December 2010 - U-Boot

[U-Boot] [PATCH] ppc4xx: Clarify comment about boot chip-select in start.S
by Stefan Roese 17 Dec '10

17 Dec '10

Ths old comment was quite screwed up. Replace it with a new version that should be a bit more descriptive. Signed-off-by: Stefan Roese <sr(a)denx.de> --- arch/powerpc/cpu/ppc4xx/start.S | 26 ++++++++++++++------------ 1 files changed, 14 insertions(+), 12 deletions(-) diff --git a/arch/powerpc/cpu/ppc4xx/start.S b/arch/powerpc/cpu/ppc4xx/start.S index 65195f5..d77272c 100644 --- a/arch/powerpc/cpu/ppc4xx/start.S +++ b/arch/powerpc/cpu/ppc4xx/start.S @@ -48,21 +48,23 @@ *------------------------------------------------------------------------------- */ -/* U-Boot - Startup Code for AMCC 4xx PowerPC based Embedded Boards - * +/* + * Startup code for IBM/AMCC PowerPC 4xx (PPC4xx) based boards * - * The processor starts at 0xfffffffc and the code is executed - * from flash/rom. - * in memory, but as long we don't jump around before relocating. - * board_init lies at a quite high address and when the cpu has - * jumped there, everything is ok. - * This works because the cpu gives the FLASH (CS0) the whole - * address space at startup, and board_init lies as a echo of - * the flash somewhere up there in the memorymap. + * The following description only applies to the NOR flash style booting. + * NAND booting is different. For more details about NAND booting on 4xx + * take a look at doc/README.nand-boot-ppc440. * - * board_init will change CS0 to be positioned at the correct - * address and (s)dram will be positioned at address 0 + * The CPU starts at address 0xfffffffc (last word in the address space). + * The U-Boot image therefore has to be located in the "upper" area of the + * flash (e.g. 512MiB - 0xfff80000 ... 0xffffffff). The default value for + * the boot chip-select (CS0) is quite big and covers this area. On the + * 405EX this is for example 0xffe00000 ... 0xffffffff. U-Boot will + * reconfigure this CS0 (and other chip-selects as well when configured + * this way) in the boot process to the "correct" values matching the + * board layout. */ + #include <asm-offsets.h> #include <config.h> #include <asm/ppc4xx.h> -- 1.7.3.2

1 1

[U-Boot] [PATCH] ppc4xx: t3corp: Add support for the Xilinx DS617 flash chip
by Stefan Roese 17 Dec '10

