1, do_can,
+	"can - CAN bus commands\n",
+	"can status [level]\n"
+	"can init [dev] [baud-index]\n"
+	"can xmit [id] [d0] [d1] ... [d7]\n"
+	"can recv, abort with CTRL-C\n"

does the help really display correctly here ? i think the "can status" line will have too many "can"'s ?
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091101/7d9b8d17/attachment.pgp

#12 lore raw

RFC 2/5 CAN device test command

Quoted 11 lines · Mike Frysinger

common/cmd_can.c line 104

Quoted 5 lines · On Sunday 01 November 2009 06:33:34 Wolfgang Grandegger wrote:
common/cmd_can.c new file +5 -0
+	if (op == 's') {
+	else if (op == 'i') {
+	else if (op == 'r') {
+	} else if (op == 'x') {
+	} else {

your if style here is inconsistent, but ignoring that, shouldnt this really be a switch() ? although, by only checking the first char, you allow people to encode typos into their commands and not realize it until some point in the future where things get stricter. i.e. people can do can ilovecandy ...

Will fix.

Quoted 11 lines
Quoted 11 lines
common/cmd_can.c new file +11 -0
+		unsigned int dev_num = 0, ibaud = 0;
+		struct can_dev *dev;
+
+		if (argc > 2)
+			dev_num = simple_strtoul (argv[2], NULL, 10);
+		if (argc > 3) {
+			ibaud = simple_strtoul (argv[3], NULL, 10);
+			if (ibaud > 2)
+				ibaud = 2;
+		}
+		dev = can_init (dev_num, ibaud);
Quoted 8 lines

common/cmd_can.c line 72

Quoted 3 lines
common/cmd_can.c new file +3 -0
+		if (!dev)
+			return 1;
+		can_dev = dev;

if i told CAN to init an unknown device, i would expect to get an error and the command state to remain in said error state until i selected a proper CAN device. otherwise, a script that didnt check the can init status would incorrectly operate on the previously selected can device.

can_init will already print an error message. But that might be changed.

Quoted 5 lines

how do other commands work ? am i complaining about common convention here ?

common/cmd_can.c line 105

Quoted 1 line
common/cmd_can.c new file +1 -0
+		printf ("Usage:\n%s\n", cmdtp->usage);

cmd_usage() ?

OK.

Quoted 9 lines

common/cmd_can.c line 118

Quoted 6 lines
common/cmd_can.c new file +6 -0
+	can, 3,	1, do_can,
+	"can - CAN bus commands\n",
+	"can status [level]\n"
+	"can init [dev] [baud-index]\n"
+	"can xmit [id] [d0] [d1] ... [d7]\n"
+	"can recv, abort with CTRL-C\n"

does the help really display correctly here ? i think the "can status" line will have too many "can"'s ?

I think the output was OK. But I will check later next week.

Note that this is a RFC trying to discuss the real requirements of a CAN interface in U-Boot. I think it would also be nice to have can_xmit() and can_recv() with a timeout parameter, e.g.:

  can_xmit(struct can_dev *dev, int timeout_us);

And maybe also a can_xmit_done() function.

Wolfgang.

#13 lore raw

RFC 2/5 CAN device test command

Quoted 7 lines · On Sunday 01 November 2009 11:24:59 Wolfgang Grandegger wrote:

Note that this is a RFC trying to discuss the real requirements of a CAN interface in U-Boot. I think it would also be nice to have can_xmit() and can_recv() with a timeout parameter, e.g.:

  can_xmit(struct can_dev *dev, int timeout_us);

And maybe also a can_xmit_done() function.

i only have a passing familiarity with CAN (havent done real work with it), but i can find someone to look over the current stuff
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091101/b78a8ed4/attachment.pgp

#14 lore raw

RFC 1/5 CAN interface library

In reply to Wolfgang Grandegger

Makefile LIBS += net/libnet.a · line 206

Quoted 7 lines · On Sunday 01 November 2009 06:33:33 Wolfgang Grandegger wrote:
Makefile +1 -0
@@ -203,6 +203,7 @@ LIBS += net/libnet.a
 LIBS += disk/libdisk.a
 LIBS += drivers/bios_emulator/libatibiosemu.a
 LIBS += drivers/block/libblock.a
+LIBS += drivers/can/libcan.a

this isnt an issue specific to CAN, but i wonder if we should switch LIBS to LIBS-y now that the top level Makefile can rely on autoconf.mk settings after the point config.mk is included. it would save time on pointlessly recursing into all the empty dirs and creating empty archives.

