[U-Boot] [PATCH v3 0/8] SPL: Port SPL framework to powerpc

[U-Boot] [PATCH v3 0/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

This patchset ports the SPL framework to powerpc. Its based on the
SPL generalization lately done by Tom Rini. The patches can be applied
on top of his 7th version located here:

http://github.com/trini/u-boot WIP/spl-improvements

Additionally, a new MPC5200 board port is included, the a3m071 board
port. This board port uses this SPL framework mainly to speed up
booting into the OS (Linux of course). Detection of Linux vs. U-Boot
booting is done here by checking the environment variable "boot_os".
If "boot_os" is set to "yes", then the OS (Linux) is booted. Otherwise
the "real" U-Boot is booted. For this env checking in the SPL, a small
restructuring of the env code has been done.

Comments welcome!

Thanks,
Stefan

Changes in v2:
- Rebased on Tom's SPL framework patches v4
- Add option to skip copying of the mkimage header
- Rebased on Tom's SPL framework patches v4
- Rebased on Tom's SPL framework patches v4
- a3m071 build: Concat SPL binary and u-boot.img directly (no padding)

Changes in v3:
- Rebased on current master plus SPL framework v7
- Don't use special MPC5xxx memcpy in SPL
- Move linker symbols to header spl.h (checkpatch)

Stefan Roese (8):
  powerpc: Extract EPAPR_MAGIC constants into processor.h
  SPL: Port SPL framework to powerpc
  env: Extract getenv_f() into separate source file
  mpc5200: Add SPL support
  mpc5200: Add a3m071 board support
  fdt: cmd_fdt: Call fdt_chosen() from "fdt boardsetup"
  Makefile: Add possibility to set entry-point for u-boot.img
  Makefile: Add target for combined spl/u-boot.bin & u-boot.img

 MAINTAINERS                             |   2 +
 Makefile                                |  14 +-
 arch/powerpc/cpu/mpc5xxx/Makefile       |   4 +
 arch/powerpc/cpu/mpc5xxx/spl_boot.c     |  79 +++++++
 arch/powerpc/cpu/mpc5xxx/start.S        |  22 ++
 arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds |  57 +++++
 arch/powerpc/include/asm/processor.h    |   6 +
 arch/powerpc/include/asm/spl.h          |  31 +++
 arch/powerpc/lib/Makefile               |  11 +
 arch/powerpc/lib/bootm.c                |   6 -
 arch/powerpc/lib/spl.c                  |  42 ++++
 board/a3m071/Makefile                   |  36 +++
 board/a3m071/README                     |  79 +++++++
 board/a3m071/a3m071.c                   | 335 +++++++++++++++++++++++++++
 board/a3m071/mt46v16m16-75.h            |  32 +++
 boards.cfg                              |   1 +
 common/Makefile                         |   4 +
 common/cmd_fdt.c                        |   4 +-
 common/cmd_nvedit.c                     |  58 -----
 common/env_getenv_f.c                   |  82 +++++++
 common/spl/spl.c                        |  10 +
 include/configs/a3m071.h                | 385 ++++++++++++++++++++++++++++++++
 22 files changed, 1234 insertions(+), 66 deletions(-)
 create mode 100644 arch/powerpc/cpu/mpc5xxx/spl_boot.c
 create mode 100644 arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds
 create mode 100644 arch/powerpc/include/asm/spl.h
 create mode 100644 arch/powerpc/lib/spl.c
 create mode 100644 board/a3m071/Makefile
 create mode 100644 board/a3m071/README
 create mode 100644 board/a3m071/a3m071.c
 create mode 100644 board/a3m071/mt46v16m16-75.h
 create mode 100644 common/env_getenv_f.c
 create mode 100644 include/configs/a3m071.h

-- 
1.7.12.1

[U-Boot] [PATCH v3 1/8] powerpc: Extract EPAPR_MAGIC constants into processor.h

[Stefan Roese]

By extracting these defines into a header, they can be re-used by other
C sources as well. This will be done by the SPL framework OS boot
support.

Signed-off-by: Stefan Roese <sr@denx.de>
---

 arch/powerpc/include/asm/processor.h | 6 ++++++
 arch/powerpc/lib/bootm.c             | 6 ------
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/powerpc/include/asm/processor.h b/arch/powerpc/include/asm/processor.h
index dc009d6..daa670b 100644
--- a/arch/powerpc/include/asm/processor.h
+++ b/arch/powerpc/include/asm/processor.h
@@ -1329,4 +1329,10 @@ void _nmask_and_or_msr(unsigned long nmask, unsigned long or_val);
 #endif
 #endif /* CONFIG_MACH_SPECIFIC */
 
+#if defined(CONFIG_MPC85xx) || defined(CONFIG_440)
+ #define EPAPR_MAGIC	(0x45504150)
+#else
+ #define EPAPR_MAGIC	(0x65504150)
+#endif
+
 #endif /* __ASM_PPC_PROCESSOR_H */
diff --git a/arch/powerpc/lib/bootm.c b/arch/powerpc/lib/bootm.c
index 53dc4df..e3fee0b 100644
--- a/arch/powerpc/lib/bootm.c
+++ b/arch/powerpc/lib/bootm.c
@@ -87,12 +87,6 @@ static void boot_jump_linux(bootm_headers_t *images)
 		 *   r8: 0
 		 *   r9: 0
 		 */
-#if defined(CONFIG_MPC85xx) || defined(CONFIG_440)
- #define EPAPR_MAGIC	(0x45504150)
-#else
- #define EPAPR_MAGIC	(0x65504150)
-#endif
-
 		debug ("   Booting using OF flat tree...\n");
 		WATCHDOG_RESET ();
 		(*kernel) ((bd_t *)of_flat_tree, 0, 0, EPAPR_MAGIC,
-- 
1.7.12.1

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

This patch enables the SPL framework to be used on powerpc platforms
and not only ARM.