17 Dec '10

The t3corp board has an Xilinx DS617 flash chip connected to the onboard FPGA. This patch adds support for these chips. Board specific flash accessor functions are needed, since the chips can only be read correctly in 16bit mode. Additionally the FPGA chip-selects are configured for device-paced transfers (ready is enabled). Signed-off-by: Stefan Roese <sr(a)denx.de> --- board/t3corp/t3corp.c | 38 ++++++++++++++++++++++++++++++++++++++ include/configs/t3corp.h | 18 ++++++++++++------ 2 files changed, 50 insertions(+), 6 deletions(-) diff --git a/board/t3corp/t3corp.c b/board/t3corp/t3corp.c index 04d6a2e..f2853e4 100644 --- a/board/t3corp/t3corp.c +++ b/board/t3corp/t3corp.c @@ -23,6 +23,7 @@ #include <libfdt.h> #include <fdt_support.h> #include <i2c.h> +#include <mtd/cfi_flash.h> #include <asm/processor.h> #include <asm/io.h> #include <asm/mmu.h> @@ -191,3 +192,40 @@ struct sdram_timing *ddr_scan_option(struct sdram_timing *default_val) { return board_scan_options; } + +/* + * Accessor functions replacing the "weak" functions in + * drivers/mtd/cfi_flash.c + * + * The NOR flash devices "behind" the FPGA's (Xilinx DS617) + * can only be read correctly in 16bit mode. We need to emulate + * 8bit and 32bit reads here in the board specific code. + */ +u8 flash_read8(void *addr) +{ + u16 val = __raw_readw((void *)((u32)addr & ~1)); + + if ((u32)addr & 1) + return val; + + return val >> 8; +} + +u32 flash_read32(void *addr) +{ + return (__raw_readw(addr) << 16) | __raw_readw((void *)((u32)addr + 2)); +} + +void flash_cmd_reset(flash_info_t *info) +{ + /* + * FLASH at address CONFIG_SYS_FLASH_BASE is a Spansion chip and + * needs the Spansion type reset commands. The other flash chip + * is located behind a FPGA (Xilinx DS617) and needs the Intel type + * reset command. + */ + if (info->start[0] == CONFIG_SYS_FLASH_BASE) + flash_write_cmd(info, 0, 0, AMD_CMD_RESET); + else + flash_write_cmd(info, 0, 0, FLASH_CMD_RESET); +} diff --git a/include/configs/t3corp.h b/include/configs/t3corp.h index d00e64e..a938b6f 100644 --- a/include/configs/t3corp.h +++ b/include/configs/t3corp.h @@ -121,11 +121,16 @@ */ #define CONFIG_SYS_FLASH_CFI /* The flash is CFI compatible */ #define CONFIG_FLASH_CFI_DRIVER /* Use common CFI driver */ -#define CONFIG_SYS_FLASH_CFI_AMD_RESET 1 /* Use AMD reset cmd */ +#define CONFIG_SYS_FLASH_CFI_WIDTH FLASH_CFI_16BIT +#define CONFIG_CFI_FLASH_USE_WEAK_ACCESSORS #define CONFIG_SYS_CFI_FLASH_STATUS_POLL /* use status poll method */ +#define CONFIG_SYS_FLASH_PROTECTION /* use hardware flash protection */ -#define CONFIG_SYS_FLASH_BANKS_LIST { CONFIG_SYS_FLASH_BASE } -#define CONFIG_SYS_MAX_FLASH_BANKS 1 /* max num of memory banks */ +#define CONFIG_SYS_FLASH_BANKS_LIST { CONFIG_SYS_FLASH_BASE, \ + (CONFIG_SYS_FPGA1_BASE + 0x01000000) } +#define CONFIG_SYS_CFI_FLASH_CONFIG_REGS { 0xffff, /* don't set */ \ + 0xbddf } /* set async read mode */ +#define CONFIG_SYS_MAX_FLASH_BANKS 2 /* max num of memory banks */ #define CONFIG_SYS_MAX_FLASH_SECT 512 /* max num of sectors p. chip*/ #define CONFIG_SYS_FLASH_ERASE_TOUT 120000 /* Timeout for Flash Erase/ms*/ @@ -356,6 +361,7 @@ "ramdisk_addr=fc200000\0" \ "pciconfighost=1\0" \ "pcie_mode=RP:RP\0" \ + "unlock=yes\0" \ "" /* @@ -424,7 +430,7 @@ EBC_BXAP_WBN_ENCODE(0) | \ EBC_BXAP_WBF_ENCODE(0) | \ EBC_BXAP_TH_ENCODE(1) | \ - EBC_BXAP_RE_DISABLED | \ + EBC_BXAP_RE_ENABLED | \ EBC_BXAP_SOR_DELAYED | \ EBC_BXAP_BEM_RW | \ EBC_BXAP_PEN_DISABLED) @@ -441,7 +447,7 @@ EBC_BXAP_WBN_ENCODE(0) | \ EBC_BXAP_WBF_ENCODE(0) | \ EBC_BXAP_TH_ENCODE(1) | \ - EBC_BXAP_RE_DISABLED | \ + EBC_BXAP_RE_ENABLED | \ EBC_BXAP_SOR_DELAYED | \ EBC_BXAP_BEM_RW | \ EBC_BXAP_PEN_DISABLED) @@ -458,7 +464,7 @@ EBC_BXAP_WBN_ENCODE(0) | \ EBC_BXAP_WBF_ENCODE(0) | \ EBC_BXAP_TH_ENCODE(1) | \ - EBC_BXAP_RE_DISABLED | \ + EBC_BXAP_RE_ENABLED | \ EBC_BXAP_SOR_DELAYED | \ EBC_BXAP_BEM_RW | \ EBC_BXAP_PEN_DISABLED) -- 1.7.3.2

