September 2012 - U-Boot

[U-Boot] [PATCH] patman: Handle checkpatch.pl not providing file/line info
by Simon Glass 09 Oct '12

09 Oct '12

Sometimes we don't get a valid filename or line number from checkpatch.pl, for example if the patch is in a bad format. Deal with this by using a default value, rather than a stack trace. Signed-off-by: Simon Glass <sjg(a)chromium.org> --- tools/patman/checkpatch.py | 5 +++-- 1 files changed, 3 insertions(+), 2 deletions(-) diff --git a/tools/patman/checkpatch.py b/tools/patman/checkpatch.py index a234277..d831087 100644 --- a/tools/patman/checkpatch.py +++ b/tools/patman/checkpatch.py @@ -145,8 +145,9 @@ def CheckPatches(verbose, args): if len(problems) != error_count + warning_count: print "Internal error: some problems lost" for item in problems: - print GetWarningMsg(col, item['type'], item['file'], - item['line'], item['msg']) + print GetWarningMsg(col, item['type'], + item.get('file', '<unknown>'), + item.get('line', 0), item['msg']) #print stdout if error_count != 0 or warning_count != 0: str = 'checkpatch.pl found %d error(s), %d warning(s)' % ( -- 1.7.7.3

2 1

[U-Boot] [PATCH] patman: Support Series-name tag to name a series
by Simon Glass 09 Oct '12

09 Oct '12

Sometimes it is possible to forget the name of the branch you used to generate an upstream series. To assist with this, add an optional patman does not use this. Signed-off-by: Simon Glass <sjg(a)chromium.org> --- tools/patman/README | 5 +++++ tools/patman/series.py | 4 ++-- 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/tools/patman/README b/tools/patman/README index 86ede78..dc3957c 100644 --- a/tools/patman/README +++ b/tools/patman/README @@ -133,6 +133,11 @@ Series-prefix: prefix Sets the subject prefix. Normally empty but it can be RFC for RFC patches, or RESEND if you are being ignored. +Series-name: name + Sets the name of the series. You don't need to have a name, and + patman does not yet use it, but it is convenient to put the branch + name here to help you keep track of multiple upstreaming efforts. + Cover-letter: This is the patch set title blah blah diff --git a/tools/patman/series.py b/tools/patman/series.py index 27528bf..67ab048 100644 --- a/tools/patman/series.py +++ b/tools/patman/series.py @@ -25,7 +25,7 @@ import gitutil import terminal # Series-xxx tags that we understand -valid_series = ['to', 'cc', 'version', 'changes', 'prefix', 'notes']; +valid_series = ['to', 'cc', 'version', 'changes', 'prefix', 'notes', 'name']; class Series(dict): """Holds information about a patch series, including all tags. @@ -76,7 +76,7 @@ class Series(dict): self[name] = value else: raise ValueError("In %s: line '%s': Unknown 'Series-%s': valid " - "options are %s" % (self.commit.hash, line, name, + "options are %s" % (commit.hash, line, name, ', '.join(valid_series))) def AddCommit(self, commit): -- 1.7.7.3

2 1

[U-Boot] [PATCH v1 0/9] block: Fix some inconsistencies in block_{read, write} implementors
by Tom Rini 09 Oct '12

