Skip to contentSkip to conversation

Add basic NVIDIA Tegra2 SoC support

PATCH Superseded Tom Warren 4 participants 26 messages Jan 21, 2011 – Jan 27, 2011

Subsystems (3)

v5 of 6 Superseded Latest is v6 →

lore
Share & tools
lore

Tom Warren

author
#0 lore raw

v5 cover Add basic NVIDIA Tegra2 SoC support

This series of patches adds preliminary/baseline support for NVIDIA's Tegra2 SoC. Basic CPU (AVP), RAM and UART init are covered so that the system (Harmony or Seaboard) can boot to the U-Boot serial cmd prompt.

Further support (for Cortex-A9 CPU(s), USB, SD/MMC, etc.) to follow.

Changes for V2:
- Coding style cleanup
- Remove mach-types.h change; wait for ARM kernel sync-up
- Move serial driver changes to separate patch
- Use board/nvidia/ instead of /board/tegra
- Remove TRUE/FALSE defines
- Use standard NS16550 register/bit defines in UART init
- Change nv-common.h config file to tegra2-common.h

Changes for V3:
- Use I/O accessors for Tegra2 HW MMIO register access
- Allow conditional compile of UARTA/UARTD code to save space

Changes for V4:
- Use address of HW structs (&pmc, etc.) in readl/writel
- Remove empty lines, fix mixed case hex #s & comments in header(s)
- Move board/nvidia/common/board.c UART code & header to arch/arm/cpu/armv7/tegra2/
- Declare internal functions as static in UART code

Changes for V5:
- Move arch/arm/cpu/armv7/uart.c & board.h to drivers/serial and rename to serial_tegra2.c
- Remove use of uart_num & UART_A/D in serial_tegra2, simplify code

Tom Warren (4):
arm: Tegra2: Add basic NVIDIA Tegra2 SoC support
serial: Add Tegra2 serial port support
arm: Tegra2: Add support for NVIDIA Harmony board
arm: Tegra2: Add support for NVIDIA Seaboard board

 MAINTAINERS                                  |    5 +
 arch/arm/cpu/armv7/tegra2/Makefile           |   48 ++++++
 arch/arm/cpu/armv7/tegra2/board.c            |   91 ++++++++++++
 arch/arm/cpu/armv7/tegra2/config.mk          |   28 ++++
 arch/arm/cpu/armv7/tegra2/lowlevel_init.S    |   65 ++++++++
 arch/arm/cpu/armv7/tegra2/sys_info.c         |   35 +++++
 arch/arm/cpu/armv7/tegra2/timer.c            |  122 +++++++++++++++
 arch/arm/include/asm/arch-tegra2/clk_rst.h   |  154 +++++++++++++++++++
 arch/arm/include/asm/arch-tegra2/pinmux.h    |   51 +++++++
 arch/arm/include/asm/arch-tegra2/pmc.h       |  124 ++++++++++++++++
 arch/arm/include/asm/arch-tegra2/sys_proto.h |   33 ++++
 arch/arm/include/asm/arch-tegra2/tegra2.h    |   49 ++++++
 arch/arm/include/asm/arch-tegra2/uart.h      |   44 ++++++
 board/nvidia/common/board.c                  |   57 +++++++
 board/nvidia/harmony/Makefile                |   50 +++++++
 board/nvidia/seaboard/Makefile               |   50 +++++++
 boards.cfg                                   |    2 +
 common/serial.c                              |    3 +-
 drivers/serial/Makefile                      |    1 +
 drivers/serial/serial_tegra2.c               |  205 ++++++++++++++++++++++++++
 drivers/serial/serial_tegra2.h               |   49 ++++++
 include/configs/harmony.h                    |   48 ++++++
 include/configs/seaboard.h                   |   44 ++++++
 include/configs/tegra2-common.h              |  160 ++++++++++++++++++++
 include/serial.h                             |    3 +-
 25 files changed, 1519 insertions(+), 2 deletions(-)
 create mode 100644 arch/arm/cpu/armv7/tegra2/Makefile
 create mode 100644 arch/arm/cpu/armv7/tegra2/board.c
 create mode 100644 arch/arm/cpu/armv7/tegra2/config.mk
 create mode 100644 arch/arm/cpu/armv7/tegra2/lowlevel_init.S
 create mode 100644 arch/arm/cpu/armv7/tegra2/sys_info.c
 create mode 100644 arch/arm/cpu/armv7/tegra2/timer.c
 create mode 100644 arch/arm/include/asm/arch-tegra2/clk_rst.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/pinmux.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/pmc.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/sys_proto.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/tegra2.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/uart.h
 create mode 100644 board/nvidia/common/board.c
 create mode 100644 board/nvidia/harmony/Makefile
 create mode 100644 board/nvidia/seaboard/Makefile
 create mode 100644 drivers/serial/serial_tegra2.c
 create mode 100644 drivers/serial/serial_tegra2.h
 create mode 100644 include/configs/harmony.h
 create mode 100644 include/configs/seaboard.h
 create mode 100644 include/configs/tegra2-common.h
Signature · 1 line
1.7.3.5
author
#1 lore raw

PATCH v5 1/4 arm: Tegra2: Add basic NVIDIA Tegra2 SoC support

---
Changes for V2:
- Coding style cleanup
- Move serial driver changes to separate patch
- Use board/nvidia instead of board/tegra
- Remove TRUE/FALSE defines
- Use standard NS16550 registers/bit defines in UART init

Changes for V3:
- Use I/O accessors for Tegra2 HW MMIO register access
- Allow conditional compile of UARTA/UARTD code to save space

Changes for V4:
- Use address of HW structs (&pmc, etc.) in readl/writel
- Remove empty lines, fix mixed case hex #s & comments in header(s)
- Move board/nvidia/common/board.c UART code & header to arch/arm/cpu/armv7/tegra2/
- Declare internal functions as static in UART code

 arch/arm/cpu/armv7/tegra2/Makefile           |   48 ++++++
 arch/arm/cpu/armv7/tegra2/board.c            |   91 +++++++++++
 arch/arm/cpu/armv7/tegra2/board.h            |   58 +++++++
 arch/arm/cpu/armv7/tegra2/config.mk          |   28 ++++
 arch/arm/cpu/armv7/tegra2/lowlevel_init.S    |   65 ++++++++
 arch/arm/cpu/armv7/tegra2/sys_info.c         |   35 ++++
 arch/arm/cpu/armv7/tegra2/timer.c            |  122 +++++++++++++++
 arch/arm/cpu/armv7/tegra2/uart.c             |  216 ++++++++++++++++++++++++++
 arch/arm/include/asm/arch-tegra2/clk_rst.h   |  154 ++++++++++++++++++
 arch/arm/include/asm/arch-tegra2/pinmux.h    |   51 ++++++
 arch/arm/include/asm/arch-tegra2/pmc.h       |  124 +++++++++++++++
 arch/arm/include/asm/arch-tegra2/sys_proto.h |   33 ++++
 arch/arm/include/asm/arch-tegra2/tegra2.h    |   49 ++++++
 arch/arm/include/asm/arch-tegra2/uart.h      |   44 ++++++
 board/nvidia/common/board.c                  |   57 +++++++
 15 files changed, 1175 insertions(+), 0 deletions(-)
 create mode 100644 arch/arm/cpu/armv7/tegra2/Makefile
 create mode 100644 arch/arm/cpu/armv7/tegra2/board.c
 create mode 100644 arch/arm/cpu/armv7/tegra2/board.h
 create mode 100644 arch/arm/cpu/armv7/tegra2/config.mk
 create mode 100644 arch/arm/cpu/armv7/tegra2/lowlevel_init.S
 create mode 100644 arch/arm/cpu/armv7/tegra2/sys_info.c
 create mode 100644 arch/arm/cpu/armv7/tegra2/timer.c
 create mode 100644 arch/arm/cpu/armv7/tegra2/uart.c
 create mode 100644 arch/arm/include/asm/arch-tegra2/clk_rst.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/pinmux.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/pmc.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/sys_proto.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/tegra2.h
 create mode 100644 arch/arm/include/asm/arch-tegra2/uart.h
 create mode 100644 board/nvidia/common/board.c
arch/arm/cpu/armv7/tegra2/Makefile new file +48 -0
new file mode 100644
index 0000000..f5b657b
@@ -0,0 +1,48 @@
+#
+# (C) Copyright 2010,2011 Nvidia Corporation.
+#
+# (C) Copyright 2000-2003
+# Wolfgang Denk, DENX Software Engineering, wd at denx.de.
+#
+# 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 $(TOPDIR)/config.mk
+
+LIB	=  $(obj)lib$(SOC).o
+
+SOBJS	:= lowlevel_init.o
+COBJS	:= board.o sys_info.o timer.o uart.o
+
+SRCS	:= $(SOBJS:.o=.S) $(COBJS:.o=.c)
+OBJS	:= $(addprefix $(obj),$(COBJS) $(SOBJS))
+
+all:	 $(obj).depend $(LIB)
+
+$(LIB):	$(OBJS)
+	$(AR) $(ARFLAGS) $@ $(OBJS)
+
+#########################################################################
+
+# defines $(obj).depend target
+include $(SRCTREE)/rules.mk
+
+sinclude $(obj).depend
+
+#########################################################################
arch/arm/cpu/armv7/tegra2/board.c new file +91 -0