Signed-off-by: Stefan Roese <sr@denx.de>
---
Changes in v2:
- Rebased on Tom's SPL framework patches v4
- Add option to skip copying of the mkimage header

 arch/powerpc/lib/Makefile |  1 +
 arch/powerpc/lib/spl.c    | 42 ++++++++++++++++++++++++++++++++++++++++++
 common/spl/spl.c          | 10 ++++++++++
 3 files changed, 53 insertions(+)
 create mode 100644 arch/powerpc/lib/spl.c

diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index 965f9ea..9bcbdde 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -50,6 +50,7 @@ COBJS-y	+= cache.o
 COBJS-y	+= extable.o
 COBJS-y	+= interrupts.o
 COBJS-$(CONFIG_CMD_KGDB) += kgdb.o
+COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o
 COBJS-y	+= time.o
 
 # Workaround for local bus unaligned access problems
diff --git a/arch/powerpc/lib/spl.c b/arch/powerpc/lib/spl.c
new file mode 100644
index 0000000..502c93b
--- /dev/null
+++ b/arch/powerpc/lib/spl.c
@@ -0,0 +1,42 @@
+/*
+ * Copyright 2012 Stefan Roese <sr@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.
+ */
+#include <common.h>
+#include <config.h>
+#include <spl.h>
+#include <image.h>
+#include <linux/compiler.h>
+
+DECLARE_GLOBAL_DATA_PTR;
+
+/*
+ * This function jumps to an image with argument. Normally an FDT or ATAGS
+ * image.
+ * arg: Pointer to paramter image in RAM
+ */
+#ifdef CONFIG_SPL_OS_BOOT
+void __noreturn jump_to_image_linux(void *arg)
+{
+	debug("Entering kernel arg pointer: 0x%p\n", arg);
+	typedef void (*image_entry_arg_t)(void *, ulong r4, ulong r5, ulong r6,
+					  ulong r7, ulong r8, ulong r9)
+		__attribute__ ((noreturn));
+	image_entry_arg_t image_entry =
+		(image_entry_arg_t)spl_image.entry_point;
+
+	image_entry(arg, 0, 0, EPAPR_MAGIC, CONFIG_SYS_BOOTMAPSZ, 0, 0);
+}
+#endif /* CONFIG_SPL_OS_BOOT */
diff --git a/common/spl/spl.c b/common/spl/spl.c
index c640f87..e568c49 100644
--- a/common/spl/spl.c
+++ b/common/spl/spl.c
@@ -74,6 +74,16 @@ __weak int spl_start_uboot(void)
 }
 #endif
 
+/*
+ * Weak default function for board specific cleanup/preparation before
+ * Linux boot. Some boards/platforms might now need it, so just provide
+ * an empty stub here.
+ */
+__weak void spl_board_prepare_for_linux(void)
+{
+	/* Nothing to do! */
+}
+
 void spl_parse_image_header(const struct image_header *header)
 {
 	u32 header_size = sizeof(struct image_header);
-- 
1.7.12.1

[U-Boot] [PATCH v3 3/8] env: Extract getenv_f() into separate source file

[Stefan Roese]

By extracting getenv_f and envmatch() from cmd_nvedit.c into a
separate file, getenv_f() can be included easily into the SPL
binary. With this, SPL boards can now use getenv_f() to read
environment variables (e.g. to detect if the OS or U-Boot shall
be executed).

In the approach this is done for env stored in NOR flash, as this
will be used by an upcoming MPC5200 board port.

Signed-off-by: Stefan Roese <sr@denx.de>
---
Changes in v3:
- Rebased on current master plus SPL framework v7

 common/Makefile       |  4 +++
 common/cmd_nvedit.c   | 58 ------------------------------------
 common/env_getenv_f.c | 82 +++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 86 insertions(+), 58 deletions(-)
 create mode 100644 common/env_getenv_f.c

diff --git a/common/Makefile b/common/Makefile
index 22e8a6f..c30099d 100644
--- a/common/Makefile
+++ b/common/Makefile
@@ -50,6 +50,7 @@ XCOBJS-$(CONFIG_ENV_IS_EMBEDDED) += env_embedded.o
 COBJS-$(CONFIG_ENV_IS_IN_EEPROM) += env_embedded.o
 XCOBJS-$(CONFIG_ENV_IS_IN_FLASH) += env_embedded.o
 COBJS-$(CONFIG_ENV_IS_IN_NVRAM) += env_embedded.o
+COBJS-y += env_getenv_f.o
 COBJS-$(CONFIG_ENV_IS_IN_FLASH) += env_flash.o
 COBJS-$(CONFIG_ENV_IS_IN_MMC) += env_mmc.o
 COBJS-$(CONFIG_ENV_IS_IN_FAT) += env_fat.o
@@ -194,6 +195,9 @@ COBJS-$(CONFIG_CMD_DFU) += cmd_dfu.o
 endif
 
 ifdef CONFIG_SPL_BUILD
+COBJS-y += env_common.o
+COBJS-y += env_getenv_f.o
+COBJS-$(CONFIG_ENV_IS_IN_FLASH) += env_flash.o
 COBJS-$(CONFIG_SPL_YMODEM_SUPPORT) += xyzModem.o
 endif
 COBJS-y += console.o
diff --git a/common/cmd_nvedit.c b/common/cmd_nvedit.c
index 3474bc6..725e082 100644
--- a/common/cmd_nvedit.c
+++ b/common/cmd_nvedit.c
@@ -563,44 +563,6 @@ char *getenv(const char *name)
 	return NULL;
 }
 
-/*
- * Look up variable from environment for restricted C runtime env.
- */
-int getenv_f(const char *name, char *buf, unsigned len)
-{
-	int i, nxt;
-
-	for (i = 0; env_get_char(i) != '\0'; i = nxt + 1) {
-		int val, n;
-
-		for (nxt = i; env_get_char(nxt) != '\0'; ++nxt) {
-			if (nxt >= CONFIG_ENV_SIZE)
-				return -1;
-		}
-
-		val = envmatch((uchar *)name, i);
-		if (val < 0)
-			continue;
-
-		/* found; copy out */
-		for (n = 0; n < len; ++n, ++buf) {
-			*buf = env_get_char(val++);
-			if (*buf == '\0')
-				return n;
-		}
-
-		if (n)
-			*--buf = '\0';
-
-		printf("env_buf [%d bytes] too small for value of \"%s\"\n",
-			len, name);
-
-		return n;
-	}
-
-	return -1;
-}
-
 /**
  * Decode the integer value of an environment variable and return it.
  *
@@ -636,26 +598,6 @@ U_BOOT_CMD(
 );
 #endif
 