09 Oct '12

Hey all, I ran into this problem while testing Pavel Herrmann's patch to move externs to <sata.h>. To fix these problems I've taken over the patch. Pavel's patch makes all of the sata drivers include <sata.h>, and this exposed a number of build problems due to type mis-match. Examining these mis-matches leads to this series. Both <ide.h> and <sata.h> set prototypes for block_dev_desc_t->block_{read,write} implementors that did not match the prototype. First we fix these implemenators (and in the case of ata_piix.[ch] fix a large number of checkpatch issues. Then we fix the prototypes in the headers. -- Tom Pavel Herrmann (1): sata.h: Make all sata/ata drivers include <sata.h> Tom Rini (8): ata_piix: Fix checkpatch issues ata_piix: sata_write must take a const void argument. dwc_ahsata: Make sata_write() comply with <part.h> fsl_sata: Make sata_{read,write}() comply with <part.h> pata_bfin: Make sata_{read,write}() comply with <part.h> sata_dwc: Make sata_write() use const void, per <part.h> sil_sata: Make sata_write() comply with <part.h> ide.h: Make ide_(read|write) match block_dev_desc_t block_(read|write) drivers/block/ata_piix.c | 455 ++++++++++++++++++++---------------------- drivers/block/ata_piix.h | 46 ++--- drivers/block/dwc_ahsata.c | 9 +- drivers/block/dwc_ahsata.h | 2 - drivers/block/fsl_sata.c | 13 +- drivers/block/pata_bfin.c | 5 +- drivers/block/pata_bfin.h | 2 - drivers/block/sata_dwc.c | 5 +- drivers/block/sata_sil.c | 7 +- drivers/block/sata_sil.h | 2 - drivers/block/sata_sil3114.c | 2 +- include/ide.h | 4 +- include/sata.h | 7 +- 13 files changed, 260 insertions(+), 299 deletions(-) -- 1.7.9.5

2 11

[U-Boot] [PATCH 0/4] Keyboard input improvements
by Simon Glass 09 Oct '12

09 Oct '12

This is a small series of patches to improve keyboard input for matrix keyboards: - Fix an fdt decoding bug which breaks seaboard - Permit key repeat/delay to be altered after init - Allow key ghosting filter to be enabled/disabled - Add a bit more debugging Since this series mostly affects tegra I suggest that it goes through the tegra tree. Simon Glass (4): input: Correct key_matrix fdt decoding input: Separate out keyboard repeat/delay from init input: Allow key ghosting filter to be disabled input: Add debugging for key matrix key codes drivers/input/input.c | 13 +++++++++---- drivers/input/key_matrix.c | 22 ++++++++++++++++------ drivers/input/tegra-kbc.c | 7 ++++--- include/input.h | 14 ++++++++++---- include/key_matrix.h | 5 ++++- 5 files changed, 43 insertions(+), 18 deletions(-) -- 1.7.7.3

2 7

[U-Boot] [PATCH v4 0/16] tegra: Add display driver and LCD support for Seaboard
by Simon Glass 09 Oct '12

