May 2010 - U-Boot

[U-Boot] [PATCH v2][mpc83xx] Removal of checkboard from spl bootstrap build for SIMPC8313
by Ron Madrid 17 May '10

17 May '10

This patch removes the checkboard function from the build of the 4k bootstrap section for the SIMPC8313 as it is not needed in the spl build. This will allow > 100 bytes of extra room for other uses. Signed-off-by: Ron Madrid <ron_madrid(a)sbcglobal.net> --- board/sheldon/simpc8313/simpc8313.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/board/sheldon/simpc8313/simpc8313.c b/board/sheldon/simpc8313/simpc8313.c index 0235545..cb30b48 100644 --- a/board/sheldon/simpc8313/simpc8313.c +++ b/board/sheldon/simpc8313/simpc8313.c @@ -33,13 +33,13 @@ DECLARE_GLOBAL_DATA_PTR; +#ifndef CONFIG_NAND_SPL int checkboard(void) { puts("Board: Sheldon Instruments SIMPC8313\n"); return 0; } -#ifndef CONFIG_NAND_SPL static struct pci_region pci_regions[] = { { bus_start: CONFIG_SYS_PCI1_MEM_BASE, -- 1.5.5.1

2 4

[U-Boot] U-boot Hangs in relocate_code function
by Gurumurthy G M 17 May '10

17 May '10

Hi Andrew, Now am able to get some messages on serial port but it is hanging after DRAM initialization i.e. during relocate_code function, below is the debug messages on serial port. I tried configuring SDRAM (DDR2) for 256 and 128 MB, but still the problem is same. I am facing one typical problem in u-boot i.e. whenever i power on the board I will get below debug messages on serial port after 5 to 8 seconds and hangs at stack pointer. U-Boot 2010.03 (May 14 2010 - 16:41:46) Linux Advanced AVN Platform CPU: Au1350, id: 0x80, rev: 0x01 DRAM: 256 MB Address value =90000000Top of RAM usable for U-Boot at: 90000000 Reserving 161k for U-Boot at: 8ffd4000 Reserving 192k for malloc() at: 8ffa4000 Reserving 36 Bytes for Board Info at: 8ffa3fdc Reserving 36 Bytes for Global Data at: 8ffa3fb8 Reserving 128k for boot params() at: 8ff83fb8 Stack Pointer at: 8ff83f98 U-Boot 2010.03 (May 14 2010 - 16:49:00) Linux Advanced AVN Platform CPU: Au1350, id: 0x80, rev: 0x01 DRAM: 128 MB Address value =88000000Top of RAM usable for U-Boot at: 88000000 Reserving 161k for U-Boot at: 87fd4000 Reserving 192k for malloc() at: 87fa4000 Reserving 36 Bytes for Board Info at: 87fa3fdc Reserving 36 Bytes for Global Data at: 87fa3fb8 Reserving 128k for boot params() at: 87f83fb8 Stack Pointer at: 87f83f98 U-Boot 2010.03 (May 14 2010 - 16:53:58) Linux Advanced AVN Platform CPU: Au1350, id: 0x80, rev: 0x01 DRAM: 256 MB Address value =90000000Top of RAM usable for U-Boot at: 90000000 Reserving 161k for U-Boot at: 8ffd4000 Reserving 192k for malloc() at: 8ffa4000 Reserving 36 Bytes for Board Info at: 8ffa3fdc Reserving 36 Bytes for Global Data at: 8ffa3fb8 Reserving 128k for boot params() at: 8ff83fb8 Stack Pointer at: 8ff83f98 Can you please share me your board config file and lowlevel_init.S file which you have ported u-boot for Au1550. I need it for the reference. I am doing DDR2 initialization in u-boot as per YAMON source code that will create any problem? please help me out to resolve the above issues. Thanks in Advance Regards, Gurumurthy Gowdar KPIT Cummins Infosystems Pvt Ltd

1 0

[U-Boot] [PATCH] arm/pxa-devel: fix and cleanup of pxa_mem_setup macro
by Mikhail Kshevetskiy 17 May '10