-
-/*
- * Match a name / name=value pair
- *
- * s1 is either a simple 'name', or a 'name=value' pair.
- * i2 is the environment index for a 'name2=value2' pair.
- * If the names match, return the index for the value2, else -1.
- */
-int envmatch(uchar *s1, int i2)
-{
-	while (*s1 == env_get_char(i2++))
-		if (*s1++ == '=')
-			return i2;
-
-	if (*s1 == '\0' && env_get_char(i2-1) == '=')
-		return i2;
-
-	return -1;
-}
-
 static int do_env_default(cmd_tbl_t *cmdtp, int __flag,
 			  int argc, char * const argv[])
 {
diff --git a/common/env_getenv_f.c b/common/env_getenv_f.c
new file mode 100644
index 0000000..7dfbfe0
--- /dev/null
+++ b/common/env_getenv_f.c
@@ -0,0 +1,82 @@
+/*
+ * (C) Copyright 2000-2010
+ * Wolfgang Denk, DENX Software Engineering, wd@denx.de.
+ *
+ * (C) Copyright 2001 Sysgo Real-Time Solutions, GmbH <www.elinos.com>
+ * Andreas Heppel <aheppel@sysgo.de>
+ *
+ * Copyright 2011 Freescale Semiconductor, Inc.
+ *
+ * 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.
+ */
+
+#include <common.h>
+#include <environment.h>
+
+/*
+ * Match a name / name=value pair
+ *
+ * s1 is either a simple 'name', or a 'name=value' pair.
+ * i2 is the environment index for a 'name2=value2' pair.
+ * If the names match, return the index for the value2, else -1.
+ */
+int envmatch(uchar *s1, int i2)
+{
+	while (*s1 == env_get_char(i2++))
+		if (*s1++ == '=')
+			return i2;
+
+	if (*s1 == '\0' && env_get_char(i2-1) == '=')
+		return i2;
+
+	return -1;
+}
+
+/*
+ * Look up variable from environment for restricted C runtime env.
+ */
+int getenv_f(const char *name, char *buf, unsigned len)
+{
+	int i, nxt;
+
+	for (i = 0; env_get_char(i) != '\0'; i = nxt + 1) {
+		int val, n;
+
+		for (nxt = i; env_get_char(nxt) != '\0'; ++nxt) {
+			if (nxt >= CONFIG_ENV_SIZE)
+				return -1;
+		}
+
+		val = envmatch((uchar *)name, i);
+		if (val < 0)
+			continue;
+
+		/* found; copy out */
+		for (n = 0; n < len; ++n, ++buf) {
+			*buf = env_get_char(val++);
+			if (*buf == '\0')
+				return n;
+		}
+
+		if (n)
+			*--buf = '\0';
+
+		printf("env_buf [%d bytes] too small for value of \"%s\"\n",
+			len, name);
+
+		return n;
+	}
+
+	return -1;
+}
-- 
1.7.12.1

[U-Boot] [PATCH v3 4/8] mpc5200: Add SPL support

[Stefan Roese]

This patch adds SPL booting support (NOR flash) for the
MPC5200 platforms.

Signed-off-by: Stefan Roese <sr@denx.de>
---
Changes in v2:
- Rebased on Tom's SPL framework patches v4

Changes in v3:
- Don't use special MPC5xxx memcpy in SPL
- Move linker symbols to header spl.h (checkpatch)

 arch/powerpc/cpu/mpc5xxx/Makefile       |  4 ++
 arch/powerpc/cpu/mpc5xxx/spl_boot.c     | 79 +++++++++++++++++++++++++++++++++
 arch/powerpc/cpu/mpc5xxx/start.S        | 22 +++++++++
 arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds | 57 ++++++++++++++++++++++++
 arch/powerpc/include/asm/spl.h          | 31 +++++++++++++
 arch/powerpc/lib/Makefile               | 10 +++++
 6 files changed, 203 insertions(+)
 create mode 100644 arch/powerpc/cpu/mpc5xxx/spl_boot.c
 create mode 100644 arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds
 create mode 100644 arch/powerpc/include/asm/spl.h

diff --git a/arch/powerpc/cpu/mpc5xxx/Makefile b/arch/powerpc/cpu/mpc5xxx/Makefile
index 1a088b7..8de2c13 100644
--- a/arch/powerpc/cpu/mpc5xxx/Makefile
+++ b/arch/powerpc/cpu/mpc5xxx/Makefile
@@ -41,6 +41,10 @@ COBJS-y += speed.o
 COBJS-$(CONFIG_CMD_USB) += usb_ohci.o
 COBJS-$(CONFIG_CMD_USB) += usb.o
 
+ifdef CONFIG_SPL_BUILD
+COBJS-y += spl_boot.o
+endif
+
 SRCS	:= $(START:.o=.S) $(SOBJS:.o=.S) $(COBJS-y:.o=.c)
 OBJS	:= $(addprefix $(obj),$(SOBJS) $(COBJS-y))
 START	:= $(addprefix $(obj),$(SSTART) $(CSTART))
diff --git a/arch/powerpc/cpu/mpc5xxx/spl_boot.c b/arch/powerpc/cpu/mpc5xxx/spl_boot.c
new file mode 100644
index 0000000..9f14127
--- /dev/null
+++ b/arch/powerpc/cpu/mpc5xxx/spl_boot.c
@@ -0,0 +1,79 @@
+/*
+ * Copyright (C) 2012 Stefan Roese <sr@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.
+ */
+
+#include <common.h>
+#include <spl.h>
+
+DECLARE_GLOBAL_DATA_PTR;
+
+/*
+ * Needed to align size SPL image to a 4-byte length
+ */
+u32 end_align __attribute__ ((section(".end_align")));
+
+/*
+ * Return selected boot device. On MPC5200 its only NOR flash right now.
+ */
+u32 spl_boot_device(void)
+{
+	return BOOT_DEVICE_NOR;
+}
+
+/*
+ * SPL version of board_init_f()
+ */
+void board_init_f(ulong bootflag)
+{
+	end_align = (u32)__spl_flash_end;
+
+	/*
+	 * First we need to initialize the SDRAM, so that the real
+	 * U-Boot or the OS (Linux) can be loaded
+	 */
+	initdram(0);
+
+	/* Clear bss */
+	memset(__bss_start, '\0', __bss_end__ - __bss_start);
+
+	/*
+	 * Init global_data pointer. Has to be done before calling
+	 * get_clocks(), as it stores some clock values into gd needed
+	 * later on in the serial driver.
+	 */
+	/* Pointer is writable since we allocated a register for it */
+	gd = (gd_t *)(CONFIG_SYS_INIT_RAM_ADDR + CONFIG_SYS_GBL_DATA_OFFSET);
+	/* Clear initial global data */
+	memset((void *)gd, 0, sizeof(gd_t));
+
+	/*
+	 * get_clocks() needs to be called so that the serial driver
+	 * works correctly
+	 */
+	get_clocks();
+
+	/*
+	 * Do rudimental console / serial setup
+	 */
+	preloader_console_init();
+
+	/*
+	 * Call board_init_r() (SPL framework version) to load and boot
+	 * real U-Boot or OS
+	 */
+	board_init_r(NULL, 0);
+	/* Does not return!!! */
+}
diff --git a/arch/powerpc/cpu/mpc5xxx/start.S b/arch/powerpc/cpu/mpc5xxx/start.S
index 51cc4e2..ad5bc0a 100644
--- a/arch/powerpc/cpu/mpc5xxx/start.S
+++ b/arch/powerpc/cpu/mpc5xxx/start.S
@@ -50,6 +50,7 @@
 #define MSR_KERNEL (MSR_FP|MSR_ME|MSR_RI)
 #endif
 
