linaro-dev November 2011

linaro-dev@lists.linaro.org

122 participants
130 discussions

Build instructions for directfb-1.6.0pre1-2011.11 ?

by Dirk Behme

I just downloaded directfb-1.6.0pre1-2011.11.tar.bz2 from [1] and wonder what are the build options for this? I.e. what are the best/tested/recommended configure/compile options for this? And maybe which compiler to use (if this matters ;) )? Are there any binary packages available, too? Many thanks and best regards Dirk [1] https://launchpad.net/linaro-multimedia-project/2011.11/2011.11/

12 years, 11 months

[PATCH 0/5] OMAP4: cache fixes for 4460

by Mans Rullgard

These patches fix and tweak various cache settings for the 4460 resulting in a speed increase exceeding 10% in some tests. Mans Rullgard (5): OMAP4: apply L2 cache lockdown workaround only on 4460 ES1.0 OMAP4: enable double linefill on 4460 OMAP4: fix PL310 prefetch offset setting OMAP4: set PL310 prefetch offset to 3 OMAP4: do not force workarounds for errata fixed in 4460 arch/arm/mach-omap2/Kconfig | 3 --- arch/arm/mach-omap2/omap4-common.c | 25 ++++++++----------------- 2 files changed, 8 insertions(+), 20 deletions(-) -- 1.7.7.4

12 years, 11 months

Explanation of the different Ubuntu images?

by Jon Medhurst (Tixy)

Does anyone know of a web-site or wiki page which explains the difference between our different Ununtu image types, i.e. developer, nano, alip, server, ubuntu-desktop. I've a fair idea myself what these are, (apart from server), but I'm looking for an existing document I can reference. -- Tixy

12 years, 11 months

Validation server work

by Zygmunt Krynicki

Hi. I'm doing some work on the validation server (validation.linaro.org). My work SHOULD NOT affect production but if I do something by accident any you spot any issues please let me know immediately. Thanks Zygmunt Krynicki

12 years, 11 months

[ACTIVITY] Multimedia WG weekly status report - wk46-47.2011 (20111114-20111125)

by Ilias Biris

Status with details on both the delivered content for 11.11 and the plans for 11.12: https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/WeeklyReport Last weekly meeting minutes: https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/Notes/2011-11-22 Highlights - Codecs + Created a wiki page to collect the status of NEON optimisation of different codecs: https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/NeonStatus. Feedback is welcome as well as adding your favourite libraries! - For LJT: + scoped out use across main, universe and multiverse of libjpeg62, libjpeg8 and -dev + ppa tom-gall/libjpeg-turbo updated with libjpeg8 compat version of libjpeg-turbo + all packages that use libjpeg-dev pushed to ppa (31 pending builds yet, 89 successful 0 failures) - UMM + Completed the xf86nouveau development of dri2video support. + Setup CI build to test CMA on snowball: https://ci.linaro.org/jenkins/view/All/job/linux-linaro-snowball-cma-test/ + Reworked debugfs and trace patches for CMA v17 Comments, clarifications needed? Please let me know! Best regards, -- Ilias Biris ilias.biris(a)linaro.org Project Manager, Linaro M: +358504839608, IRC: ibiris Skype: ilias_biris Linaro.org│ Open source software for ARM SoCs

12 years, 11 months

launchpad: ddeb packages in ppa problem

by Riku Voipio

Hi, I got the following errors from my recent uploads to linaro-maintainers/staging-overlay ppa: Rejected: Orphaned debug packages: zlib1g-udeb-dbgsym 1:1.2.3.4.dfsg-3ubuntu3linaro1 (i386) So it seems ddebs have been enabled (thanks!), but seems broken - I didn't do anything udeb related while packaging, and if udebs are being dropped automatically, seems so should be done to the udebs dbgsym package. Cheers, Riku