17 May '10

* strict following to section 6.4.10 of Intel PXA27xx Developer's Manual. * use r7 to store CONFIG_SYS_MDREFR_VAL as r6 is used in pxa_wait_ticks WARNING: This macro do not assume the values for K0DB4, KxDB2, KxFREE and APD bits of CONFIG_SYS_MDREFR_VAL as it was done early on many pxa platforms. All pxa developers that plan to use this macro should check the validity of their MDREFR values. Signed-off-by: Mikhail Kshevetskiy <mikhail.kshevetskiy(a)gmail.com> --- arch/arm/include/asm/arch-pxa/macro.h | 81 +++++++++++++++++++++++---------- 1 files changed, 56 insertions(+), 25 deletions(-) diff --git a/arch/arm/include/asm/arch-pxa/macro.h b/arch/arm/include/asm/arch-pxa/macro.h index 1f1759b..26d7a8d 100644 --- a/arch/arm/include/asm/arch-pxa/macro.h +++ b/arch/arm/include/asm/arch-pxa/macro.h @@ -102,7 +102,11 @@ /* * This macro sets up the Memory controller of the PXA2xx CPU * - * Clobbered regs: r3, r4, r5 + * Clobbered regs: r3, r4, r5, r6, r7 + * + * See section 6.4.10 of Intel PXA2xx Processor Developer's Manual + * http://www.marvell.com/products/processors/applications/pxa_family/pxa_27x_… + * http://www.marvell.com/products/processors/applications/pxa_family/PXA3xx_D… */ .macro pxa_mem_setup /* This comes handy when setting MDREFR */ @@ -149,27 +153,38 @@ */ /* - * Before accessing MDREFR we need a valid DRI field, so we set - * this to power on defaults + DRI field. + * Before accessing MDREFR we need a valid DRI field. + * Also we must properly configure MDREFR[K0DB2] and MDREFR[K0DB4]. + * Optionaly we can set MDREFR[KxFREE] bits. + * So we set MDREFR to power on defaults + (DRI, K0DB2, K0DB4, KxFREE) + * fields from the config. + * + * WARNING: K0DB2 and K0DB4 bits are usually set, while KxFREE bits + * are usually unser. */ ldr r5, [r3, #MDREFR_OFFSET] - bic r5, r5, #0x0ff - bic r5, r5, #0xf00 /* MDREFR user config with zeroed DRI */ - - ldr r4, =CONFIG_SYS_MDREFR_VAL - mov r6, r4 - lsl r4, #20 - lsr r4, #20 /* Get a valid DRI field */ + ldr r4, =( 0xFFF | MDREFR_K0DB4 | MDREFR_K0DB2 ) + orr r4, #( MDREFR_K0FREE | MDREFR_K1FREE | MDREFR_K2FREE ) + bic r5, r5, r4 /* clear DRI, K0DB2, K0DB4, KxFREE fields */ - orr r5, r5, r4 /* MDREFR user config with correct DRI */ + /* + * r3 is busy with MEMC_BASE, + * r4, r5, r6 used in pxa_wait_ticks and other places, + * so use r7 to store user specified MDREFR_VAL + */ + ldr r7, =CONFIG_SYS_MDREFR_VAL + and r4, r7, r4 + orr r5, r5, r4 /* use custom DRI, K0DB2, K0DB4, KxFREE */ orr r5, #MDREFR_K0RUN orr r5, #MDREFR_SLFRSH bic r5, #MDREFR_APD bic r5, #MDREFR_E1PIN + bic r5, #MDREFR_K1RUN + bic r5, #MDREFR_K2RUN str r5, [r3, #MDREFR_OFFSET] - ldr r4, [r3, #MDREFR_OFFSET] + ldr r5, [r3, #MDREFR_OFFSET] /* * 5) Initialize Synchronous Static Memory (Flash/Peripherals) @@ -184,16 +199,29 @@ write32rb (MEMC_BASE + SXCNFG_OFFSET), CONFIG_SYS_SXCNFG_VAL /* - * 6) Initialize SDRAM + * 6) Initialize SDRAM, + * also we must properly set MDREFR[K1DB2] and MDREFR[K2DB2] + * + * WARNING: K1DB2 and K2DB2 bits are usually set */ - bic r6, #MDREFR_SLFRSH - str r6, [r3, #MDREFR_OFFSET] - ldr r4, [r3, #MDREFR_OFFSET] + and r4, r7, #( MDREFR_K1DB2 | MDREFR_K2DB2 ) + ldr r5, [r3, #MDREFR_OFFSET] + bic r5, #( MDREFR_K1DB2 | MDREFR_K2DB2 ) + orr r5, r5, r4 - orr r6, #MDREFR_E1PIN - str r6, [r3, #MDREFR_OFFSET] - ldr r4, [r3, #MDREFR_OFFSET] + orr r5, #MDREFR_K1RUN + orr r5, #MDREFR_K2RUN + str r5, [r3, #MDREFR_OFFSET] + ldr r5, [r3, #MDREFR_OFFSET] + + bic r5, #MDREFR_SLFRSH + str r5, [r3, #MDREFR_OFFSET] + ldr r5, [r3, #MDREFR_OFFSET] + + orr r5, #MDREFR_E1PIN + str r5, [r3, #MDREFR_OFFSET] + ldr r5, [r3, #MDREFR_OFFSET] /* * 7) Write MDCNFG with MDCNFG:DEx deasserted (set to 0), to configure @@ -246,14 +274,17 @@ ldr r4, [r3, #MDMRS_OFFSET] /* - * 11) Enable APD + * 11) Optionaly enable MDREFR[APD] + * + * WARNING: APD bit is usually set. */ - ldr r4, [r3, #MDREFR_OFFSET] - and r6, r6, #MDREFR_APD - orr r4, r4, r6 - str r4, [r3, #MDREFR_OFFSET] - ldr r4, [r3, #MDREFR_OFFSET] + and r7, #MDREFR_APD + ldr r5, [r3, #MDREFR_OFFSET] + bic r5, #MDREFR_APD + orr r5, r5, r7 + str r5, [r3, #MDREFR_OFFSET] + ldr r5, [r3, #MDREFR_OFFSET] .endm /* -- 1.7.1