+#ifndef CONFIG_SPL_BUILD
 /*
  * Set up GOT: Global Offset Table
  *
@@ -68,6 +69,7 @@
 	GOT_ENTRY(__bss_end__)
 	GOT_ENTRY(__bss_start)
 	END_GOT
+#endif
 
 /*
  * Version string
@@ -84,6 +86,18 @@ version_string:
 	. = EXC_OFF_SYS_RESET
 	.globl	_start
 _start:
+
+#if defined(CONFIG_SPL) && !defined(CONFIG_SPL_BUILD)
+	/*
+	 * This is the entry of the real U-Boot from a board port
+	 * that supports SPL booting on the MPC5200. We only need
+	 * to call board_init_f() here. Everything else has already
+	 * been done in the SPL u-boot version.
+	 */
+	GET_GOT			/* initialize GOT access		*/
+	bl	board_init_f	/* run 1st part of board init code (in Flash)*/
+	/* NOTREACHED - board_init_f() does not return */
+#else
 	mfmsr	r5			/* save msr contents		*/
 
 	/* Move CSBoot and adjust instruction pointer                   */
@@ -152,7 +166,9 @@ lowboot_reentry:
 	/* Be careful to keep code relocatable !			*/
 	/*--------------------------------------------------------------*/
 
+#ifndef CONFIG_SPL_BUILD
 	GET_GOT			/* initialize GOT access		*/
+#endif
 
 	/* r3: IMMR */
 	bl	cpu_init_f	/* run low-level CPU init code (in Flash)*/
@@ -160,7 +176,9 @@ lowboot_reentry:
 	bl	board_init_f	/* run 1st part of board init code (in Flash)*/
 
 	/* NOTREACHED - board_init_f() does not return */
+#endif
 
+#ifndef CONFIG_SPL_BUILD
 /*
  * Vector Table
  */
@@ -333,6 +351,7 @@ int_return:
 	lwz	r1,GPR1(r1)
 	SYNC
 	rfi
+#endif /* CONFIG_SPL_BUILD */
 
 /*
  * This code initialises the MPC5xxx processor core
@@ -522,6 +541,7 @@ get_pvr:
 	mfspr	r3, PVR
 	blr
 
+#ifndef CONFIG_SPL_BUILD
 /*------------------------------------------------------------------------------*/
 
 /*
@@ -759,3 +779,5 @@ trap_init:
 
 	mtlr	r4			/* restore link register    */
 	blr
+
+#endif /* CONFIG_SPL_BUILD */
diff --git a/arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds b/arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds
new file mode 100644
index 0000000..cab9b92
--- /dev/null
+++ b/arch/powerpc/cpu/mpc5xxx/u-boot-spl.lds
@@ -0,0 +1,57 @@
+/*
+ * Copyright 2012 Stefan Roese <sr@denx.de>
+ *
+ * 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
+ */
+
+MEMORY
+{
+	sdram : ORIGIN = CONFIG_SPL_BSS_START_ADDR,
+		LENGTH = CONFIG_SPL_BSS_MAX_SIZE
+	flash : ORIGIN = CONFIG_SPL_TEXT_BASE,
+		LENGTH = CONFIG_SYS_SPL_MAX_LEN
+}
+
+OUTPUT_ARCH(powerpc)
+ENTRY(_start)
+SECTIONS
+{
+	.text :
+	{
+		__start = .;
+		arch/powerpc/cpu/mpc5xxx/start.o	(.text)
+		*(.text*)
+	} > flash
+
+	. = ALIGN(4);
+	.data : { *(SORT_BY_ALIGNMENT(.data*)) } > flash
+
+	. = ALIGN(4);
+	.rodata : { *(SORT_BY_ALIGNMENT(.rodata*)) } > flash
+
+	. = ALIGN(4);
+	.end_align : { *(.end_align*) } > flash
+	__spl_flash_end = .;
+
+	.bss :
+	{
+		. = ALIGN(4);
+		__bss_start = .;
+		*(.bss*)
+		. = ALIGN(4);
+		__bss_end__ = .;
+	} > sdram
+}
diff --git a/arch/powerpc/include/asm/spl.h b/arch/powerpc/include/asm/spl.h
new file mode 100644
index 0000000..f43bc23
--- /dev/null
+++ b/arch/powerpc/include/asm/spl.h
@@ -0,0 +1,31 @@
+/*
+ * (C) Copyright 2012
+ * Texas Instruments, <www.ti.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	_ASM_SPL_H_
+#define	_ASM_SPL_H_
+
+#define BOOT_DEVICE_NOR		1
+
+/* Linker symbols */
+extern char __bss_start[], __bss_end__[];
+
+#endif
diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
index 9bcbdde..cdb7a31 100644
--- a/arch/powerpc/lib/Makefile
+++ b/arch/powerpc/lib/Makefile
@@ -44,7 +44,9 @@ SOBJS-y	+= ticks.o
 SOBJS-y	+= reloc.o
 
