common: imx: Implement generic u-boot.nand target
Subsystems (3)
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main Makefile and into the CPU's Makefile (very similar to what u-boot.imx does now). The main Makefile shall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the main Makefile.
To demonstrate this implementation, add u-boot.nand target for i.MX53.
--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)
Makefile +18 -0
@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@
@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb
arch/arm/imx-common/Makefile +6 -0
@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \ + dd if=/dev/zero bs=1015 count=1 2>/dev/null ) | \ + cat - $< > $@ + $(obj)SPL: $(OBJTREE)/spl/u-boot-spl.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX_CONFIG)).cfgtmp $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SPL_TEXT_BASE) -d $< $@
Signature · 1 line
1.7.10.4
PATCH 2/4 imx: Align the imximage header and payload to multiples of 4k
The MX53 ROM loads the data from NAND in multiples of pages and supports maximum page size of 4k. Thus, align the image and header to 4k to be safe from ROM bugs.
--- tools/imximage.c | 11 +++++++---- tools/imximage.h | 3 ++- 2 files changed, 9 insertions(+), 5 deletions(-)
tools/imximage.c +7 -4
@@ -518,11 +518,14 @@ static void imximage_set_header(void *ptr, struct stat *sbuf, int ifd,/* * ROM bug alert -* mx53 only loads 512 byte multiples. -* The remaining fraction of a block bytes would -* not be loaded. +* +* MX53 only loads 512 byte multiples in case of SD boot. +* MX53 only loads NAND page multiples in case of NAND boot and +* supports up to 4096 byte large pages, thus align to 4096. +* +* The remaining fraction of a block bytes would not be loaded! */ - *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,512 ); + *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,4096 ); }int imximage_check_params (struct mkimage_params *params)
tools/imximage.h +2 -1
@@ -151,13 +151,14 @@ typedef struct {dcd_v2_t dcd_table; } imx_header_v2_t; +/* The header must be aligned to 4k on MX53 for NAND boot */ struct imx_header {union {imx_header_v1_t hdr_v1;imx_header_v2_t hdr_v2; } header;uint32_t flash_offset; -}; +} __attribute__((aligned (4096 ))); typedef void (*set_dcd_val_t)(struct imx_header *imxhdr, char *name, int lineno,
Signature · 1 line
1.7.10.4
PATCH 2/4 imx: Align the imximage header and payload to multiples of 4k
Quoted 44 lines · Marek Vasut
The MX53 ROM loads the data from NAND in multiples of pages and supports maximum page size of 4k. Thus, align the image and header to 4k to be safe from ROM bugs.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- tools/imximage.c | 11 +++++++---- tools/imximage.h | 3 ++- 2 files changed, 9 insertions(+), 5 deletions(-)tools/imximage.c +7
-4@@ -518,11 +518,14 @@ static void imximage_set_header(void *ptr, struct stat *sbuf, int ifd,/* * ROM bug alert -* mx53 only loads 512 byte multiples. -* The remaining fraction of a block bytes would -* not be loaded. +* +* MX53 only loads 512 byte multiples in case of SD boot. +* MX53 only loads NAND page multiples in case of NAND boot and +* supports up to 4096 byte large pages, thus align to 4096. +* +* The remaining fraction of a block bytes would not be loaded! */ - *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,512 ); + *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,4096 ); }int
imximage_check_params(struct mkimage_params *params)tools/imximage.h +0
-0@@ -151,13 +151,14 @@ typedef struct {dcd_v2_t dcd_table; } imx_header_v2_t;
tools/imximage.h typedef struct { · line 161
Quoted 10 lines
tools/imximage.h +2
-1+/* The header must be aligned to 4k on MX53 for NAND boot */ struct imx_header {union {imx_header_v1_t hdr_v1;imx_header_v2_t hdr_v2; } header;uint32_t flash_offset; -}; +} __attribute__((aligned (4096 )));
So what is the exact rule for the image start page in the FCB? Is it just 0x400 + 4096 + start page * page size? This does not seem to match table 7-12, unless 0 has a special meaning?
Quoted 7 lines
tools/imximage.h +0
-1typedef void (*set_dcd_val_t)(struct imx_header *imxhdr, char *name, int lineno, --1.7.10.4
Best regards,
Beno?t
PATCH 2/4 imx: Align the imximage header and payload to multiples of 4k
Dear Beno?t Th?baudeau,
Quoted 67 lines · Marek Vasut
Quoted 66 lines
The MX53 ROM loads the data from NAND in multiples of pages and supports maximum page size of 4k. Thus, align the image and header to 4k to be safe from ROM bugs.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>---
tools/imximage.c | 11 +++++++---- tools/imximage.h | 3 ++- 2 files changed, 9 insertions(+), 5 deletions(-)tools/imximage.c +7
-4@@ -518,11 +518,14 @@ static void imximage_set_header(void *ptr, struct stat *sbuf, int ifd,/* * ROM bug alert -* mx53 only loads 512 byte multiples. -* The remaining fraction of a block bytes would -* not be loaded. +* +* MX53 only loads 512 byte multiples in case of SD boot. +* MX53 only loads NAND page multiples in case of NAND boot and +* supports up to 4096 byte large pages, thus align to 4096. +* +* The remaining fraction of a block bytes would not be loaded! */ - *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,512 ); + *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,4096 ); }int imximage_check_params (struct mkimage_params *params)tools/imximage.h +1
-0@@ -151,13 +151,14 @@ typedef struct {dcd_v2_t dcd_table; } imx_header_v2_t; +/* The header must be aligned to 4k on MX53 for NAND boot */ struct imx_header {union {imx_header_v1_t hdr_v1;imx_header_v2_t hdr_v2;tools/imximage.h +0
-0} header;uint32_t flash_offset;
Quoted 6 lines
tools/imximage.h typedef struct { · line 161
Quoted 2 lines
tools/imximage.h +1
-1-}; +} __attribute__((aligned(4096)));So what is the exact rule for the image start page in the FCB? Is it just
0x400+ 4096 + start page * page size? This does not seem to match table 7-12, unless 0 has a special meaning?
I believe it's '0 == offset 4096 in NAND', do you happen to have board with 4k page NAND to verify this?
Quoted 9 lines
Quoted 6 lines
typedef void (*set_dcd_val_t)(struct
imx_header*imxhdr,char *name, int lineno,
--
1.7.10.4Best regards,
Beno?t
Best regards,
Marek Vasut
PATCH 2/4 imx: Align the imximage header and payload to multiples of 4k
Dear Marek Vasut,
Quoted 69 lines · Marek Vasut
Dear Beno?t Th?baudeau,
Quoted 67 lines · Marek Vasut
Quoted 66 lines
The MX53 ROM loads the data from NAND in multiples of pages and supports maximum page size of 4k. Thus, align the image and header to 4k to be safe from ROM bugs.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>---
tools/imximage.c | 11 +++++++---- tools/imximage.h | 3 ++- 2 files changed, 9 insertions(+), 5 deletions(-)tools/imximage.c +7
-4@@ -518,11 +518,14 @@ static void imximage_set_header(void *ptr, struct stat *sbuf, int ifd,/* * ROM bug alert -* mx53 only loads 512 byte multiples. -* The remaining fraction of a block bytes would -* not be loaded. +* +* MX53 only loads 512 byte multiples in case of SD boot. +* MX53 only loads NAND page multiples in case of NAND boot and +* supports up to 4096 byte large pages, thus align to 4096. +* +* The remaining fraction of a block bytes would not be loaded! */ - *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,512 ); + *header_size_ptr =ROUND (sbuf->st_size + imxhdr->flash_offset,4096 ); }int imximage_check_params (struct mkimage_params *params)tools/imximage.h +1
-0@@ -151,13 +151,14 @@ typedef struct {dcd_v2_t dcd_table; } imx_header_v2_t; +/* The header must be aligned to 4k on MX53 for NAND boot */ struct imx_header {union {imx_header_v1_t hdr_v1;imx_header_v2_t hdr_v2;tools/imximage.h +0
-0} header;uint32_t flash_offset;
Quoted 8 lines
Quoted 6 lines
tools/imximage.h typedef struct { · line 161
Quoted 2 lines
tools/imximage.h +1
-1-}; +} __attribute__((aligned(4096)));So what is the exact rule for the image start page in the FCB? Is it just
0x400+ 4096 + start page * page size? This does not seem to match table 7-12, unless 0 has a special meaning?I believe it's '0 == offset 4096 in NAND',
You set the i.MX header to a size of 4096, then you add the 1024-byte FCB, so we're already at 5120.
Quoted 3 lines
do you happen to have board with
4k
page NAND to verify this?
No. :(
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
typedef void (*set_dcd_val_t)(struct
imx_header*imxhdr,char *name, int lineno,
--
1.7.10.4
Best regards,
Beno?t
PATCH 3/4 mx5: Add NAND clock handling
In reply to Marek VasutAugment the MX5 clock code with function to enable and configure NFC clock. This is necessary to get NFC working on MX5.
--- arch/arm/cpu/armv7/mx5/clock.c | 12 ++++++++++-- arch/arm/include/asm/arch-mx5/clock.h | 1 + 2 files changed, 11 insertions(+), 2 deletions(-)
arch/arm/cpu/armv7/mx5/clock.c +10 -2
@@ -739,10 +739,11 @@ static int config_core_clk(u32 ref, u32 freq)static int config_nfc_clk (u32 nfc_clk) {u32 parent_rate =get_emi_slow_clk (); -u32 div = parent_rate / nfc_clk; +u32 div; -if (nfc_clk <=0 ) +if (nfc_clk ==0 )return -EINVAL ; + div = parent_rate / nfc_clk;if (div ==0 ) div++;if (parent_rate / div >NFC_CLK_MAX )
@@ -755,6 +756,13 @@ static int config_nfc_clk(u32 nfc_clk)return 0 ; } +void enable_nfc_clk (unsigned char enable) +{ +unsigned int cg = enable ?MXC_CCM_CCGR_CG_ON :MXC_CCM_CCGR_CG_OFF ; + +clrsetbits_le32 (&mxc_ccm->CCGR5,MXC_CCM_CCGR_CG_MASK <<20 , cg <<20 ); +} +/* Config main_bus_clock for periphs */ static int config_periph_clk (u32 ref,u32 freq) {
arch/arm/include/asm/arch-mx5/clock.h +1 -0
@@ -68,5 +68,6 @@ void set_usboh3_clk(void);void enable_usboh3_clk (unsigned char enable);void mxc_set_sata_internal_clock (void );int enable_i2c_clk (unsigned char enable,unsigned i2c_num); +void enable_nfc_clk (unsigned char enable);#endif /* __ASM_ARCH_CLOCK_H */
Signature · 1 line
1.7.10.4
PATCH 3/4 mx5: Add NAND clock handling
Dear Marek Vasut,
Quoted 31 lines · Marek Vasut
Augment the MX5 clock code with function to enable and configure NFC clock. This is necessary to get NFC working on MX5.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- arch/arm/cpu/armv7/mx5/clock.c | 12 ++++++++++-- arch/arm/include/asm/arch-mx5/clock.h | 1 + 2 files changed, 11 insertions(+), 2 deletions(-)arch/arm/cpu/armv7/mx5/clock.c +3
-2@@ -739,10 +739,11 @@ static int config_core_clk(u32 ref, u32 freq)static int config_nfc_clk (u32 nfc_clk) {u32 parent_rate =get_emi_slow_clk (); -u32 div = parent_rate / nfc_clk; +u32 div; -if (nfc_clk <=0 ) +if (nfc_clk ==0 )return -EINVAL ; + div = parent_rate / nfc_clk;if (div ==0 ) div++;if (parent_rate / div >NFC_CLK_MAX )
arch/arm/cpu/armv7/mx5/clock.c static int config_nfc_clk(u32 nfc_clk) · line 763
Quoted 8 lines
arch/arm/cpu/armv7/mx5/clock.c +5
-0@@ -755,6 +756,13 @@ static int config_nfc_clk(u32 nfc_clk)return 0 ; } +void enable_nfc_clk (unsigned char enable) +{ +unsigned int cg = enable ?MXC_CCM_CCGR_CG_ON :MXC_CCM_CCGR_CG_OFF ; + +clrsetbits_le32 (&mxc_ccm->CCGR5,MXC_CCM_CCGR_CG_MASK <<20 , cg <<20 );
Please use MXC_CCM_CCGR5_EMI_ENFC(MXC_CCM_CCGR_CG_MASK) and MXC_CCM_CCGR_CG_MASK(cg).