91 lines not on this page · raw message

arch/arm/cpu/armv7/tegra2/board.h new file +58 -0

58 lines not on this page · raw message

arch/arm/cpu/armv7/tegra2/config.mk new file +28 -0
new file mode 100644
index 0000000..96c0795
@@ -0,0 +1,28 @@
+#
+# (C) Copyright 2010,2011
+# NVIDIA Corporation <www.nvidia.com>
+#
+# (C) Copyright 2002
+# Gary Jennejohn, DENX Software Engineering, <[email protected]>
+#
+# 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
+#
+
+# Use ARMv4 for Tegra2 - initial code runs on the AVP, which is an ARM7TDI.
+PLATFORM_CPPFLAGS += -march=armv4
arch/arm/cpu/armv7/tegra2/lowlevel_init.S new file +65 -0

65 lines not on this page · raw message

arch/arm/cpu/armv7/tegra2/sys_info.c new file +35 -0
new file mode 100644
index 0000000..6d11dc1
@@ -0,0 +1,35 @@
+/*
+ * (C) Copyright 2010,2011
+ * NVIDIA Corporation <www.nvidia.com>
+ *
+ * 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>
+
+#ifdef CONFIG_DISPLAY_CPUINFO
+/* Print CPU information */
+int print_cpuinfo(void)
+{
+	puts("TEGRA2\n");
+
+	/* TBD: Add printf of major/minor rev info, stepping, etc. */
+	return 0;
+}
+#endif	/* CONFIG_DISPLAY_CPUINFO */
arch/arm/cpu/armv7/tegra2/timer.c new file +122 -0

122 lines not on this page · raw message

arch/arm/cpu/armv7/tegra2/uart.c new file +216 -0

216 lines not on this page · raw message

arch/arm/include/asm/arch-tegra2/clk_rst.h new file +154 -0

154 lines not on this page · raw message

arch/arm/include/asm/arch-tegra2/pinmux.h new file +51 -0

51 lines not on this page · raw message

arch/arm/include/asm/arch-tegra2/pmc.h new file +124 -0

124 lines not on this page · raw message

arch/arm/include/asm/arch-tegra2/sys_proto.h new file +33 -0
new file mode 100644
index 0000000..70e63a8
@@ -0,0 +1,33 @@
+/*
+ * (C) Copyright 2010,2011
+ * NVIDIA Corporation <www.nvidia.com>
+ *
+ * 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 _SYS_PROTO_H_
+#define _SYS_PROTO_H_
+
+typedef struct {
+	char *board_string;
+} tegra2_sysinfo;
+
+void invalidate_dcache(void);
+
+#endif
arch/arm/include/asm/arch-tegra2/tegra2.h new file +49 -0

49 lines not on this page · raw message

arch/arm/include/asm/arch-tegra2/uart.h new file +44 -0

44 lines not on this page · raw message

board/nvidia/common/board.c new file +57 -0

57 lines not on this page · raw message

Signature · 1 line
1.7.3.5
#2 lore raw

PATCH v5 1/4 arm: Tegra2: Add basic NVIDIA Tegra2 SoC support

Dear Tom Warren,

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

...

...

Quoted 2 lines
arch/arm/include/asm/arch-tegra2/clk_rst.h new file +2 -0
+/* Clock/Reset Controller (CLK_RST_CONTROLLER_) regs */
+typedef volatile struct clk_rst_ctlr {

...

Quoted 2 lines
arch/arm/include/asm/arch-tegra2/pinmux.h new file +2 -0
+/* APB MISC Pin Mux and Tristate (APB_MISC_PP_) registers */
+typedef volatile struct pinmux_tri_ctlr {

...

Quoted 2 lines
arch/arm/include/asm/arch-tegra2/pmc.h new file +2 -0
+/* Power Management Controller (APBDEV_PMC_) registers */
+typedef volatile struct pmc_ctlr {

...

Quoted 4 lines
arch/arm/include/asm/arch-tegra2/tegra2.h new file +4 -0
+#ifndef __ASSEMBLY__
+typedef volatile struct timerus {
+	unsigned int cntr_1us;
+} timerus_t;

...

arch/arm/include/asm/arch-tegra2/uart.h line 28

Quoted 2 lines
arch/arm/include/asm/arch-tegra2/uart.h new file +2 -0
+/* UART registers */
+typedef volatile struct uart_ctlr {

Please drop the "typedef" and the "volatile" from all these (and all other, similar) declarations.

See linux/Documentation/CodingStyle:

It's a _mistake_ to use typedef for structures and pointers.

See Documentation/volatile-considered-harmful.txt

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
I am a computer. I am dumber than any human and smarter than any  ad-
ministrator.
author
#3 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

In reply to Tom Warren

---
Changes for V2:
- Move serial driver to separate patch

Changes for V5:
- Move arch/arm/cpu/armv7/uart.c & board.h to drivers/serial and rename to serial_tegra2.c
- Remove use of uart_num & UART_A/D in serial_tegra2, simplify code

 arch/arm/cpu/armv7/tegra2/Makefile |    2 +-
 arch/arm/cpu/armv7/tegra2/board.c  |    2 +-
 arch/arm/cpu/armv7/tegra2/board.h  |   58 ----------
 arch/arm/cpu/armv7/tegra2/uart.c   |  216 ------------------------------------
 common/serial.c                    |    3 +-
 drivers/serial/Makefile            |    1 +
 drivers/serial/serial_tegra2.c     |  205 ++++++++++++++++++++++++++++++++++
 drivers/serial/serial_tegra2.h     |   49 ++++++++
 include/serial.h                   |    3 +-
 9 files changed, 261 insertions(+), 278 deletions(-)
 delete mode 100644 arch/arm/cpu/armv7/tegra2/board.h
 delete mode 100644 arch/arm/cpu/armv7/tegra2/uart.c
 create mode 100644 drivers/serial/serial_tegra2.c
 create mode 100644 drivers/serial/serial_tegra2.h
arch/arm/cpu/armv7/tegra2/Makefile +1 -1
index f5b657b..687c887 100644
@@ -28,7 +28,7 @@ include $(TOPDIR)/config.mk
 LIB	=  $(obj)lib$(SOC).o
 
 SOBJS	:= lowlevel_init.o
-COBJS	:= board.o sys_info.o timer.o uart.o
+COBJS	:= board.o sys_info.o timer.o
 
 SRCS	:= $(SOBJS:.o=.S) $(COBJS:.o=.c)
 OBJS	:= $(addprefix $(obj),$(COBJS) $(SOBJS))
arch/arm/cpu/armv7/tegra2/board.c +1 -1
index 816a8cd..1e92d98 100644
@@ -25,7 +25,7 @@
 #include <asm/io.h>
 #include <asm/arch/sys_proto.h>
 #include <asm/arch/tegra2.h>
-#include "board.h"
+#include <asm/arch/pmc.h>
 