 COBJS-$(CONFIG_BAT_RW) += bat_rw.o
+ifndef CONFIG_SPL_BUILD
 COBJS-y	+= board.o
+endif
 COBJS-y	+= bootm.o
 COBJS-y	+= cache.o
 COBJS-y	+= extable.o
@@ -53,6 +55,11 @@ COBJS-$(CONFIG_CMD_KGDB) += kgdb.o
 COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o
 COBJS-y	+= time.o
 
+# Don't include the MPC5xxx special memcpy into the
+# SPL U-Boot image. memcpy is used in the SPL NOR
+# flash driver. And we need the real, fast memcpy
+# here. We have no problems with unaligned access.
+ifndef CONFIG_SPL_BUILD
 # Workaround for local bus unaligned access problems
 # on MPC512x and MPC5200
 ifdef CONFIG_MPC512X
@@ -63,6 +70,7 @@ ifdef CONFIG_MPC5200
 $(obj)ppcstring.o: AFLAGS += -Dmemcpy=__memcpy
 COBJS-y += memcpy_mpc5200.o
 endif
+endif
 
 COBJS	+= $(sort $(COBJS-y))
 
@@ -75,12 +83,14 @@ TARGETS += $(LIB)
 all: $(TARGETS)
 
 $(LIB):	$(obj).depend $(OBJS)
+ifndef CONFIG_SPL_BUILD
 	@if ! $(CROSS_COMPILE)readelf -S $(OBJS) | grep -q '\.fixup.*PROGBITS';\
 	then \
 		echo "ERROR: Your compiler doesn't generate .fixup sections!";\
 		echo "       Upgrade to a recent toolchain."; \
 		exit 1; \
 	fi;
+endif
 	$(call cmd_link_o_target, $(OBJS))
 
 $(LIBGCC): $(obj).depend $(LGOBJS)
-- 
1.7.12.1

[U-Boot] [PATCH v3 3/8] env: Extract getenv_f() into separate source file

[Stefan Roese]

On 09/26/2012 11:01 AM, Stefan Roese wrote:
> By extracting getenv_f and envmatch() from cmd_nvedit.c into a
> separate file, getenv_f() can be included easily into the SPL
> binary. With this, SPL boards can now use getenv_f() to read
> environment variables (e.g. to detect if the OS or U-Boot shall
> be executed).
> 
> In the approach this is done for env stored in NOR flash, as this
> will be used by an upcoming MPC5200 board port.

This patch is not needed as it seems. I'll post a new version in a short
while that only enables the needed files for SPL_BUILD.

Thanks,
Stefan

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Scott Wood]

On 09/26/2012 04:01:08 AM, Stefan Roese wrote:
> This patch enables the SPL framework to be used on powerpc platforms
> and not only ARM.
> 
> Signed-off-by: Stefan Roese <sr@denx.de>
> ---
> Changes in v2:
> - Rebased on Tom's SPL framework patches v4
> - Add option to skip copying of the mkimage header
> 
>  arch/powerpc/lib/Makefile |  1 +
>  arch/powerpc/lib/spl.c    | 42  
> ++++++++++++++++++++++++++++++++++++++++++
>  common/spl/spl.c          | 10 ++++++++++
>  3 files changed, 53 insertions(+)
>  create mode 100644 arch/powerpc/lib/spl.c
> 
> diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
> index 965f9ea..9bcbdde 100644
> --- a/arch/powerpc/lib/Makefile
> +++ b/arch/powerpc/lib/Makefile
> @@ -50,6 +50,7 @@ COBJS-y	+= cache.o
>  COBJS-y	+= extable.o
>  COBJS-y	+= interrupts.o
>  COBJS-$(CONFIG_CMD_KGDB) += kgdb.o
> +COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o
>  COBJS-y	+= time.o

Won't this build spl.o into the main U-Boot as well?

> +/*
> + * This function jumps to an image with argument. Normally an FDT or  
> ATAGS
> + * image.
> + * arg: Pointer to paramter image in RAM
> + */
> +#ifdef CONFIG_SPL_OS_BOOT
> +void __noreturn jump_to_image_linux(void *arg)
> +{
> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
> +	typedef void (*image_entry_arg_t)(void *, ulong r4, ulong r5,  
> ulong r6,
> +					  ulong r7, ulong r8, ulong r9)
> +		__attribute__ ((noreturn));
> +	image_entry_arg_t image_entry =
> +		(image_entry_arg_t)spl_image.entry_point;
> +
> +	image_entry(arg, 0, 0, EPAPR_MAGIC, CONFIG_SYS_BOOTMAPSZ, 0, 0);
> +}

At what point does the image get cache-flushed?

-Scott

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Tom Rini]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 09/28/12 16:13, Scott Wood wrote:
> On 09/26/2012 04:01:08 AM, Stefan Roese wrote:
>> This patch enables the SPL framework to be used on powerpc
>> platforms and not only ARM.
>> 
>> Signed-off-by: Stefan Roese <sr@denx.de> --- Changes in v2: -
>> Rebased on Tom's SPL framework patches v4 - Add option to skip
>> copying of the mkimage header
>> 
>> arch/powerpc/lib/Makefile |  1 + arch/powerpc/lib/spl.c    | 42 
>> ++++++++++++++++++++++++++++++++++++++++++ common/spl/spl.c
>> | 10 ++++++++++ 3 files changed, 53 insertions(+) create mode
>> 100644 arch/powerpc/lib/spl.c
>> 
>> diff --git a/arch/powerpc/lib/Makefile
>> b/arch/powerpc/lib/Makefile index 965f9ea..9bcbdde 100644 ---
>> a/arch/powerpc/lib/Makefile +++ b/arch/powerpc/lib/Makefile @@
>> -50,6 +50,7 @@ COBJS-y    += cache.o COBJS-y    += extable.o 
>> COBJS-y    += interrupts.o COBJS-$(CONFIG_CMD_KGDB) += kgdb.o 
>> +COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o COBJS-y    += time.o
> 
> Won't this build spl.o into the main U-Boot as well?

True and probably uncaught as PowerPC uses -ffunction-sections /
- --gc-sections on main U-Boot.

