November 2012 - U-Boot

[U-Boot] Kirkwood register and interrupt issues
by Bernhard Schuster 08 Nov '12

08 Nov '12

Dear mailinglist subscribers, I am currently trying to write a suspend (via wait-for-interrupt, cp15 instruction) function in the u-boot context which will recover on a gpio edge event. Thus I am trying to enable interrupts via the interrupt function the linux kernel uses (see end of this mail). Unfortunately the cp15 instruction never recovered. Now I am stuck with this, I try to write some registers to configure timers (to force create cyclic interrupts) and unmask all kind of interrupts, but the intended values actually never get written to the registers. I am really puzzled. Any help, demo implementation or tipp is really appreciated. Best regards Bernhard post scriptum: ============================================================= code (chunk from init function): struct kwcpu_registers *cpureg = (struct kwcpu_registers*)(KW_REGISTER(0x20100)); writel (&cpureg->ahb_mbus_mask_irq, 0xFFFFFFFF); writel (&cpureg->ahb_mbus_cause_irq, 0); struct kwgpio_registers *gpio0reg = (struct kwgpio_registers*)(KW_GPIO0_BASE); struct kwgpio_registers *gpio1reg = (struct kwgpio_registers*)(KW_GPIO1_BASE); //unmask edge fun, disable level stuff writel (&gpio0reg->irq_mask, 0xFFFFFFFF); writel (&gpio0reg->irq_level, 0); writel (&gpio1reg->irq_mask, 0xFFFFFFFF); writel (&gpio1reg->irq_level, 0); struct kwtimer_registers *timer = (struct kwtimer_registers*)(KW_TIMER_BASE); writel (&timer->t0, 0x000FFFFF); writel (&timer->reload0, 0x00000FFF); writel (&timer->t1, 0x000FFFFF); writel (&timer->reload1, 0x00000FFF); writel (&timer->ctl, 0x0F); //enable both timers, no watchdog printf ("cpu config 0x%08x\n", readl(&cpureg->config)); printf ("cpu ctrl_stat 0x%08x\n", readl(&cpureg->ctrl_stat)); printf ("mbus mask 0x%08x\n", readl(&cpureg->ahb_mbus_mask_irq)); printf ("mbus cause 0x%08x\n", readl(&cpureg->ahb_mbus_cause_irq)); printf ("timer 0 RELOAD VAL 0x%08x\n", readl(&timer->reload0)); printf ("timer 1 RELOAD VAL 0x%08x\n", readl(&timer->reload1)); printf ("GPIO 0 Edge 0x%08x\n", readl(&gpio0reg->irq_mask)); printf ("GPIO 0 Level 0x%08x\n", readl(&gpio0reg->irq_level)); printf ("GPIO 1 Edge 0x%08x\n", readl(&gpio1reg->irq_mask)); printf ("GPIO 1 Level 0x%08x\n", readl(&gpio1reg->irq_level)); printf ("0x%08x (low interrupt cause)\n", (u32) readl (KW_REGISTER(0x20200))); printf ("0x%08x (high interrupt cause)\n", (u32) readl (KW_REGISTER(0x20210))); output from the above indented chunk: cpu config 0x00100036 cpu ctrl_stat 0x00000001 mbus mask 0x00000000 mbus cause 0x00000002 //mbus interrupt source detected timer 0 RELOAD VAL 0xffffffff //just wrong, we did write 0x00000FFF to that one timer 1 RELOAD VAL 0x00000000 //just wrong 0x00000FFF GPIO 0 Edge 0x00000000 // wrong GPIO 0 Level 0x00000000 // wrong GPIO 1 Edge 0x00000000 // wrong GPIO 1 Level 0x00000000 // wrong 0x00800000 (low interrupt cause) // spi interrupt detected, not expected 0x00000000 (high interrupt cause) ============================================================= /* * Enable IRQs */ inline void arch_local_irq_enable(void) { unsigned long temp; __asm__ __volatile__( "mrs %0, cpsr @ arch_local_irq_enable\n" "bic %0, %0, #128\n" "msr cpsr_c, %0" : "=r" (temp) : : "memory", "cc"); } /* * Disable IRQs */ inline void arch_local_irq_disable(void) { unsigned long temp; __asm__ __volatile__( "mrs %0, cpsr @ arch_local_irq_disable\n" "orr %0, %0, #128\n" "msr cpsr_c, %0" : "=r" (temp) : : "memory", "cc"); }