1 1

[U-Boot] [PATCH] AT91RM9200: fix AT91_PMC_MCKR_MDIV_* defines for this CPU
by clagix＠gmail.com 17 Dec '10

17 Dec '10

The AT91RM9200 has a different bit assignment in PMC MCKR register than it's successors (verified by Atmel's manuals on AT91RM9200 and AT91SAM926). AT91RM9200 supports the additional divider of 3. Signed-off-by: Guido Classen <clagix(a)gmail.com> --- u-boot-git/arch/arm/include/asm/arch-at91/at91_pmc.h 2010-11-12 18:30:31.000000000 +0100 +++ u-boot-ccm2200/arch/arm/include/asm/arch-at91/at91_pmc.h 2010-11-16 22:00:56.000000000 +0100 @@ -91,10 +91,19 @@ #define AT91_PMC_MCKR_PRES_64 0x00000018 #define AT91_PMC_MCKR_PRES_MASK 0x0000001C +#ifdef CONFIG_AT91RM9200 +#define AT91_PMC_MCKR_MDIV_1 0x00000000 +#define AT91_PMC_MCKR_MDIV_2 0x00000100 +#define AT91_PMC_MCKR_MDIV_3 0x00000200 +#define AT91_PMC_MCKR_MDIV_4 0x00000300 +#define AT91_PMC_MCKR_MDIV_MASK 0x00000300 +#else #define AT91_PMC_MCKR_MDIV_1 0x00000000 #define AT91_PMC_MCKR_MDIV_2 0x00000100 #define AT91_PMC_MCKR_MDIV_4 0x00000200 #define AT91_PMC_MCKR_MDIV_MASK 0x00000300 +#endif #define AT91_PMC_MCKR_PLLADIV_1 0x00001000 #define AT91_PMC_MCKR_PLLADIV_2 0x00002000

3 3

[U-Boot] [PATCH] AT91: fix TOP9000 build problem and change CONFIG_SYS_TEXT_BASE
by Reinhard Meyer 17 Dec '10

17 Dec '10

Signed-off-by: Reinhard Meyer <u-boot(a)emk-elektronik.de> --- board/emk/top9000/Makefile | 6 +++--- board/emk/top9000/top9000.c | 2 +- include/configs/top9000.h | 4 ++-- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/board/emk/top9000/Makefile b/board/emk/top9000/Makefile index 9b28048..3ac6f14 100644 --- a/board/emk/top9000/Makefile +++ b/board/emk/top9000/Makefile @@ -26,9 +26,9 @@ include $(TOPDIR)/config.mk -LIB = $(obj)lib$(BOARD).a +LIB = $(obj)lib$(BOARD).o -COBJS-y += top9000.o +COBJS-y += $(BOARD).o COBJS-$(CONFIG_ATMEL_SPI) += spi.o SRCS := $(SOBJS:.o=.S) $(COBJS-y:.o=.c) @@ -36,7 +36,7 @@ OBJS := $(addprefix $(obj),$(COBJS-y)) SOBJS := $(addprefix $(obj),$(SOBJS)) $(LIB): $(obj).depend $(OBJS) $(SOBJS) - $(AR) $(ARFLAGS) $@ $(OBJS) $(SOBJS) + $(call cmd_link_o_target, $(OBJS) $(SOBJS)) clean: rm -f $(SOBJS) $(OBJS) diff --git a/board/emk/top9000/top9000.c b/board/emk/top9000/top9000.c index b0fe2d6..73dd706 100644 --- a/board/emk/top9000/top9000.c +++ b/board/emk/top9000/top9000.c @@ -120,7 +120,7 @@ int board_mmc_getcd(u8 *cd, struct mmc *mmc) int board_early_init_f(void) { - at91_shdwn_t *shdwn = (at91_shdwn_t *)AT91_SHDWN_BASE; + struct at91_shdwn *shdwn = (struct at91_shdwn *)AT91_SHDWN_BASE; /* * make sure the board can be powered on by diff --git a/include/configs/top9000.h b/include/configs/top9000.h index ff3933b..0b23a5d 100644 --- a/include/configs/top9000.h +++ b/include/configs/top9000.h @@ -40,9 +40,9 @@ /* * Warning: changing CONFIG_SYS_TEXT_BASE requires - * adapting the initial boot program + * adapting the initial boot program. */ -#define CONFIG_SYS_TEXT_BASE 0x21f00000 /* 31 MB into RAM */ +#define CONFIG_SYS_TEXT_BASE 0x20000000 /* start of SDRAM */ /* Command line configuration */ #include <config_cmd_default.h> -- 1.5.6.5