- -- 
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/
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=VYvH
-----END PGP SIGNATURE-----

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

On 09/29/2012 01:13 AM, Scott Wood wrote:
> On 09/26/2012 04:01:08 AM, Stefan Roese wrote:
>> This patch enables the SPL framework to be used on powerpc platforms
>> and not only ARM.
>>
>> Signed-off-by: Stefan Roese <sr@denx.de>
>> ---
>> Changes in v2:
>> - Rebased on Tom's SPL framework patches v4
>> - Add option to skip copying of the mkimage header
>>
>>  arch/powerpc/lib/Makefile |  1 +
>>  arch/powerpc/lib/spl.c    | 42  
>> ++++++++++++++++++++++++++++++++++++++++++
>>  common/spl/spl.c          | 10 ++++++++++
>>  3 files changed, 53 insertions(+)
>>  create mode 100644 arch/powerpc/lib/spl.c
>>
>> diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
>> index 965f9ea..9bcbdde 100644
>> --- a/arch/powerpc/lib/Makefile
>> +++ b/arch/powerpc/lib/Makefile
>> @@ -50,6 +50,7 @@ COBJS-y	+= cache.o
>>  COBJS-y	+= extable.o
>>  COBJS-y	+= interrupts.o
>>  COBJS-$(CONFIG_CMD_KGDB) += kgdb.o
>> +COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o
>>  COBJS-y	+= time.o
> 
> Won't this build spl.o into the main U-Boot as well?

Yes. I'll fix it in the next patchset version.

>> +/*
>> + * This function jumps to an image with argument. Normally an FDT or  
>> ATAGS
>> + * image.
>> + * arg: Pointer to paramter image in RAM
>> + */
>> +#ifdef CONFIG_SPL_OS_BOOT
>> +void __noreturn jump_to_image_linux(void *arg)
>> +{
>> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
>> +	typedef void (*image_entry_arg_t)(void *, ulong r4, ulong r5,  
>> ulong r6,
>> +					  ulong r7, ulong r8, ulong r9)
>> +		__attribute__ ((noreturn));
>> +	image_entry_arg_t image_entry =
>> +		(image_entry_arg_t)spl_image.entry_point;
>> +
>> +	image_entry(arg, 0, 0, EPAPR_MAGIC, CONFIG_SYS_BOOTMAPSZ, 0, 0);
>> +}
> 
> At what point does the image get cache-flushed?

Not at all right now. MPC5200 has dcache disabled, at least in the SPL.
Other PowerPC architectures might add a cache flush here if needed at
some time. Okay?

Thanks,
Stefan

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Scott Wood]

On 10/02/2012 05:20:54 AM, Stefan Roese wrote:
> On 09/29/2012 01:13 AM, Scott Wood wrote:
> > On 09/26/2012 04:01:08 AM, Stefan Roese wrote:
> >> This patch enables the SPL framework to be used on powerpc  
> platforms
> >> and not only ARM.
> >>
> >> Signed-off-by: Stefan Roese <sr@denx.de>
> >> ---
> >> Changes in v2:
> >> - Rebased on Tom's SPL framework patches v4
> >> - Add option to skip copying of the mkimage header
> >>
> >>  arch/powerpc/lib/Makefile |  1 +
> >>  arch/powerpc/lib/spl.c    | 42
> >> ++++++++++++++++++++++++++++++++++++++++++
> >>  common/spl/spl.c          | 10 ++++++++++
> >>  3 files changed, 53 insertions(+)
> >>  create mode 100644 arch/powerpc/lib/spl.c
> >>
> >> diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
> >> index 965f9ea..9bcbdde 100644
> >> --- a/arch/powerpc/lib/Makefile
> >> +++ b/arch/powerpc/lib/Makefile
> >> @@ -50,6 +50,7 @@ COBJS-y	+= cache.o
> >>  COBJS-y	+= extable.o
> >>  COBJS-y	+= interrupts.o
> >>  COBJS-$(CONFIG_CMD_KGDB) += kgdb.o
> >> +COBJS-$(CONFIG_SPL_FRAMEWORK) += spl.o
> >>  COBJS-y	+= time.o
> >
> > Won't this build spl.o into the main U-Boot as well?
> 
> Yes. I'll fix it in the next patchset version.
> 
> >> +/*
> >> + * This function jumps to an image with argument. Normally an FDT  
> or
> >> ATAGS
> >> + * image.
> >> + * arg: Pointer to paramter image in RAM
> >> + */
> >> +#ifdef CONFIG_SPL_OS_BOOT
> >> +void __noreturn jump_to_image_linux(void *arg)
> >> +{
> >> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
> >> +	typedef void (*image_entry_arg_t)(void *, ulong r4, ulong r5,
> >> ulong r6,
> >> +					  ulong r7, ulong r8, ulong r9)
> >> +		__attribute__ ((noreturn));
> >> +	image_entry_arg_t image_entry =
> >> +		(image_entry_arg_t)spl_image.entry_point;
> >> +
> >> +	image_entry(arg, 0, 0, EPAPR_MAGIC, CONFIG_SYS_BOOTMAPSZ, 0, 0);
> >> +}
> >
> > At what point does the image get cache-flushed?
> 
> Not at all right now. MPC5200 has dcache disabled, at least in the  
> SPL.
> Other PowerPC architectures might add a cache flush here if needed at
> some time. Okay?

Or they might forget to do so and have weird bugs.

Why not just call flush_cache() on the image now?  Non-SPL does this in  
common code, not even PPC-specific.

-Scott

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

On 10/02/2012 10:08 PM, Scott Wood wrote:
>>>> +void __noreturn jump_to_image_linux(void *arg)
>>>> +{
>>>> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
>>>> +	typedef void (*image_entry_arg_t)(void *, ulong r4, ulong r5,
>>>> ulong r6,
>>>> +					  ulong r7, ulong r8, ulong r9)
>>>> +		__attribute__ ((noreturn));
>>>> +	image_entry_arg_t image_entry =
>>>> +		(image_entry_arg_t)spl_image.entry_point;
>>>> +
>>>> +	image_entry(arg, 0, 0, EPAPR_MAGIC, CONFIG_SYS_BOOTMAPSZ, 0, 0);
>>>> +}
>>>
>>> At what point does the image get cache-flushed?
>>
>> Not at all right now. MPC5200 has dcache disabled, at least in the  
>> SPL.
>> Other PowerPC architectures might add a cache flush here if needed at
>> some time. Okay?
> 
> Or they might forget to do so and have weird bugs.
> 
> Why not just call flush_cache() on the image now?  Non-SPL does this in  
> common code, not even PPC-specific.