1 0

[U-Boot] [PATCH v5 00/26] pmic: Redesign PMIC framework to support multiple instances of devices
by Lukasz Majewski 08 Nov '12

08 Nov '12

PMIC framework has been redesigned to support multiple instances of power related devices (e.g. fuel gauge, PMICs, chargers, micro USB IC, battery). Due to that, code at other architectures and boards have been adjusted properly. New power_init_board() method at ./lib/board.c file has been introduced. It is meant to be an architecture dependent function to support advanced power management. Since PMIC framework uses lists internally to link different devices, its initialization must be done just after malloc initialization. Please consider commits from this patch set as the example of advanced power management for a particular HW (Trats board in this case). In the new approach PMICs are selected with their names (e.g. 'pmic dump MAX8997_PMIC') Presented patch set is a first step to change 'pmic' command to more generic (i.e. power) to provide control for multiple devices. Moreover device's battery is treated as an oridinary "power" device (like PMIC). Due to that, framework unification is possible. For even more versalite design, each power device instance can specify its parent. For trats one can build a following scheme (it can de different for other boards): ----------------- --------| BAT |------------ | | | | | ----------------- | | | | \|/ \|/ \|/ ----------- ----------------- --------- |FG | |MUIC | |CHRG | | | | | | | ----------- ----------------- --------- Finally, this patch series also comprises of a "namespace" cleanup - an attempt to rename "pmic" to "power" to better reflect generality of this framework. Test HW: - Exynos4210 Trats development board - Exynos3 (C110) Goni development board Lukasz Majewski (26): pmic:i2c: Handle PMIC I2C transmission comprising of two bytes pmic:i2c: Add I2C sensor byte order (big/little) to PMIC framework pmic:max8997: Switch the MAX8997 PMIC to be used with multibus I2C pmic: Extend PMIC framework to support multiple instances of PMIC devices pmic: Introduce power_init_board() method at ./lib/board.c file pmic: Enable power_board_init() support at TRATS pmic:chrg: Common information about charger and battery (power_chrg.h) pmic: Move pmic related code to ./drivers/power directory pmic: Extend struct pmic to support battery and charger related operations pmic:battery: Support for Trats Battery at PMIC framework pmic:muic: Support for MUIC built into MAX8997 device pmic:fuel-gauge: Support for MAX17042 fuel-gauge pmic:max8997: Function for calculating LDO internal register value arm:trats:pmic: Default PMIC(MAX8997) initialization for Samsung's TRATS board arm:trats:pmic: Enable MUIC (MAX8997) at Samsung's TRATS board arm:trats:pmic: Enable fuel-gauge (MAX17042) at Samsung's TRATS board arm:trats:pmic: Enable battery support at Samsung's TRATS board pmic:max8997: Support for MAX8997 internal charger control arm:trats:pmic: Power consumption reduction state for Samsung's TRATS board arm:trats:pmic: Support for charging battery at Samsung's TRATS board pmic: Extend PMIC framework to support battery related commands power:pmic: Rename ./drivers/power/pmic_* to ./drivers/power/power_* files power:pmic: Rename CONFIG_PMIC* defines to CONFIG_POWER power:pmic: Rename CONFIG_DIALOG_PMIC defines to CONFIG_DIALOG_POWER arm:goni:pmic: Adjust GONI target platform board to new PMIC framework arm:universal_c210:pmic: Adjust C210 Universal target platform board to new PMIC framework Makefile | 5 +- arch/arm/lib/board.c | 8 + board/davedenx/qong/qong.c | 6 +- board/freescale/mx31pdk/mx31pdk.c | 6 +- board/freescale/mx35pdk/mx35pdk.c | 8 +- board/freescale/mx51evk/mx51evk.c | 6 +- board/freescale/mx53evk/mx53evk.c | 6 +- board/freescale/mx53loco/mx53loco.c | 10 +- board/genesi/mx51_efikamx/efikamx.c | 6 +- board/hale/tt01/tt01.c | 6 +- board/samsung/goni/goni.c | 16 +- board/samsung/trats/trats.c | 247 +++++++++++++++++++- board/samsung/universal_c210/universal.c | 19 +- board/ttcontrol/vision2/vision2.c | 6 +- drivers/misc/Makefile | 7 - drivers/misc/pmic_core.c | 147 ------------ drivers/misc/pmic_i2c.c | 98 -------- drivers/misc/pmic_max8997.c | 43 ---- drivers/power/Makefile | 12 +- drivers/power/battery/Makefile | 47 ++++ drivers/power/battery/bat_trats.c | 94 ++++++++ drivers/power/fuel_gauge/Makefile | 47 ++++ drivers/power/fuel_gauge/fg_max17042.c | 244 +++++++++++++++++++ drivers/power/pmic/Makefile | 49 ++++ drivers/power/pmic/muic_max8997.c | 84 +++++++ drivers/power/pmic/pmic_max8997.c | 117 +++++++++ drivers/{misc => power/pmic}/pmic_max8998.c | 10 +- drivers/power/power_core.c | 230 ++++++++++++++++++ .../{misc/pmic_dialog.c => power/power_dialog.c} | 8 +- drivers/{misc/pmic_fsl.c => power/power_fsl.c} | 16 +- drivers/power/power_i2c.c | 125 ++++++++++ drivers/{misc/pmic_spi.c => power/power_spi.c} | 4 +- drivers/rtc/mc13xxx-rtc.c | 6 +- include/configs/imx31_litekit.h | 6 +- include/configs/mx31ads.h | 6 +- include/configs/mx31pdk.h | 6 +- include/configs/mx35pdk.h | 6 +- include/configs/mx51_efikamx.h | 6 +- include/configs/mx51evk.h | 6 +- include/configs/mx53evk.h | 6 +- include/configs/mx53loco.h | 8 +- include/configs/qong.h | 6 +- include/configs/s5p_goni.h | 6 +- include/configs/s5pc210_universal.h | 6 +- include/configs/trats.h | 14 +- include/configs/tt01.h | 6 +- include/configs/vision2.h | 6 +- include/pmic.h | 72 ------ include/power/battery.h | 38 +++ include/power/fg_battery_cell_params.h | 90 +++++++ include/power/max17042_fg.h | 74 ++++++ include/power/max8997_muic.h | 61 +++++ include/{ => power}/max8997_pmic.h | 26 ++- include/{ => power}/max8998_pmic.h | 0 include/power/pmic.h | 109 +++++++++ include/power/power_chrg.h | 43 ++++ 56 files changed, 1859 insertions(+), 491 deletions(-) delete mode 100644 drivers/misc/pmic_core.c delete mode 100644 drivers/misc/pmic_i2c.c delete mode 100644 drivers/misc/pmic_max8997.c create mode 100644 drivers/power/battery/Makefile create mode 100644 drivers/power/battery/bat_trats.c create mode 100644 drivers/power/fuel_gauge/Makefile create mode 100644 drivers/power/fuel_gauge/fg_max17042.c create mode 100644 drivers/power/pmic/Makefile create mode 100644 drivers/power/pmic/muic_max8997.c create mode 100644 drivers/power/pmic/pmic_max8997.c rename drivers/{misc => power/pmic}/pmic_max8998.c (89%) create mode 100644 drivers/power/power_core.c rename drivers/{misc/pmic_dialog.c => power/power_dialog.c} (89%) rename drivers/{misc/pmic_fsl.c => power/power_fsl.c} (86%) create mode 100644 drivers/power/power_i2c.c rename drivers/{misc/pmic_spi.c => power/power_spi.c} (97%) delete mode 100644 include/pmic.h create mode 100644 include/power/battery.h create mode 100644 include/power/fg_battery_cell_params.h create mode 100644 include/power/max17042_fg.h create mode 100644 include/power/max8997_muic.h rename include/{ => power}/max8997_pmic.h (88%) rename include/{ => power}/max8998_pmic.h (100%) create mode 100644 include/power/pmic.h create mode 100644 include/power/power_chrg.h -- 1.7.2.3