Quoted 17 lines
drivers/can/Makefile new file +14 -0
@@ -0,0 +1,47 @@
+include $(TOPDIR)/config.mk
+
+LIB 	:= $(obj)libcan.a
+
+COBJS-$(CONFIG_CAN)	+= can.o
+
+COBJS	:= $(COBJS-y)
+SRCS 	:= $(COBJS:.o=.c)
+OBJS 	:= $(addprefix $(obj),$(COBJS))
+
+all:	$(LIB)
+
+$(LIB):	$(obj).depend $(OBJS)
+	$(AR) $(ARFLAGS) $@ $(OBJS)

drivers/can/Makefile line 24

Quoted 10 lines
drivers/can/Makefile new file +10 -0
+
+
+#####################################################
+
+# defines $(obj).depend target
+include $(SRCTREE)/rules.mk
+
+sinclude $(obj).depend
+
+####################################################

also not specific to CAN, but i think it's time we start creating .mk files for subdirs to include

drivers/can/can.c

Quoted 4 lines
drivers/can/can.c new file +1 -0
 
+static char *baudrates[] = { "125K", "250K", "500K" };

so we're restricting ourselves to these three speeds ? i have passing familiarity with CAN, but i didnt think the protocol was restricted to specific speeds.

Quoted 1 line

+int can_register (struct can_dev* can_dev)

no space before the paren, and the * is cuddled on the wrong side of the space. seems like a lot of this code suffers from these two issues.

drivers/can/can.c line 45

Quoted 11 lines
drivers/can/can.c new file +11 -0
+{
+	struct can_dev* dev;
+
+	can_dev->next = NULL;
+	if (!can_devs)
+		can_devs = can_dev;
+	else {
+		for (dev = can_devs; dev->next; dev = dev->next)
+			    ;
+		dev->next = can_dev;
+	}

invert the if logic and i think the code would look "nicer" -- use braces on the first branch instead of the second.

Quoted 18 lines
drivers/can/can.c new file +18 -0
+struct can_dev *can_init (int dev_num, int ibaud)
+{
+	struct can_dev *dev;
+	int i;
+
+	if (!can_devs) {
+		puts ("No CAN devices registered\n");
+		return NULL;
+	}
+
+	/* Advance to selected device */
+	for (i = 0, dev = can_devs; dev; i++, dev = dev->next) {
+		if (i == dev_num)
+			break;
+	}
+
+	if (!dev) {
+		printf ("CAN device %d does not exist\n", dev_num);

drivers/can/can.c line 79

Quoted 10 lines
drivers/can/can.c new file +10 -0
+		return NULL;
+	}
+
+	printf ("Initializing CAN%d at 0x%08x with baudrate %s\n",
+		i, dev->base, baudrates[ibaud]);
+
+	dev->init (dev, ibaud);
+
+	return dev;
+}

wonder if we should have a generic device list code base since this looks similar to a lot of other u-boot device lists ...

include/can.h line 2

Quoted 5 lines
include/can.h new file +2 -0
@@ -0,0 +1,70 @@
+/*
+ * (C) Copyright 2007-2009, Wolfgang Grandegger <[email protected]>

have you really been working on this stuff since 2007 ?

include/can.h line 42

Quoted 2 lines
include/can.h new file +2 -0
+struct can_dev {
+	char *name;

const ?
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091101/747d1ff9/attachment.pgp

#15 lore raw

RFC 1/5 CAN interface library

Quoted 1 line · Mike Frysinger

On Sunday 01 November 2009 06:33:33 Wolfgang Grandegger wrote:

[snip]

Quoted 8 lines

drivers/can/can.c

Quoted 4 lines
drivers/can/can.c new file +1 -0
 
+static char *baudrates[] = { "125K", "250K", "500K" };

so we're restricting ourselves to these three speeds ? i have passing familiarity with CAN, but i didnt think the protocol was restricted to specific speeds.

Well, this is an RFC and as I wrote in the introduction some features need to be added or extended, especially for CAN device configuration. My idea is to have a more complete default bit-timing table, which board specific code may overwrite using, for example:

   sja1000_register(&my_sja1000, &my_config_opts);

This would also allow to set the CAN clock, cdr and ocr registers.

Quoted 4 lines
Quoted 1 line

+int can_register (struct can_dev* can_dev)

no space before the paren, and the * is cuddled on the wrong side of the space. seems like a lot of this code suffers from these two issues.

U-Boot coding style requires a space after the function name and before "(". But the "*" is misplaced, of course.

Quoted 6 lines
Quoted 6 lines
drivers/can/can.c new file +6 -0
+{
+	struct can_dev* dev;
+
+	can_dev->next = NULL;
+	if (!can_devs)
+		can_devs = can_dev;
Quoted 8 lines

drivers/can/can.c line 45

Quoted 5 lines
drivers/can/can.c new file +5 -0
+	else {
+		for (dev = can_devs; dev->next; dev = dev->next)
+			    ;
+		dev->next = can_dev;
+	}

invert the if logic and i think the code would look "nicer" -- use braces on the first branch instead of the second.

OK.

Quoted 23 lines
Quoted 23 lines
drivers/can/can.c new file +23 -0
+struct can_dev *can_init (int dev_num, int ibaud)
+{
+	struct can_dev *dev;
+	int i;
+
+	if (!can_devs) {
+		puts ("No CAN devices registered\n");
+		return NULL;
+	}
+
+	/* Advance to selected device */
+	for (i = 0, dev = can_devs; dev; i++, dev = dev->next) {
+		if (i == dev_num)
+			break;
+	}
+
+	if (!dev) {
+		printf ("CAN device %d does not exist\n", dev_num);
+		return NULL;
+	}
+
+	printf ("Initializing CAN%d at 0x%08x with baudrate %s\n",
+		i, dev->base, baudrates[ibaud]);
Quoted 8 lines

drivers/can/can.c line 79

Quoted 5 lines
drivers/can/can.c new file +5 -0
+
+	dev->init (dev, ibaud);
+
+	return dev;
+}

wonder if we should have a generic device list code base since this looks similar to a lot of other u-boot device lists ...