 DECLARE_GLOBAL_DATA_PTR;
 
arch/arm/cpu/armv7/tegra2/board.h deleted +0 -58

58 lines not on this page · raw message

arch/arm/cpu/armv7/tegra2/uart.c deleted +0 -216

216 lines not on this page · raw message

common/serial.c +2 -1
index 051ae4e..8ebf9a5 100644
@@ -41,7 +41,8 @@ struct serial_device *__default_serial_console (void)
 #elif defined(CONFIG_4xx) \
    || defined(CONFIG_MB86R0x) || defined(CONFIG_MPC5xxx) \
    || defined(CONFIG_MPC83xx) || defined(CONFIG_MPC85xx) \
-   || defined(CONFIG_MPC86xx) || defined(CONFIG_SYS_SC520)
+   || defined(CONFIG_MPC86xx) || defined(CONFIG_SYS_SC520) \
+   || defined(CONFIG_TEGRA2)
 #if defined(CONFIG_CONS_INDEX) && defined(CONFIG_SYS_NS16550_SERIAL)
 #if (CONFIG_CONS_INDEX==1)
 	return &eserial1_device;
drivers/serial/Makefile +1 -0
index 7d221fc..5a6011e 100644
@@ -55,6 +55,7 @@ COBJS-$(CONFIG_S3C24X0_SERIAL) += serial_s3c24x0.o
 COBJS-$(CONFIG_S3C44B0_SERIAL) += serial_s3c44b0.o
 COBJS-$(CONFIG_XILINX_UARTLITE) += serial_xuartlite.o
 COBJS-$(CONFIG_SCIF_CONSOLE) += serial_sh.o
+COBJS-$(CONFIG_TEGRA2) += serial_tegra2.o
 COBJS-$(CONFIG_USB_TTY) += usbtty.o
 
 COBJS	:= $(sort $(COBJS-y))
drivers/serial/serial_tegra2.c new file +205 -0

205 lines not on this page · raw message

drivers/serial/serial_tegra2.h new file +49 -0

49 lines not on this page · raw message

include/serial.h +2 -1
index 15ab73c..f21d961 100644
@@ -27,7 +27,8 @@ extern struct serial_device * default_serial_console (void);
     defined(CONFIG_405EP) || defined(CONFIG_405EZ) || defined(CONFIG_405EX) || \
     defined(CONFIG_MB86R0x) || defined(CONFIG_MPC5xxx) || \
     defined(CONFIG_MPC83xx) || defined(CONFIG_MPC85xx) || \
-    defined(CONFIG_MPC86xx) || defined(CONFIG_SYS_SC520)
+    defined(CONFIG_MPC86xx) || defined(CONFIG_SYS_SC520) || \
+    defined(CONFIG_TEGRA2)
 extern struct serial_device serial0_device;
 extern struct serial_device serial1_device;
 #if defined(CONFIG_SYS_NS16550_SERIAL)
Signature · 1 line
1.7.3.5
#4 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Hi Tom,

Quoted 10 lines · Tom Warren

---
Changes for V2:
- Move serial driver to separate patch

Changes for V5:
- Move arch/arm/cpu/armv7/uart.c & board.h to drivers/serial and rename to serial_tegra2.c
- Remove use of uart_num & UART_A/D in serial_tegra2, simplify code

Quoted 14 lines
 arch/arm/cpu/armv7/tegra2/Makefile |    2 +-
 arch/arm/cpu/armv7/tegra2/board.c  |    2 +-
 arch/arm/cpu/armv7/tegra2/board.h  |   58 ----------
 arch/arm/cpu/armv7/tegra2/uart.c   |  216 ------------------------------------
 common/serial.c                    |    3 +-
 drivers/serial/Makefile            |    1 +
 drivers/serial/serial_tegra2.c     |  205 ++++++++++++++++++++++++++++++++++
 drivers/serial/serial_tegra2.h     |   49 ++++++++
 include/serial.h                   |    3 +-
 9 files changed, 261 insertions(+), 278 deletions(-)
 delete mode 100644 arch/arm/cpu/armv7/tegra2/board.h
 delete mode 100644 arch/arm/cpu/armv7/tegra2/uart.c
 create mode 100644 drivers/serial/serial_tegra2.c
 create mode 100644 drivers/serial/serial_tegra2.h

It looks like arch/arm/cpu/armv7/tegra2/board.h and arch/arm/cpu/armv7/tegra2/uart.c are added in the first patch, then moved in this patch. It'd be ideal to just add them once in the proper location.

On a side note, if you pass "git format-patch" the -M and -C options it will make pretty diffs that only show what lines changed during a move. In the case that you do move files in the future its nice to use those options to ease review.

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+	/* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+	init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+	init_uart();
+#endif
+}

How about:

#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD)
	init_uart();

#endif

Best,
Peter

author
#5 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Peter,

Quoted 33 lines · Peter Tyser

Hi Tom,

Quoted 19 lines · Tom Warren

---
Changes for V2:
? ? ? - Move serial driver to separate patch

Changes for V5:
? ? ? - Move arch/arm/cpu/armv7/uart.c & board.h to drivers/serial and ? ? ? ? ? ? ? rename to serial_tegra2.c
? ? ? - Remove use of uart_num & UART_A/D in serial_tegra2, simplify code

?arch/arm/cpu/armv7/tegra2/Makefile | ? ?2 +-
?arch/arm/cpu/armv7/tegra2/board.c ?| ? ?2 +-
?arch/arm/cpu/armv7/tegra2/board.h ?| ? 58 ----------
?arch/arm/cpu/armv7/tegra2/uart.c ? | ?216 ------------------------------------
?common/serial.c ? ? ? ? ? ? ? ? ? ?| ? ?3 +-
?drivers/serial/Makefile ? ? ? ? ? ?| ? ?1 +
?drivers/serial/serial_tegra2.c ? ? | ?205 ++++++++++++++++++++++++++++++++++
?drivers/serial/serial_tegra2.h ? ? | ? 49 ++++++++
?include/serial.h ? ? ? ? ? ? ? ? ? | ? ?3 +-
Quoted 5 lines

?9 files changed, 261 insertions(+), 278 deletions(-) ?delete mode 100644 arch/arm/cpu/armv7/tegra2/board.h ?delete mode 100644 arch/arm/cpu/armv7/tegra2/uart.c ?create mode 100644 drivers/serial/serial_tegra2.c ?create mode 100644 drivers/serial/serial_tegra2.h

It looks like arch/arm/cpu/armv7/tegra2/board.h and arch/arm/cpu/armv7/tegra2/uart.c are added in the first patch, then moved in this patch. ?It'd be ideal to just add them once in the proper location.

Quoted 5 lines

On a side note, if you pass "git format-patch" the -M and -C options it will make pretty diffs that only show what lines changed during a move. In the case that you do move files in the future its nice to use those options to ease review.

I'll use those options in the future (thanks!) to keep things cleaner. Hopefully we can just go with this set of patches so I can get to the other, more interesting code (drivers, A9 CPU init, etc.).

Quoted 17 lines

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+ ? ? /* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+ ? ? init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+ ? ? init_uart();
+#endif
+}

How about:
#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD) ? ? ? ?init_uart();
#endif

That won't work for a board like Harmony where the developer wants both UARTs active, since uart_init is only called once.

Quoted 3 lines

Best,
Peter

Thanks,
Tom

Quoted 2 lines
#6 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

<snip>

Quoted 5 lines
Quoted 5 lines

It looks like arch/arm/cpu/armv7/tegra2/board.h and arch/arm/cpu/armv7/tegra2/uart.c are added in the first patch, then moved in this patch. It'd be ideal to just add them once in the proper location.

Quoted 8 lines
Quoted 5 lines

On a side note, if you pass "git format-patch" the -M and -C options it will make pretty diffs that only show what lines changed during a move. In the case that you do move files in the future its nice to use those options to ease review.

I'll use those options in the future (thanks!) to keep things cleaner. Hopefully we can just go with this set of patches so I can get to the other, more interesting code (drivers, A9 CPU init, etc.).

Its up to the ARM maintainer and Wolfgang to decide if adding code in one patch just to move it around in the 2nd is acceptable. I'm guess it won't be acceptable since its considered "bad form", and its unclear what patch in this series contains what functionality. Eg this isn't really meant to "Add Tegra2 serial port support", it moves existing serial port code around? And more? Its not really just adding serial port support as the patch title states in any case.