Quoted 18 lines
arch/arm/cpu/armv7/mx5/clock.c +2
-0+} +/* Config main_bus_clock for periphs */ static int config_periph_clk (u32 ref,u32 freq) {arch/arm/include/asm/arch-mx5/clock.h +1
-1@@ -68,5 +68,6 @@ void set_usboh3_clk(void);void enable_usboh3_clk (unsigned char enable);void mxc_set_sata_internal_clock (void );int enable_i2c_clk (unsigned char enable,unsigned i2c_num); +void enable_nfc_clk (unsigned char enable);#endif /* __ASM_ARCH_CLOCK_H */ --
Quoted 3 lines
1.7.10.4
Best regards,
Beno?t
PATCH 4/4 mx5: Add support for DENX M53EVK
In reply to Marek VasutAdd basic support for the DENX M53EVK board. Currently supported is the MMC, Ethernet, I2C.
--- board/denx/m53evk/Makefile | 40 ++++ board/denx/m53evk/imximage.cfg | 108 +++++++++++ board/denx/m53evk/m53evk.c | 404 ++++++++++++++++++++++++++++++++++++++++ boards.cfg | 1 + include/configs/m53evk.h | 226 ++++++++++++++++++++++ 5 files changed, 779 insertions(+) create mode 100644 board/denx/m53evk/Makefile create mode 100644 board/denx/m53evk/imximage.cfg create mode 100644 board/denx/m53evk/m53evk.c create mode 100644 include/configs/m53evk.h
board/denx/m53evk/Makefile new file +40 -0
@@ -0,0 +1,40 @@ +# +# DENX M53EVK +# Copyright (C) 2012-2013 Marek Vasut <[email protected]> +# +# 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 = $(obj)lib$(BOARD).o + +COBJS := m53evk.o + +SRCS := $(COBJS:.o=.c) +OBJS := $(addprefix $(obj),$(COBJS)) + +$(LIB): $(obj).depend $(OBJS) + $(call cmd_link_o_target, $(OBJS)) + +######################################################################### + +# defines $(obj).depend target +include $(SRCTREE)/rules.mk + +sinclude $(obj).depend + +#########################################################################
board/denx/m53evk/imximage.cfg new file +108 -0
108 lines not on this page · raw message
board/denx/m53evk/m53evk.c new file +404 -0
404 lines not on this page · raw message
boards.cfg +1 -0
@@ -239,6 +239,7 @@ am335x_evm_uart3 arm armv7 am335x ti am335x_evm_uart4 arm armv7 am335x ti am33xx am335x_evm:SERIAL5,CONS_INDEX=5 am335x_evm_uart5 arm armv7 am335x ti am33xx am335x_evm:SERIAL6,CONS_INDEX=6 highbank arm armv7 highbank - highbank +m53evk arm armv7 m53evk denx mx5 m53evk:IMX_CONFIG=board/denx/m53evk/imximage.cfg mx51_efikamx arm armv7 mx51_efikamx genesi mx5 mx51_efikamx:MACH_TYPE=MACH_TYPE_MX51_EFIKAMX,IMX_CONFIG=board/genesi/mx51_efikamx/imximage_mx.cfg mx51_efikasb arm armv7 mx51_efikamx genesi mx5 mx51_efikamx:MACH_TYPE=MACH_TYPE_MX51_EFIKASB,IMX_CONFIG=board/genesi/mx51_efikamx/imximage_sb.cfg mx51evk arm armv7 mx51evk freescale mx5 mx51evk:IMX_CONFIG=board/freescale/mx51evk/imximage.cfg
include/configs/m53evk.h new file +226 -0
226 lines not on this page · raw message
Signature · 1 line
1.7.10.4
PATCH 4/4 mx5: Add support for DENX M53EVK
Dear Marek Vasut,
Quoted 848 lines · Marek Vasut
Add basic support for the DENX M53EVK board. Currently supported is the MMC, Ethernet, I2C.
Signed-off-by Marek Vasut <[email protected]>Cc Andreas Widder <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>Cc Wolfgang Denk <[email protected]>--- board/denx/m53evk/Makefile | 40 ++++ board/denx/m53evk/imximage.cfg | 108 +++++++++++ board/denx/m53evk/m53evk.c | 404 ++++++++++++++++++++++++++++++++++++++++ boards.cfg | 1 + include/configs/m53evk.h | 226 ++++++++++++++++++++++ 5 files changed, 779 insertions(+) create mode 100644 board/denx/m53evk/Makefile create mode 100644 board/denx/m53evk/imximage.cfg create mode 100644 board/denx/m53evk/m53evk.c create mode 100644 include/configs/m53evk.hboard/denx/m53evk/Makefile new file +40
-0@@ -0,0 +1,40 @@ +# +# DENX M53EVK +# Copyright (C) 2012-2013 Marek Vasut <[email protected]> +# +# 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 = $(obj)lib$(BOARD).o + +COBJS := m53evk.o + +SRCS := $(COBJS:.o=.c) +OBJS := $(addprefix $(obj),$(COBJS)) + +$(LIB): $(obj).depend $(OBJS) + $(call cmd_link_o_target, $(OBJS)) + +######################################################################### + +# defines $(obj).depend target +include $(SRCTREE)/rules.mk + +sinclude $(obj).depend + +#########################################################################board/denx/m53evk/imximage.cfg new file +108
-0@@ -0,0 +1,108 @@ +/* + * DENX M53 DRAM init values + * Copyright (C) 2012-2013 Marek Vasut <[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. 51 Franklin Street Fifth Floor Boston, + * MA 02110-1301 USA + * + * Refer docs/README.imxmage for more details about how-to configure + * and create imximage boot image + * + * The syntax is taken as close as possible with the kwbimage + */ +/* image version */ +IMAGE_VERSION 2 + +/* + * Boot Device : one of + * spi, sd, nand + */ +BOOT_FROM nand + +/* + * Device Configuration Data (DCD) + * + * Each entry must have the format: + * Addr-type Address Value + * + * where: + * Addr-type register length (1,2 or 4 bytes) + * Address absolute address of the register + * value value to be stored in the register + */ +DATA 4 0x53fa86f4 0x00000000 /* GRP_DDRMODE_CTL */ +DATA 4 0x53fa8714 0x00000000 /* GRP_DDRMODE */ +DATA 4 0x53fa86fc 0x00000000 /* GRP_DDRPKE */ +DATA 4 0x53fa8724 0x04000000 /* GRP_DDR_TYPE */ + +DATA 4 0x53fa872c 0x00300000 /* GRP_B3DS */ +DATA 4 0x53fa8554 0x00300000 /* DRAM_DQM3 */ +DATA 4 0x53fa8558 0x00300040 /* DRAM_SDQS3 */ + +DATA 4 0x53fa8728 0x00300000 /* GRP_B2DS */ +DATA 4 0x53fa8560 0x00300000 /* DRAM_DQM2 */ +DATA 4 0x53fa8568 0x00300040 /* DRAM_SDQS2 */ + +DATA 4 0x53fa871c 0x00300000 /* GRP_B1DS */ +DATA 4 0x53fa8594 0x00300000 /* DRAM_DQM1 */ +DATA 4 0x53fa8590 0x00300040 /* DRAM_SDQS1 */ + +DATA 4 0x53fa8718 0x00300000 /* GRP_B0DS */ +DATA 4 0x53fa8584 0x00300000 /* DRAM_DQM0 */ +DATA 4 0x53fa857c 0x00300040 /* DRAM_SDQS0 */ + +DATA 4 0x53fa8578 0x00300000 /* DRAM_SDCLK_0 */ +DATA 4 0x53fa8570 0x00300000 /* DRAM_SDCLK_1 */ + +DATA 4 0x53fa8574 0x00300000 /* DRAM_CAS */ +DATA 4 0x53fa8588 0x00300000 /* DRAM_RAS */ +DATA 4 0x53fa86f0 0x00300000 /* GRP_ADDDS */ +DATA 4 0x53fa8720 0x00300000 /* GRP_CTLDS */ + +DATA 4 0x53fa8564 0x00300040 /* DRAM_SDODT1 */ +DATA 4 0x53fa8580 0x00300040 /* DRAM_SDODT0 */ + +/* ESDCTL */ +DATA 4 0x63fd9088 0x32383535 +DATA 4 0x63fd9090 0x40383538 +DATA 4 0x63fd907c 0x0136014d +DATA 4 0x63fd9080 0x01510141 + +DATA 4 0x63fd9018 0x00011740 +DATA 4 0x63fd9000 0xc3190000 +DATA 4 0x63fd900c 0x555952e3 +DATA 4 0x63fd9010 0xb68e8b63 +DATA 4 0x63fd9014 0x01ff00db +DATA 4 0x63fd902c 0x000026d2 +DATA 4 0x63fd9030 0x009f0e21 +DATA 4 0x63fd9008 0x12273030 +DATA 4 0x63fd9004 0x0002002d +DATA 4 0x63fd901c 0x00008032 +DATA 4 0x63fd901c 0x00008033 +DATA 4 0x63fd901c 0x00028031 +DATA 4 0x63fd901c 0x092080b0 +DATA 4 0x63fd901c 0x04008040 +DATA 4 0x63fd901c 0x0000803a +DATA 4 0x63fd901c 0x0000803b +DATA 4 0x63fd901c 0x00028039 +DATA 4 0x63fd901c 0x09208138 +DATA 4 0x63fd901c 0x04008048 +DATA 4 0x63fd9020 0x00001800 +DATA 4 0x63fd9040 0x04b80003 +DATA 4 0x63fd9058 0x00022227 +DATA 4 0x63fd901c 0x00000000board/denx/m53evk/m53evk.c new file +404
-0@@ -0,0 +1,404 @@ +/* +* DENX M53 module +* +* Copyright (C) 2012-2013 Marek Vasut <[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 +*/ + +#include <common.h> +#include <asm/io.h> +#include <asm/arch/imx-regs.h> +#include <asm/arch/mx5x_pins.h> +#include <asm/arch/sys_proto.h> +#include <asm/arch/crm_regs.h> +#include <asm/arch/clock.h> +#include <asm/arch/iomux.h> +#include <asm/arch/clock.h> +#include <asm/errno.h> +#include <netdev.h> +#include <i2c.h> +#include <mmc.h> +#include <fsl_esdhc.h> +#include <asm/gpio.h> +#include <usb/ehci-fsl.h> + +DECLARE_GLOBAL_DATA_PTR ; + +int dram_init (void ) +{ +u32 size1, size2; + + size1 =get_ram_size ((void *)PHYS_SDRAM_1 ,PHYS_SDRAM_1_SIZE ); + size2 =get_ram_size ((void *)PHYS_SDRAM_2 ,PHYS_SDRAM_2_SIZE ); + + gd->ram_size = size1 + size2; + +return 0 ; +} +void dram_init_banksize (void ) +{ + gd->bd->bi_dram[0 ].start =PHYS_SDRAM_1 ; + gd->bd->bi_dram[0 ].size =PHYS_SDRAM_1_SIZE ; + + gd->bd->bi_dram[1 ].start =PHYS_SDRAM_2 ; + gd->bd->bi_dram[1 ].size =PHYS_SDRAM_2_SIZE ; +} + +u32 get_board_rev (void ) +{ +struct iim_regs *iim = (struct iim_regs *)IMX_IIM_BASE ; +struct fuse_bank *bank = &iim->bank[0 ]; +struct fuse_bank0_regs *fuse = + (struct fuse_bank0_regs *)bank->fuse_regs; +int rev =readl (&fuse->gp[6 ]); + +return (get_cpu_rev () & ~(0xF <<8 )) | (rev &0xF ) <<8 ; +} + +static void setup_iomux_uart (void ) +{ +mxc_request_iomux (MX53_PIN_ATA_BUFFER_EN ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DMARQ ,IOMUX_CONFIG_ALT3 ); + +mxc_iomux_set_pad (MX53_PIN_ATA_BUFFER_EN , +PAD_CTL_SRE_FAST |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DMARQ , +PAD_CTL_SRE_FAST |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE ); + +mxc_iomux_set_input (MX53_UART2_IPP_UART_RXD_MUX_SELECT_INPUT ,0x3 ); +} + +#ifdef CONFIG_USB_EHCI_MX5 +int board_ehci_hcd_init (int port) +{ +if (port ==0 ) { +/* USB OTG PWRON */ +mxc_request_iomux (MX53_PIN_GPIO_4 ,IOMUX_CONFIG_ALT1 ); +mxc_iomux_set_pad (MX53_PIN_GPIO_4 , +PAD_CTL_PKE_ENABLE | +PAD_CTL_100K_PD | +PAD_CTL_DRV_HIGH + ); +gpio_direction_output (IOMUX_TO_GPIO (MX53_PIN_GPIO_4 ),0 ); + +/* USB OTG Over Current */ +mxc_request_iomux (MX53_PIN_GPIO_18 ,IOMUX_CONFIG_ALT1 ); +mxc_iomux_set_input (MX53_USBOH3_IPP_IND_OTG_OC_SELECT_INPUT ,1 ); + }else if (port ==1 ) { +/* USB Host PWRON */ +mxc_request_iomux (MX53_PIN_GPIO_2 ,IOMUX_CONFIG_ALT1 ); +mxc_iomux_set_pad (MX53_PIN_GPIO_2 , +PAD_CTL_PKE_ENABLE | +PAD_CTL_100K_PD | +PAD_CTL_DRV_HIGH + ); +gpio_direction_output (IOMUX_TO_GPIO (MX53_PIN_GPIO_2 ),0 ); + +/* USB Host Over Current */ +mxc_request_iomux (MX53_PIN_GPIO_3 ,IOMUX_CONFIG_ALT6 ); +mxc_iomux_set_input (MX53_USBOH3_IPP_IND_UH1_OC_SELECT_INPUT ,1 ); + } + +return 0 ; +} +#endif + +static void setup_iomux_fec (void ) +{ +/* MDIO IOMUX */ +mxc_request_iomux (MX53_PIN_FEC_MDIO ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_MDC ,IOMUX_CONFIG_ALT0 ); + +/* FEC 0 IOMUX */ +mxc_request_iomux (MX53_PIN_FEC_CRS_DV ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_REF_CLK ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_RX_ER ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_TX_EN ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_RXD0 ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_RXD1 ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_TXD0 ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_FEC_TXD1 ,IOMUX_CONFIG_ALT0 ); + +/* FEC 1 IOMUX */ +mxc_request_iomux (MX53_PIN_KEY_COL0 ,IOMUX_CONFIG_ALT6 );/* RXD3 */ +mxc_request_iomux (MX53_PIN_KEY_ROW0 ,IOMUX_CONFIG_ALT6 );/* TX_ER */ +mxc_request_iomux (MX53_PIN_KEY_COL1 ,IOMUX_CONFIG_ALT6 );/* RX_CLK */ +mxc_request_iomux (MX53_PIN_KEY_ROW1 ,IOMUX_CONFIG_ALT6 );/* COL */ +mxc_request_iomux (MX53_PIN_KEY_COL2 ,IOMUX_CONFIG_ALT6 );/* RXD2 */ +mxc_request_iomux (MX53_PIN_KEY_ROW2 ,IOMUX_CONFIG_ALT6 );/* TXD2 */ +mxc_request_iomux (MX53_PIN_KEY_COL3 ,IOMUX_CONFIG_ALT6 );/* CRS */ +mxc_request_iomux (MX53_PIN_GPIO_19 ,IOMUX_CONFIG_ALT6 );/* TXD3 */ + +/* MDIO PADs */ +mxc_iomux_set_pad (MX53_PIN_FEC_MDIO , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_22K_PU |PAD_CTL_ODE_OPENDRAIN_ENABLE ); +mxc_iomux_set_input (MX53_FEC_FEC_MDI_SELECT_INPUT ,0x1 ); +mxc_iomux_set_pad (MX53_PIN_FEC_MDC ,PAD_CTL_DRV_HIGH ); + +/* FEC 0 PADs */ +mxc_iomux_set_pad (MX53_PIN_FEC_CRS_DV , +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_FEC_REF_CLK , +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_FEC_RX_ER , +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_FEC_TX_EN ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_FEC_RXD0 , +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_FEC_RXD1 , +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_FEC_TXD0 ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_FEC_TXD1 ,PAD_CTL_DRV_HIGH ); + +/* FEC 1 PADs */ +mxc_iomux_set_pad (MX53_PIN_KEY_COL0 ,/* RXD3 */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_KEY_ROW0 ,/* TX_ER */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_KEY_COL1 ,/* RX_CLK */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_KEY_ROW1 ,/* COL */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_KEY_COL2 ,/* RXD2 */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_KEY_ROW2 ,/* TXD2 */ +PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_KEY_COL3 ,/* CRS */ +PAD_CTL_HYS_ENABLE |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_GPIO_19 ,/* TXD3 */ +PAD_CTL_DRV_HIGH ); +} + +#ifdef CONFIG_FSL_ESDHC +struct fsl_esdhc_cfg esdhc_cfg = { +MMC_SDHC1_BASE_ADDR , +}; + +int board_mmc_getcd (struct mmc *mmc) +{ +mxc_request_iomux (MX53_PIN_GPIO_1 ,IOMUX_CONFIG_ALT1 ); +gpio_direction_input (IMX_GPIO_NR (1 ,1 )); + +return !gpio_get_value (IMX_GPIO_NR (1 ,1 )); +} + +int board_mmc_init (bd_t *bis) +{ + esdhc_cfg.sdhc_clk =mxc_get_clock (MXC_ESDHC_CLK ); + +mxc_request_iomux (MX53_PIN_SD1_CMD ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_SD1_CLK ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_SD1_DATA0 , +IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_SD1_DATA1 , +IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_SD1_DATA2 , +IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_SD1_DATA3 , +IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_EIM_DA13 , +IOMUX_CONFIG_ALT1 ); + +mxc_iomux_set_pad (MX53_PIN_SD1_CMD , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_100K_PU ); +mxc_iomux_set_pad (MX53_PIN_SD1_CLK , +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_47K_PU | +PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_SD1_DATA0 , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_47K_PU ); +mxc_iomux_set_pad (MX53_PIN_SD1_DATA1 , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_47K_PU ); +mxc_iomux_set_pad (MX53_PIN_SD1_DATA2 , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_47K_PU ); +mxc_iomux_set_pad (MX53_PIN_SD1_DATA3 , +PAD_CTL_HYS_ENABLE |PAD_CTL_DRV_HIGH | +PAD_CTL_PUE_PULL |PAD_CTL_PKE_ENABLE | +PAD_CTL_HYS_ENABLE |PAD_CTL_47K_PU ); + +/* GPIO 2_31 is SD power */ +mxc_request_iomux (MX53_PIN_EIM_EB3 ,IOMUX_CONFIG_ALT1 ); +gpio_direction_output (IMX_GPIO_NR (2 ,31 ),0 ); + +return fsl_esdhc_initialize (bis, &esdhc_cfg); +} +#endif + +static void setup_iomux_i2c (void ) +{ +mxc_request_iomux (MX53_PIN_EIM_D16 , +IOMUX_CONFIG_ALT5 |IOMUX_CONFIG_SION ); +mxc_request_iomux (MX53_PIN_EIM_EB2 , +IOMUX_CONFIG_ALT5 |IOMUX_CONFIG_SION ); + +mxc_iomux_set_pad (MX53_PIN_EIM_D16 , +PAD_CTL_SRE_FAST |PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE | +PAD_CTL_PUE_PULL | +PAD_CTL_ODE_OPENDRAIN_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_EIM_EB2 , +PAD_CTL_SRE_FAST |PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE | +PAD_CTL_PUE_PULL | +PAD_CTL_ODE_OPENDRAIN_ENABLE ); + +mxc_iomux_set_input (MX53_I2C2_IPP_SDA_IN_SELECT_INPUT ,0x1 ); +mxc_iomux_set_input (MX53_I2C2_IPP_SCL_IN_SELECT_INPUT ,0x1 ); +} + +static void setup_iomux_nand (void ) +{ +mxc_request_iomux (MX53_PIN_NANDF_WE_B ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_RE_B ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_CLE ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_ALE ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_WP_B ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_RB0 ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_NANDF_CS0 ,IOMUX_CONFIG_ALT0 ); +mxc_request_iomux (MX53_PIN_ATA_DATA0 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA1 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA2 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA3 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA4 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA5 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA6 ,IOMUX_CONFIG_ALT3 ); +mxc_request_iomux (MX53_PIN_ATA_DATA7 ,IOMUX_CONFIG_ALT3 ); + +mxc_iomux_set_pad (MX53_PIN_NANDF_WE_B ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_NANDF_RE_B ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_NANDF_CLE ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_NANDF_ALE ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_NANDF_WP_B ,PAD_CTL_PUE_PULL | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_NANDF_RB0 ,PAD_CTL_PUE_PULL | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_NANDF_CS0 ,PAD_CTL_DRV_HIGH ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA0 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA1 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA2 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA3 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA4 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA5 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA6 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +mxc_iomux_set_pad (MX53_PIN_ATA_DATA7 ,PAD_CTL_DRV_HIGH | +PAD_CTL_100K_PU |PAD_CTL_PKE_ENABLE ); +} + +static void m53_set_clock (uint32_t cpuclk,uint32_t dramclk) +{ +int ret; +const uint32_t ref_clk =MXC_HCLK ; + + ret =mxc_set_clock (ref_clk, cpuclk,MXC_ARM_CLK ); +if (ret) +printf ("CPU: Switch CPU clock to %dMHz failed\n" , cpuclk); + + ret =mxc_set_clock (ref_clk, dramclk,MXC_PERIPH_CLK ); +if (ret) { +printf ("CPU: Switch peripheral clock to %dMHz failed\n" , + dramclk); + } + + ret =mxc_set_clock (ref_clk, dramclk,MXC_DDR_CLK ); +if (ret) +printf ("CPU: Switch DDR clock to %dMHz failed\n" , dramclk); +} + +int board_early_init_f (void ) +{ +setup_iomux_uart (); +setup_iomux_fec (); +setup_iomux_i2c (); +setup_iomux_nand (); + +return 0 ; +} + +int print_cpuinfo (void ) +{ +u32 cpurev; + + cpurev =get_cpu_rev (); +printf ("CPU: Freescale i.MX%x family rev%d.%d at %d MHz\n" , + (cpurev &0xFF000 ) >>12 , + (cpurev &0x000F0 ) >>4 , + (cpurev &0x0000F ) >>0 , +mxc_get_clock (MXC_ARM_CLK ) /1000000 ); +printf ("Reset cause: %s\n" ,get_reset_cause ()); +return 0 ; +} + +/* +* Do not overwrite the console +* Use always serial for U-Boot console +*/ +int overwrite_console (void ) +{ +return 1 ; +} + +int board_init (void ) +{ + gd->bd->bi_boot_params =PHYS_SDRAM_1 +0x100 ; + +m53_set_clock (800 ,400 ); + +mxc_set_sata_internal_clock (); + +mxc_set_clock (0 ,33 ,MXC_NFC_CLK ); +enable_nfc_clk (1 ); + +return 0 ; +} + +int board_late_init (void ) +{ +print_cpuinfo (); + +return 0 ; +} + +int checkboard (void ) +{ +puts ("Board: DENX M53EVK\n" ); + +return 0 ; +}boards.cfg +0