09 Oct '12

This series adds support for the Tegra2x's display peripheral. This supports the LCD display on Seaboard and we use this to enable console output in U-Boot on the LCD. Configuration is via the device tree. Proposed bindings are included in this series, taken from pwm bindings that should be in linux-next, a Tegra video binding that might be accepted in devicetree-discuss and a proposed video mode binding posted to dri-devel. While I agree EDID is convenient for machines I would prefer to provide a user-friendly way of selecting LCD settings as well, with EDID more as a fallback and auto-detection when available. To improve performance two optimisations are offered: 1. The LCD frame buffer is cached, with the cache being flushed after each newline sent to putc(), and in a few other situations. This dramatically increases performance (around 10x). This requires a few additions to the ARM cache support. 2. The console supports scrolling in steps of more than 1 line. This speeds up scrolling output considerably, particularly commands like 'printenv' which display a lot of output, and particular when the dcache is off. This requires a new CONFIG and a change to the console_scrollup() function. Changes in v2: - Add new patch to use const in pinmux_config_pingroup/table() - Use const where possible in funcmux - Add nvidia prefix to device tree properties - Update LCD driver to deal with new fdt bindings - Use a more generic config CONFIG_LCD_ALIGNMENT for lcd alignment - Put the LCD cache flush logic into lcd_putc() instead of lcd_puts() - Align tegra display using new CONFIG_LCD_ALIGNMENT feature - Update seaboard LCD definitions for new fdt binding Changes in v3: - Remove LPW1 pin which is not needed by display - Add new commit for pwm binding and node - Use displaymode binding for fdt - Bring in proposed tegra display controller binding - Add new panel binding to fit with tegra display controller binding - Rename pwfm driver to pwm - Use new proposed upstream pwm binding - Decode fdt node within the pwm driver - Introduce concept of a pwm channel, rather than separate peripherals - Separate display driver and LCD driver more in fdt - Add probe function to read in fdt parameters in display driver - Move some fdt decode code from LCD driver to SOC display driver - Rename fdt config structures - Use new pwm binding from pre-linux-next - Adjust LCD driver to use new SOC display driver structures - Add separate call to pwm_init() in board_init() - Put the LCD cache flush logic back into lcd_puts() - Handle a cached frame buffer out of normal U-Boot memory - Use new upstream proposed LCD definitions - Fix tiny bug in mult-line lcd scrolling Changes in v4: - Moved panel information into panel node (was in rgb node) - Add CONFIG_PWM_TEGRA to control inclusion of PWM support - Add a little more debugging to the PWM driver - Move code to set up PWM into the PWM commit - Rebase on top of new tegra common directory structure - Adjust fdt binding according to Thierry and Stephen's comments - Remove LCD frame buffer address property hack - Rebase on top of new tegra common directory structure - Change LCD panel GPIOs to be entirely optional - Only indicate that LCD init is complete when all stages complete - Change order of LCD init so that PWM comes first - Use system bit values for enum dcache_option instead of arbitrary numbers - Remove system.h header which is included anyway by most archs - Use lcd_puts() for echoing user input so lcd cached data is flushed - Adjust LCD fdt nodes for new binding - Use CONFIG_CONSOLE_SCROLL_LINES instead of CONSOLE_SCROLL_LINES - Put default CONFIG_CONSOLE_SCROLL_LINES at top of file - Add CONFIG_BOARD_LATE_INIT to enable LCD - Enable separate CONFIG_PWM_TEGRA option Mayuresh Kulkarni (1): tegra: Enable display/lcd support on Seaboard Simon Glass (14): tegra: Use const for pinmux_config_pingroup/table() tegra: Add display support to funcmux tegra: fdt: Add pwm binding and node tegra: fdt: Add LCD definitions for Tegra tegra: Add support for PWM tegra: Add LCD driver tegra: Add LCD support to Nvidia boards arm: Add control over cachability of memory regions lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment lcd: Add support for flushing LCD fb from dcache after update tegra: Align LCD frame buffer to section boundary tegra: Support control of cache settings for LCD tegra: fdt: Add LCD definitions for Seaboard lcd: Add CONSOLE_SCROLL_LINES option to speed console Wei Ni (1): tegra: Add SOC support for display/lcd README | 16 + arch/arm/cpu/armv7/cache_v7.c | 11 + arch/arm/cpu/armv7/tegra20/Makefile | 2 + arch/arm/cpu/armv7/tegra20/display.c | 409 ++++++++++++++++++ arch/arm/cpu/armv7/tegra20/pwm.c | 101 +++++ arch/arm/cpu/tegra20-common/funcmux.c | 37 ++ arch/arm/cpu/tegra20-common/pinmux.c | 4 +- arch/arm/dts/tegra20.dtsi | 96 +++++ arch/arm/include/asm/arch-tegra20/dc.h | 545 ++++++++++++++++++++++++ arch/arm/include/asm/arch-tegra20/display.h | 152 +++++++ arch/arm/include/asm/arch-tegra20/pinmux.h | 4 +- arch/arm/include/asm/arch-tegra20/pwm.h | 75 ++++ arch/arm/include/asm/system.h | 30 ++ arch/arm/lib/cache-cp15.c | 50 ++- board/nvidia/common/board.c | 26 ++- board/nvidia/dts/tegra20-seaboard.dts | 31 ++ common/lcd.c | 85 +++- common/main.c | 12 +- doc/device-tree-bindings/pwm/tegra20-pwm.txt | 18 + doc/device-tree-bindings/video/displaymode.txt | 42 ++ doc/device-tree-bindings/video/tegra20-dc.txt | 82 ++++ drivers/video/Makefile | 1 + drivers/video/tegra.c | 372 ++++++++++++++++ include/configs/seaboard.h | 13 +- include/configs/tegra20-common.h | 3 + include/fdtdec.h | 2 + include/lcd.h | 11 + lib/fdtdec.c | 2 + 28 files changed, 2197 insertions(+), 35 deletions(-) create mode 100644 arch/arm/cpu/armv7/tegra20/display.c create mode 100644 arch/arm/cpu/armv7/tegra20/pwm.c create mode 100644 arch/arm/include/asm/arch-tegra20/dc.h create mode 100644 arch/arm/include/asm/arch-tegra20/display.h create mode 100644 arch/arm/include/asm/arch-tegra20/pwm.h create mode 100644 doc/device-tree-bindings/pwm/tegra20-pwm.txt create mode 100644 doc/device-tree-bindings/video/displaymode.txt create mode 100644 doc/device-tree-bindings/video/tegra20-dc.txt create mode 100644 drivers/video/tegra.c -- 1.7.7.3