1 1

Re: [U-Boot] NEC OHCI chip bug
by gavinguo＠andestech.com 17 Dec '10

17 Dec '10

> not sure if the PCI bus number plays a role here. In our case the PCI > window size was not set correctly so that the OHCI access to the PCI > bus/RAM was refused. I found it strange that 2 devices are detected, and we can get descriptor from hub descriptor. And the "Unrecoverable Error" is occurred when getting device descriptor from keyboard. So whether this still relate to PCI bridge setting problem? Regards, Gavin Guo

2 1

[U-Boot] TFTP Issue
by T K, Sunil Kumar 17 Dec '10

17 Dec '10

Hi All, When I start for TFTP I see below shown messages on my screen though I had set serverip properly. Can anyone please comment on the issue. ===================================== Retry count exceeded; starting again BOOTP broadcast 1 BOOTP broadcast 2 BOOTP broadcast 3 BOOTP broadcast 4 BOOTP broadcast 5 ===================================== Thanks in advance! Best Regards, Sunil

2 1

[U-Boot] [PATCH] tools: fw_printenv supports mmc device
by Donghwa Lee 17 Dec '10

17 Dec '10

I modified fw_printenv tools to use /dev/mmcblk0 node. Original fw_printenv tool can be access MTD devices, but, in some cases, environment variables can be stored other memory devices, for example, mmc devices. So, I modified a few code to use /dev/mmcblk0. Signed-off-by: Donghwa Lee <dh09.lee(a)samsung.com> Signed-off-by: Kyungmin Park <kyungmin.park(a)samsung.com> --- tools/env/fw_env.c | 13 ++++++++++--- tools/env/fw_env.config | 7 +++++-- 2 files changed, 15 insertions(+), 5 deletions(-) diff --git a/tools/env/fw_env.c b/tools/env/fw_env.c index 8ff7052..5a707f6 100644 --- a/tools/env/fw_env.c +++ b/tools/env/fw_env.c @@ -855,6 +855,7 @@ static int flash_write_buf (int dev, int fd, void *buf, size_t count, } erase.start = blockstart; +#ifndef CONFIG_ENV_IS_IN_MMC ioctl (fd, MEMUNLOCK, &erase); if (ioctl (fd, MEMERASE, &erase) != 0) { @@ -863,7 +864,7 @@ static int flash_write_buf (int dev, int fd, void *buf, size_t count, strerror (errno)); return -1; } - +#endif if (lseek (fd, blockstart, SEEK_SET) == -1) { fprintf (stderr, "Seek error on %s: %s\n", @@ -880,8 +881,9 @@ static int flash_write_buf (int dev, int fd, void *buf, size_t count, return -1; } +#ifndef CONFIG_ENV_IS_IN_MMC ioctl (fd, MEMLOCK, &erase); - +#endif processed += blocklen; block_seek = 0; blockstart += blocklen; @@ -964,9 +966,10 @@ static int flash_write (int fd_current, int fd_target, int dev_target) static int flash_read (int fd) { - struct mtd_info_user mtdinfo; int rc; +#ifndef CONFIG_ENV_IS_IN_MMC + struct mtd_info_user mtdinfo; rc = ioctl (fd, MEMGETINFO, &mtdinfo); if (rc < 0) { perror ("Cannot get MTD information"); @@ -982,6 +985,10 @@ static int flash_read (int fd) rc = flash_read_buf (dev_current, fd, environment.image, CONFIG_ENV_SIZE, DEVOFFSET (dev_current), mtdinfo.type); +#else + rc = flash_read_buf (dev_current, fd, environment.image, CONFIG_ENV_SIZE, + DEVOFFSET (dev_current), 0); +#endif return (rc != CONFIG_ENV_SIZE) ? -1 : 0; } diff --git a/tools/env/fw_env.config b/tools/env/fw_env.config index c8f12cf..228ded8 100644 --- a/tools/env/fw_env.config +++ b/tools/env/fw_env.config @@ -4,8 +4,11 @@ # Notice, that the "Number of sectors" is ignored on NOR. # MTD device name Device offset Env. size Flash sector size Number of sectors -/dev/mtd1 0x0000 0x4000 0x4000 -/dev/mtd2 0x0000 0x4000 0x4000 +#/dev/mtd1 0x0000 0x4000 0x4000 +#/dev/mtd2 0x0000 0x4000 0x4000 # NAND example #/dev/mtd0 0x4000 0x4000 0x20000 2 + +# MMC device name Device offset Env. size Flash sector size Number of sectors +/dev/mmcblk0 0x7000 0x1000 0x1000 -- 1.6.0.4