12 years, 11 months

[ACTIVITY] Graphics weekly status report - wk46-47 (20111114-20111125)

by Ilias Biris

Status report in detail - contains the delivered content for 11.11 and the plans for 11.12 as of now: https://wiki.linaro.org/WorkingGroups/Middleware/Graphics/WeeklyReport Last weekly meeting notes: https://wiki.linaro.org/WorkingGroups/Middleware/Graphics/Notes/2011-11-23 Highlights: - Based on the meeting last week with DX side: no changed lately on NUX. openGLES code has been merged, and believed to be functional and working. At the moment doing more testing on NUX. There is a shift in DX to accept branches after more testing, much of which is automated. + In the future there is more need for testing for functional verification and also regression testing + Specification related to the tests was requested from DX. Note that the tests for the graphics stack have not been decided yet. + For the Linaro side what is important is that DX team want to have a test for every major feature/release related to openGLES. New features should be accompanied by test code, covering the functionality as we well as regression cases. + Admittedly there is still a definition pending on DX side about what coverage to provide, not every submission will need a new test necessarily. This is still being decided at the DX team level, and they will communicate with Linaro on the process (still undefined on wk47) o This delay in communicating the test specification can potentially cause delays on Graphics WG - Daily benchmarks running on LAVA: + http://validation.linaro.org/lava-server/dashboard/streams/anonymous/linaro… + Dealing with some final issues: Having some boot issues, probably kernel related, which will need to be cleared this week - Starting in 11.12 to analyse the power management work for Graphics - Moved all Launchpad projects for Graphics WG to use trunk as their active series of development Any questions or clarification needed? Please let me know. -- Ilias Biris ilias.biris(a)linaro.org Project Manager, Linaro M: +358504839608, IRC: ibiris Skype: ilias_biris Linaro.org│ Open source software for ARM SoCs

12 years, 11 months

11.11 Postmortem and 11.12 Plan

by Zach Pfeffer

The Linaro Android Platform 11.11 Cycle 11.11 started the last week of October after Linaro 11.10 was released on Oct 27th. The Android team had the RC for the 24th done on the 21st so that Bernhard Rosenkränzer and I could focus on ELC-E in Prague and the team could focus on getting ready for Linaro Connect Q4.11 in Orlando, Oct 31st through November 4th. The theme for 11.10 had been "make it shiny" as we wanted to put our best foot forward at ELC-E and during Connect. Prague went well. Bernhard and I did our presentation and ran the booth with Ricardo Salveti. We had a bit of excitement when one of our evening hacking sessions turned into a travel scramble as Arnd Bergman and others tried to find alternative flights after the French Air Strike. ELC-E Presentation, Linaro's Android Platform http://www.youtube.com/watch?v=zWuGDX6Sz2Y elinux.org tutorial http://elinux.org/Android_Tutorials_Unbox_to_App Everyone got to Connect okay and we set off on a very exciting week of planning, hacking, arguing and "socializing." We ran sessions on device tree, sched_mc, build improvements, automated testing, educational outreach, handling binaries, benchmarking, future plans, QA and gave a tutorial on how to use Linaro's Android distribution. Amit Pundir also started off his Linaro career with a great porting Android to new platforms session. Bernhard made great strides in supporting a dual toolchain, Chao Yang debugged libpng issues, Tony Månsson worked on Powertutor extensions, Frans Gifford got the CI loop working with each target, Prashanth Srinivasan onboarded by helping do and test