2 1

[U-Boot] [PATCH] ppc4xx: Fix building PLU405 board
by Matthias Fuchs 17 May '10

17 May '10

Due to some overlapping sections it's time to update TEXT_BASE for this board. Signed-off-by: Matthias Fuchs <matthias.fuchs(a)esd.eu> --- board/esd/plu405/config.mk | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/board/esd/plu405/config.mk b/board/esd/plu405/config.mk index 0fb4efa..0a4dbaa 100644 --- a/board/esd/plu405/config.mk +++ b/board/esd/plu405/config.mk @@ -25,4 +25,4 @@ # esd PLU405 boards # -TEXT_BASE = 0xFFFA0000 +TEXT_BASE = 0xFFF80000 -- 1.6.1

2 1

[U-Boot] flash_real_protect() and CFI_CMDSET_AMD_STANDARD
by Timur Tabi 17 May '10

17 May '10

I'm trying to add support for AMD manufacturer ID to flash_real_protect(). The code currently does this: case CFI_CMDSET_AMD_EXTENDED: case CFI_CMDSET_AMD_STANDARD: /* U-Boot only checks the first byte */ if (info->manufacturer_id == (uchar)ATM_MANUFACT) { ... } break; Unfortunately, I don't really understand CFI programming that well, but from my research, it appears to me that AMD chips that implement the AMD standard command set will work with this code: case CFI_CMDSET_INTEL_PROG_REGIONS: case CFI_CMDSET_INTEL_STANDARD: case CFI_CMDSET_INTEL_EXTENDED: flash_write_cmd (info, sector, 0, FLASH_CMD_CLEAR_STATUS); flash_write_cmd (info, sector, 0, FLASH_CMD_PROTECT); if (prot) flash_write_cmd (info, sector, 0, FLASH_CMD_PROTECT_SET); else flash_write_cmd (info, sector, 0, FLASH_CMD_PROTECT_CLEAR); break; So I don't understand the need for Rafael Campos' patch: http://lists.denx.de/pipermail/u-boot/2008-July/037419.html It looks as if Rafael removed support for hardware flash protection from all AMD standard/extended flash chips that are not made by Atmel. -- Timur Tabi Linux kernel developer at Freescale

2 1

[U-Boot] Can we do this
by Mr Ming Yang 17 May '10

17 May '10

Please I would like you to keep this proposal as a top secret and delete it if you are not interested and get back to me if you are ineterested for details as regards to the transfer of $24,500,000.000 to you. this money initially belongs to a client who died and had no next of kin in his account-opening package. in other to achieve this, I shall require your full name, and telephone number to reach you.Most importantly, a confirmation of acceptance from you after which I shall furnish you with the full details of this transaction Reply via this email only mrmingyanghsk1(a)w.cn regards, Mr Ming Yang

1 0

Re: [U-Boot] [PATCH 4/4] Environment in MMC
by Stefano Babic 17 May '10

17 May '10

Sudhakar Rajashekhara wrote: > Hi Alagu, > Hi, > On Wed, May 12, 2010 at 15:08:27, Alagu Sankar wrote: >> This patch is to save environment data to mmc card. It uses interfaces defined >> in generic MMC framework. This is enabled with CONFIG_ENV_IS_IN_MMC option. >> Based on the earlier patch from Terry Lv at Freescale Should be not better you send your comments to Terry as to generate a new patch? His patch had already passed some reviews and he can integrate your comments. I miss here what you changed respect Terry, and you miss some comments in the previous reviews, for example regarding the usage of the ALIGN macro. Why do you need two patches for the same purpose ? > Why this has been moved up? If there is any genuine reason for this, then > you can mention it in patch description. That is correct. The mmc must be initialized before env_init to get the environment working. Best regards, Stefano Babic -- ===================================================================== DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office(a)denx.de =====================================================================

3 4

[U-Boot] [PATCH 1/3] The BUILD_BUG_ON macros moved from ubi_uboot.h to common.h
by Michael Zaidman 16 May '10