6 39

[U-Boot] [PATCH] fix IDE_BUS(dev) macro
by Luka Perkov 08 Oct '12

08 Oct '12

IDE_BUS assumes that each bus has two devices and thus returns the first bus even when the second one should be probed. Signed-off-by: Simon Baatz <gmbnomis(a)gmail.com> Tested-by: Luka Perkov <uboot(a)lukaperkov.net> --- Simon discovered this while adding support for new board IB NAS6210. More info can be found here: http://lists.denx.de/pipermail/u-boot/2012-April/122525.html When this is commited I will do a coding style cleanup. There are tabs after few "#define" parts in include/ide.h. include/ide.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/ide.h b/include/ide.h index 8ecc9dd..385e909 100644 --- a/include/ide.h +++ b/include/ide.h @@ -24,7 +24,7 @@ #ifndef _IDE_H #define _IDE_H -#define IDE_BUS(dev) (dev >> 1) +#define IDE_BUS(dev) (dev / (CONFIG_SYS_IDE_MAXDEVICE / CONFIG_SYS_IDE_MAXBUS)) #define ATA_CURR_BASE(dev) (CONFIG_SYS_ATA_BASE_ADDR+ide_bus_offset[IDE_BUS(dev)])

11 23

[U-Boot] [PATCH v3 0/23] tegra: warmboot (suspend / resume) support
by Simon Glass 06 Oct '12