2 37

[U-Boot] [PATCH 1/4] microblaze: Fix compilation warning in ext2_find_next_zero_bit
by Michal Simek 08 Nov '12

08 Nov '12

ext2_find_next_zero_bit must be also static if __swab32 is also static. Warning: include/asm/bitops.h:369:22: warning: '__fswab32' is static but used in inline function 'ext2_find_next_zero_bit' which is not static [enabled by default] Signed-off-by: Michal Simek <monstr(a)monstr.eu> --- arch/microblaze/include/asm/bitops.h | 3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git a/arch/microblaze/include/asm/bitops.h b/arch/microblaze/include/asm/bitops.h index e8c835f..eafa2b5 100644 --- a/arch/microblaze/include/asm/bitops.h +++ b/arch/microblaze/include/asm/bitops.h @@ -319,7 +319,8 @@ extern __inline__ int ext2_test_bit(int nr, const volatile void * addr) #define ext2_find_first_zero_bit(addr, size) \ ext2_find_next_zero_bit((addr), (size), 0) -extern __inline__ unsigned long ext2_find_next_zero_bit(void *addr, unsigned long size, unsigned long offset) +static inline unsigned long ext2_find_next_zero_bit(void *addr, + unsigned long size, unsigned long offset) { unsigned long *p = ((unsigned long *) addr) + (offset >> 5); unsigned long result = offset & ~31UL; -- 1.7.0.4

3 12

[U-Boot] S-o-b entries for bcm2708 SDHCI driver in the downstream R.Pi Linux kernel
by Stephen Warren 08 Nov '12