-0@@ -239,6 +239,7 @@ am335x_evm_uart3 arm armv7am335x ti
am335x_evm_uart4arm armv7 am335x ti
am33xx am335x_evm:SERIAL5,CONS_INDEX=5
am335x_evm_uart5arm armv7 am335x ti
am33xx am335x_evm:SERIAL6,CONS_INDEX=6
highbank arm armv7 highbank -
highbank
+m53evk arm armv7 m53evk
denx mx5 m53evk:IMX_CONFIG=board/denx/m53evk/imximage.cfg
mx51_efikamxarm armv7mx51_efikamx
genesi mx5boards.cfg +0
-0mx51_efikamx:MACH_TYPE=MACH_TYPE_MX51_EFIKAMX,IMX_CONFIG=board/genesi/mx51_efikamx/imximage_mx.cfg mx51_efikasb arm armv7 mx51_efikamx genesi mx5 mx51_efikamx:MACH_TYPE=MACH_TYPE_MX51_EFIKASB,IMX_CONFIG=board/genesi/mx51_efikamx/imximage_sb.cfg mx51evk arm armv7 mx51evk freescale mx5 mx51evk:IMX_CONFIG=board/freescale/mx51evk/imximage.cfginclude/configs/m53evk.h new file +226
-1@@ -0,0 +1,226 @@ +/* +* DENX M53 configuration +* Copyright (C) 2012-2013 Marek Vasut <[email protected]> +* +* 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 __M53EVK_CONFIG_H__ +#define __M53EVK_CONFIG_H__ + +#define CONFIG_MX53 +#define CONFIG_MXC_GPIO +#define CONFIG_SYS_HZ 1000 + +#include <asm/arch/imx-regs.h> + +#define CONFIG_BOARD_LATE_INIT +#define CONFIG_BOARD_EARLY_INIT_F +#define CONFIG_REVISION_TAG +#define CONFIG_SYS_NO_FLASH + +/* +* U-Boot Commands +*/ +#include <config_cmd_default.h> +#define CONFIG_DISPLAY_BOARDINFO +#define CONFIG_DOS_PARTITION + +#define CONFIG_CMD_DHCP +#define CONFIG_CMD_EXT2 +#define CONFIG_CMD_FAT +#define CONFIG_CMD_FAT +#define CONFIG_CMD_I2C +#define CONFIG_CMD_MII +#define CONFIG_CMD_MMC +#define CONFIG_CMD_NAND +#define CONFIG_CMD_NET +#define CONFIG_CMD_PING +#define CONFIG_CMD_SATA +#define CONFIG_CMD_USB + +/* +* Memory configurations +*/ +#define CONFIG_NR_DRAM_BANKS 2 +#define PHYS_SDRAM_1 CSD0_BASE_ADDR +#define PHYS_SDRAM_1_SIZE (512 * 1024 * 1024) +#define PHYS_SDRAM_2 CSD1_BASE_ADDR +#define PHYS_SDRAM_2_SIZE (512 * 1024 * 1024) +#define PHYS_SDRAM_SIZE (PHYS_SDRAM_1_SIZE + PHYS_SDRAM_2_SIZE) +#define CONFIG_SYS_MALLOC_LEN (10 * 1024 * 1024) +#define CONFIG_SYS_MEMTEST_START 0x70010000 +#define CONFIG_SYS_MEMTEST_END 0x70020000 + +#define CONFIG_SYS_SDRAM_BASE (PHYS_SDRAM_1) +#define CONFIG_SYS_INIT_RAM_ADDR (IRAM_BASE_ADDR) +#define CONFIG_SYS_INIT_RAM_SIZE (IRAM_SIZE) + +#define CONFIG_SYS_INIT_SP_OFFSET \ + (CONFIG_SYS_INIT_RAM_SIZE - GENERATED_GBL_DATA_SIZE) +#define CONFIG_SYS_INIT_SP_ADDR \ + (CONFIG_SYS_INIT_RAM_ADDR + CONFIG_SYS_INIT_SP_OFFSET) + +#define CONFIG_SYS_TEXT_BASE 0x77800000 + +/* +* U-Boot general configurations +*/ +#define CONFIG_SYS_LONGHELP +#define CONFIG_SYS_PROMPT "=> " +#define CONFIG_SYS_CBSIZE 1024/* Console I/O buffer size */ +#define CONFIG_SYS_PBSIZE \ + (CONFIG_SYS_CBSIZE + sizeof(CONFIG_SYS_PROMPT) + 16) +/* Print buffer size */ +#define CONFIG_SYS_MAXARGS 32/* Max number of command args */ +#define CONFIG_SYS_BARGSIZE CONFIG_SYS_CBSIZE +/* Boot argument buffer size */ +#define CONFIG_VERSION_VARIABLE /* U-BOOT version */ +#define CONFIG_AUTO_COMPLETE /* Command auto complete */ +#define CONFIG_CMDLINE_EDITING /* Command history etc */ +#define CONFIG_SYS_HUSH_PARSER + +/* +* Serial Driver +*/ +#define CONFIG_MXC_UART +#define CONFIG_MXC_UART_BASE UART2_BASE +#define CONFIG_CONS_INDEX 1 +#define CONFIG_BAUDRATE 115200 + +/* +* MMC Driver +*/ +#ifdef CONFIG_CMD_MMC +#define CONFIG_MMC +#define CONFIG_GENERIC_MMC +#define CONFIG_FSL_ESDHC +#define CONFIG_SYS_FSL_ESDHC_ADDR 0 +#define CONFIG_SYS_FSL_ESDHC_NUM 1 +#endif + +/* +* NAND +*/ +#define CONFIG_ENV_SIZE (16 * 1024) +#ifdef CONFIG_CMD_NAND +#define CONFIG_SYS_MAX_NAND_DEVICE 1 +#define CONFIG_SYS_NAND_BASE NFC_BASE_ADDR_AXI +#define CONFIG_NAND_MXC +#define CONFIG_MXC_NAND_REGS_BASE NFC_BASE_ADDR_AXI +#define CONFIG_MXC_NAND_IP_REGS_BASE NFC_BASE_ADDR +#define CONFIG_SYS_NAND_LARGEPAGE +#define CONFIG_MXC_NAND_HWECC +#define CONFIG_SYS_NAND_USE_FLASH_BBT + +/* Environment is in NAND */ +#define CONFIG_ENV_IS_IN_NAND +#define CONFIG_ENV_SIZE_REDUND CONFIG_ENV_SIZE +#define CONFIG_ENV_SECT_SIZE (128 * 1024) +#define CONFIG_ENV_RANGE (512 * 1024) +#define CONFIG_ENV_OFFSET 0x100000 +#define CONFIG_ENV_OFFSET_REDUND \ + (CONFIG_ENV_OFFSET + CONFIG_ENV_RANGE) + +#define CONFIG_CMD_UBI +#define CONFIG_CMD_UBIFS +#define CONFIG_CMD_MTDPARTS +#define CONFIG_RBTREE +#define CONFIG_LZO +#define CONFIG_MTD_DEVICE +#define CONFIG_MTD_PARTITIONS +#define MTDIDS_DEFAULT "nand0=mxc-nand" +#define MTDPARTS_DEFAULT \ + "mtdparts=mxc-nand:" \ + "1m(bootloader)ro," \ + "512k(environment)," \ + "512k(redundant-environment)," \ + "4m(kernel)," \ + "128k(fdt)," \ + "8m(ramdisk)," \ + "-(filesystem)" +#else +#define CONFIG_ENV_IS_NOWHERE +#endif + +/* +* Ethernet on SOC (FEC) +*/ +#ifdef CONFIG_CMD_NET +#define CONFIG_FEC_MXC +#define CONFIG_ETHPRIME "FEC0" +#define IMX_FEC_BASE FEC_BASE_ADDR +#define CONFIG_FEC_MXC_PHYADDR 0x0 +#define CONFIG_MII +#define CONFIG_DISCOVER_PHY +#define CONFIG_FEC_XCV_TYPE RMII +#define CONFIG_PHYLIB +#define CONFIG_PHY_MICREL +#define CONFIG_ARP_TIMEOUT 200UL +#endif + +/* +* I2C +*/ +#ifdef CONFIG_CMD_I2C +#define CONFIG_HARD_I2C +#define CONFIG_I2C_MXC +#define CONFIG_SYS_I2C_BASE I2C2_BASE_ADDR +#define CONFIG_SYS_I2C_SPEED 100000 +#endif + +/* +* USB +*/ +#ifdef CONFIG_CMD_USB +#define CONFIG_USB_EHCI +#define CONFIG_USB_EHCI_MX5 +#define CONFIG_USB_STORAGE +#define CONFIG_USB_HOST_ETHER +#define CONFIG_USB_ETHER_ASIX +#define CONFIG_USB_ETHER_SMSC95XX +#define CONFIG_MXC_USB_PORT 1 +#define CONFIG_MXC_USB_PORTSC (PORT_PTS_UTMI | PORT_PTS_PTW) +#define CONFIG_MXC_USB_FLAGS 0 +#endif + +/* +* SATA +*/ +#ifdef CONFIG_CMD_SATA +#define CONFIG_DWC_AHSATA +#define CONFIG_SYS_SATA_MAX_DEVICE 1 +#define CONFIG_DWC_AHSATA_PORT_ID 0 +#define CONFIG_DWC_AHSATA_BASE_ADDR SATA_BASE_ADDR +#define CONFIG_LBA48 +#define CONFIG_LIBATA +#endif + +/* +* Boot Linux +*/ +#define CONFIG_CMDLINE_TAG +#define CONFIG_INITRD_TAG +#define CONFIG_SETUP_MEMORY_TAGS +#define CONFIG_BOOTDELAY 3 +#define CONFIG_BOOTFILE "uImage" +#define CONFIG_BOOTARGS "console=ttymxc1,115200" +#define CONFIG_BOOTCOMMAND "run bootcmd_net" +#define CONFIG_LOADADDR 0x70800000 +#define CONFIG_SYS_LOAD_ADDR CONFIG_LOADADDR +#define CONFIG_OF_LIBFDT + +#endif /* __M53EVK_CONFIG_H__ */ --
Quoted 3 lines
1.7.10.4
Looks good, but I don't know the hardware details of this board.
The MAINTAINERS file should be updated.
Best regards,
Beno?t
Fabio Estevam
reviewerPATCH 4/4 mx5: Add support for DENX M53EVK
In reply to Marek Vasutboard/denx/m53evk/m53evk.c line 366
Quoted 12 lines · Marek Vasut
board/denx/m53evk/m53evk.c new file +12
-0+int print_cpuinfo (void ) +{ +u32 cpurev; + + cpurev =get_cpu_rev (); +printf ("CPU: Freescale i.MX%x family rev%d.%d at %d MHz\n" , + (cpurev &0xFF000 ) >>12 , + (cpurev &0x000F0 ) >>4 , + (cpurev &0x0000F ) >>0 , +mxc_get_clock (MXC_ARM_CLK ) /1000000 ); +printf ("Reset cause: %s\n" ,get_reset_cause ()); +return 0 ;
Do you need to define print_cpuinfo locally?
I know I did the same on mx53loco, and the reason is that due to the PMIC I need to read the CPU frequency after the PMIC has raised the core voltage and bumped to 1GHz.
I do not see PMIC definitions in this board, so you probably you can just remove this local definition and add
#define CONFIG_DISPLAY_CPUINFO into your m53evk.h board instead.
Fabio Estevam
reviewerPATCH 4/4 mx5: Add support for DENX M53EVK
In reply to Marek VasutHi Marek,
board/denx/m53evk/imximage.cfg line 35
Quoted 5 lines · Marek Vasut
board/denx/m53evk/imximage.cfg new file +5
-0+/* + * Boot Device : one of + * spi, sd, nand + */ +BOOT_FROM nand
Just curious: which method did you use to flash the final binary into the NAND? Did you boot the kernel and used mtd-utils?
Maybe we could add such info later into a README?
PATCH 4/4 mx5: Add support for DENX M53EVK
Dear Fabio Estevam,
Quoted 12 lines
Hi Marek,
board/denx/m53evk/imximage.cfg line 35
Quoted 5 lines · Marek Vasut
board/denx/m53evk/imximage.cfg new file +5
-0+/* + * Boot Device : one of + * spi, sd, nand + */ +BOOT_FROM nandJust curious: which method did you use to flash the final binary into the NAND? Did you boot the kernel and used mtd-utils?
Quoted 1 line
Maybe we could add such info later into a README?
I just wrote the resulting u-boot.nand to the 0x0 of the NAND.
Best regards,
Marek Vasut
PATCH 4/4 mx5: Add support for DENX M53EVK
Dear Marek Vasut,
Quoted 7 lines · In message <[email protected]> you wrote:
Quoted 5 lines
Just curious: which method did you use to flash the final binary into the NAND? Did you boot the kernel and used mtd-utils?
Maybe we could add such info later into a README?
I just wrote the resulting u-boot.nand to the
0x0of the NAND.
This is not exactly a very precise instruction...