Quoted 17 lines
Quoted 14 lines

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+     /* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+     init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+     init_uart();
+#endif
+}

How about:

Quoted 3 lines
#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD)
       init_uart();

#endif

Quoted 2 lines

That won't work for a board like Harmony where the developer wants both UARTs active, since uart_init is only called once.

Why should init_uart() be called two times? It looks to initialize both ports, meaning it should only be called once, right?

Also, just noticed:
+static void init_uart(void)
+{
+#if CONFIG_TEGRA2_ENABLE_UARTA
+ uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTA_BASE;
+
+ uart_clock_init();
+
+ /* Enable UARTA - uses config 0 */
+ pin_mux_uart();
+
+ setup_uart(uart);
+#endif /* CONFIG_TEGRA2_ENABLE_UARTD */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+ uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTD_BASE;
+

Have you compiled with both UARTA and UARTD enabled? Redeclaring 'uart' in the middle of the function should throw an error or warning.

+ uart_clock_init();
+
+ /* Enable UARTD - uses config 0 */
+ pin_mux_uart();
+
+ setup_uart(uart);
+#endif /* CONFIG_TEGRA2_ENABLE_UARTD */
+}

Best,
Peter

author
#7 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Peter,

Quoted 16 lines · Peter Tyser

<snip>

Quoted 5 lines
Quoted 5 lines

It looks like arch/arm/cpu/armv7/tegra2/board.h and arch/arm/cpu/armv7/tegra2/uart.c are added in the first patch, then moved in this patch. ?It'd be ideal to just add them once in the proper location.

Quoted 8 lines
Quoted 5 lines

On a side note, if you pass "git format-patch" the -M and -C options it will make pretty diffs that only show what lines changed during a move. In the case that you do move files in the future its nice to use those options to ease review.

I'll use those options in the future (thanks!) to keep things cleaner. Hopefully we can just go with this set of patches so I can get to the other, more interesting code (drivers, A9 CPU init, etc.).

Quoted 7 lines

Its up to the ARM maintainer and Wolfgang to decide if adding code in one patch just to move it around in the 2nd is acceptable. ?I'm guess it won't be acceptable since its considered "bad form", and its unclear what patch in this series contains what functionality. ?Eg this isn't really meant to "Add Tegra2 serial port support", it moves existing serial port code around? ?And more? ?Its not really just adding serial port support as the patch title states in any case.

I see what you're talking about now - I've got uart.c in 2 patch files - created in 0001 and then moved in 0002. My bad - that wasn't the intent, just what happened when I applied my V4 patches to a new branch to get the V5 patchset built. I'll fix it and resubmit.

As to 0002 not adding serial port support for Tegra2, that's all it does - adds TEGRA2 defines to serial.h/serial.c for the eserial* tables, and then adds code to turn on Tegra2-specific UART HW. If I remove any mention of uart.c in patch 0001 (add basic Tegra2 support), then does patch 0002 make sense to you?

Quoted 18 lines
Quoted 18 lines
Quoted 18 lines

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+ ? ? /* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+ ? ? init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+ ? ? init_uart();
+#endif
+}

How about:
#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD) ? ? ? ?init_uart();
#endif

Quoted 5 lines
Quoted 2 lines

That won't work for a board like Harmony where the developer wants both UARTs active, since uart_init is only called once.

Why should init_uart() be called two times? ?It looks to initialize both ports, meaning it should only be called once, right?

Correct, again (need more coffee!) Your if defined construct wouldn't work as written, though, because CONFIG_TEGRA2_ENABLE_UARTx is always defined (as 0 or 1). I'd have to rework it.

Quoted 18 lines

Also, just noticed:

drivers/serial/serial_tegra2.c new file +15 -0
+static void init_uart(void)
+{
+#if CONFIG_TEGRA2_ENABLE_UARTA
+ ? ? ? ? ? ? ? uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTA_BASE;
+
+ ? ? ? ? ? ? ? uart_clock_init();
+
+ ? ? ? ? ? ? ? /* Enable UARTA - uses config 0 */
+ ? ? ? ? ? ? ? pin_mux_uart();
+
+ ? ? ? ? ? ? ? setup_uart(uart);
+#endif /* CONFIG_TEGRA2_ENABLE_UARTD */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+ ? ? ? ? ? ? ? uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTD_BASE;
+
Quoted 2 lines

Have you compiled with both UARTA and UARTD enabled? ?Redeclaring 'uart' in the middle of the function should throw an error or warning.

I'd tested with both enabled earlier, but maybe not since the rewrite. I'll check & resubmit.

Quoted 10 lines

+ ? ? ? ? ? ? ? uart_clock_init();

drivers/serial/serial_tegra2.c new file +7 -0
+
+ ? ? ? ? ? ? ? /* Enable UARTD - uses config 0 */
+ ? ? ? ? ? ? ? pin_mux_uart();
+
+ ? ? ? ? ? ? ? setup_uart(uart);
+#endif /* CONFIG_TEGRA2_ENABLE_UARTD */
+}
Quoted 2 lines

Best,
Peter

Thanks, again, for your thoroughness!
Tom

Quoted 2 lines
#8 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Dear Tom Warren,

Quoted 4 lines · In message <[email protected]> you…

Correct, again (need more coffee!) Your if defined construct wouldn't work as written, though, because CONFIG_TEGRA2_ENABLE_UARTx is always defined (as 0 or 1). I'd have to rework it.

Please try and avoid relying on specific values.
Rather use "#if defined".

Best regards,

Wolfgang Denk

Signature · 6 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
"We have the right to survive!"
"Not be killing others."
	-- Deela and Kirk, "Wink of An Eye", stardate 5710.5
#9 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

In reply to Tom Warren

<snip>

Quoted 10 lines

I see what you're talking about now - I've got uart.c in 2 patch files - created in 0001 and then moved in 0002. My bad - that wasn't the intent, just what happened when I applied my V4 patches to a new branch to get the V5 patchset built. I'll fix it and resubmit.

As to 0002 not adding serial port support for Tegra2, that's all it does - adds TEGRA2 defines to serial.h/serial.c for the eserial* tables, and then adds code to turn on Tegra2-specific UART HW. If I remove any mention of uart.c in patch 0001 (add basic Tegra2 support), then does patch 0002 make sense to you?

Yeah, that'd make more sense. Patch 2 would just contain:

 common/serial.c                    |    3 +-
 drivers/serial/Makefile            |    1 +
 drivers/serial/serial_tegra2.c     |  205 ++++++++++++++++++++++++++++++++++
 drivers/serial/serial_tegra2.h     |   49 ++++++++
 include/serial.h                   |    3 +-
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines
Quoted 17 lines

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+     /* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+     init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+     init_uart();
+#endif
+}

How about:

#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD)
       init_uart();

#endif

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

That won't work for a board like Harmony where the developer wants both UARTs active, since uart_init is only called once.

Why should init_uart() be called two times? It looks to initialize both ports, meaning it should only be called once, right?

Correct, again (need more coffee!) Your if defined construct wouldn't work as written, though, because CONFIG_TEGRA2_ENABLE_UARTx is always defined (as 0 or 1). I'd have to rework it.

You could also just get rid of uart_init() altogether and rename init_uart() to uart_init(). That would get rid of some idefs and simplify the flow.

Best,
Peter

author
#10 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Peter,

Quoted 13 lines · Peter Tyser

<snip>

Quoted 10 lines

I see what you're talking about now - I've got uart.c in 2 patch files - created in 0001 and then moved in 0002. My bad - that wasn't the intent, just what happened when I applied my V4 patches to a new branch to get the V5 patchset built. ?I'll fix it and resubmit.

As to 0002 not adding serial port support for Tegra2, that's all it does - adds TEGRA2 defines to serial.h/serial.c for the eserial* tables, and then adds code to turn on Tegra2-specific UART HW. ?If I remove any mention of uart.c in patch 0001 (add basic Tegra2 support), then does patch 0002 make sense to you?