3 4

[U-Boot] [ANNOUNCE] Public x-loader git tree
by Gadiyar, Anand 17 Dec '10

17 Dec '10

Hi all, As all OMAP folks know, x-loader is a first-stage bootloader for OMAPs, derived from the u-boot base code. There have been several forks of the x-loader code within and outside of TI. X-loader has no upstream path as it is currently used only for TI-products. To avoid duplication of efforts, it's high time the code from these different trees was consolidated. Several contributors to x-loader have expressed a desire to have the code developed collaboratively in public. To this end, we have set up an x-loader git tree on gitorious, and seeded it with Steve Sakoman's x-loader tree [1] as of 15 December 2010. (Thanks Steve for unifying so much of the forked code, and getting rid of the dependency on u-boot, and more!). The tree is available at: http://gitorious.org/x-loader/x-loader And a wiki page at: http://gitorious.org/x-loader/pages/Home In addition, a mailing list has been set up at: http://groups.google.com/group/x-loader (Thanks Nishanth Menon for setting these up). Deva and I have volunteered to maintain the x-loader code - everyone is welcome to contribute and help make it better. (This is our first attempt at maintaining a software project - so any help is appreciated). Currently, thanks mainly to Steve's efforts, the tree builds for and works on the following platforms: the OMAP3 EVM, the Overo and Beagle (both 35xx and 37xx), and the Pandaboard. There are also configs for the 1710 h3, 2420 h4, 2430 SDP, 3430 SDP and 3430 Labrador boards - however these don't currently compile. They will be removed soon unless someone steps in to help bring them into shape and maintain them. (We can discuss this on the x-loader mailing list above ;) ). Patches are welcome for any new platforms that would like to be supported in this repository. If there are forked versions out there, let us know and we'll try and integrate them here. Feel free to use this tree, test it, report issues, and help keep the code in shape. - Anand and Deva [1] Steve Sakoman's x-loader tree: http://www.sakoman.com/cgi-bin/gitweb.cgi?p=x-loader.git;a=summary