builds and Abhishek Paliwal worked to unify our QA tests with each test description. We capped the week off with demos including, Vishal Bhoj's getting a Linphone based Video Conferencing App working, Noritsuna Imamura's Cluster Android App Engine and Botao Sun showing WiFi on Samsung Origen, Woot! Jon Medhurst (Tixy) was also able to make substantial progress getting Android running on Vexpress with Mathieu Poirier's help and Michael Edwards offered to help with our SGX upgrade. Linaro Connect: Demo Friday III http://www.youtube.com/watch?v=3lRxiTA6A10 Linaro Connect Q4.11 - Panda cloud cluster http://www.youtube.com/watch?v=pSGNf0KZ71c Linaro Android Tech Lead Zach Pfeffer at Linaro Connect http://www.youtube.com/watch?v=yVZ1kblmQMY Porting Android to New Platforms https://docs.google.com/a/linaro.org/viewer?a=v&pid=explorer&chrome=true&sr… After connect we set out to finish out 11.11, knowing that Ice Cream Sandwich (ICS) would land at anytime. Bernhard even had a cron job looking for new commits at AOSP. Even thought we live in the future we can't predict it and Patrik Ryd set out to upgrade our baseline to 2.3.7. The upgrade plus our usual work of upgrading to the latest Linaro toolchain and adding support for each board feature kept us busy until, Nov 14th's email about ICS dropping: http://groups.google.com/group/android-building/browse_thread/thread/4f85d9… Alexander Sack started syncing the code the minute is was ready, I started a build and VIshal and I started getting AOSP to compile on Pandaboard. Vishal fixed a few issues and we tried the AOSP first thing the next morning. AOSP running on Pandaboard http://www.youtube.com/user/pfefferzlinaro?feature=mhee#p/u/4/eaVszdsZ8aY Vishal also started looking into getting this running using just software GL and had a prototype working soon after that showed Panda running with a Linaro kernel Android 4.0.1 (ICS) with the Linaro Kernel http://www.youtube.com/watch?v=-LtjvekCDKM With this proof on concept Mathieu was able to get Snowball running ICS ICS running on Snowball http://www.youtube.com/watch?v=CvRg6OX0DL8 Alexander then challenged the group to get each board switched over and to release a preliminary release on Thanksgiving. With the wind at our backs the Linaro Android team achieved this. We had a few snags, the largest was related to the enormous strain that the ICS code base put on the existing infrastructure. On Friday Nov 18th, as we were getting ready to start cutting 2.3.7 releases, we had 2 coincident events: a cross git change that had to be reverted and a build system that couldn't sync the code base because of the expanded size of ICS in the git mirror. After a stand-up meeting Paul Sokolovsky saved the day be launching his seeded build mechanism that not only allowed the builds to go through, but cut down on the sync time. This change brought about a paradigm shift in build performance and reliability. 11.11 was an incredible cycle. The lessons learned were: 1. Gerrit doesn't handle cross git changes so it should be extended to 2. We need to ensure that each target in LAVA is stable 3. If things are starting to get chaotic then anyone can yell stopthepresses in #linaro-android and we'll all stop and regroup For 11.12 the high-level plan is to: 1. Compile ICS with Linaro's 4.6 Toolchain and send the source modifications to Google 2. Integrate our work on 2.3.7 including, libjpeg-turbo and ds-5 into ICS and send the changes to Google 3. Get PoCs for each binary, get them into linaro-android, enabling graphics and multimedia 11.12 starts tomorrow, Nov 28th, the 11.12 RCs will come out Dec. 20th and be cut on the 18th. Its going to be a great December! -- Zach Pfeffer Android Platform Team Lead, Linaro Platform Teams Linaro.org | Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