How exactly do I write to NAND on a virgin (or bricked) system? I think Fabio's request makes a lot of sense - yes, you know how to do this, but I don't, and neither does the average user. Can we please document this? With sufficient detail everybody understands which tools you used to do this, and what the exact steps where to perform the operation?
Thanks.
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 Einstein argued that there must be simplified explanations of nature, because God is not capricious or arbitrary. No such faith comforts the software engineer. - Fred Brooks, Jr.
PATCH 4/4 mx5: Add support for DENX M53EVK
Dear Wolfgang Denk,
Quoted 11 lines
Dear Marek Vasut,
Quoted 6 lines · In message <[email protected]> you wrote:
Quoted 4 lines
Just curious: which method did you use to flash the final binary into the NAND? Did you boot the kernel and used mtd-utils?
Maybe we could add such info later into a README?
I just wrote the resulting u-boot.nand to the
0x0of the NAND.This is not exactly a very precise instruction...
Use simple "nand write" to flash the resulting image to address 0x0 of the nand, that's all.
Quoted 3 lines
How exactly do I write to NAND on a virgin (or bricked) system? I think Fabio's request makes a lot of sense - yes, you know how to do this, but I don't, and neither does the average user.
This will change anyway, as will the rest of the patchset as I will need to rework it on top of Benoit's patches.
Quoted 8 lines
Can we
please document this? With sufficient detail everybody understands which tools you used to do this, and what the exact steps where to perform the operation?Thanks.
Wolfgang Denk
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
In reply to Marek VasutDear Marek Vasut,
Quoted 24 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +0
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 480
Quoted 10 lines
Makefile +8
-0$(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common
What about calling it rather CONFIG_NAND_TGT_PATH? TRG looks more like trigger.
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 483
Quoted 3 lines
Makefile +3
-0+endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend
^
$(obj)u-boot.bin already depends on depend through $(obj)u-boot, so it's useless here.
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 5 lines
Makefile +5
-0+ if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand
^
$(OBJTREE)/Quoted 18 lines
Makefile +2
-0+ $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 57
Quoted 8 lines
arch/arm/imx-common/Makefile +5
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \ + dd if=/dev/zero bs=1015 count=1 2>/dev/null ) | \ + cat - $< > $@
Is that all? According to 7.5.2.2 (i.MX53 RM), the boot ROM switches to serial mode if anything goes wrong with the NAND. Hence, for reliable NAND boot, you have to choose either DBBT or SPL (or both, but that would be waste).
Populating the DBBT would be complicated, so I'd go for SPL. You could just use my u-boot-with-nand-spl.imx and change the dummy header to a true FCB with the fingerprint like you did in your header above. u-boot.nand then becomes useless.
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 61
Quoted 5 lines
arch/arm/imx-common/Makefile +1
-0+ $(obj)SPL: $(OBJTREE)/spl/u-boot-spl.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX_CONFIG)).cfgtmp $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SPL_TEXT_BASE) -d $< $@
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Beno?t Th?baudeau,
Quoted 49 lines
Dear Marek Vasut,
Quoted 34 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>---
Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +4
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementationof +# the actual u-boot.nand generator below.
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 480
Quoted 3 lines
Makefile +3
-0+# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-commonWhat about calling it rather
CONFIG_NAND_TGT_PATH? TRG looks more like trigger.Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 483
Quoted 3 lines
Makefile +3
-0+endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend^
Quoted 4 lines
$(obj)u-boot.bin already depends on depend through $(obj)u-boot, so it's useless here.
Makefile
Quoted 1 line
Makefile +1
-0+ if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then
\
Quoted 1 line
Makefile
Quoted 1 line
Makefile +1
-0+ echo "This CPU does not support u-boot.nand target!" ;
\
Quoted 1 line
Makefile
Quoted 1 line
Makefile +1
-0+ exit 1 ;
\
Quoted 35 lines
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 2 lines
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand^ $(OBJTREE)/Quoted 29 lines
Makefile +2
-0+ $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nandMakefile +0
-0@rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile +0
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@
Quoted 7 lines
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 57
Quoted 5 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx
arch/arm/imx-common/Makefile +4
-0+ ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \ + dd if=/dev/zero bs=1015 count=1 2>/dev/null ) | \ + cat - $< > $@Is that all?
That's all for now, that's how it boots and how others do it as well. In the long run though, I'd prefer to bend mxsboot to generate the DBBT.
Quoted 12 lines
According to 7.5.2.2 (i.MX53 RM), the boot ROM switches to serial mode if anything goes wrong with the NAND. Hence, for reliable NAND boot, you have to choose either DBBT or SPL (or both, but that would be waste).
Populating the DBBT would be complicated, so I'd go for SPL. You could just use my u-boot-with-nand-spl.imx and change the dummy header to a true FCB with the fingerprint like you did in your header above. u-boot.nand then becomes useless.
Quoted 2 lines
+
Quoted 8 lines
arch/arm/imx-common/Makefile
Quoted 5 lines
$(obj)SPL: $(OBJTREE)/spl/u-boot-spl.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX_CONFIG)).cfgtmp
arch/arm/imx-common/Makefile +0
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SPL_TEXT_BASE) -d $< $@Best regards,
Beno?t
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Marek Vasut,
Quoted 103 lines · Marek Vasut
Dear Beno?t Th?baudeau,
Quoted 49 lines
Dear Marek Vasut,
Quoted 34 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>---
Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +4
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementationof +# the actual u-boot.nand generator below.
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 480
Quoted 3 lines
Makefile +3
-0+# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-commonWhat about calling it rather
CONFIG_NAND_TGT_PATH? TRG looks more like trigger.Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 483
Quoted 3 lines
Makefile +3
-0+endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend^Quoted 4 lines
$(obj)u-boot.bin already depends on depend through $(obj)u-boot, so it's useless here.
Makefile
Quoted 1 line
Makefile +1
-0+ if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then\
Quoted 1 line
Makefile
Quoted 1 line
Makefile +1
-0+ echo "This CPU does not support u-boot.nand target!" ;\
Quoted 1 line
Makefile
Quoted 1 line
Makefile +1
-0+ exit 1 ;\
Quoted 35 lines
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 2 lines
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand^ $(OBJTREE)/Quoted 29 lines
Makefile +2
-0+ $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nandMakefile +0
-0@rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile +0
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Quoted 7 lines
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 57
Quoted 5 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx
arch/arm/imx-common/Makefile +4
-0+ ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \ + dd if=/dev/zero bs=1015 count=1 2>/dev/null ) | \ + cat - $< > $@Is that all?
Quoted 2 lines
That's all for now, that's how it boots and how others do it as well. In the long run though, I'd prefer to bend mxsboot to generate the DBBT.
After a quick look at mxsboot, it seems to just generate an empty DBBT, so this does not help for boot reliability.
Quoted 10 lines
Quoted 10 lines
According to 7.5.2.2 (i.MX53 RM), the boot ROM switches to serial mode if anything goes wrong with the NAND. Hence, for reliable NAND boot, you have to choose either DBBT or SPL (or both, but that would be waste).
Populating the DBBT would be complicated, so I'd go for SPL. You could just use my u-boot-with-nand-spl.imx and change the dummy header to a true FCB with the fingerprint like you did in your header above. u-boot.nand then becomes useless.
Quoted 7 lines
Quoted 7 lines
arch/arm/imx-common/Makefile
Quoted 7 lines
arch/arm/imx-common/Makefile +1
-0+ $(obj)SPL: $(OBJTREE)/spl/u-boot-spl.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX_CONFIG)).cfgtmp $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SPL_TEXT_BASE) -d $< $@
Best regards,
Beno?t
[U-Boot] patch error
In reply to Benoît ThébaudeauDear Beno?t Th?baude,
I git the u-boot-imx from denx.And I patch your patches,but I get error
as:
git apply U-Boot-v7-14-19-imx-Fix-automatic-make-targets-for-imx-images.patch
error:
error: patch failed: Makefile:486
error: Makefile: patch does not apply
How can i do?
thanks
wanxs
PATCH 1/4 common: imx: Implement generic u-boot.nand target
In reply to Marek VasutQuoted 49 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target
out of the mainMakefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only contain path to which the
u-boot.nand target is delegated. Hopefully this will not produce too much
bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +0
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common$(obj)u-boot.imx
Makefile +4
-0+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to animplementation of
Makefile +8
-0+# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ;then \
Makefile +1
-0+ echo "This CPU does not support u-boot.nandtarget!" ; \
Makefile +1
-0+ exit 1; \
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 2 lines
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand
We already have CONFIG_SPL_TARGET. Why do we need a new mechanism specific to NAND, with ifdefs in the main Makefile?
If we do have something specific to NAND, it should be u-boot-nand.bin to match what nand_spl produced, and it should just be a generic rule that links to what CONFIG_SPL_TARGET produced, if the right CONFIG symbol is set by the board config to say that this is a NAND SPL. Or is someone going to say that they want to generate multiple different SPL images from the same build? :-P
-Scott
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Scott Wood,
Quoted 51 lines · Marek Vasut
Quoted 50 lines
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target
out of the mainMakefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only contain path to which the
u-boot.nand target is delegated. Hopefully this will not produce too much
bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>---
Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +0
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common$(obj)u-boot.imx
Makefile +4
-0+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to animplementation of
Makefile +8
-0+# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ;then \
Makefile +1
-0+ echo "This CPU does not support u-boot.nand
Quoted 8 lines
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 5 lines
target!" ; \
Makefile +1
-0+ exit 1; \
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nandWe already have
CONFIG_SPL_TARGET. Why do we need a new mechanism specific to NAND, with ifdefs in the mainMakefile?
This board doesn't use SPL at all.
Quoted 6 lines
If we do have something specific to NAND, it should be u-boot-nand.bin to match what
nand_splproduced, and it should just be a generic rule that links to whatCONFIG_SPL_TARGETproduced, if the right CONFIG symbol is set by the board config to say that this is a NAND SPL. Or is someone going to say that they want to generate multiple different SPL images from the same build? :-P
I'd like to generate a bootable NAND image on mx53 and on mx23/28, thus I need generic target. And I need it flexible enough, since on mx23/28 it munges u- boot.sb into the resulting image while on mx53 it munges u-boot.imx into the resulting image.
Quoted 1 line
-Scott
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 16 lines · Marek Vasut
Dear Scott Wood,
Quoted 5 lines · Marek Vasut
Quoted 4 lines
+$(obj)u-boot.nand: $(obj)u-boot.bin depend
(quoted diff) +1
-0+ if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ;then \
(quoted diff) +1
-0+ echo "This CPU does not support u-boot.nandQuoted 8 lines
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 5 lines
target!" ; \
(quoted diff) +1
-0+ exit 1; \
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nandWe already have
CONFIG_SPL_TARGET. Why do we need a new mechanism specific to NAND, with ifdefs in the mainMakefile?
Quoted 1 line
This board doesn't use SPL at all.
So maybe we need a more general (but optional) CONFIG_BUILD_TARGET.
Quoted 4 lines
Quoted 1 line
If we do have something specific to NAND, it should be
u-boot-nand.bin
Quoted 1 line
to match what
nand_splproduced, and it should just be a genericrule
Quoted 8 lines
Quoted 2 lines
that links to what
CONFIG_SPL_TARGETproduced, if the right CONFIG symbol is set by the board config to say that this is a NAND SPL.Or
Quoted 1 line
is someone going to say that they want to generate multiple
different
Quoted 1 line
SPL images from the same build? :-P
I'd like to generate a bootable NAND image on mx53 and on mx23/28,
From the same build, or just in general?
Quoted 5 lines
thus I need generic target. And I need it flexible enough, since on mx23/28 it munges u-
boot.sb into the resulting image while on mx53 it munges u-boot.imx into the
resulting image.
So each one would set the appropriate CONFIG_BUILD_TARGET for whatever needs to get built, and then something like CONFIG_NAND_IMAGE could hold the image name that should be linked to produce a standard u-boot-nand.bin output.
-Scott
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Scott Wood,
Quoted 17 lines · Marek Vasut
Quoted 16 lines
Dear Scott Wood,
Quoted 13 lines · Marek Vasut
Makefile $(obj)u-boot.img: $(obj)u-boot.bin · line 488
Quoted 9 lines
+$(obj)u-boot.nand: $(obj)u-boot.bin depend
(quoted diff) +1
-0+ if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ;then \
(quoted diff) +1
-0+ echo "This CPU does not support u-boot.nandtarget!" ; \
(quoted diff) +1
-0+ exit 1; \
Makefile +2
-0+ fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nandWe already have
CONFIG_SPL_TARGET. Why do we need a new mechanism specific to NAND, with ifdefs in the mainMakefile?