06 Oct '12

This series adds support for warm boot, allowing the device to suspend and resume. U-Boot sets up some 'warm boot' code in a special area such that the SOC can find it on a resume. This code is responsible for setting up memory and clocked and then allowing the OS to continue where it left off. Also included here is support for the EMC, which allows setting the memory timings correctly in U-Boot for maximum speed operation. Note: sign-offs are needed from Jimmy Zhang, Wei Ni and Yen Lin. Changes in v2: - Add check of undocumented values in hidrev register - Add debug() output to EMC - Add new fdtdec_locate_array() function - Add patch to find DVC bus number - Check error return from pmu_set_nominal() in debug mode - Move EMC tables to device tree - Move structs shared between A9 and AVP into warmboot.h header file - New patch to add low-level clock functions - Rely on compiler to optimise out debug_print_vector() - Remove unused crypto code - Remove unused crypto code from crypto.c - Removed check for nominal voltage (not needed as it is done just before) - Split PMU code into separate TPS6586X driver - Tidy SDRAM range check in warmboot_prepare_code() - Tidy whitespace problems - Use const for sbox arrays - Use low-level clock functions in warmboot code instead of register access Changes in v3: - Add apb_misc.h header file in new patch - Add better error reporting when EMC setup fails - Add new fdtdec_next_compatible_subnode() patch - Add new patch to put abs() in common.h - Fix CONFIG_TEGRA_I2C define - Move abs() macro into common.h - Support nvidia,use-ram-code binding option for EMC - Try to return a useful error code when EMC config fails Jimmy Zhang (3): tegra: Add EMC support for optimal memory timings tegra: Add PMU to manage power supplies tegra: Add EMC settings for Seaboard Simon Glass (14): fdt: Add function to locate an array in the device tree fdt: Add function to return next compatible subnode Add abs() macro to return absolute value i2c: Add TPS6586X driver tegra: Move ap20.h header into arch location tegra: Add functions to access low-level Osc/PLL details tegra: Add tegra_get_chip_type() to detect SKU tegra: Add header file for APB_MISC register tegra: Set up PMU for Nvidia boards tegra: Set up warmboot code on Nvidia boards fdt: tegra: Add EMC node to device tree tegra: i2c: Add function to find DVC bus tegra: fdt: Add EMC data for Tegra2 Seaboard tegra: Enable LP0 on Seaboard Wei Ni (1): tegra: Turn off power detect in board init Yen Lin (5): Add AES crypto library tegra: Add crypto library for warmboot code tegra: Add flow, gp_padctl, fuse, sdram headers tegra: Add warmboot implementation tegra: Setup PMC scratch info from ap20 setup arch/arm/cpu/armv7/tegra2/Makefile | 4 + arch/arm/cpu/armv7/tegra2/ap20.c | 44 ++- arch/arm/cpu/armv7/tegra2/board.c | 4 +- arch/arm/cpu/armv7/tegra2/clock.c | 32 + arch/arm/cpu/armv7/tegra2/crypto.c | 230 ++++++++ arch/arm/cpu/armv7/tegra2/crypto.h | 36 ++ arch/arm/cpu/armv7/tegra2/emc.c | 286 ++++++++++ arch/arm/cpu/armv7/tegra2/pmu.c | 70 +++ arch/arm/cpu/armv7/tegra2/warmboot.c | 382 +++++++++++++ arch/arm/cpu/armv7/tegra2/warmboot_avp.c | 250 ++++++++ arch/arm/cpu/armv7/tegra2/warmboot_avp.h | 81 +++ arch/arm/dts/tegra20.dtsi | 7 + .../tegra2 => include/asm/arch-tegra2}/ap20.h | 7 + arch/arm/include/asm/arch-tegra2/apb_misc.h | 36 ++ arch/arm/include/asm/arch-tegra2/clk_rst.h | 3 + arch/arm/include/asm/arch-tegra2/clock.h | 22 + arch/arm/include/asm/arch-tegra2/emc.h | 113 ++++ arch/arm/include/asm/arch-tegra2/flow.h | 36 ++ arch/arm/include/asm/arch-tegra2/fuse.h | 39 ++ arch/arm/include/asm/arch-tegra2/gp_padctrl.h | 73 +++ arch/arm/include/asm/arch-tegra2/pmu.h | 30 + arch/arm/include/asm/arch-tegra2/sdram_param.h | 148 +++++ arch/arm/include/asm/arch-tegra2/tegra2.h | 25 + arch/arm/include/asm/arch-tegra2/tegra_i2c.h | 7 + arch/arm/include/asm/arch-tegra2/warmboot.h | 150 +++++ board/nvidia/common/Makefile | 1 + board/nvidia/common/board.c | 41 ++- board/nvidia/common/emc.c | 53 ++ board/nvidia/common/emc.h | 29 + board/nvidia/dts/tegra2-seaboard.dts | 72 +++ drivers/i2c/tegra_i2c.c | 14 + drivers/power/Makefile | 1 + drivers/power/tps6586x.c | 280 +++++++++ include/aes.h | 70 +++ include/common.h | 13 + include/configs/seaboard.h | 8 + include/configs/tegra2-common.h | 17 + include/fdtdec.h | 38 ++ include/tps6586x.h | 68 +++ lib/Makefile | 1 + lib/aes.c | 598 ++++++++++++++++++++ lib/fdtdec.c | 28 + 42 files changed, 3443 insertions(+), 4 deletions(-) create mode 100644 arch/arm/cpu/armv7/tegra2/crypto.c create mode 100644 arch/arm/cpu/armv7/tegra2/crypto.h create mode 100644 arch/arm/cpu/armv7/tegra2/emc.c create mode 100644 arch/arm/cpu/armv7/tegra2/pmu.c create mode 100644 arch/arm/cpu/armv7/tegra2/warmboot.c create mode 100644 arch/arm/cpu/armv7/tegra2/warmboot_avp.c create mode 100644 arch/arm/cpu/armv7/tegra2/warmboot_avp.h rename arch/arm/{cpu/armv7/tegra2 => include/asm/arch-tegra2}/ap20.h (96%) create mode 100644 arch/arm/include/asm/arch-tegra2/apb_misc.h create mode 100644 arch/arm/include/asm/arch-tegra2/emc.h create mode 100644 arch/arm/include/asm/arch-tegra2/flow.h create mode 100644 arch/arm/include/asm/arch-tegra2/fuse.h create mode 100644 arch/arm/include/asm/arch-tegra2/gp_padctrl.h create mode 100644 arch/arm/include/asm/arch-tegra2/pmu.h create mode 100644 arch/arm/include/asm/arch-tegra2/sdram_param.h create mode 100644 arch/arm/include/asm/arch-tegra2/warmboot.h create mode 100644 board/nvidia/common/emc.c create mode 100644 board/nvidia/common/emc.h create mode 100644 drivers/power/tps6586x.c create mode 100644 include/aes.h create mode 100644 include/tps6586x.h create mode 100644 lib/aes.c -- 1.7.7.3