Quoted 7 lines

Yeah, that'd make more sense. ?Patch 2 would just contain:

?common/serial.c ? ? ? ? ? ? ? ? ? ?| ? ?3 +-
?drivers/serial/Makefile ? ? ? ? ? ?| ? ?1 +
?drivers/serial/serial_tegra2.c ? ? | ?205 ++++++++++++++++++++++++++++++++++
?drivers/serial/serial_tegra2.h ? ? | ? 49 ++++++++
?include/serial.h ? ? ? ? ? ? ? ? ? | ? ?3 +-

Exactly what I was thinking. I'll try to get it right in patch V6.

Quoted 22 lines
Quoted 22 lines
Quoted 22 lines
Quoted 17 lines
Quoted 17 lines

<snip>

+void uart_init(void)

drivers/serial/serial_tegra2.c line 205

Quoted 9 lines
drivers/serial/serial_tegra2.c new file +9 -0
+{
+ ? ? /* Init each UART - there may be more than 1 on a board/build */
+#if (CONFIG_TEGRA2_ENABLE_UARTA)
+ ? ? init_uart();
+#endif
+#if (CONFIG_TEGRA2_ENABLE_UARTD)
+ ? ? init_uart();
+#endif
+}

How about:
#if defined(CONFIG_TEGRA2_ENABLE_UARTA) || defined(CONFIG_TEGRA2_ENABLE_UARTD) ? ? ? ?init_uart();
#endif

Quoted 2 lines

That won't work for a board like Harmony where the developer wants both UARTs active, since uart_init is only called once.

Why should init_uart() be called two times? ?It looks to initialize both ports, meaning it should only be called once, right?

Quoted 7 lines
Quoted 3 lines

Correct, again (need more coffee!) ?Your if defined construct wouldn't work as written, though, because CONFIG_TEGRA2_ENABLE_UARTx is always defined (as 0 or 1). I'd have to rework it.

You could also just get rid of uart_init() altogether and rename init_uart() to uart_init(). ?That would get rid of some idefs and simplify the flow.

Yeah, I saw that as I was cleaning up the indentation & reworking the code to compile
with both UARTs defined. I'll get rid of uart_init (renamed to init_uart). Thanks.

Quoted 3 lines

Best,
Peter

Thanks,

Tom

Quoted 2 lines
#11 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

In reply to Peter Tyser

Dear Peter Tyser,

Quoted 7 lines · In message <1295891506.2045.146.camel@petert> you wrote:
Quoted 4 lines

I'll use those options in the future (thanks!) to keep things cleaner. Hopefully we can just go with this set of patches so I can get to the other, more interesting code (drivers, A9 CPU init, etc.).

Its up to the ARM maintainer and Wolfgang to decide if adding code in one patch just to move it around in the 2nd is acceptable. I'm guess it

No, it is not acceptable. This should be fixed. Thanks for pointing out.

Quoted 17 lines

Also, just noticed:

drivers/serial/serial_tegra2.c new file +15 -0
+static void init_uart(void)
+{
+#if CONFIG_TEGRA2_ENABLE_UARTA
+               uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTA_BASE;
+
+               uart_clock_init();
+
+               /* Enable UARTA - uses config 0 */
+               pin_mux_uart();
+
+               setup_uart(uart);
+#endif /* CONFIG_TEGRA2_ENABLE_UARTD */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+               uart_ctlr *const uart = (uart_ctlr *)NV_PA_APB_UARTD_BASE;
+
Quoted 2 lines

Have you compiled with both UARTA and UARTD enabled? Redeclaring 'uart' in the middle of the function should throw an error or warning.

Even if the compiler should accept it (which I hpe not), then I will not accept this ;-) Thanks for pointing out.

Best regards,

Wolfgang Denk

Signature · 7 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 an old custom among my people. When a woman  saves  a  man's
life, he is grateful.
	-- Nona, the Kanuto which woman, "A Private Little War",
	   stardate 4211.8.
#12 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

In reply to Tom Warren
Quoted 2 lines · Tom Warren

---

[ snip ]

Quoted 18 lines
drivers/serial/serial_tegra2.c new file +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+	pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+	u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+		reg = readl(&pmt->pmt_ctl_c);
+		reg &= 0xFFF0FFFF;	/* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+		writel(reg, &pmt->pmt_ctl_c);
+
+		reg = readl(&pmt->pmt_tri_a);
+		reg &= ~Z_IRRX;		/* Z_IRRX = normal (0) */
+		reg &= ~Z_IRTX;		/* Z_IRTX = normal (0) */
+		writel(reg, &pmt->pmt_tri_a);

drivers/serial/serial_tegra2.c line 133

Quoted 10 lines
drivers/serial/serial_tegra2.c new file +10 -0
+#endif	/* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+		reg = readl(&pmt->pmt_ctl_b);
+		reg &= 0xFFFFFFF3;	/* GMC_SEL [3:2] = 00, UARTD */
+		writel(reg, &pmt->pmt_ctl_b);
+
+		reg = readl(&pmt->pmt_tri_a);
+		reg &= ~Z_GMC;		/* Z_GMC = normal (0) */
+		writel(reg, &pmt->pmt_tri_a);
+#endif	/* CONFIG_TEGRA2_ENABLE_UARTD */

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

(1) http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/92887/focus=93233

Signature · 2 lines
Sincerely yours,
Mike.
author
#13 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Mike,

Quoted 35 lines · Mike Rapoport
Quoted 2 lines · Tom Warren

---

[ snip ]

Quoted 18 lines
(quoted diff) +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+ ? ? pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+ ? ? u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_c);
+ ? ? ? ? ? ? reg &= 0xFFF0FFFF; ? ? ?/* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_c);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_IRRX; ? ? ? ? /* Z_IRRX = normal (0) */
+ ? ? ? ? ? ? reg &= ~Z_IRTX; ? ? ? ? /* Z_IRTX = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);

Review

Quoted 10 lines
(quoted diff) +10 -0
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_b);
+ ? ? ? ? ? ? reg &= 0xFFFFFFF3; ? ? ?/* GMC_SEL [3:2] = 00, UARTD */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_b);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_GMC; ? ? ? ? ?/* Z_GMC = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTD */
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

If there are examples that you approve of in any extant U-Boot code (omap, ti91, davinci, etc.), please point them out. I'd really like to finalize this patch series.

Quoted 6 lines

(1) http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/92887/focus=93233

--
Sincerely yours,
Mike.

Thanks,

Tom

Quoted 1 line
#14 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Quoted 46 lines · Tom Warren

Mike,

Quoted 35 lines · Mike Rapoport
Quoted 2 lines · Tom Warren

---

[ snip ]

Quoted 18 lines
drivers/serial/serial_tegra2.c new file +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+     pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+     u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+             reg = readl(&pmt->pmt_ctl_c);
+             reg &= 0xFFF0FFFF;      /* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+             writel(reg, &pmt->pmt_ctl_c);
+
+             reg = readl(&pmt->pmt_tri_a);
+             reg &= ~Z_IRRX;         /* Z_IRRX = normal (0) */
+             reg &= ~Z_IRTX;         /* Z_IRTX = normal (0) */
+             writel(reg, &pmt->pmt_tri_a);

drivers/serial/serial_tegra2.c line 133

Quoted 10 lines
drivers/serial/serial_tegra2.c new file +10 -0
+#endif       /* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+             reg = readl(&pmt->pmt_ctl_b);
+             reg &= 0xFFFFFFF3;      /* GMC_SEL [3:2] = 00, UARTD */
+             writel(reg, &pmt->pmt_ctl_b);
+
+             reg = readl(&pmt->pmt_tri_a);
+             reg &= ~Z_GMC;          /* Z_GMC = normal (0) */
+             writel(reg, &pmt->pmt_tri_a);
+#endif       /* CONFIG_TEGRA2_ENABLE_UARTD */
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

Quoted 3 lines

If there are examples that you approve of in any extant U-Boot code (omap, ti91, davinci, etc.), please point them out. I'd really like to finalize this patch series.

I think OMAP is a good example. There's set_muxconf_regs function in each board file that is responsible for configuration of all the pins.

Quoted 10 lines
Quoted 6 lines

(1) http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/92887/focus=93233

--
Sincerely yours,
Mike.