12 years, 11 months

[PATCH] backlight : add lp855x backlight driver

by Kim, Milo

Hello, Backlight Class Maintainer I send the patch for new lp855x backlight driver. This patch provides Texas Instruments LP855x backlight driver. * Supported Backlight ICs LP8550, LP8551, LP8552 and LP8556 * Consolidated Driver Just one driver file can support several LP855x backlight ICs. * Driver Description 1) Brightness control Brightness can be controlled by pwm input or i2c command. lp855x driver support both cases. PWM are platform specific data, so the driver provides pwm functions in lp855x_platform_data structure. 2) Debugfs nodes : For debug purpose, 3 files are exported in the debugfs. a) bl_ctl_mode : backlight control mode pwm based or register based b) chip_id : lp855x chip id lp8550/lp8551/lp8552/lp8556 c) registers we can control the lp855x registers via the debugfs. read/write/dump mode are supported. 3) Platform data for lp855x a) name : backlight driver name b) mode : brightness control by pwm or lp855x register c) device_control : value of DEVICE CONTROL register d) initial_brightness : initial value of backlight brightness e) max_brightness : maximum value of backlight brightness f) pwm_data : platform specific pwm generation functions. Only valid when brightness is pwm input mode. g) load_new_rom_data : 0 : use default configuration data 1 : update values of eeprom or eprom registers on loading driver h) size_program : total size of lp855x_rom_data i) rom_data : list of new eeprom/eprom registers example 1) lp8552 platform data : i2c register mode with new eeprom data static struct lp855x_rom_data lp8552_eeprom_arr[] = { {0xa5, 0x4f}, /* EN_VSYNC=0 */ }; static struct lp855x_platform_data lp8552_pdata = { .name = "lcd-backlight", .mode = REGISTER_BASED, .device_control = I2C_CONFIG(LP8552), .initial_brightness = INITIAL_BRT, .max_brightness = MAX_BRT, .load_new_rom_data = 1, .size_program = ARRAY_SIZE(lp8552_eeprom_arr), .rom_data = lp8552_eeprom_arr, }; example 2) lp8556 platform data : pwm input mode static struct lp855x_platform_data lp8556_pdata = { .name = "lcd-backlight", .mode = PWM_BASED, .device_control = PWM_CONFIG(LP8556), .initial_brightness = INITIAL_BRT, .max_brightness = MAX_BRT, .pwm_data = { .pwm_set_intensity = platform_pwm_set_intensity, .pwm_get_intensity = platform_pwm_get_intensity, }, }; * Test Environment OMAP3530 beagelborad with LP8551/LP8552/LP8556 ICs >From 903ca54f2da059321f6f1519af79aeec19aff577 Mon Sep 17 00:00:00 2001 From: Milo(Woogyom) Kim <milo.kim(a)ti.com> Date: Mon, 28 Nov 2011 10:48:07 +0900 Subject: [PATCH] backlight : add lp855x backlight driver Signed-off-by: Milo(Woogyom) Kim <milo.kim(a)ti.com> --- drivers/video/backlight/Kconfig | 6 + drivers/video/backlight/Makefile | 1 + drivers/video/backlight/lp855x_bl.c | 473 +++++++++++++++++++++++++++++++++++ include/linux/lp855x.h | 123 +++++++++ 4 files changed, 603 insertions(+), 0 deletions(-) create mode 100755 drivers/video/backlight/lp855x_bl.c create mode 100644 include/linux/lp855x.h diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig index 278aeaa..9e7ba76 100644 --- a/drivers/video/backlight/Kconfig +++ b/drivers/video/backlight/Kconfig @@ -342,6 +342,12 @@ config BACKLIGHT_AAT2870 If you have a AnalogicTech AAT2870 say Y to enable the backlight driver. +config BACKLIGHT_LP855X + tristate "Backlight driver for Texas Instruments LP855X" + depends on BACKLIGHT_CLASS_DEVICE && I2C + help + This supports TI LP8550, LP8551, LP8552 and LP8556 backlight driver. + endif # BACKLIGHT_CLASS_DEVICE endif # BACKLIGHT_LCD_SUPPORT diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile index fdd1fc4..91ae232 100644 --- a/drivers/video/backlight/Makefile +++ b/drivers/video/backlight/Makefile @@ -39,4 +39,5 @@ obj-$(CONFIG_BACKLIGHT_ADP8870) += adp8870_bl.o obj-$(CONFIG_BACKLIGHT_88PM860X) += 88pm860x_bl.o obj-$(CONFIG_BACKLIGHT_PCF50633) += pcf50633-backlight.o obj-$(CONFIG_BACKLIGHT_AAT2870) += aat2870_bl.o +obj-$(CONFIG_BACKLIGHT_LP855X) += lp855x_bl.o diff --git a/drivers/video/backlight/lp855x_bl.c b/drivers/video/backlight/lp855x_bl.c new file mode 100755 index 0000000..2bae816 --- /dev/null +++ b/drivers/video/backlight/lp855x_bl.c @@ -0,0 +1,473 @@ +/* + * TI LP855x Backlight Driver + * + * Copyright (C) 2011 Texas Instruments + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/i2c.h> +#include <linux/backlight.h> +#include <linux/err.