8 53

[U-Boot] [PATCH] Add a new "ini" command
by Joe Hershberger 06 Oct '12

06 Oct '12

This allows you to read ini-formatted data from anywhere and then import one of the sections into the environment This is based on rev 16 at http://code.google.com/p/inih/ Signed-off-by: Joe Hershberger <joe.hershberger(a)ni.com> --- common/Makefile | 1 + common/cmd_ini.c | 256 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 257 insertions(+) create mode 100644 common/cmd_ini.c diff --git a/common/Makefile b/common/Makefile index 3d62775..e296368 100644 --- a/common/Makefile +++ b/common/Makefile @@ -100,6 +100,7 @@ COBJS-$(CONFIG_CMD_GPIO) += cmd_gpio.o COBJS-$(CONFIG_CMD_I2C) += cmd_i2c.o COBJS-$(CONFIG_CMD_IDE) += cmd_ide.o COBJS-$(CONFIG_CMD_IMMAP) += cmd_immap.o +COBJS-$(CONFIG_CMD_INI) += cmd_ini.o COBJS-$(CONFIG_CMD_IRQ) += cmd_irq.o COBJS-$(CONFIG_CMD_ITEST) += cmd_itest.o COBJS-$(CONFIG_CMD_JFFS2) += cmd_jffs2.o diff --git a/common/cmd_ini.c b/common/cmd_ini.c new file mode 100644 index 0000000..0f0e5cd --- /dev/null +++ b/common/cmd_ini.c @@ -0,0 +1,256 @@ +/* + * inih -- simple .INI file parser + * + * inih is released under the New BSD license (see LICENSE.txt). Go to the + * project home page for more info: + * + * http://code.google.com/p/inih/ + */ + +#include <common.h> +#include <command.h> +#include <environment.h> +#include <linux/ctype.h> +#include <linux/string.h> + +#ifdef CONFIG_INI_MAX_LINE +#define MAX_LINE CONFIG_INI_MAX_LINE +#else +#define MAX_LINE 200 +#endif + +#ifdef CONFIG_INI_MAX_SECTION +#define MAX_SECTION CONFIG_INI_MAX_SECTION +#else +#define MAX_SECTION 50 +#endif + +#ifdef CONFIG_INI_MAX_NAME +#define MAX_NAME CONFIG_INI_MAX_NAME +#else +#define MAX_NAME 50 +#endif + +/* Strip whitespace chars off end of given string, in place. Return s. */ +static char *rstrip(char *s) +{ + char *p = s + strlen(s); + + while (p > s && isspace(*--p)) + *p = '\0'; + return s; +} + +/* Return pointer to first non-whitespace char in given string. */ +static char *lskip(const char *s) +{ + while (*s && isspace(*s)) + s++; + return (char *)s; +} + +/* Return pointer to first char c or ';' comment in given string, or pointer to + null at end of string if neither found. ';' must be prefixed by a whitespace + character to register as a comment. */ +static char *find_char_or_comment(const char *s, char c) +{ + int was_whitespace = 0; + + while (*s && *s != c && !(was_whitespace && *s == ';')) { + was_whitespace = isspace(*s); + s++; + } + return (char *)s; +} + +/* Version of strncpy that ensures dest (size bytes) is null-terminated. */ +static char *strncpy0(char *dest, const char *src, size_t size) +{ + strncpy(dest, src, size); + dest[size - 1] = '\0'; + return dest; +} + +/* Emulate the behavior of fgets but on memory */ +static char *memgets(char *str, int num, char **mem, size_t *memsize) +{ + char *end; + int len; + int newline = 1; + + end = memchr(*mem, '\n', *memsize); + if (end == NULL) { + if (*memsize == 0) + return NULL; + end = *mem + *memsize; + newline = 0; + } + len = min((end - *mem) + newline, num); + memcpy(str, *mem, len); + if (len < num) + str[len] = '\0'; + + /* prepare the mem vars for the next call */ + *memsize -= (end - *mem) + newline; + *mem += (end - *mem) + newline; + + return str; +} + +/* Parse given INI-style file. May have [section]s, name=value pairs + (whitespace stripped), and comments starting with ';' (semicolon). Section + is "" if name=value pair parsed before any section heading. name:value + pairs are also supported as a concession to Python's ConfigParser. + + For each name=value pair parsed, call handler function with given user + pointer as well as section, name, and value (data only valid for duration + of handler call). Handler