Thanks,

Tom

Quoted 1 line
Signature · 2 lines
Sincerely yours,
Mike.
author
#15 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Mike,

Quoted 58 lines · Mike Rapoport
Quoted 46 lines · Tom Warren

Mike,

Quoted 35 lines · Mike Rapoport
Quoted 2 lines · Tom Warren

---

[ snip ]

Quoted 18 lines
(quoted diff) +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+ ? ? pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+ ? ? u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_c);
+ ? ? ? ? ? ? reg &= 0xFFF0FFFF; ? ? ?/* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_c);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_IRRX; ? ? ? ? /* Z_IRRX = normal (0) */
+ ? ? ? ? ? ? reg &= ~Z_IRTX; ? ? ? ? /* Z_IRTX = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);

Review

Quoted 10 lines
(quoted diff) +10 -0
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_b);
+ ? ? ? ? ? ? reg &= 0xFFFFFFF3; ? ? ?/* GMC_SEL [3:2] = 00, UARTD */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_b);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_GMC; ? ? ? ? ?/* Z_GMC = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTD */
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

Quoted 3 lines

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Quoted 7 lines
Quoted 4 lines

If there are examples that you approve of in any extant U-Boot code (omap, ti91, davinci, etc.), please point them out. ?I'd really like to finalize this patch series.

I think OMAP is a good example. There's set_muxconf_regs function in each board file that is responsible for configuration of all the pins.

I'll take a look. Thanks.

Quoted 17 lines
Quoted 11 lines
Quoted 7 lines

(1) http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/92887/focus=93233

--
Sincerely yours,
Mike.

Thanks,

Tom

Quoted 1 line

--
Sincerely yours,
Mike.

author
#16 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Mike et al,

Quoted 61 lines · Tom Warren

Mike,

Quoted 58 lines · Mike Rapoport
Quoted 56 lines · Tom Warren

Mike,

Quoted 35 lines · Mike Rapoport
Quoted 2 lines · Tom Warren

---

[ snip ]

Quoted 18 lines
(quoted diff) +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+ ? ? pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+ ? ? u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_c);
+ ? ? ? ? ? ? reg &= 0xFFF0FFFF; ? ? ?/* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_c);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_IRRX; ? ? ? ? /* Z_IRRX = normal (0) */
+ ? ? ? ? ? ? reg &= ~Z_IRTX; ? ? ? ? /* Z_IRTX = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);

Review

Quoted 10 lines
(quoted diff) +10 -0
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_ctl_b);
+ ? ? ? ? ? ? reg &= 0xFFFFFFF3; ? ? ?/* GMC_SEL [3:2] = 00, UARTD */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_ctl_b);
+
+ ? ? ? ? ? ? reg = readl(&pmt->pmt_tri_a);
+ ? ? ? ? ? ? reg &= ~Z_GMC; ? ? ? ? ?/* Z_GMC = normal (0) */
+ ? ? ? ? ? ? writel(reg, &pmt->pmt_tri_a);
+#endif ? ? ? /* CONFIG_TEGRA2_ENABLE_UARTD */
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

Quoted 6 lines
Quoted 3 lines

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Actually, I think it makes more sense to move pinmux_init_uart and clock_init_uart to board/nvidia/common/board.c.
These routines are more board-specific than SoC-specific - they depend on how the UART is routed on a board.
Let me do that & gen up a new patchset.

Quoted 27 lines
Quoted 8 lines
Quoted 5 lines

If there are examples that you approve of in any extant U-Boot code (omap, ti91, davinci, etc.), please point them out. ?I'd really like to finalize this patch series.

I think OMAP is a good example. There's set_muxconf_regs function in each board file that is responsible for configuration of all the pins.

I'll take a look. Thanks.

Quoted 17 lines
Quoted 11 lines
Quoted 7 lines

(1) http://thread.gmane.org/gmane.comp.boot-loaders.u-boot/92887/focus=93233

--
Sincerely yours,
Mike.

Thanks,

Tom

Quoted 1 line

--
Sincerely yours,
Mike.

#17 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Quoted 53 lines
Quoted 53 lines
Quoted 53 lines
Quoted 39 lines
Quoted 31 lines

[ snip ]

Quoted 18 lines
drivers/serial/serial_tegra2.c new file +18 -0
+/*
+ * Routine: pin_mux_uart
+ * Description: setup the pin muxes/tristate values for a UART
+ */
+static void pin_mux_uart(void)
+{
+     pinmux_tri_ctlr *const pmt = (pinmux_tri_ctlr *)NV_PA_APB_MISC_BASE;
+     u32 reg;
+
+#if CONFIG_TEGRA2_ENABLE_UARTA
+             reg = readl(&pmt->pmt_ctl_c);
+             reg &= 0xFFF0FFFF;      /* IRRX_/IRTX_SEL [19:16] = 00 UARTA */
+             writel(reg, &pmt->pmt_ctl_c);
+
+             reg = readl(&pmt->pmt_tri_a);
+             reg &= ~Z_IRRX;         /* Z_IRRX = normal (0) */
+             reg &= ~Z_IRTX;         /* Z_IRTX = normal (0) */
+             writel(reg, &pmt->pmt_tri_a);

drivers/serial/serial_tegra2.c line 133

Quoted 10 lines
drivers/serial/serial_tegra2.c new file +10 -0
+#endif       /* CONFIG_TEGRA2_ENABLE_UARTA */
+#if CONFIG_TEGRA2_ENABLE_UARTD
+             reg = readl(&pmt->pmt_ctl_b);
+             reg &= 0xFFFFFFF3;      /* GMC_SEL [3:2] = 00, UARTD */
+             writel(reg, &pmt->pmt_ctl_b);
+
+             reg = readl(&pmt->pmt_tri_a);
+             reg &= ~Z_GMC;          /* Z_GMC = normal (0) */
+             writel(reg, &pmt->pmt_tri_a);
+#endif       /* CONFIG_TEGRA2_ENABLE_UARTD */
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

Quoted 8 lines
Quoted 3 lines

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Actually, I think it makes more sense to move pinmux_init_uart and clock_init_uart to board/nvidia/common/board.c.
These routines are more board-specific than SoC-specific - they depend on how the UART is routed on a board.
Let me do that & gen up a new patchset.

You previously mentioned that "To date, all of our Tegra boards use these pinmux options for both UARTs. If a board vendor chooses to use different pinmuxes, then they can override these funcs in their board files, or use their own code triggered by their own defines. But according to our HW guys, the vast majority will use these pins"

If the vast majority of boards really will use the same pinmuxing, it would be nice to provide that common muxing as a default which can be overridden. Perhaps moving pinmux_init_uart() and uart_clock_init() into armv7/tegra/*, and making them weak functions would make everyone happy.

Or could you make default defines that board's could override as needed? eg in pin_mux_uart():
#ifndef CONFIG_TEGRA2_PMT_CTLC_MASK
#define CONFIG_TEGRA2_PMC_CTLC_MASK 0xfff0ffff
#endif
#ifndef CONFIG_TEGRA2_PMT_TRI_A_MASK
#define CONFIG_TEGRA2_PMC_TRI_A_MASK ~(Z_IRRX | Z_IRTX) #endif

pin_mux_uart() {
	reg = readl(&pmt->pmt_ctl_c);
	reg &= CONFIG_TEGRA2_PMT_CTLC_MASK;
	writel(reg, &pmt->pmt_ctl_c);

	reg = readl(&pmt->pmt_tri_a);
	reg &= CONFIG_TEGRA2_PMC_TRI_A_MASK;
	writel(reg, &pmt->pmt_tri_a);
}

Or make the functions table driven so each board would define a pinmux/clock configuration table?

I think its worthwhile to try and reduce duplicated code whenever possible.

Best,
Peter

#18 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Quoted 37 lines · Peter Tyser
Quoted 22 lines
Quoted 22 lines
Quoted 22 lines
Quoted 8 lines
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

Quoted 8 lines
Quoted 3 lines

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Actually, I think it makes more sense to move pinmux_init_uart and clock_init_uart to board/nvidia/common/board.c.
These routines are more board-specific than SoC-specific - they depend on how the UART is routed on a board.
Let me do that & gen up a new patchset.