Okay, probably better to add this code now. But shouldn't we add this
code to the common SPL framework code then? Right before calling
jump_to_image_linux()?

Tom? Should I prepare a patch adding this cache flushing to
common/spl/spl.c?

Thanks,
Stefan

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Scott Wood]

On 10/04/2012 02:36:34 AM, Stefan Roese wrote:
> On 10/02/2012 10:08 PM, Scott Wood wrote:
> >>>> +void __noreturn jump_to_image_linux(void *arg)
> >>>> +{
> >>>> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
> >>>> +	typedef void (*image_entry_arg_t)(void *, ulong r4,  
> ulong r5,
> >>>> ulong r6,
> >>>> +					  ulong r7, ulong r8,  
> ulong r9)
> >>>> +		__attribute__ ((noreturn));
> >>>> +	image_entry_arg_t image_entry =
> >>>> +		(image_entry_arg_t)spl_image.entry_point;
> >>>> +
> >>>> +	image_entry(arg, 0, 0, EPAPR_MAGIC,  
> CONFIG_SYS_BOOTMAPSZ, 0, 0);
> >>>> +}
> >>>
> >>> At what point does the image get cache-flushed?
> >>
> >> Not at all right now. MPC5200 has dcache disabled, at least in the
> >> SPL.
> >> Other PowerPC architectures might add a cache flush here if needed  
> at
> >> some time. Okay?
> >
> > Or they might forget to do so and have weird bugs.
> >
> > Why not just call flush_cache() on the image now?  Non-SPL does  
> this in
> > common code, not even PPC-specific.
> 
> Okay, probably better to add this code now. But shouldn't we add this
> code to the common SPL framework code then? Right before calling
> jump_to_image_linux()?

Sure, I didn't mean it should go here (it needs to be somewhere that  
knows the image start/end, not just the entry point).  This is just the  
patch that prompted me to ask the question.

-Scott

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

On 10/04/2012 07:14 PM, Scott Wood wrote:
> On 10/04/2012 02:36:34 AM, Stefan Roese wrote:
>> On 10/02/2012 10:08 PM, Scott Wood wrote:
>>>>>> +void __noreturn jump_to_image_linux(void *arg)
>>>>>> +{
>>>>>> +	debug("Entering kernel arg pointer: 0x%p\n", arg);
>>>>>> +	typedef void (*image_entry_arg_t)(void *, ulong r4,  
>> ulong r5,
>>>>>> ulong r6,
>>>>>> +					  ulong r7, ulong r8,  
>> ulong r9)
>>>>>> +		__attribute__ ((noreturn));
>>>>>> +	image_entry_arg_t image_entry =
>>>>>> +		(image_entry_arg_t)spl_image.entry_point;
>>>>>> +
>>>>>> +	image_entry(arg, 0, 0, EPAPR_MAGIC,  
>> CONFIG_SYS_BOOTMAPSZ, 0, 0);
>>>>>> +}
>>>>>
>>>>> At what point does the image get cache-flushed?
>>>>
>>>> Not at all right now. MPC5200 has dcache disabled, at least in the
>>>> SPL.
>>>> Other PowerPC architectures might add a cache flush here if needed  
>> at
>>>> some time. Okay?
>>>
>>> Or they might forget to do so and have weird bugs.
>>>
>>> Why not just call flush_cache() on the image now?  Non-SPL does  
>> this in
>>> common code, not even PPC-specific.
>>
>> Okay, probably better to add this code now. But shouldn't we add this
>> code to the common SPL framework code then? Right before calling
>> jump_to_image_linux()?
> 
> Sure, I didn't mean it should go here (it needs to be somewhere that  
> knows the image start/end, not just the entry point).  This is just the  
> patch that prompted me to ask the question.

I see.

Tom, whats your input on this? Do you see any problems about putting a
flush_cache() into the common SPL framework code? Are any of the ARM
platforms currently using this framework already running with d-cache
enabled?

Should I post a patch?

Thanks,
Stefan

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Tom Rini]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/05/12 06:15, Stefan Roese wrote:
> On 10/04/2012 07:14 PM, Scott Wood wrote:
>> On 10/04/2012 02:36:34 AM, Stefan Roese wrote:
>>> On 10/02/2012 10:08 PM, Scott Wood wrote:
>>>>>>> +void __noreturn jump_to_image_linux(void *arg) +{ +
>>>>>>> debug("Entering kernel arg pointer: 0x%p\n", arg); +
>>>>>>> typedef void (*image_entry_arg_t)(void *, ulong r4,
>>> ulong r5,
>>>>>>> ulong r6, +					  ulong r7, ulong r8,
>>> ulong r9)
>>>>>>> +		__attribute__ ((noreturn)); +	image_entry_arg_t
>>>>>>> image_entry = +
>>>>>>> (image_entry_arg_t)spl_image.entry_point; + +
>>>>>>> image_entry(arg, 0, 0, EPAPR_MAGIC,
>>> CONFIG_SYS_BOOTMAPSZ, 0, 0);
>>>>>>> +}
>>>>>> 
>>>>>> At what point does the image get cache-flushed?
>>>>> 
>>>>> Not at all right now. MPC5200 has dcache disabled, at least
>>>>> in the SPL. Other PowerPC architectures might add a cache
>>>>> flush here if needed
>>> at
>>>>> some time. Okay?
>>>> 
>>>> Or they might forget to do so and have weird bugs.
>>>> 
>>>> Why not just call flush_cache() on the image now?  Non-SPL
>>>> does
>>> this in
>>>> common code, not even PPC-specific.
>>> 
>>> Okay, probably better to add this code now. But shouldn't we
>>> add this code to the common SPL framework code then? Right
>>> before calling jump_to_image_linux()?
>> 
>> Sure, I didn't mean it should go here (it needs to be somewhere
>> that knows the image start/end, not just the entry point).  This
>> is just the patch that prompted me to ask the question.
> 
> I see.
> 
> Tom, whats your input on this? Do you see any problems about
> putting a flush_cache() into the common SPL framework code? Are any
> of the ARM platforms currently using this framework already running
> with d-cache enabled?

On ARM, jump_to_image_linux() calls cleanup_before_linux() to take
care of flushing, etc.  So the PowerPC jump_to_image_linux should take
care of what it needs to take care of.