should return nonzero on success, zero on error. + + Returns 0 on success, line number of first error on parse error (doesn't + stop on first error). +*/ +int ini_parse(char *filestart, size_t filelen, + int (*handler)(void *, char *, char *, char *), void *user) +{ + /* Uses a fair bit of stack (use heap instead if you need to) */ + char line[MAX_LINE]; + char section[MAX_SECTION] = ""; + char prev_name[MAX_NAME] = ""; + + char *curmem = filestart; + char *start; + char *end; + char *name; + char *value; + size_t memleft = filelen; + int lineno = 0; + int error = 0; + + /* Scan through file line by line */ + while (memgets(line, sizeof(line), &curmem, &memleft) != NULL) { + lineno++; + start = lskip(rstrip(line)); + + if (*start == ';' || *start == '#') { + /* + * Per Python ConfigParser, allow '#' comments at start + * of line + */ + } +#if CONFIG_INI_ALLOW_MULTILINE + else if (*prev_name && *start && start > line) { + /* + * Non-blank line with leading whitespace, treat as + * continuation of previous name's value (as per Python + * ConfigParser). + */ + if (!handler(user, section, prev_name, start) && !error) + error = lineno; + } +#endif + else if (*start == '[') { + /* A "[section]" line */ + end = find_char_or_comment(start + 1, ']'); + if (*end == ']') { + *end = '\0'; + strncpy0(section, start + 1, sizeof(section)); + *prev_name = '\0'; + } else if (!error) { + /* No ']' found on section line */ + error = lineno; + } + } else if (*start && *start != ';') { + /* Not a comment, must be a name[=:]value pair */ + end = find_char_or_comment(start, '='); + if (*end != '=') + end = find_char_or_comment(start, ':'); + if (*end == '=' || *end == ':') { + *end = '\0'; + name = rstrip(start); + value = lskip(end + 1); + end = find_char_or_comment(value, '\0'); + if (*end == ';') + *end = '\0'; + rstrip(value); + /* Strip double-quotes */ + if (value[0] == '"' && + value[strlen(value)-1] == '"') { + value[strlen(value)-1] = '\0'; + value += 1; + } + + /* + * Valid name[=:]value pair found, call handler + */ + strncpy0(prev_name, name, sizeof(prev_name)); + if (!handler(user, section, name, value) && + !error) + error = lineno; + } else if (!error) + /* No '=' or ':' found on name[=:]value line */ + error = lineno; + } + } + + return error; +} + +static int ini_handler(void *user, char *section, char *name, char *value) +{ + char *requested_section = (char *)user; +#ifdef CONFIG_INI_CASE_INSENSITIVE + int i; + + for (i = 0; i < strlen(requested_section); i++) + requested_section[i] = tolower(requested_section[i]); + for (i = 0; i < strlen(section); i++) + section[i] = tolower(section[i]); +#endif + + if (!strcmp(section, requested_section)) { +#ifdef CONFIG_INI_CASE_INSENSITIVE + for (i = 0; i < strlen(name); i++) + name[i] = tolower(name[i]); + for (i = 0; i < strlen(value); i++) + value[i] = tolower(value[i]); +#endif + setenv(name, value); + printf("ini: Imported %s as %s\n", name, value); + } + + /* success */ + return 1; +} + +int do_ini(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) +{ + const char *section; + char *file_address; + size_t file_size; + + if (argc == 1) + return CMD_RET_USAGE; + + section = argv[1]; + if (argc < 3) + file_address = (char *)simple_strtoul(getenv("loadaddr"), + NULL, 16); + else + file_address = (char *)simple_strtoul(argv[2], NULL, 16); + if (argc < 4) + file_size = (size_t)simple_strtoul(getenv("filesize"), + NULL, 16); + else + file_size = (size_t)simple_strtoul(argv[3], NULL, 16); + + ini_parse(file_address, file_size, ini_handler, (void *)section); + + /* success */ + return 0; +} + +U_BOOT_CMD( + ini, 4, 0, do_ini, + "parse an ini file in memory and merge the specified section into the env", + "section [[file-address] file-size]" +); -- 1.7.11.5