You previously mentioned that "To date, all of our Tegra boards use these pinmux options for both UARTs. If a board vendor chooses to use different pinmuxes, then they can override these funcs in their board files, or use their own code triggered by their own defines. But according to our HW guys, the vast majority will use these pins"

Quoted 5 lines

If the vast majority of boards really will use the same pinmuxing, it would be nice to provide that common muxing as a default which can be overridden. Perhaps moving pinmux_init_uart() and uart_clock_init() into armv7/tegra/*, and making them weak functions would make everyone happy.

My point was that pin muxing belongs to the board code rather than to the driver. Driver should just assume that pins are configured elsewhere and it does not need to deal with pin muxing at all.
Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

Quoted 27 lines

Or could you make default defines that board's could override as needed? eg in pin_mux_uart():
#ifndef CONFIG_TEGRA2_PMT_CTLC_MASK
#define CONFIG_TEGRA2_PMC_CTLC_MASK 0xfff0ffff
#endif
#ifndef CONFIG_TEGRA2_PMT_TRI_A_MASK
#define CONFIG_TEGRA2_PMC_TRI_A_MASK ~(Z_IRRX | Z_IRTX) #endif

pin_mux_uart() {
	reg = readl(&pmt->pmt_ctl_c);
	reg &= CONFIG_TEGRA2_PMT_CTLC_MASK;
	writel(reg, &pmt->pmt_ctl_c);

	reg = readl(&pmt->pmt_tri_a);
	reg &= CONFIG_TEGRA2_PMC_TRI_A_MASK;
	writel(reg, &pmt->pmt_tri_a);
}

Or make the functions table driven so each board would define a pinmux/clock configuration table?

I think its worthwhile to try and reduce duplicated code whenever possible.

Best,
Peter

Signature · 2 lines
Sincerely yours,
Mike.
#19 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Quoted 44 lines · Mike Rapoport
Quoted 37 lines · Peter Tyser
Quoted 22 lines
Quoted 22 lines
Quoted 22 lines
Quoted 8 lines
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

Quoted 8 lines
Quoted 3 lines

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Actually, I think it makes more sense to move pinmux_init_uart and clock_init_uart to board/nvidia/common/board.c.
These routines are more board-specific than SoC-specific - they depend on how the UART is routed on a board.
Let me do that & gen up a new patchset.

You previously mentioned that "To date, all of our Tegra boards use these pinmux options for both UARTs. If a board vendor chooses to use different pinmuxes, then they can override these funcs in their board files, or use their own code triggered by their own defines. But according to our HW guys, the vast majority will use these pins"

Quoted 5 lines

If the vast majority of boards really will use the same pinmuxing, it would be nice to provide that common muxing as a default which can be overridden. Perhaps moving pinmux_init_uart() and uart_clock_init() into armv7/tegra/*, and making them weak functions would make everyone happy.

Quoted 3 lines

My point was that pin muxing belongs to the board code rather than to the driver. Driver should just assume that pins are configured elsewhere and it does not need to deal with pin muxing at all.

I understand your point, but think if 9/10 boards use the same pin muxing its good to provide a default so we don't have 9 chunks of duplicate code floating around in board/*. What's the harm in providing a default? It can be overridden if needed.

Quoted 3 lines

Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

I can see that point but its a different discussion. I don't know enough about the Tegra2 to comment on this, but it seems like a good idea based on previous experiences with boards with similar pinmuxing (eg mpc8260). In my last email I mentioned a table-driven approach (similar to the mpc8260 implementation?) which sounds somewhat like you're proposing.

Best,
Peter

#20 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Quoted 7 lines · Peter Tyser
Quoted 5 lines · Mike Rapoport

On 01/26/11 00:24, Peter Tyser wrote:

Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

Quoted 6 lines

I can see that point but its a different discussion. I don't know enough about the Tegra2 to comment on this, but it seems like a good idea based on previous experiences with boards with similar pinmuxing (eg mpc8260). In my last email I mentioned a table-driven approach (similar to the mpc8260 implementation?) which sounds somewhat like you're proposing.

Yes, I was thinking about table-driven approach, though slightly more complicated than in mpc8260 :)
I think we can mostly reuse the Linux implementation of Tegra2 pinmux (1 and 2):
However, the pinmuxing API can be added after the basic Tegra2 support is merged, IMO.

(1)
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=arch/arm/mach-tegra/pinmux.c;h=f80d507671bc1b22921ab2821c8a2cde427f1b5f;hb=HEAD
(2)
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=arch/arm/mach-tegra/pinmux-t2-tables.c;h=a6ea34e782dc339091b49e54871f953048651ad3;hb=HEAD

Quoted 3 lines

Best,
Peter

Signature · 2 lines
Sincerely yours,
Mike.
author
#21 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

In reply to Mike Rapoport

Mike,

Quoted 44 lines · Mike Rapoport
Quoted 37 lines · Peter Tyser
Quoted 22 lines
Quoted 22 lines
Quoted 22 lines
Quoted 8 lines
Quoted 8 lines

As I've already pointed out (1) this covers only one possibility of UART pin muxing options. I agree that it makes sense when this is a part of the board-specific code. However, forcing specific pins configuration in the generic driver seems problematic to me, even if most Tegra2 boards use the same configuration.
As a side note, I'd prefer to have all the pin multiplexing defined in one place in board file rather than scattered among different drivers.

Quoted 10 lines

Alright. I'd like to get this wrapped up and accepted before the merge window closes so I can get on with the important bits (drivers, etc.). What would you like
here? A generic function like 'pinmux_set_config(reg, val, bit)' that lives in the board
files and gets called from the driver code with specifics of that board's/drivers pinmux
config? Or do you see the pinmux code living entirely in the board files, and being done
as part of board init? That's where it was before, BTW.

I think that the pinmux code should live entirely in the board file and performed as part of board init. The drivers than can assume that the pins are configured properly.

Quoted 8 lines
Quoted 3 lines

OK. I'll move the clock/pinmux init code into armv7/tegra/board.c (since it's SoC-centric),
and call it during board_init().

Actually, I think it makes more sense to move pinmux_init_uart and clock_init_uart to board/nvidia/common/board.c.
These routines are more board-specific than SoC-specific - they depend on how the UART is routed on a board.
Let me do that & gen up a new patchset.

You previously mentioned that "To date, all of our Tegra boards use these pinmux options for both UARTs. ?If a board vendor chooses to use different pinmuxes, then they can override these funcs in their board files, or use their own code triggered by their own defines. But according to our HW guys, the vast majority will use these pins"

Quoted 5 lines

If the vast majority of boards really will use the same pinmuxing, it would be nice to provide that common muxing as a default which can be overridden. ?Perhaps moving pinmux_init_uart() and uart_clock_init() into armv7/tegra/*, and making them weak functions would make everyone happy.

Quoted 3 lines

My point was that pin muxing belongs to the board code rather than to the driver. Driver should just assume that pins are configured elsewhere and it does not need to deal with pin muxing at all.

I understand that point - sorry if I wasn't clear. No objection to having pinmux code in board files.

Quoted 4 lines

Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

I see a couple of reasons not to do it that way. First, I don't know at this time what all the pinmux settings will be, since I haven't ported all the periph driver code yet. It's vastly different from what's acceptable in U-Boot, and will all need significant rewrite. It'd take me a week to gather all that info, and I'm not at full BW on this project (one of 4 on my plate right now).
Second, I've been chastised before for including code/features in this initial patchset that aren't needed or used. I'm trying to keep the code as simple as possible to make it easier on reviewers and get through the review in as short a time as possible. This has already dragged on far longer than I thought it would.
I'm willing to change the pinmux code to make it as generic as possible, but only if there's a consensus on the list that it has to be that way to get accepted & pushed.