Quoted 3 lines
Quoted 1 line
This board doesn't use SPL at all.
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.
Can you elaborate?
Quoted 15 lines
Quoted 15 lines
Quoted 1 line
If we do have something specific to NAND, it should be
u-boot-nand.bin
Quoted 1 line
to match what
nand_splproduced, and it should just be a genericrule
Quoted 2 lines
that links to what
CONFIG_SPL_TARGETproduced, if the right CONFIG symbol is set by the board config to say that this is a NAND SPL.Or
Quoted 1 line
is someone going to say that they want to generate multiple
Quoted 7 lines
Quoted 5 lines
different
Quoted 1 line
SPL images from the same build? :-P
I'd like to generate a bootable NAND image on mx53 and on mx23/28,
From the same build, or just in general?
What do you mean by 'from the same build' ? From a build for the particular board running the SOC=mxs .
Quoted 10 lines
Quoted 5 lines
thus I need generic target. And I need it flexible enough, since on mx23/28 it munges u-
boot.sb into the resulting image while on mx53 it munges u-boot.imx into the
resulting image.So each one would set the appropriate
CONFIG_BUILD_TARGETfor whatever needs to get built, and then something likeCONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.
Yea, sounds reasonable. But why call it CONFIG_ , it can't be stored in the board.h files, it has to be somewhere in the Makefile hierarchy.
Quoted 1 line
-Scott
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 5 lines · Marek Vasut
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).
Quoted 9 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.
Why can't it go in the board.h files?
-Scott
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Scott Wood,
Quoted 16 lines · Scott Wood
Quoted 5 lines · Marek Vasut
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 6 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.
Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Beno?t Th?baudeau,
Quoted 27 lines
Dear Scott Wood,
Quoted 18 lines · Scott Wood
Quoted 5 lines · Marek Vasut
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 8 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?
Quoted 9 lines
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
I'm not against SPL, but then we're starting to drift away from the whole idea of generating u-boot-nand.bin or similar image. Being able to generate u-boot- nand.bin or u-boot-sd.bin etc ... on a per-CPU basis (since this is CPU specific) is the ultimate goal here, whatever is embedded in the image.
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 39 lines · Marek Vasut
Dear Beno?t Th?baudeau,
Quoted 27 lines
Dear Scott Wood,
Quoted 18 lines · Scott Wood
Quoted 5 lines · Marek Vasut
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 8 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?Quoted 9 lines
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
Quoted 6 lines
I'm not against SPL, but then we're starting to drift away from the whole idea
of generating u-boot-nand.bin or similar image. Being able to generate u-boot-
nand.bin or u-boot-sd.bin etc ... on a per-CPU basis (since this is CPU specific) is the ultimate goal here, whatever is embedded in the image.
OK, I didn't know that this was your goal here. If the contents of the image do not matter, then my u-boot-with-nand-spl.imx could be renamed into your u-boot-nand.bin with the appropriate FCB header, and CONFIG_SPL_TARGET could be changed to something more generic as Scott explained.
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 47 lines · Beno??t Th??baudeau
Quoted 39 lines · Marek Vasut
Dear Beno??t Th??baudeau,
Quoted 27 lines
Dear Scott Wood,
Quoted 18 lines · Scott Wood
Quoted 5 lines · Marek Vasut
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 8 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?Quoted 9 lines
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
Quoted 6 lines
I'm not against SPL, but then we're starting to drift away from the whole idea
of generating u-boot-nand.bin or similar image. Being able to generate u-boot-
nand.bin or u-boot-sd.bin etc ... on a per-CPU basis (since this is CPU specific) is the ultimate goal here, whatever is embedded in the image.
Quoted 4 lines
OK, I didn't know that this was your goal here. If the contents of the image do not matter, then my u-boot-with-nand-spl.imx could be renamed into your u-boot-nand.bin with the appropriate FCB header, and
CONFIG_SPL_TARGETcould be changed to something more generic as Scott explained.
I wonder how the rules start looking. In the end, some way to say "Here is the image to write to NAND, called u-boot-nand.bin" which will have whatever board needs (say spl/u-boot-spl.bin + some header attached followed by pad, followed by u-boot.img). And also allow for u-boot-nand.bin to be what someone else needs (say a different header on u-boot-spl.bin), etc.
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130227/2d294fef/attachment.pgp>
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Dear Tom Rini,
Quoted 55 lines · Beno??t Th??baudeau
Quoted 48 lines · Marek Vasut
Quoted 40 lines
Dear Beno??t Th??baudeau,
Quoted 28 lines
Dear Scott Wood,
Quoted 19 lines · Scott Wood
Quoted 6 lines · Marek Vasut
Dear Scott Wood,
Quoted 2 lines
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 8 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?Quoted 9 lines
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
Quoted 6 lines
I'm not against SPL, but then we're starting to drift away from the whole idea
of generating u-boot-nand.bin or similar image. Being able to generate u-boot-
nand.bin or u-boot-sd.bin etc ... on a per-CPU basis (since this is CPU specific) is the ultimate goal here, whatever is embedded in the image.Quoted 5 lines
OK, I didn't know that this was your goal here. If the contents of the image do not matter, then my u-boot-with-nand-spl.imx could be renamed into your u-boot-nand.bin with the appropriate FCB header, and
CONFIG_SPL_TARGETcould be changed to something more generic as Scott explained.
Quoted 6 lines
I wonder how the rules start looking. In the end, some way to say "Here is the image to write to NAND, called u-boot-nand.bin" which will have whatever board needs (say spl/u-boot-spl.bin + some header attached followed by pad, followed by u-boot.img). And also allow for u-boot-nand.bin to be what someone else needs (say a different header on u-boot-spl.bin), etc.
They'd be CPU specific rules, so that depends on the CPU really.
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 63 lines · Marek Vasut
Dear Tom Rini,
Quoted 61 lines
Quoted 58 lines · Beno??t Th??baudeau
Quoted 45 lines · Marek Vasut
Dear Beno??t Th??baudeau,
Quoted 32 lines
Dear Scott Wood,
Quoted 22 lines · Scott Wood
Quoted 6 lines · Marek Vasut
Dear Scott Wood,
Quoted 2 lines
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific.
Basically a way for a board config file to add to
$(ALL-y).Quoted 10 lines
Quoted 2 lines
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 4 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that
should be linked to produce a standard
u-boot-nand.bin output.Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?Quoted 10 lines
We could do all that, but should we? As I said to Marek, I think that it's a big mistake to omit the SPL here. The only other solution to get a reliable boot would be the DBBT, but it's very hard to use in real life, away from a production line. The SPL is really easy to enable here, and it's only a matter of time before someone gets bitten by this lack of reliability, so why not just do things right? The boot time and footprint of an SPL would really be negligible, and it's not because other implementations omit both SPL and a valid DBBT that U-Boot should do the same.
Quoted 6 lines
I'm not against SPL, but then we're starting to drift away from the whole idea of generating u-boot-nand.bin or similar image. Being able to generate u-boot- nand.bin or u-boot-sd.bin etc ... on a per-CPU basis (since this is CPU specific) is the ultimate goal here, whatever is embedded in the image.
OK, I didn't know that this was your goal here. If the contents of the image do not matter, then my u-boot-with-nand-spl.imx could be renamed into your u-boot-nand.bin with the appropriate FCB header, and
CONFIG_SPL_TARGETcould be changed to something more generic as Scott explained.
Quoted 8 lines
Quoted 6 lines
I wonder how the rules start looking. In the end, some way to say "Here is the image to write to NAND, called u-boot-nand.bin" which will have whatever board needs (say spl/u-boot-spl.bin + some header attached followed by pad, followed by u-boot.img). And also allow for u-boot-nand.bin to be what someone else needs (say a different header on u-boot-spl.bin), etc.
They'd be CPU specific rules, so that depends on the CPU really.
Yes, they'd have to have a level of granularity (and cross-sharability) to them. Which is why I wonder how it would all look, since it needs to look on the clean side. I'm expressing a desire to use these rules on non-Freescale parts right away, so long as the infrastructure is done right.
- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRL6exAAoJENk4IS6UOR1WR60QAJ8UGeXt1xe/bAOYQkacmgSK g03ZEWFIRZiQIogcnR7hzk2eiIb2zqCPHRtpnEIumOVpaOMISN/DKKeI5XbyQIiA GOW8UgTnqQ0Inr6Uoo2gwRqsPS3G9qDVPd3ZeZXo1rU48h+j3jAH5feelcwmVaFx koef5+5GpQUsaah203w5g6PSDz/+r2unud5ZptK+Fzw4yQmBHgnOET99tlRtg/pQ v0pOKDDtwh+Y/xyhV56iZrchnGBIfL05w1O2Urb+SDFvwZLyNjO4bmh4ZuNH+v1c z2dxDWo+V57hfqKiqeQwT3Bz55WiUMGamMSW13xyK0vuNZz+m66PpF9tOdUUizlt lkn2U1O++KPi+POXWzu9ILD00U2m5K5fHhMmfVCP+hJuJu26ada5rFaufkYqDU0p ksxcPmvH+9gz4swc+fNGONQGNcGLde2x1nboLVYn8Gfh0Ckcc8Bhh6aJt4VOU0ID LLVay1vbq5uCe1MPX2uHkZ5ZZrgms5L8ZFwWcjG2K/BnbA7Rg4pri/SOICiH27P9 B8qKb8myhKrtup91wHnCeCKi7PSY69kZRpKevkrFirdq4rY5XDVBlSeCGlfvMk49 i5gi6vov1BclBRIgIh1+fUIN8R6Cg7Eo30j9ptitzM9R6V7sLTu3CMl+GvCTHWwN E44z2bb+9GJyB1M/+ZVL
=SxUx
-----END PGP SIGNATURE-----
PATCH 1/4 common: imx: Implement generic u-boot.nand target
In reply to Scott WoodDear Scott Wood,
Quoted 18 lines · Marek Vasut
Quoted 5 lines
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a way for a board config file to add to $(ALL-y).Quoted 8 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 3 lines
needs to get built, and then something like
CONFIG_NAND_IMAGEcould hold the image name that should be linked to produce a standard u-boot-nand.bin output.
Quoted 5 lines
Quoted 3 lines
Yea, sounds reasonable. But why call it
CONFIG_, it can't be stored in theboard.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?
How would a config file contain the CPU-specific portions of a Makefile?
Quoted 1 line
-Scott
Best regards,
Marek Vasut
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 24 lines · Marek Vasut
Dear Scott Wood,
Quoted 4 lines · Marek Vasut
Quoted 3 lines
Dear Scott Wood,
Quoted 1 line
So maybe we need a more general (but optional)
CONFIG_BUILD_TARGET.Quoted 4 lines
Quoted 2 lines
Can you elaborate?
Same as
CONFIG_SPL_TARGET, but not SPL-specific. Basically a wayfor a
Quoted 7 lines
board config file to add to $(ALL-y).
Quoted 5 lines
Quoted 1 line
So each one would set the appropriate
CONFIG_BUILD_TARGETforwhatever
Quoted 1 line
needs to get built, and then something like
CONFIG_NAND_IMAGEcould
Quoted 4 lines
Quoted 4 lines
Quoted 2 lines
hold the image name that should be linked to produce a standard u-boot-nand.bin output.
Yea, sounds reasonable. But why call it
CONFIG_, it can't be
Quoted 8 lines
stored
Quoted 4 lines
Quoted 2 lines
in the
board.hfiles, it has to be somewhere in theMakefilehierarchy.Why can't it go in the
board.hfiles?How would a config file contain the CPU-specific portions of a
Makefile?
The target rule itself would need to go somewhere in a makefile, but just pointing at which target to use, and which image to link, can go in config.h -- just like CONFIG_SPL_TARGET does.
-Scott
PATCH 1/4 common: imx: Implement generic u-boot.nand target
In reply to Marek VasutHi Marek,
Quoted 60 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno?t Th?baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +18
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 55
Quoted 6 lines
arch/arm/imx-common/Makefile +3
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \
^
It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because /bin/sh (set to dash) is invoked by default on my machine here. It would work with the /bin/bash set by the main Makefile for SHELL, but this is not passed to the sub-make.