Do we already have a generic interface?

Quoted 7 lines

include/can.h line 2

Quoted 5 lines
include/can.h new file +2 -0
@@ -0,0 +1,70 @@
+/*
+ * (C) Copyright 2007-2009, Wolfgang Grandegger <[email protected]>

have you really been working on this stuff since 2007 ?

The code was written in 2007. "2007, 2009" is more appropriate.

Quoted 4 lines

include/can.h line 42

Quoted 2 lines
include/can.h new file +2 -0
+struct can_dev {
+	char *name;

const ?

OK.

Wolfgang.

#16 lore raw

RFC 1/5 CAN interface library

Quoted 11 lines · On Sunday 01 November 2009 11:16:59 Wolfgang Grandegger wrote:
Quoted 9 lines · Mike Frysinger

drivers/can/can.c

Quoted 4 lines · On Sunday 01 November 2009 06:33:33 Wolfgang Grandegger wrote:
drivers/can/can.c new file +1 -0
 
+static char *baudrates[] = { "125K", "250K", "500K" };

so we're restricting ourselves to these three speeds ? i have passing familiarity with CAN, but i didnt think the protocol was restricted to specific speeds.

Quoted 8 lines

Well, this is an RFC and as I wrote in the introduction some features need to be added or extended, especially for CAN device configuration. My idea is to have a more complete default bit-timing table, which board specific code may overwrite using, for example:

   sja1000_register(&my_sja1000, &my_config_opts);

This would also allow to set the CAN clock, cdr and ocr registers.

this makes sense if the device supports a limited number of baud rates. but what if the baud rate is arbitrary (between two limits) ?

Quoted 7 lines
Quoted 4 lines
Quoted 1 line

+int can_register (struct can_dev* can_dev)

no space before the paren, and the * is cuddled on the wrong side of the space. seems like a lot of this code suffers from these two issues.

U-Boot coding style requires a space after the function name and before "(". But the "*" is misplaced, of course.

it's (thankfully) been changing to Linux kernel style

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

+struct can_dev *can_init (int dev_num, int ibaud)

wonder if we should have a generic device list code base since this looks similar to a lot of other u-boot device lists ...

Do we already have a generic interface?

i dont think so

Quoted 9 lines
Quoted 7 lines

include/can.h line 2

Quoted 5 lines
include/can.h new file +2 -0
@@ -0,0 +1,70 @@
+/*
+ * (C) Copyright 2007-2009, Wolfgang Grandegger <[email protected]>

have you really been working on this stuff since 2007 ?

The code was written in 2007. "2007, 2009" is more appropriate.

the intro made it sound like this was a recent development (i.e. last few days/weeks). if code is actually from 2007, then the range is fine.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091101/6367ebef/attachment.pgp

#17 lore raw

RFC 1/5 CAN interface library

Quoted 15 lines · Mike Frysinger
Quoted 14 lines · On Sunday 01 November 2009 11:16:59 Wolfgang Grandegger wrote:
Quoted 8 lines · Mike Frysinger

drivers/can/can.c

Quoted 4 lines · On Sunday 01 November 2009 06:33:33 Wolfgang Grandegger wrote:
drivers/can/can.c new file +1 -0
 
+static char *baudrates[] = { "125K", "250K", "500K" };

so we're restricting ourselves to these three speeds ? i have passing familiarity with CAN, but i didnt think the protocol was restricted to specific speeds.

Well, this is an RFC and as I wrote in the introduction some features need to be added or extended, especially for CAN device configuration. My idea is to have a more complete default bit-timing table, which board specific code may overwrite using, for example:

Quoted 6 lines
Quoted 3 lines
   sja1000_register(&my_sja1000, &my_config_opts);

This would also allow to set the CAN clock, cdr and ocr registers.

this makes sense if the device supports a limited number of baud rates. but what if the baud rate is arbitrary (between two limits) ?

Board specific code can define what ever table it likes, including non-standard bit-rate and bit-timings. Nevertheless, for most CAN controllers the default bit-timing parameters are just fine.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

+int can_register (struct can_dev* can_dev)

no space before the paren, and the * is cuddled on the wrong side of the space. seems like a lot of this code suffers from these two issues.

U-Boot coding style requires a space after the function name and before "(". But the "*" is misplaced, of course.

it's (thankfully) been changing to Linux kernel style

I really appreciate that.

Quoted 6 lines
Quoted 4 lines
Quoted 3 lines
Quoted 1 line

+struct can_dev *can_init (int dev_num, int ibaud)

wonder if we should have a generic device list code base since this looks similar to a lot of other u-boot device lists ...

Do we already have a generic interface?

i dont think so

Nor do I.

Wolfgang.