Quoted 24 lines
Quoted 24 lines

Or could you make default defines that board's could override as needed? eg in pin_mux_uart():
#ifndef CONFIG_TEGRA2_PMT_CTLC_MASK
#define CONFIG_TEGRA2_PMC_CTLC_MASK 0xfff0ffff
#endif
#ifndef CONFIG_TEGRA2_PMT_TRI_A_MASK
#define CONFIG_TEGRA2_PMC_TRI_A_MASK ~(Z_IRRX | Z_IRTX) #endif
pin_mux_uart() {
? ? ? reg = readl(&pmt->pmt_ctl_c);
? ? ? reg &= CONFIG_TEGRA2_PMT_CTLC_MASK;
? ? ? writel(reg, &pmt->pmt_ctl_c);

? ? ? reg = readl(&pmt->pmt_tri_a);
? ? ? reg &= CONFIG_TEGRA2_PMC_TRI_A_MASK;
? ? ? writel(reg, &pmt->pmt_tri_a);
}

Or make the functions table driven so each board would define a pinmux/clock configuration table?

I think its worthwhile to try and reduce duplicated code whenever possible.

Quoted 8 lines
Quoted 3 lines

Best,
Peter

--
Sincerely yours,
Mike.

Thanks,
Tom

#22 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Tom,

Quoted 13 lines · Tom Warren

Mike,

Quoted 3 lines · Mike Rapoport

My point was that pin muxing belongs to the board code rather than to the driver. Driver should just assume that pins are configured elsewhere and it does not need to deal with pin muxing at all.

I understand that point - sorry if I wasn't clear. No objection to having pinmux code in board files.

Quoted 4 lines

Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

Quoted 14 lines

I see a couple of reasons not to do it that way. First, I don't know at this time what all the pinmux settings will be, since I haven't ported all the periph driver code yet. It's vastly different from what's acceptable in U-Boot, and will all need significant rewrite. It'd take me a week to gather all that info, and I'm not at full BW on this project (one of 4 on my plate right now).
Second, I've been chastised before for including code/features in this initial patchset that aren't needed or used. I'm trying to keep the code as simple as possible to make it easier on reviewers and get through the review in as short a time as possible. This has already dragged on far longer than I thought it would.
I'm willing to change the pinmux code to make it as generic as possible, but only if there's a consensus on the list that it has to be that way to get accepted & pushed.

I'm Ok with pinmux_init_uart in the board code for now. I think that the generic pinmux functionality can be added afterwards.

Quoted 3 lines
Quoted 1 line

Thanks,
Tom

Signature · 2 lines
Sincerely yours,
Mike.
author
#23 lore raw

PATCH v5 2/4 serial: Add Tegra2 serial port support

Mike,

Quoted 31 lines · Mike Rapoport

Tom,

Quoted 13 lines · Tom Warren

Mike,

Quoted 3 lines · Mike Rapoport

My point was that pin muxing belongs to the board code rather than to the driver. Driver should just assume that pins are configured elsewhere and it does not need to deal with pin muxing at all.

I understand that point - sorry if I wasn't clear. No objection to having pinmux code in board files.

Quoted 4 lines

Moreover, I'd prefer to see pinmux_board_init or something similar that configures all the pins at once rather than collection of pinmux_init_uart, pinmux_init_sdmmc, pinmux_init_gmi etc that will grow as more drivers are added.

Quoted 14 lines

I see a couple of reasons not to do it that way. First, I don't know at this time what all the pinmux settings will be, since I haven't ported all the periph driver code yet. It's vastly different from what's acceptable in U-Boot, and will all need significant rewrite. It'd take me a week to gather all that info, and I'm not at full BW on this project (one of 4 on my plate right now).
Second, I've been chastised before for including code/features in this initial patchset that aren't needed or used. ?I'm trying to keep the code as simple as possible to make it easier on reviewers and get through the review in as short a time as possible. This has already dragged on far longer than I thought it would.
I'm willing to change the pinmux code to make it as generic as possible, but only if there's a consensus on the list that it has to be that way to get accepted & pushed.

Quoted 3 lines

I'm Ok with pinmux_init_uart in the board code for now. I think that the generic pinmux functionality can be added afterwards.

Great. My main goal is to get the baseline Tegra2 support in upstream U-Boot. I'm happy to refine the pinmux approach in the next patchset. Either a generic pinmux as you propose, or a set of weak pinmux configs that can be overridden by another board design, etc. as Peter proposes.

Quoted 9 lines
Quoted 3 lines
Quoted 1 line

Thanks,
Tom

--
Sincerely yours,
Mike.

Thanks,
Tom

author
#24 lore raw

PATCH v5 3/4 arm: Tegra2: Add support for NVIDIA Harmony board

In reply to Tom Warren

---
Changes for V2:
- Use board/nvidia/ instead of /board/tegra
- Change nv-common.h config file to tegra2-common.h

 MAINTAINERS                     |    4 +
 board/nvidia/harmony/Makefile   |   50 ++++++++++++
 boards.cfg                      |    1 +
 include/configs/harmony.h       |   48 ++++++++++++
 include/configs/tegra2-common.h |  160 +++++++++++++++++++++++++++++++++++++++
 5 files changed, 263 insertions(+), 0 deletions(-)
 create mode 100644 board/nvidia/harmony/Makefile
 create mode 100644 include/configs/harmony.h
 create mode 100644 include/configs/tegra2-common.h
MAINTAINERS +4 -0
index ba83f71..b5cff19 100644
@@ -841,6 +841,10 @@ Prafulla Wadaskar <[email protected]>
 	rd6281a		ARM926EJS (Kirkwood SoC)
 	sheevaplug	ARM926EJS (Kirkwood SoC)
 
+Tom Warren <[email protected]>
+
+	harmony		Tegra2 (ARM7 & A9 Dual Core)
+
 Matthias Weisser <[email protected]>
 
 	jadecpu		ARM926EJS (MB86R01 SoC)
board/nvidia/harmony/Makefile new file +50 -0

50 lines not on this page · raw message

boards.cfg +1 -0
index 94b8745..ee7c4b7 100644
@@ -122,6 +122,7 @@ omap4_panda                  arm         armv7       panda               ti
 omap4_sdp4430                arm         armv7       sdp4430             ti             omap4
 s5p_goni                     arm         armv7       goni                samsung        s5pc1xx
 smdkc100                     arm         armv7       smdkc100            samsung        s5pc1xx
+harmony                      arm         armv7       harmony             nvidia         tegra2
 actux1                       arm         ixp
 actux2                       arm         ixp
 actux3                       arm         ixp
include/configs/harmony.h new file +48 -0

48 lines not on this page · raw message

include/configs/tegra2-common.h new file +160 -0

160 lines not on this page · raw message

Signature · 1 line
1.7.3.5
author
#25 lore raw

PATCH v5 4/4 arm: Tegra2: Add support for NVIDIA Seaboard board

In reply to Tom Warren

---
Changes for V2:
- Remove mach-types.h change; wait for ARM kernel sync-up
- Use board/nvidia instead of board/tegra

 MAINTAINERS                    |    1 +
 board/nvidia/seaboard/Makefile |   50 ++++++++++++++++++++++++++++++++++++++++
 boards.cfg                     |    1 +
 include/configs/seaboard.h     |   44 +++++++++++++++++++++++++++++++++++
 4 files changed, 96 insertions(+), 0 deletions(-)
 create mode 100644 board/nvidia/seaboard/Makefile
 create mode 100644 include/configs/seaboard.h
MAINTAINERS +1 -0
index b5cff19..f4795d3 100644
@@ -844,6 +844,7 @@ Prafulla Wadaskar <[email protected]>
 Tom Warren <[email protected]>
 
 	harmony		Tegra2 (ARM7 & A9 Dual Core)
+	seaboard	Tegra2 (ARM7 & A9 Dual Core)
 
 Matthias Weisser <[email protected]>
 
board/nvidia/seaboard/Makefile new file +50 -0

50 lines not on this page · raw message

boards.cfg +1 -0
index ee7c4b7..3c4c249 100644
@@ -123,6 +123,7 @@ omap4_sdp4430                arm         armv7       sdp4430             ti
 s5p_goni                     arm         armv7       goni                samsung        s5pc1xx
 smdkc100                     arm         armv7       smdkc100            samsung        s5pc1xx
 harmony                      arm         armv7       harmony             nvidia         tegra2
+seaboard                     arm         armv7       seaboard            nvidia         tegra2
 actux1                       arm         ixp
 actux2                       arm         ixp
 actux3                       arm         ixp
include/configs/seaboard.h new file +44 -0

44 lines not on this page · raw message

Signature · 1 line
1.7.3.5