diff options
Diffstat (limited to 'doc')
-rw-r--r-- | doc/README.JFFS2 | 43 | ||||
-rw-r--r-- | doc/README.JFFS2_NAND | 20 | ||||
-rw-r--r-- | doc/README.at91-soc | 48 | ||||
-rw-r--r-- | doc/README.imx31 | 29 | ||||
-rw-r--r-- | doc/README.mpc83xxads | 1 | ||||
-rw-r--r-- | doc/README.nand | 2 | ||||
-rw-r--r-- | doc/mkimage.1 | 2 |
7 files changed, 7 insertions, 138 deletions
diff --git a/doc/README.JFFS2 b/doc/README.JFFS2 index f0e9bc1..604e5b9 100644 --- a/doc/README.JFFS2 +++ b/doc/README.JFFS2 @@ -19,52 +19,15 @@ more or less a bubble sort. That algorithm is known to be O(n^2), thus you should really consider if you can avoid it! -There is two ways for JFFS2 to find the disk. The default way uses -the flash_info structure to find the start of a JFFS2 disk (called -partition in the code) and you can change where the partition is with -two defines. +There only one way for JFFS2 to find the disk. It uses the flash_info +structure to find the start of a JFFS2 disk (called partition in the code) +and you can change where the partition is with two defines. CONFIG_SYS_JFFS2_FIRST_BANK defined the first flash bank to use CONFIG_SYS_JFFS2_FIRST_SECTOR defines the first sector to use - - -The second way is to define CONFIG_SYS_JFFS_CUSTOM_PART and implement the -jffs2_part_info(int part_num) function in your board specific files. -In this mode CONFIG_SYS_JFFS2_FIRST_BANK and CONFIG_SYS_JFFS2_FIRST_SECTOR is not -used. - -The input is a partition number starting with 0. -Return a pointer to struct part_info or NULL for error; - -Ex jffs2_part_info() for one partition. ---- -#if defined CONFIG_SYS_JFFS_CUSTOM_PART -#include <jffs2/jffs2.h> - -static struct part_info part; - -struct part_info* -jffs2_part_info(int part_num) -{ - if(part_num==0){ - if(part.usr_priv==(void*)1) - return ∂ - - memset(&part, 0, sizeof(part)); - part.offset=(char*)0xFF800000; - part.size=1024*1024*8; - - /* Mark the struct as ready */ - part.usr_priv=(void*)1; - - return ∂ - } - return 0; -} -#endif --- TODO. diff --git a/doc/README.JFFS2_NAND b/doc/README.JFFS2_NAND index 09788d5..92fa0f6 100644 --- a/doc/README.JFFS2_NAND +++ b/doc/README.JFFS2_NAND @@ -2,23 +2,7 @@ JFFS2 NAND support: To enable, use the following #define in the board configuration file: -#define CONFIG_JFFS2_NAND 1 +#define CONFIG_JFFS2_NAND Configuration of partitions is similar to how this is done in U-Boot -for JFFS2 on top NOR flash. If a single partition is used, it can be -configured using the following #defines in the configuration file: - -#define CONFIG_JFFS2_NAND_DEV 0 /* nand device jffs2 lives on */ -#define CONFIG_JFFS2_NAND_OFF 0 /* start of jffs2 partition */ -#define CONFIG_JFFS2_NAND_SIZE 2*1024*1024 /* size of jffs2 partition */ - -If more than a single partition is desired, the user can define a -CONFIG_SYS_JFFS_CUSTOM_PART macro and implement a - - struct part_info* jffs2_part_info(int part_num) - -function in a board-specific module. An example of such function is -available in common/cmd_jffs2.c - -The default configuration for the DAVE board has a single JFFS2 -partition of 2 MB size. +for JFFS2 on top NOR flash. diff --git a/doc/README.at91-soc b/doc/README.at91-soc deleted file mode 100644 index 9a9f74e..0000000 --- a/doc/README.at91-soc +++ /dev/null @@ -1,48 +0,0 @@ - New C structure AT91 SoC access -================================= - -The goal --------- - -Currently the at91 arch uses hundreds of address defines and special -at91_xxxx_write/read functions to access the SOC. -The u-boot project perferred method is to access memory mapped hw -regisister via a c structure. - -e.g. old - - *AT91C_PIOA_IDR = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - *AT91C_PIOC_PUDR = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - *AT91C_PIOC_PER = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - *AT91C_PIOC_OER = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - *AT91C_PIOC_PIO = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - - at91_sys_write(AT91_RSTC_CR, - AT91_RSTC_KEY | AT91_RSTC_PROCRST | AT91_RSTC_PERRST); - -e.g new - pin = AT91_PMX_AA_TWD | AT91_PMX_AA_TWCK; - writel(pin, &pio->pioa.idr); - writel(pin, &pio->pioa.pudr); - writel(pin, &pio->pioa.per); - writel(pin, &pio->pioa.oer); - writel(pin, &pio->pioa.sodr); - - writel(AT91_RSTC_KEY | AT91_RSTC_CR_PROCRST | - AT91_RSTC_CR_PERRST, &rstc->cr); - -The method for updating ------------------------- - -1. add's the temporary CONFIG_AT91_LEGACY to all at91 board configs -2. Display a compile time warning, if the board has not been converted -3. add new structures for SoC access -4. Convert arch, driver and boards file to new SoC -5. remove legacy code, if all boards and drives are ready - -2013-10-30 Andreas Bießmann <andreas@biessmann.org>: - -The goal is almost reached, we could remove the CONFIG_AT91_LEGACY switch but -remain the CONFIG_ATMEL_LEGACY switch until the GPIO disaster is fixed. The -AT91 spi driver has also some CONFIG_ATMEL_LEGACY stuff left, so another point -to fix until this README can be removed. diff --git a/doc/README.imx31 b/doc/README.imx31 deleted file mode 100644 index 91ef766..0000000 --- a/doc/README.imx31 +++ /dev/null @@ -1,29 +0,0 @@ -U-Boot for Freescale i.MX31 - -This file contains information for the port of U-Boot to the Freescale -i.MX31 SoC. - -1. CONFIGURATION OPTIONS/SETTINGS ---------------------------------- - -1.1 Configuration of MC13783 SPI bus ------------------------------------- - The power management companion chip MC13783 is connected to the - i.MX31 via an SPI bus. Use the following configuration options - to setup the bus and chip select used for a particular board. - - CONFIG_MC13783_SPI_BUS -- defines the SPI bus the MC13783 is connected to. - Note that 0 is CSPI1, 1 is CSPI2 and 2 is CSPI3. - CONFIG_MC13783_SPI_CS -- define the chip select the MC13783 s connected to. - -1.2 Timer precision -------------------- - CONFIG_MX31_TIMER_HIGH_PRECISION - - Enable higher precision timer. The low-precision timer - (default) provides approximately 4% error, whereas the - high-precision timer is about 0.4% accurate. The extra - accuracy is achieved at the cost of higher computational - overhead, which, in places where time is measured, should - not be critical, so, it should be safe to enable this - option. diff --git a/doc/README.mpc83xxads b/doc/README.mpc83xxads index 7a8b706..372fdd9 100644 --- a/doc/README.mpc83xxads +++ b/doc/README.mpc83xxads @@ -50,7 +50,6 @@ Freescale MPC83xx ADS Boards CONFIG_MPC83xx MPC83xx family CONFIG_MPC8349 MPC8349 specific - CONFIG_MPC8349ADS MPC8349ADS board specific CONFIG_TSEC_ENET Use on-chip 10/100/1000 ethernet diff --git a/doc/README.nand b/doc/README.nand index f1c20ff..2295bb2 100644 --- a/doc/README.nand +++ b/doc/README.nand @@ -160,7 +160,7 @@ Configuration Options: /* * devnum is the device number to be used in nand commands - * and in mtd->name. Must be less than CONFIG_SYS_NAND_MAX_DEVICE. + * and in mtd->name. Must be less than CONFIG_SYS_MAX_NAND_DEVICE. */ if (nand_register(devnum, mtd)) error out diff --git a/doc/mkimage.1 b/doc/mkimage.1 index ffa7d60..e883f07 100644 --- a/doc/mkimage.1 +++ b/doc/mkimage.1 @@ -70,7 +70,7 @@ Set compression type. Pass \-h as the compression to see the list of supported compression type. .TP -.BI "\-a [" "load addess" "]" +.BI "\-a [" "load address" "]" Set load address with a hex number. .TP |