- -- 
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/
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=5wlk
-----END PGP SIGNATURE-----

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Scott Wood]

On 10/05/2012 10:22:40 AM, Tom Rini wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 10/05/12 06:15, Stefan Roese wrote:
> > On 10/04/2012 07:14 PM, Scott Wood wrote:
> >> On 10/04/2012 02:36:34 AM, Stefan Roese wrote:
> >>> On 10/02/2012 10:08 PM, Scott Wood wrote:
> >>>>>>> +void __noreturn jump_to_image_linux(void *arg) +{ +
> >>>>>>> debug("Entering kernel arg pointer: 0x%p\n", arg); +
> >>>>>>> typedef void (*image_entry_arg_t)(void *, ulong r4,
> >>> ulong r5,
> >>>>>>> ulong r6, +					  ulong r7,  
> ulong r8,
> >>> ulong r9)
> >>>>>>> +		__attribute__ ((noreturn)); +	 
> image_entry_arg_t
> >>>>>>> image_entry = +
> >>>>>>> (image_entry_arg_t)spl_image.entry_point; + +
> >>>>>>> image_entry(arg, 0, 0, EPAPR_MAGIC,
> >>> CONFIG_SYS_BOOTMAPSZ, 0, 0);
> >>>>>>> +}
> >>>>>>
> >>>>>> At what point does the image get cache-flushed?
> >>>>>
> >>>>> Not at all right now. MPC5200 has dcache disabled, at least
> >>>>> in the SPL. Other PowerPC architectures might add a cache
> >>>>> flush here if needed
> >>> at
> >>>>> some time. Okay?
> >>>>
> >>>> Or they might forget to do so and have weird bugs.
> >>>>
> >>>> Why not just call flush_cache() on the image now?  Non-SPL
> >>>> does
> >>> this in
> >>>> common code, not even PPC-specific.
> >>>
> >>> Okay, probably better to add this code now. But shouldn't we
> >>> add this code to the common SPL framework code then? Right
> >>> before calling jump_to_image_linux()?
> >>
> >> Sure, I didn't mean it should go here (it needs to be somewhere
> >> that knows the image start/end, not just the entry point).  This
> >> is just the patch that prompted me to ask the question.
> >
> > I see.
> >
> > Tom, whats your input on this? Do you see any problems about
> > putting a flush_cache() into the common SPL framework code? Are any
> > of the ARM platforms currently using this framework already running
> > with d-cache enabled?
> 
> On ARM, jump_to_image_linux() calls cleanup_before_linux() to take
> care of flushing, etc.  So the PowerPC jump_to_image_linux should take
> care of what it needs to take care of.

It seems that most implementations of cleanup_before_linux() -- which  
is quite misnamed as we need to do this stuff no matter what we're  
loading -- disable and flush the entire cache.  Are there any  
implementations that just flush the range that the image was loaded  
into?  Is that information available at that point?

What if we're loading main U-Boot and not Linux?  Does  
cleanup_before_linux() still get called?

Normal U-Boot does this from generic code, why not SPL?  It should  
become a no-op on platforms that don't need it.

-Scott

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Stefan Roese]

On 10/05/2012 05:22 PM, Tom Rini wrote:

<snip>

>>>> Okay, probably better to add this code now. But shouldn't we
>>>> add this code to the common SPL framework code then? Right
>>>> before calling jump_to_image_linux()?
>>>
>>> Sure, I didn't mean it should go here (it needs to be somewhere
>>> that knows the image start/end, not just the entry point).  This
>>> is just the patch that prompted me to ask the question.
>>
>> I see.
>>
>> Tom, whats your input on this? Do you see any problems about
>> putting a flush_cache() into the common SPL framework code? Are any
>> of the ARM platforms currently using this framework already running
>> with d-cache enabled?
> 
> On ARM, jump_to_image_linux() calls cleanup_before_linux() to take
> care of flushing, etc.  So the PowerPC jump_to_image_linux should take
> care of what it needs to take care of.

Hmmm. Why not move this cleanup_xxx stuff into the common code as well?
Is it really that platform specific? Cache flushing is a common problem.

So basically a "+1" for Scotts comments. But I would suggest to address
this in a follow up patch, to consolidate all this cleanup/cache_flush
stuff. Okay?

Thanks,
Stefan

[U-Boot] [PATCH v3 2/8] SPL: Port SPL framework to powerpc

[Tom Rini]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 10/05/12 10:03, Stefan Roese wrote:
> On 10/05/2012 05:22 PM, Tom Rini wrote:
> 
> <snip>
> 
>>>>> Okay, probably better to add this code now. But shouldn't
>>>>> we add this code to the common SPL framework code then?
>>>>> Right before calling jump_to_image_linux()?
>>>> 
>>>> Sure, I didn't mean it should go here (it needs to be
>>>> somewhere that knows the image start/end, not just the entry
>>>> point).  This is just the patch that prompted me to ask the
>>>> question.
>>> 
>>> I see.
>>> 
>>> Tom, whats your input on this? Do you see any problems about 
>>> putting a flush_cache() into the common SPL framework code? Are
>>> any of the ARM platforms currently using this framework already
>>> running with d-cache enabled?
>> 
>> On ARM, jump_to_image_linux() calls cleanup_before_linux() to
>> take care of flushing, etc.  So the PowerPC jump_to_image_linux
>> should take care of what it needs to take care of.
> 
> Hmmm. Why not move this cleanup_xxx stuff into the common code as
> well? Is it really that platform specific? Cache flushing is a
> common problem.
> 
> So basically a "+1" for Scotts comments. But I would suggest to
> address this in a follow up patch, to consolidate all this
> cleanup/cache_flush stuff. Okay?

OK, so my quick check of things shows that in SPL, arm isn't ever
enabling the dcache (at least CONFIG_SPL_FRAMEWORK), we just make sure
that yes, really, we're good to go for Linux kernel booting.  That's
why we haven't had this as a general issue yet.  So in sum, yes,
please come up with patches to get things in a more consistent state.
 It should probably be re-naming cleanup_before_linux as well and
calling that as in some cases we need to do a little more than
i/d_disable, flush (see the v7 one).  Thanks!

- -- 
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://www.enigmail.net/
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=3AZY
-----END PGP SIGNATURE-----