So what do you think we should do:
1) Add "export SHELL" to the main Makefile?
2) Move the SHELL assignment from the main Makefile to the top-level config.mk?
3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make?
4) Use printf instead of echo?
5) Something else?
I like 1). Do you see possible side effects?
I will add a patch for that in my v8 once we have made a choice if this choice is a global fix (i.e. 1 or 2).
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 61
Quoted 7 lines
arch/arm/imx-common/Makefile +3
-0+ dd if=/dev/zero bs=1015 count=1 2>/dev/null ) | \ + cat - $< > $@ + $(obj)SPL: $(OBJTREE)/spl/u-boot-spl.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX_CONFIG)).cfgtmp $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SPL_TEXT_BASE) -d $< $@
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 57 lines · Beno?t Th?baudeau
Hi Marek,
Quoted 46 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno?t Th?baudeau<[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano Babic <[email protected]>
---Makefile| 18 ++++++++++++++++++arch/arm/imx-common/Makefile| 6 ++++++ 2 files changed, 24
insertions(+)Makefile index 41054b7..8b1010a 100644 renamed from Makefile +0
-1--- a/Makefile +++ b/Makefile @@ -470,6 +470,23 @@$(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef
CONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU
does not support u-boot.nand target!" ; \ + exit 1 ; \ +
fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand +
$(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n
$(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ @@ -857,6 +874,7 @@ clobber:
tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f
$(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f
$(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb(quoted diff) +0
-0b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@
Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne
'\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.
Quoted 7 lines
So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?I like 1). Do you see possible side effects?
It looks like in these cases the kernel does $(shell echo -...), or is that also not working on your setup?
- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRLppcAAoJENk4IS6UOR1WiXoP/1wy8CU/bEgreOP/aPu2xsEA 7oh6CRRVLqGhioTo766IjaaDWD21uNmUpI3VK8wF8FBrPmmgS+IU9icYjd39ZsoL m0riocPX3TS6SJFqs/xJpgsCFSCnw7MAjz8Eo/RphV8pohMnsqUrKX4TiPiZKxeS 2wky4E9PcMDdf07ahJEUC3Y73BNBXrml+MGh59sKrAKpvES7OutbUHCHD+ZeX3WC Tk3nbFis3Sb+VY8vnMllyqVyCZQAI+MhJxm9zHTW2GzkX3/FsQ8tUND+DT8JmKv9 RQMkVCY92cdO5+0y+8RGO7WUwY0YghMdzfIkkr3ZQGlu9ccghcwvm21d2zVt7zCa UfRyl1NK3xgFIfAsyBWVF8q75WkQUVT06YJo2SJTfcvIa/hlGH6FvN9fODcxeris WAIan5Z3haZ7EiIGjn/KHq4dHYhS7c3iKqsi4FIQcoGsbwI9oFVqIQ98JW1UbWpS 174DvASSobba487fuAYTKgCwa2Qm9LHmy7XvJfb7CQTjrJuzsuRBWuhy3DD70AOC Mo/Ewq4b1juvS/vTwT+d/eJcnBklDYvvJ31FMuf+P+EBwyz3AmIAmzBKRhRP37SV EIKjB7GKHSuxrvJV6SjPAJW6suY4TiDSUh/myeXnGs/1+KDeGj91I5Dja4Cidl0o TpP52BmyQrPVs2hS1PpC
=UQTu
-----END PGP SIGNATURE-----
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Hi Tom,
Quoted 67 lines · Tom Rini
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1Quoted 63 lines · Beno?t Th?baudeau
Hi Marek,
Quoted 46 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno?t Th?baudeau<[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano Babic <[email protected]>
---Makefile| 18 ++++++++++++++++++arch/arm/imx-common/Makefile| 6 ++++++ 2 files changed, 24
insertions(+)Makefile index 41054b7..8b1010a 100644 renamed from Makefile +0
-1--- a/Makefile +++ b/Makefile @@ -470,6 +470,23 @@$(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef
CONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU
does not support u-boot.nand target!" ; \ + exit 1 ; \ +
fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand +
$(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n
$(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ @@ -857,6 +874,7 @@ clobber:
tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f
$(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f
$(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb(quoted diff) +0
-0b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@
Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne
'\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?
Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
It looks like in these cases the kernel does $(shell echo -...), or is that also not working on your setup?
I'll test that tomorrow, but I don't see how it could work better since $(shell ...) will very likely invoke $(SHELL) -c "...".
/bin/echo works though, but is this portable enough for U-Boot?
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Hi Tom,
Quoted 72 lines · Beno?t Th?baudeau
Hi Tom,
Quoted 69 lines · Tom Rini
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1Quoted 64 lines · Beno?t Th?baudeau
Hi Marek,
Quoted 46 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno?t Th?baudeau<[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano Babic <[email protected]>
---Makefile| 18 ++++++++++++++++++arch/arm/imx-common/Makefile| 6 ++++++ 2 files changed, 24
insertions(+)Makefile index 41054b7..8b1010a 100644 renamed from Makefile +0
-1--- a/Makefile +++ b/Makefile @@ -470,6 +470,23 @@$(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef
CONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU
does not support u-boot.nand target!" ; \ + exit 1 ; \ +
fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand +
$(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n
$(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ @@ -857,6 +874,7 @@ clobber:
tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f
$(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f
$(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb(quoted diff) +0
-0b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@
Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne
'\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?I like 1). Do you see possible side effects?
Quoted 7 lines
Quoted 2 lines
It looks like in these cases the kernel does $(shell echo -...), or is that also not working on your setup?
I'll test that tomorrow, but I don't see how it could work better since $(shell ...) will very likely invoke $(SHELL) -c "...".
/bin/echoworks though, but is this portable enough for U-Boot?
No, your suggestion does not work on my setup.
I've looked at what Linux does, and this is not what you said: When it uses $(shell echo -...), this is not to make the arguments work, but this is because $(shell ...) is required in the context, e.g. for make variable assignments or conditional parts of Makefiles.
U-Boot sets SHELL in its main Makefile and does not export it, whereas Linux sets CONFIG_SHELL and exports it, using it explicitly.
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
In reply to Benoît ThébaudeauQuoted 75 lines · Beno??t Th??baudeau
Hi Marek,
Quoted 60 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno??t Th??baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +18
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 55
Quoted 6 lines
arch/arm/imx-common/Makefile +3
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.
Quoted 8 lines
So what do you think we should do:
1) Add "export SHELL" to the mainMakefile?
2) Move the SHELL assignment from the mainMakefileto the top-level config.mk?
3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make?
4) Use printf instead of echo?
5) Something else?I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (cf7a7b99), so what's going on?
Signature · 8 lines
Tom -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130228/49a9b571/attachment.pgp>
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Hi Tom,
On Thursday, February 28, 2013 3:03:09 PM, Tom Rini write:
Quoted 90 lines · Beno??t Th??baudeau
Quoted 89 lines
Hi Marek,
Quoted 64 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target
out of the mainMakefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only contain path to which the
u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno??t Th??baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +4
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementationof
Makefile +14
-0+# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 55
Quoted 6 lines
arch/arm/imx-common/Makefile +3
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text,
so the FCB is broken. This is because/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do:
1) Add "export SHELL" to the mainMakefile?
2) Move the SHELL assignment from the mainMakefileto the top-level config.mk?
3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make?
4) Use printf instead of echo?
5) Something else?
Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (
cf7a7b99), so what's going on?
Indeed! But I get anyway SHELL set to /bin/bash in /Makefile, and to /bin/sh in /arch/arm/imx-common/makefile.
I don't see SHELL being set anywhere else, and $(MAKE) SHELL=$(SHELL) works.
That's really weird. It's like SHELL export was ignored, or applied before the assignment. I will further dig into this.
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Hi Tom,
Quoted 111 lines · Beno?t Th?baudeau
Hi Tom,
On Thursday, February 28, 2013 3:03:09 PM, Tom Rini write:
Quoted 94 lines · Beno??t Th??baudeau
Quoted 93 lines
Hi Marek,
Quoted 68 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target
out of the mainMakefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only contain path to which
the
u-boot.nand target is delegated. Hopefully this will not produce too much
bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]>Cc Beno??t Th??baudeau <[email protected]>Cc Fabio Estevam <[email protected]>Cc Stefano Babic <[email protected]>--- Makefile | 18 ++++++++++++++++++ arch/arm/imx-common/Makefile | 6 ++++++ 2 files changed, 24 insertions(+)Makefile +4
-0@@ -470,6 +470,23 @@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx: $(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx +# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to animplementation
ofMakefile +14
-0+# the actual u-boot.nand generator below. +# +ifdef CONFIG_MX53 +CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif + +$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [ "X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo "This CPU does not support u-boot.nand target!" ; \ + exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtbarch/arm/imx-common/Makefile +0
-0@@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin$(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX
arch/arm/imx-common/Makefile $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX · line 55
Quoted 6 lines
arch/arm/imx-common/Makefile +3
-0$(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@ +$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text,
so the FCB is broken. This is because/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do:
1) Add "export SHELL" to the mainMakefile?
2) Move the SHELL assignment from the mainMakefileto the top-level config.mk?
3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make?
4) Use printf instead of echo?
5) Something else?Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (
cf7a7b99), so what's going on?Indeed! But I get anyway SHELL set to
/bin/bashin /Makefile, and to/bin/shin
/arch/arm/imx-common/makefile.I don't see SHELL being set anywhere else, and $(MAKE) SHELL=$(SHELL) works.
That's really weird. It's like SHELL export was ignored, or applied before the
assignment. I will further dig into this.
PATCH 1/4 common: imx: Implement generic u-boot.nand target
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 99 lines · Beno?t Th?baudeau
Hi Tom,
Quoted 78 lines · Beno?t Th?baudeau
Hi Tom,
On Thursday, February 28, 2013 3:03:09 PM, Tom Rini write:
Quoted 70 lines
Quoted 68 lines · Beno??t Th??baudeau
Hi Marek,
Quoted 51 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only
contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno??tTh??baudeau <[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano Babic
<[email protected]> ---Makefile| 18
++++++++++++++++++arch/arm/imx-common/Makefile| 6
++++++ 2 files changed, 24 insertions(+)Makefile index 41054b7..8b1010a renamed from Makefile +0
-0100644 --- a/Makefile +++ b/Makefile @@ -470,6 +470,23 @@
$(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef
CONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo
"This CPU does not support u-boot.nand target!" ; \ +
exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH)
$(obj)u-boot.nand + $(obj)u-boot.kwb:
$(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Makefile index 41054b7..8b1010a renamed from Makefile +0
-0@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb@rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f
$(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb diff --git a/arch/arm/imx-common/Makefile b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo
-ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (
cf7a7b99), so what's going on?Quoted 8 lines
Indeed! But I get anyway SHELL set to
/bin/bashin /Makefile, and to/bin/shin /arch/arm/imx-common/makefile.I don't see SHELL being set anywhere else, and $(MAKE) SHELL=$(SHELL) works.
That's really weird. It's like SHELL export was ignored, or applied before the assignment. I will further dig into this.
From "5.3.2 Choosing the Shell" in the make manual, it seems impossible to force make to use SHELL from the environment, even with export. The note about SHELL in "5.7.2 Communicating Variables to a Sub-make" can be misleading however.
Without "export SHELL", the recipes are executed with the set SHELL, but the default value of SHELL is in their environment.
Quoted 8 lines
With "export SHELL", the recipes are executed with the set SHELL, which is also in their environment. But since the sub-make ignores its environment for SHELL, it does not help. That's probably why Linux uses
CONFIG_SHELL(which means nothing special to make) instead of SHELL.So 1) is not a solution, and 4) also does not work. So we are left with 2), 3) or 5).
How about 3?
- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRL4QiAAoJENk4IS6UOR1W9HMP/RvkNPQoxdf9JIxcK0U7vBZC 3JDTnzCheGxI58GE58c2NvK9Swj6h2Zc31zlacYOUH2Xgb7TbM3UyrgD/e+RxrgW 0/6tVPkKyUGouJA+56OO/mcFpmiyFsiJyCKxAMUroBMZrBvXaPnOZYX5T+vQ7SI8 N2RqtFzmkQlSJdNR7Pxy4OC8cFLCUe5I0AffFFOgDBDSEqc7emaK8e4Y+mRIUg9+ Tz9Eai24GvF/5mYqbeenLXtbGnZZBxbYKcD7gUH61DuUlNHeW7kY0jaWwoSLT/yS mdqy7nsJJenJgs6bjpDzLtL3pdF5Bmlu0BAedMa4U4cfKEYQAAG181CP4AGAA80Y DE4lj5SXP89nNSiqqNrAdiXjniBpkvcUWZR1oMOgQcAMRnYODSUES8ktCMEAHboO u3jtVTYWdH6kh4zGo2dY6g/gtYIbXrKtHpARZXeJ2YYnhXPlt641M1HNV1kkHr0x BIOXGpDsAdIxSZceki7Mm8yLBz0SPiDpkXOpUaUbGAVjsQT+qbRZLazOMzkD/vYB dkH7HnxXS2orQ3ZJXmShBivMHHOCE1FaGPP+p98y9aZiRuH6e4lg02ft7Oa9p1JE 6hTspGCv+RmsvGYHkkitxjZ8hmuxw+Qg6GxcDsnbAl87qi5gnOeg7QI6G2tGzYZU CQ0S0MwlUHF8Gk8qQx1v
=qSEN
-----END PGP SIGNATURE-----
PATCH 1/4 common: imx: Implement generic u-boot.nand target
Quoted 109 lines · Tom Rini
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1Quoted 105 lines · Beno?t Th?baudeau
Hi Tom,
Quoted 78 lines · Beno?t Th?baudeau
Hi Tom,
On Thursday, February 28, 2013 3:03:09 PM, Tom Rini write:
Quoted 70 lines
Quoted 68 lines · Beno??t Th??baudeau
Hi Marek,
Quoted 51 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a small amount of churn across all CPU models. The idea is to delegate the u-boot.nand target out of the main
Makefileand into the CPU'sMakefile(very similar to what
u-boot.imx does now). The mainMakefileshall only
contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno??tTh??baudeau <[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano Babic
<[email protected]> ---Makefile| 18
++++++++++++++++++arch/arm/imx-common/Makefile| 6
++++++ 2 files changed, 24 insertions(+)Makefile index 41054b7..8b1010a renamed from Makefile +0
-0100644 --- a/Makefile +++ b/Makefile @@ -470,6 +470,23 @@
$(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that needs u-boot.nand must add a path to an implementation of +# the actual u-boot.nand generator below. +# +ifdef
CONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo
"This CPU does not support u-boot.nand target!" ; \ +
exit 1 ; \ + fi + $(MAKE) -C $(CONFIG_NAND_TRG_PATH)
$(obj)u-boot.nand + $(obj)u-boot.kwb:
$(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Makefile index 41054b7..8b1010a renamed from Makefile +0
-0@@ -857,6 +874,7 @@ clobber: tidy @rm -f $(obj)u-boot.kwb@rm -f $(obj)u-boot.pbl @rm -f $(obj)u-boot.imx + @rm -f
$(obj)u-boot.nand @rm -f $(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb diff --git a/arch/arm/imx-common/Makefile b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ + echo
-ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (
cf7a7b99), so what's going on?Quoted 8 lines
Indeed! But I get anyway SHELL set to
/bin/bashin /Makefile, and to/bin/shin /arch/arm/imx-common/makefile.I don't see SHELL being set anywhere else, and $(MAKE) SHELL=$(SHELL) works.
That's really weird. It's like SHELL export was ignored, or applied before the assignment. I will further dig into this.
From "5.3.2 Choosing the Shell" in the make manual, it seems impossible to force make to use SHELL from the environment, even with export. The note about SHELL in "5.7.2 Communicating Variables to a Sub-make" can be misleading however.
Without "export SHELL", the recipes are executed with the set SHELL, but the default value of SHELL is in their environment.
With "export SHELL", the recipes are executed with the set SHELL, which is also in their environment. But since the sub-make ignores its environment for SHELL, it does not help. That's probably why Linux uses
CONFIG_SHELL(which means nothing special to make) instead of SHELL.
Quoted 4 lines
Quoted 2 lines
So 1) is not a solution, and 4) also does not work. So we are left with 2), 3) or 5).
How about 3?
3) pros:
- limited change without any side effect
3) cons:
- the recipe works or not depending on how the sub-make is invoked
- the same issue might happen again with another recipe and remain unnoticed for some time before causing trouble
That's why I'd probably prefer 2), but the risk is side effects in sub-makes.
Best regards,
Beno?t
PATCH 1/4 common: imx: Implement generic u-boot.nand target
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 118 lines · Beno?t Th?baudeau
Quoted 112 lines · Tom Rini
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Quoted 109 lines · Beno?t Th?baudeau
Hi Tom,
Quoted 82 lines · Beno?t Th?baudeau
Hi Tom,
On Thursday, February 28, 2013 3:03:09 PM, Tom Rini write:
Quoted 74 lines
Quoted 72 lines · Beno??t Th??baudeau
Hi Marek,
Quoted 52 lines · Marek Vasut
Implement u-boot.nand target that can be reused with a
small amount of churn across all CPU models. The idea
is to delegate the u-boot.nand target out of the mainMakefileand into the CPU'sMakefile(very similar to what u-boot.imx does now). The mainMakefileshall only contain path to which the u-boot.nand target is delegated. Hopefully this will not produce too much bloat in the mainMakefile.To demonstrate this implementation, add u-boot.nand target for i.MX53.
Signed-off-by Marek Vasut <[email protected]> Cc: Beno??tTh??baudeau <[email protected]> Cc: Fabio Estevam <[email protected]> Cc: Stefano
Babic <[email protected]> ---Makefile| 18
++++++++++++++++++arch/arm/imx-common/Makefile| 6
++++++ 2 files changed, 24 insertions(+)Makefile index 41054b7..8b1010a renamed from Makefile +0
-0100644 --- a/Makefile +++ b/Makefile @@ -470,6 +470,23
@@ $(obj)u-boot.img: $(obj)u-boot.bin $(obj)u-boot.imx:
$(obj)u-boot.bin depend $(MAKE) -C $(SRCTREE)/arch/arm/imx-common $(obj)u-boot.imx+# +# Generic u-boot.nand target. +# +# Every CPU that
needs u-boot.nand must add a path to an
implementation of +# the actual u-boot.nand generator below. +# +ifdefCONFIG_MX53+CONFIG_NAND_TRG_PATH := $(SRCTREE)/arch/arm/imx-common +endif +
+$(obj)u-boot.nand: $(obj)u-boot.bin depend + if [
"X$(CONFIG_NAND_TRG_PATH)X" = "XX" ] ; then \ + echo
"This CPU does not support u-boot.nand target!" ; \ +
exit 1 ; \ + fi + $(MAKE) -C
$(CONFIG_NAND_TRG_PATH) $(obj)u-boot.nand + $(obj)u-boot.kwb: $(obj)u-boot.bin $(obj)tools/mkimage -n $(CONFIG_SYS_KWD_CONFIG) -T kwbimage \ -a $(CONFIG_SYS_TEXT_BASE) -e $(CONFIG_SYS_TEXT_BASE) -d
$< $@ @@ -857,6 +874,7 @@ clobber: tidy @rm -f
$(obj)u-boot.kwb @rm -f $(obj)u-boot.pbl @rm -f
$(obj)u-boot.imx + @rm -f $(obj)u-boot.nand @rm -f
$(obj)u-boot.ubl @rm -f $(obj)u-boot.ais @rm -f $(obj)u-boot.dtb diff --git a/arch/arm/imx-common/Makefile b/arch/arm/imx-common/Makefile index 5d5c5b2..71ea36f 100644 --- a/arch/arm/imx-common/Makefile +++ b/arch/arm/imx-common/Makefile @@ -50,6 +50,12 @@ $(obj)u-boot.imx: $(OBJTREE)/u-boot.bin $(OBJTREE)/$(patsubst "%",%,$(CONFIG_IMX $(OBJTREE)/tools/mkimage -n $(filter-out %.bin,$^) -T imximage \ -e $(CONFIG_SYS_TEXT_BASE) -d $< $@Quoted 2 lines
+$(obj)u-boot.nand: $(obj)u-boot.imx + ( \ +
echo -ne '\x00\x00\x00\x00\x46\x43\x42\x20\x01' ; \^ It does not work in my environment (Ubuntu 12.10). -ne is interpreted as text, so the FCB is broken. This is because
/bin/sh(set to dash) is invoked by default on my machine here. It would work with the/bin/bashset by the mainMakefilefor SHELL, but this is not passed to the sub-make.So what do you think we should do: 1) Add "export SHELL" to the main
Makefile? 2) Move the SHELL assignment from the mainMakefileto the top-level config.mk? 3) Set "SHELL=$(SHELL)" on the command line when invoking the sub-make? 4) Use printf instead of echo? 5) Something else?Quoted 4 lines
Quoted 1 line
I like 1). Do you see possible side effects?
Wait! We do 1 already and have for years (
cf7a7b99), so what's going on?Quoted 8 lines
Indeed! But I get anyway SHELL set to
/bin/bashin /Makefile, and to/bin/shin /arch/arm/imx-common/makefile.I don't see SHELL being set anywhere else, and $(MAKE) SHELL=$(SHELL) works.
That's really weird. It's like SHELL export was ignored, or applied before the assignment. I will further dig into this.
From "5.3.2 Choosing the Shell" in the make manual, it seems impossible to force make to use SHELL from the environment, even with export. The note about SHELL in "5.7.2 Communicating
Variables to a Sub-make" can be misleading however.Without "export SHELL", the recipes are executed with the set SHELL, but the default value of SHELL is in their environment.
With "export SHELL", the recipes are executed with the set SHELL, which is also in their environment. But since the sub-make ignores its environment for SHELL, it does not help. That's probably why Linux uses
CONFIG_SHELL(which means nothing special to make) instead of SHELL.Quoted 4 lines
Quoted 2 lines
So 1) is not a solution, and 4) also does not work. So we are left with 2), 3) or 5).
How about 3?
Quoted 8 lines
3) pros: - limited change without any side effect
3) cons: - the recipe works or not depending on how the sub-make is invoked - the same issue might happen again with another recipe and remain unnoticed for some time before causing trouble
That's why I'd probably prefer 2), but the risk is side effects in sub-makes.
Yeah, it would be good to not have to whack every place we do $(MAKE) now with passing SHELL around, to avoid new problems in the future. So lets try 2 and test it out a bit.
- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRL4j8AAoJENk4IS6UOR1WV9QQAI5F2A9kqoAjUpeW+EJUu2tC fAzE6hmOHD2dqglYy08c1kSdD/a4PupILsoX52LHpPO7nedea1fQEgu74I0tOTiI pRWMZRGLgB584BJvb3R1dyYvzbKeSPCc7CshKt4iNfduLbLSEE8wBzbFa8hGa5Xi S2LMfeOVf4KX3gBPV+qMERpjcqKAPbrysLLKuV5W2WSBGiwWoHndWxFDqJ56CLRE GejSv7EhbEEk7WtiwY9maiNlQXOpVcZI5ne9fFwIfRiJ1qddWlfGvbmI8Mi5gac6 J/xMF7cBVT+qpz30Fj9sQz4LQKLgYqIVOo1jm/aKKiQfv67uHdzHM6als0p7mzSj gYV0LSTUr2CDFEpVZ1wZRlnFg4H7xrxkHfQFffu1MP0BXhPVJQRHCI/73BU9yiZY WntxbQbgyS7CDFuecNlVxpIZZc+vY4jr7gCRvhEmDa7LJtpNBbF5v4tIaH1sm4ex VpsgcT93fVkNy7PDkv1JgTH1A5AqI9Yw5WWmyycH6EHF4QxxYtmU7vl2/ot6tS+F rclLnsCrPuQFTOu57AJmomVcW97ULai2CLaxAqwMm+575RfHaHjT8Uiw6XkxezfA TFTOFXipSpsNIkYWtG0Y8gT2mhWyOOKNqB62x1tyYv6ypiKn/KGDM3CbIeJPgK9v 96hWXW+AA9LkH/GR7H9r
=9TFg
-----END PGP SIGNATURE-----