6 9

[U-Boot] [PATCH 1/2] GPIO: pca953x: fix spelling in help
by Laurence Withers 06 Oct '12

06 Oct '12

Signed-off-by: Laurence Withers <lwithers(a)guralp.com> --- drivers/gpio/pca953x.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/gpio/pca953x.c b/drivers/gpio/pca953x.c index 359fdee..64c7797 100644 --- a/drivers/gpio/pca953x.c +++ b/drivers/gpio/pca953x.c @@ -287,7 +287,7 @@ U_BOOT_CMD( " - set pin as output and drive low or high\n" "pca953x invert pin 0|1\n" " - disable/enable polarity inversion for reads\n" - "pca953x intput pin\n" + "pca953x input pin\n" " - set pin as input and read value" ); -- 1.7.2.5

2 3

[U-Boot] [PATCH v2] ml507: Fix Xilinx uartlite driver hang
by Rommel Custodio 05 Oct '12

05 Oct '12

The default configuration for ml507 will generate a hang() in the Xilinx uartlite driver. userial_ports[] in drivers/serial/serial_xuartlite.c does not get initialized properly. CONFIG_SERIAL_BASE is unused. XILINX_UARTLITE_BASEADDR is used instead. Signed-off-by: Rommel Custodio <sessyargc+uboot(a)gmail.com> --- Changes since v1: - fixed the path, tabs got mangled causing the patch to fail include/configs/xilinx-ppc.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/configs/xilinx-ppc.h b/include/configs/xilinx-ppc.h index 1235c37..2bdaa05 100644 --- a/include/configs/xilinx-ppc.h +++ b/include/configs/xilinx-ppc.h @@ -118,7 +118,7 @@ /* serial communication */ #ifdef XPAR_UARTLITE_0_BASEADDR #define CONFIG_XILINX_UARTLITE -#define CONFIG_SERIAL_BASE XPAR_UARTLITE_0_BASEADDR +#define XILINX_UARTLITE_BASEADDR XPAR_UARTLITE_0_BASEADDR #define CONFIG_BAUDRATE XPAR_UARTLITE_0_BAUDRATE #define CONFIG_SYS_BAUDRATE_TABLE { CONFIG_BAUDRATE } #else -- 1.7.9.5

2 1