16 May '10

The BUILD_BUG_ON macros made availible for rest of the u-boot code by moving it from ubi_uboot.h to common.h. In Linux kernel this widely used macros is defined in kernel.h file. Signed-off-by: Michael Zaidman <michael.zaidman(a)gmail.com> --- include/common.h | 3 +++ include/ubi_uboot.h | 3 --- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/include/common.h b/include/common.h index a133e34..7c22e69 100644 --- a/include/common.h +++ b/include/common.h @@ -136,6 +136,9 @@ typedef volatile unsigned char vu_char; #define BUG_ON(condition) do { if (unlikely((condition)!=0)) BUG(); } while(0) #endif /* BUG */ +/* Force a compilation error if condition is true */ +#define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)])) + typedef void (interrupt_handler_t)(void *); #include <asm/u-boot.h> /* boot information for Linux kernel */ diff --git a/include/ubi_uboot.h b/include/ubi_uboot.h index 60f6a5f..20e3da5 100644 --- a/include/ubi_uboot.h +++ b/include/ubi_uboot.h @@ -191,9 +191,6 @@ static inline long IS_ERR(const void *ptr) return IS_ERR_VALUE((unsigned long)ptr); } -/* Force a compilation error if condition is true */ -#define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)])) - /* module */ #define THIS_MODULE 0 #define try_module_get(...) 1 -- 1.6.3.3

2 7

[U-Boot] [PATCH] MVBLM7, MVSMR: fix Makefile (cleanup bootscript.img)
by Wolfgang Denk 15 May '10

15 May '10

Fix MVBLM7 and MVSMR Makefiles for correct out-of-tree building (create "bootscript.img" in build directory instead of source directory) and cleanup (remove "bootscript.img" when cleaning up). Signed-off-by: Wolfgang Denk <wd(a)denx.de> Cc: Andre Schwarz <andre.schwarz(a)matrix-vision.de> --- board/matrix_vision/mvblm7/Makefile | 4 ++-- board/matrix_vision/mvsmr/Makefile | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/board/matrix_vision/mvblm7/Makefile b/board/matrix_vision/mvblm7/Makefile index b3e3e0b..12c7cb6 100644 --- a/board/matrix_vision/mvblm7/Makefile +++ b/board/matrix_vision/mvblm7/Makefile @@ -32,13 +32,13 @@ SOBJS := $(addprefix $(obj),$(SOBJS)) $(LIB): $(obj).depend $(OBJS) $(AR) $(ARFLAGS) $@ $(OBJS) - @mkimage -T script -C none -n M7_script -d bootscript bootscript.img + @mkimage -T script -C none -n M7_script -d bootscript $(obj)bootscript.img clean: rm -f $(SOBJS) $(OBJS) distclean: clean - rm -f $(LIB) core *.bak $(obj).depend + rm -f $(LIB) core *.bak $(obj).depend $(obj)bootscript.img ######################################################################### diff --git a/board/matrix_vision/mvsmr/Makefile b/board/matrix_vision/mvsmr/Makefile index b179e6d..2817fe0 100644 --- a/board/matrix_vision/mvsmr/Makefile +++ b/board/matrix_vision/mvsmr/Makefile @@ -36,13 +36,13 @@ SOBJS := $(addprefix $(obj),$(SOBJS)) $(LIB): $(obj).depend $(OBJS) $(AR) $(ARFLAGS) $@ $(OBJS) - @mkimage -T script -C none -n mvSMR_Script -d bootscript bootscript.img + @mkimage -T script -C none -n mvSMR_Script -d bootscript $(obj)bootscript.img clean: rm -f $(SOBJS) $(OBJS) distclean: clean - rm -f $(LIB) core *.bak $(obj).depend + rm -f $(LIB) core *.bak $(obj).depend $(obj)bootscript.img ######################################################################### -- 1.6.6.1

1 1

[U-Boot] [PATCH] Makefile: fix out-of-tree building uf "u-boot.img" target
by Wolfgang Denk 15 May '10

15 May '10

Signed-off-by: Wolfgang Denk <wd(a)denx.de> --- Makefile | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/Makefile b/Makefile index 82cbbf4..1445e8b 100644 --- a/Makefile +++ b/Makefile @@ -313,7 +313,7 @@ $(obj)u-boot.ldr.srec: $(obj)u-boot.ldr $(OBJCOPY) ${OBJCFLAGS} -O srec $< $@ -I binary $(obj)u-boot.img: $(obj)u-boot.bin - ./tools/mkimage -A $(ARCH) -T firmware -C none \ + $(obj)tools/mkimage -A $(ARCH) -T firmware -C none \ -a $(TEXT_BASE) -e 0 \ -n $(shell sed -n -e 's/.*U_BOOT_VERSION//p' $(VERSION_FILE) | \ sed -e 's/"[ ]*$$/ for $(BOARD) board"/') \ -- 1.6.6.1

1 0