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/lp855x.h> + +#define BRIGHTNESS_CTRL (0x00) +#define DEVICE_CTRL (0x01) + +#ifdef CONFIG_DEBUG_FS +struct debug_dentry { + struct dentry *dir; + struct dentry *reg; + struct dentry *chip; + struct dentry *blmode; +}; +#endif + +struct lp855x { + const char *chipid; + struct i2c_client *client; + struct backlight_device *bl; + struct device *dev; + struct mutex xfer_lock; + struct lp855x_platform_data *pdata; +#ifdef CONFIG_DEBUG_FS + struct debug_dentry dd; +#endif +}; + +static int lp855x_i2c_read(struct lp855x *lp, u8 reg, u8 *data, u8 len) +{ + s32 ret; + + mutex_lock(&lp->xfer_lock); + ret = i2c_smbus_read_i2c_block_data(lp->client, reg, len, data); + mutex_unlock(&lp->xfer_lock); + + return (ret != len) ? -EIO : 0; +} + +static int lp855x_i2c_write(struct lp855x *lp, u8 reg, u8 *data, u8 len) +{ + s32 ret; + + mutex_lock(&lp->xfer_lock); + ret = i2c_smbus_write_i2c_block_data(lp->client, reg, len, data); + mutex_unlock(&lp->xfer_lock); + + return ret; +} + +static inline int lp855x_read_byte(struct lp855x *lp, u8 reg, u8 *data) +{ + return lp855x_i2c_read(lp, reg, data, 1); +} + +static inline int lp855x_write_byte(struct lp855x *lp, u8 reg, u8 data) +{ + u8 written = data; + return lp855x_i2c_write(lp, reg, &written, 1); +} + +#ifdef CONFIG_DEBUG_FS +static int lp855x_dbg_open(struct inode *inode, struct file *file) +{ + file->private_data = inode->i_private; + return 0; +} + +static ssize_t lp855x_help_register(struct file *file, char __user *userbuf, + size_t count, loff_t *ppos) +{ + char buf[320]; + unsigned int len; + const char *help = "\n How to read/write LP855x registers\n\n \ + (example) To read 0x00 register,\n \ + echo 0x00 r > /sys/kernel/debug/lp855x/registers\n \ + To write 0xff into 0x1 address,\n \ + echo 0x00 0xff w > /sys/kernel/debug/lp855x/registers \n \ + To dump values from 0x00 to 0x06 address,\n \ + echo 0x00 0x06 d > /sys/kernel/debug/lp855x/registers\n"; + + len = snprintf(buf, sizeof(buf), "%s\n", help); + if (len > sizeof(buf)) + len = sizeof(buf); + + return simple_read_from_buffer(userbuf, count, ppos, buf, len); +} + +static char *lp855x_parse_register_cmd(const char *cmd, u8 *byte) +{ + char tmp[10]; + char *blank; + unsigned long arg; + + blank = strchr(cmd, ' '); + memset(tmp, 0x0, sizeof(tmp)); + memcpy(tmp, cmd, blank - cmd); + + if (strict_strtol(tmp, 16, &arg) < 0) + return NULL; + + *byte = arg; + return blank; +} + +static ssize_t lp855x_ctrl_register(struct file *file, + const char __user *userbuf, size_t count, + loff_t *ppos) +{ + char mode, buf[20]; + char *pos, *pos2; + u8 i, arg1, arg2, val; + struct lp855x *lp = file->private_data; + + if (copy_from_user(buf, userbuf, min(count, sizeof(buf)))) + return -EFAULT; + + mode = buf[count - 2]; + switch (mode) { + case 'r': + if (!lp855x_parse_register_cmd(buf, &arg1)) + return -EINVAL; + + lp855x_read_byte(lp, arg1, &val); + dev_info(lp->dev, "Read [0x%.2x] = 0x%.2x\n", arg1, val); + break; + case 'w': + pos = lp855x_parse_register_cmd(buf, &arg1); + if (!pos) + return -EINVAL; + pos2 = lp855x_parse_register_cmd(pos + 1, &arg2); + if (!pos2) + return -EINVAL; + + lp855x_write_byte(lp, arg1, arg2); + dev_info(lp->dev, "Written [0x%.2x] = 0x%.2x\n", arg1, arg2); + break; + case 'd': + pos = lp855x_parse_register_cmd(buf, &arg1); + if (!pos) + return -EINVAL; + pos2 = lp855x_parse_register_cmd(pos + 1, &arg2); + if (!pos2) + return -EINVAL; + + for (i = arg1; i <= arg2; i++) { + lp855x_read_byte(lp, i, &val); + dev_info(lp->dev, "Read [0x%.2x] = 0x%.2x\n", i, val); + } + break; + default: + break; + } + + return count; +} + +static ssize_t lp855x_get_chipid(struct file *file, char __user *userbuf, + size_t count, loff_t *ppos) +{ + struct lp855x *lp = file->private_data; + char buf[10]; + unsigned int len; + + len = snprintf(buf, sizeof(buf), "%s\n", lp->chipid); + + if (len > sizeof(buf)) + len = sizeof(buf); + + return simple_read_from_buffer(userbuf, count, ppos, buf, len); +} + +static ssize_t lp855x_get_bl_mode(struct file *file, char __user *userbuf, + size_t count, loff_t *ppos) +{ + char buf[20]; + unsigned int len; + char *strmode = NULL; + struct lp855x *lp = file->private_data; + enum lp855x_brightness_ctrl_mode mode = lp->pdata->mode; + + if (mode == PWM_BASED) + strmode = "pwm based"; + else if (mode == REGISTER_BASED) + strmode = "register based"; + + len = snprintf(buf, sizeof(buf), "%s\n", strmode); + + if (len > sizeof(buf)) + len = sizeof(buf); + + return simple_read_from_buffer(userbuf, count, ppos, buf, len); +} + +#define LP855X_DBG_ENTRY(name, pread, pwrite) \ +static const struct file_operations dbg_##name##_fops = { \ + .open = lp855x_dbg_open, \ + .read = pread, \ + .write = pwrite, \ + .owner = THIS_MODULE, \ + .llseek = default_llseek, \ +} + +LP855X_DBG_ENTRY(registers, lp855x_help_register, lp855x_ctrl_register); +LP855X_DBG_ENTRY(chip, lp855x_get_chipid, NULL); +LP855X_DBG_ENTRY(blmode, lp855x_get_bl_mode, NULL); + +static void lp855x_create_debugfs(struct lp855x *lp) +{ + struct debug_dentry *dd = &lp->dd; + + dd->dir = debugfs_create_dir("lp855x", NULL); + + dd->reg = debugfs_create_file("registers", S_IWUSR | S_IRUGO, + dd->dir, lp, &dbg_registers_fops); + + dd->chip = debugfs_create_file("chip_id", S_IRUGO, + dd->dir, lp, &dbg_chip_fops); + + dd->blmode = debugfs_create_file("bl_ctl_mode", S_IRUGO, + dd->dir, lp, &dbg_blmode_fops); +} + +static void lp855x_remove_debugfs(struct lp855x *lp) +{ + struct debug_dentry *dd = &lp->dd; + + debugfs_remove(dd->blmode); + debugfs_remove(dd->chip); + debugfs_remove(dd->reg); + debugfs_remove(dd->dir); +} +#else +static inline void lp855x_create_debugfs(struct lp855x *lp) +{ + return; +} + +static inline void lp855x_remove_debugfs(struct lp855x *lp) +{ + return; +} +#endif + +static int lp855x_is_valid_rom_area(struct lp855x *lp, u8 addr) +{ + const char *id = lp->chipid; + u8 start, end; + + if (strstr(id, "lp8550") || strstr(id, "lp8551") + || strstr(id, "lp8552")) { + start = EEPROM_START; + end = EEPROM_END; + } else if (strstr(id, "lp8556")) { + start = EPROM_START; + end = EPROM_END; + } + + return (addr >= start && addr <= end) ? 1 : 0; +} + +static void lp855x_init_device(struct lp855x *lp) +{ + u8 val, addr; + int i, ret; + struct lp855x_platform_data *pd = lp->pdata; + + if (pd->load_new_rom_data && pd->size_program) { + for (i = 0; i < pd->size_program; i++) { + addr = pd->rom_data[i].addr; + val = pd->rom_data[i].val; + if (!lp855x_is_valid_rom_area(lp, addr)) + continue; + + ret |= lp855x_write_byte(lp, addr, val); + } + } + + val = pd->initial_brightness; + ret |= lp855x_write_byte(lp, BRIGHTNESS_CTRL, val); + + val = pd->device_control; + ret |= lp855x_write_byte(lp, DEVICE_CTRL, val); + + if (ret) + dev_err(lp->dev, "i2c write err\n"); +} + +static int lp855x_bl_update_status(struct backlight_device *bl) +{ + struct lp855x *lp = bl_get_data(bl); + enum lp855x_brightness_ctrl_mode mode = lp->pdata->mode; + + if (bl->props.state & BL_CORE_SUSPENDED) + bl->props.brightness = 0; + + if (mode == PWM_BASED) { + struct lp855x_pwm_data *pd = &lp->pdata->pwm_data; + int br = bl->props.brightness; + int max_br = bl->props.max_brightness; + + if (pd->pwm_set_intensity) + pd->pwm_set_intensity(br, max_br); + + } else if (mode == REGISTER_BASED) { + u8 val = bl->props.brightness; + lp855x_write_byte(lp, BRIGHTNESS_CTRL, val); + } + + return (bl->props.brightness); +} + +static int lp855x_bl_get_brightness(struct backlight_device *bl) +{ + struct lp855x *lp = bl_get_data(bl); + enum lp855x_brightness_ctrl_mode mode = lp->pdata->mode; + + if (mode == PWM_BASED) { + struct lp855x_pwm_data *pd = &lp->pdata->pwm_data; + int max_br = bl->props.max_brightness; + + if (pd->pwm_get_intensity) + bl->props.brightness = pd->pwm_get_intensity(max_br); + + } else if (mode == REGISTER_BASED) { + u8 val; + + lp855x_read_byte(lp, BRIGHTNESS_CTRL, &val); + bl->props.brightness = val; + } + + return (bl->props.brightness); +} + +static const struct backlight_ops lp855x_bl_ops = { + .options = BL_CORE_SUSPENDRESUME, + .update_status = lp855x_bl_update_status, + .get_brightness = lp855x_bl_get_brightness, +}; + +static int lp855x_backlight_register(struct lp855x *lp) +{ + struct backlight_device *bl; + struct backlight_properties props; + const char *name = lp->pdata->name; + + if (!name) + return -ENODEV; + + props.brightness = lp->pdata->initial_brightness; + props.max_brightness = + (lp->pdata->max_brightness < lp->pdata->initial_brightness) ? + 255 : lp->pdata->max_brightness; + + bl = backlight_device_register(name, lp->dev, lp, + &lp855x_bl_ops, &props); + if (IS_ERR(bl)) + return -EIO; + + lp->bl = bl; + + return 0; +} + +static void lp855x_backlight_unregister(struct lp855x *lp) +{ + if (lp->bl) + backlight_device_unregister(lp->bl); +} + +static int lp855x_probe(struct i2c_client *cl, const struct i2c_device_id *id) +{ + struct lp855x *lp; + struct lp855x_platform_data *pdata = cl->dev.platform_data; + int ret; + + if (!i2c_check_functionality(cl->adapter, I2C_FUNC_SMBUS_I2C_BLOCK)) + goto err_io; + + lp = kzalloc(sizeof(struct lp855x), GFP_KERNEL); + if (!lp) + goto err_mem; + + lp->client = cl; + lp->dev = &cl->dev; + lp->pdata = pdata; + lp->chipid = id->name; + i2c_set_clientdata(cl, lp); + + mutex_init(&lp->xfer_lock); + + lp855x_init_device(lp); + ret = lp855x_backlight_register(lp); + if (ret) + goto err_dev; + + backlight_update_status(lp->bl); + lp855x_create_debugfs(lp); + + return ret; + +err_io: + return -EIO; +err_mem: + return -ENOMEM; +err_dev: + dev_err(lp->dev, "can not register backlight device. errcode = %d\n", + ret); + kfree(lp); + return ret; +} + +static int __devexit lp855x_remove(struct i2c_client *cl) +{ + struct lp855x *lp = i2c_get_clientdata(cl); + + lp->bl->props.brightness = 0; + backlight_update_status(lp->bl); + lp855x_remove_debugfs(lp); + lp855x_backlight_unregister(lp); + kfree(lp); + + return 0; +} + +static const struct i2c_device_id lp855x_ids[] = { + {"lp8550", LP8550}, + {"lp8551", LP8551}, + {"lp8552", LP8552}, + {"lp8556", LP8556}, +}; + +static struct i2c_driver lp855x_driver = { + .driver = { + .name = "lp855x", + }, + .probe = lp855x_probe, + .remove = __devexit_p(lp855x_remove), + .id_table = lp855x_ids, +}; + +static int __init lp855x_init(void) +{ + return i2c_add_driver(&lp855x_driver); +} + +static void __exit lp855x_exit(void) +{ + i2c_del_driver(&lp855x_driver); +} + +module_init(lp855x_init); +module_exit(lp855x_exit); + +MODULE_DESCRIPTION("Texas Instruments LP855x Backlight driver"); +MODULE_AUTHOR("Milo Kim <milo.kim(a)ti.com>, Dainel Jeong <daniel.jeong(a)ti.com>"); +MODULE_LICENSE("GPL"); diff --git a/include/linux/lp855x.h b/include/linux/lp855x.h new file mode 100644 index 0000000..5078558 --- /dev/null +++ b/include/linux/lp855x.h @@ -0,0 +1,123 @@ +/* + * lp855x.h - TI LP8556 Backlight Driver + * + * Copyright (C) 2011 Texas Instruments + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + */ + +#ifndef _LP855X_H +#define _LP855X_H + +#define BL_CTL_SHFT (0) +#define BRT_MODE_SHFT (1) +#define BRT_MODE_MASK (0x06) + +/* Enable backlight. Only valid when BRT_MODE=10(I2C only) */ +#define ENABLE_BL (1) +#define DISABLE_BL (0) + +#define I2C_CONFIG(id) id ## _I2C_CONFIG +#define PWM_CONFIG(id) id ## _PWM_CONFIG + +/* DEVICE CONTROL register - LP8550 */ +#define LP8550_PWM_CONFIG (LP8550_PWM_ONLY << BRT_MODE_SHFT) +#define LP8550_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ + (LP8550_I2C_ONLY << BRT_MODE_SHFT)) + +/* DEVICE CONTROL register - LP8551 */ +#define LP8551_PWM_CONFIG LP8550_PWM_CONFIG +#define LP8551_I2C_CONFIG LP8550_I2C_CONFIG + +/* DEVICE CONTROL register - LP8552 */ +#define LP8552_PWM_CONFIG LP8550_PWM_CONFIG +#define LP8552_I2C_CONFIG LP8550_I2C_CONFIG + +/* DEVICE CONTROL register - LP8556 */ +#define LP8556_PWM_CONFIG (LP8556_PWM_ONLY << BRT_MODE_SHFT) +#define LP8556_COMB1_CONFIG (LP8556_COMBINED1 << BRT_MODE_SHFT) +#define LP8556_I2C_CONFIG ((ENABLE_BL << BL_CTL_SHFT) | \ + (LP8556_I2C_ONLY << BRT_MODE_SHFT)) +#define LP8556_COMB2_CONFIG (LP8556_COMBINED2 << BRT_MODE_SHFT) + +/* ROM area boundary */ +#define EEPROM_START (0xA0) +#define EEPROM_END (0xA7) +#define EPROM_START (0xA0) +#define EPROM_END (0xAF) + +enum lp855x_chip_id { + LP8550, + LP8551, + LP8552, + LP8556, +}; + +enum lp855x_brightness_ctrl_mode { + PWM_BASED = 1, + REGISTER_BASED, +}; + +enum lp8550_brighntess_source { + LP8550_PWM_ONLY, + LP8550_I2C_ONLY = 2, +}; + +enum lp8551_brighntess_source { + LP8551_PWM_ONLY = LP8550_PWM_ONLY, + LP8551_I2C_ONLY = LP8550_I2C_ONLY, +}; + +enum lp8552_brighntess_source { + LP8552_PWM_ONLY = LP8550_PWM_ONLY, + LP8552_I2C_ONLY = LP8550_I2C_ONLY, +}; + +enum lp8556_brightness_source { + LP8556_PWM_ONLY, + LP8556_COMBINED1, /* pwm + i2c before the shaper block */ + LP8556_I2C_ONLY, + LP8556_COMBINED2, /* pwm + i2c after the shaper block */ +}; + +struct lp855x_pwm_data { + void (*pwm_set_intensity) (int brightness, int max_brightness); + int (*pwm_get_intensity) (int max_brightness); +}; + +struct lp855x_rom_data { + u8 addr; + u8 val; +}; + +/** + * struct lp855x_platform_data + * @name : backlight driver name + * @mode : brightness control by pwm or lp855x register + * @device_control : value of DEVICE CONTROL register + * @initial_brightness : initial value of backlight brightness + * @max_brightness : maximum value of backlight brightness + * @pwm_data : platform specific pwm generation functions. + Only valid when mode is PWM_BASED. + * @load_new_rom_data : + 0 : use default configuration data + 1 : update values of eeprom or eprom registers on loading driver + * @size_program : total size of lp855x_rom_data + * @rom_data : list of new eeprom/eprom registers + */ +struct lp855x_platform_data { + const char *name; + enum lp855x_brightness_ctrl_mode mode; + u8 device_control; + int initial_brightness; + int max_brightness; + struct lp855x_pwm_data pwm_data; + u8 load_new_rom_data; + int size_program; + struct lp855x_rom_data *rom_data; +}; + +#endif -- 1.7.4.1 Best Regards Milo (Woogyom) Kim Texas Instruments Incorporated Software Engineer Mobile Lighting & Power Office : +82-2-560-6963

12 years, 11 months

Was: Track the Linaro Roadmap on status.linaro.org

by Fathi Boudra

On 26 November 2011 12:37, Raffaele Recalcati <lamiaposta71(a)gmail.com> wrote: > 1 download monthly release for each official board (ubuntu, android) > 2 linaro-media-create it > 3 simple benchmark tests (boot is clean, video decoding, QT or Android > gui is responsive, browser) and report it > 4 keep a SD card for android and ubuntu for this month and the > previous (only 4*2*2=16 SD cards needed) in order to see quickly the > differences. You can do all of this right now. Though, we don't publish an analysis between releases as you suggest on step 4. > If not why? This can give an help to industrial application of Linaro. > > Thx, > Raffaele Cheers -- Fathi Boudra Linaro Release Manager | Validation Project Manager Linaro.org | Open source software for ARM SoCs

12 years, 11 months

← Newer
1
2
3
4
5
6
7
...
13
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-dev November 2011