3 2

[U-Boot] Pull request: nand flash
by Scott Wood 16 Dec '10

16 Dec '10

The following changes since commit f8689b9eb3a7f6925cd50404a12479889188c510: Wolfgang Denk (1): Merge branch 'master' of git://git.denx.de/u-boot-mpc83xx are available in the git repository at: git://git.denx.de/u-boot-nand-flash.git master Scott Wood (1): powerpc/nand spl: link libgcc nand_spl/board/amcc/acadia/Makefile | 2 +- nand_spl/board/amcc/bamboo/Makefile | 2 +- nand_spl/board/amcc/canyonlands/Makefile | 2 +- nand_spl/board/amcc/kilauea/Makefile | 2 +- nand_spl/board/amcc/sequoia/Makefile | 2 +- nand_spl/board/freescale/mpc8313erdb/Makefile | 2 +- nand_spl/board/freescale/mpc8315erdb/Makefile | 2 +- nand_spl/board/sheldon/simpc8313/Makefile | 2 +- 8 files changed, 8 insertions(+), 8 deletions(-)

2 1

[U-Boot] [PULL] [NEXT] u-boot-arm.git next
by Wolfgang Denk 16 Dec '10

16 Dec '10

The following changes since commit 006915fbb0e3c1c9927fe32c4e92cb011f8499e7: Merge branch 'master' of ../master into next (2010-12-16 23:00:53 +0100) are available in the git repository at: git://git.denx.de/u-boot-arm.git next Prafulla Wadaskar (7): arm: Add Support for Marvell ARMADA 100 Familiy SoCs gpio: Add Multi-Function-Pin configuration driver for Marvell SoCs add Multi Function Pin configuration support for ARMADA100 Serial: ns16550: Add support for CONFIG_SYS_NS16550_IER macro Serial: Add UART support for Marvell ARMADA 100 SoCs. mv-common.h: Add support for ARMADA100 Platforms Armada100: Add Board Support for Marvell Aspenite-DB Simon Kagstrom (1): MAINTAINERS: Transfer openrd_base maintainership to Prafulla Wadaskar MAINTAINERS | 6 +- MAKEALL | 1 + arch/arm/cpu/arm926ejs/armada100/Makefile | 46 +++++ arch/arm/cpu/arm926ejs/armada100/cpu.c | 92 ++++++++++ arch/arm/cpu/arm926ejs/armada100/dram.c | 131 ++++++++++++++ arch/arm/cpu/arm926ejs/armada100/timer.c | 207 +++++++++++++++++++++++ arch/arm/include/asm/arch-armada100/armada100.h | 121 +++++++++++++ arch/arm/include/asm/arch-armada100/cpu.h | 53 ++++++ arch/arm/include/asm/arch-armada100/mfp.h | 67 ++++++++ board/Marvell/aspenite/Makefile | 52 ++++++ board/Marvell/aspenite/aspenite.c | 53 ++++++ boards.cfg | 1 + drivers/gpio/Makefile | 1 + drivers/gpio/mvmfp.c | 87 ++++++++++ drivers/serial/ns16550.c | 8 +- drivers/serial/serial.c | 5 +- include/configs/aspenite.h | 63 +++++++ include/configs/mv-common.h | 66 ++++++-- include/mvmfp.h | 100 +++++++++++ 19 files changed, 1136 insertions(+), 24 deletions(-) create mode 100644 arch/arm/cpu/arm926ejs/armada100/Makefile create mode 100644 arch/arm/cpu/arm926ejs/armada100/cpu.c create mode 100644 arch/arm/cpu/arm926ejs/armada100/dram.c create mode 100644 arch/arm/cpu/arm926ejs/armada100/timer.c create mode 100644 arch/arm/include/asm/arch-armada100/armada100.h create mode 100644 arch/arm/include/asm/arch-armada100/cpu.h create mode 100644 arch/arm/include/asm/arch-armada100/mfp.h create mode 100644 board/Marvell/aspenite/Makefile create mode 100644 board/Marvell/aspenite/aspenite.c create mode 100644 drivers/gpio/mvmfp.c create mode 100644 include/configs/aspenite.h create mode 100644 include/mvmfp.h Best regards, Wolfgang Denk -- 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(a)denx.de The only way to learn a new programming language is by writing pro- grams in it. - Brian Kernighan

1 1