08 Nov '12

(Sorry for the resend; I sent this to the wrong U-Boot mailing list the first time around) Dom, ddv2005, Gray, I'm trying to get Signed-off-by lines for all code in the Raspberry Pi Linux kernel trees related to the SDHCI driver (or just everything). The list of relevant commits for SDHCI is below. Can you please check Documentation/SubmittingPatches in the kernel source tree, and if you're able to, provide your Signed-of-by lines for all of the patches I list below that were either authored by you, or passed through you (the latter I assume mainly in the case of Dom who applied these to git). Related, do you know who the following person is: Grigori Goronzy <greg@blackbox>; he's the author of at least one patch, but the email address is invalid. Dom, it'd be great if you could obtain the appropriate S-o-b lines from all contributors to your kernel trees, then rebase these branches and add the appropriate S-o-bs to all the patches. If in the future you make sure that all the git metadata and S-o-b entries are present and valid in all the commits (full name and valid email address), we won't have to go back to people and validate this after the fact again. Note: the patch list below is every patch that touches the driver. It may be possible to analyze each patch and find only those that touched the specific subset of code that Oleksandr used for U-Boot. I can give that a try if getting S-o-bs for all the patches proves problematic. Of course, that will then simply defer the issue until someone attempts to upstream the whole downstream Linux SDHCI driver into the upstream Linux kernel... Thanks very much! rpi-3.2.27: > 118e2d3 Allow the number of cycles delay between sdcard peripheral writes to be specified on command line with sdhci-bcm2708.cycle_delay > 7042895 Enable low latency mode by default in sdcard driver. Can be disabled with sdhci-bcm2708.enable_llm=0 > 307d9fb Fix build for non-preempt case. Fix warning > e886b45 Add low-latency mode to sdcard driver. Enable with sdhci-bcm2708.enable_llm=1. Thanks ddv2005. > 2c671b9 Fix spinlock recursion in sdhci-bcm2708.c > 147779e Add module parameter for missing_status quirk. sdhci-bcm2708.missing_status=0 may improve interrupt latency > 31c350f sdhci-bcm2708: use multiblock-type transfers for single blocks > 3a47543 sdhci-bcm2708: use extension FIFO to buffer DMA transfers > 0bf2669 Add sync_after_dma module parameter > f1da243 Use ndelay rather than udelay. Thanks lb Who is lb? > 63aa167 Fix bool/int error, part 2 > d34d49e Fix bool/int error > d7afba1 sdhci-bcm2708: raise DMA sync timeout > 84042ac Allow emmc clock to be specified as command line parameter > 70b125d sdhci-bcm2708: assume 50 MHz eMMC clock > d7156c5 sdhci-bcm2708: add allow_highspeed parameter > 8f82daa sdhci-bcm2708: add additional quirks > 6ab6d02 sdhci-bcm2708: remove custom clock handling > 8390fb8 sdhci-bcm2708: speed up DMA sync > 2d05297 sdcard patch improvements from naren Who is naren? > e9382f7 possible fix for sdcard missing status. Thank naren > 2456b9f Replace #if 1 sections by local configuration #defines (tidy) > 8a93b2a Fix problem with dma that could corrupt sdcard, especially when L2 enabled > 5961574 Fix for mmc timeouts when erasing multiple blocks > 031ee3e Main bcm2708 linux port Are these actually all authored by who git says they are? I notice that the two commits I called out above have a git authorship field of Dom Cobley, even though the patch subject implies someone else wrote them. rpi-3.6.y: > a2386d3 Fix up issues with rebase > f34c715 Allow the number of cycles delay between sdcard peripheral writes to be specified on command line with sdhci-bcm2708.cycle_delay > f0dbf1e Enable low latency mode by default in sdcard driver. Can be disabled with sdhci-bcm2708.enable_llm=0 > da819bb Fix build for non-preempt case. Fix warning > f26afe4 Add low-latency mode to sdcard driver. Enable with sdhci-bcm2708.enable_llm=1. Thanks ddv2005. > 641bd2c Fix spinlock recursion in sdhci-bcm2708.c > ee410b8 Add module parameter for missing_status quirk. sdhci-bcm2708.missing_status=0 may improve interrupt latency > e1ccc32 sdhci-bcm2708: use multiblock-type transfers for single blocks > 34eec60 sdhci-bcm2708: use extension FIFO to buffer DMA transfers > 1245667 Add sync_after_dma module parameter > 0b047fb Use ndelay rather than udelay. Thanks lb > 234c311 Fix bool/int error, part 2 > 27f5be9 Fix bool/int error > ab619f6 sdhci-bcm2708: raise DMA sync timeout > 1b89d0c Allow emmc clock to be specified as command line parameter > 5109353 sdhci-bcm2708: assume 50 MHz eMMC clock > ce8e5cb sdhci-bcm2708: add allow_highspeed parameter > aa86832 sdhci-bcm2708: add additional quirks > e2076c4 sdhci-bcm2708: remove custom clock handling > f354b15 sdhci-bcm2708: speed up DMA sync > 9c9780e sdcard patch improvements from naren > d47f5ca possible fix for sdcard missing status. Thank naren > 1ba126b Replace #if 1 sections by local configuration #defines (tidy) > 7a69d91 Fix problem with dma that could corrupt sdcard, especially when L2 enabled > f5fe324 Fix for mmc timeouts when erasing multiple blocks > f5b930b Main bcm2708 linux port

2 4

Re: [U-Boot] [PATCH 00/46] Enhance spear support
by Stefan Roese 08 Nov '12

08 Nov '12

Hi Vipin, On 11/02/2012 06:39 PM, Vipin Kumar wrote: > This patchset is split in several pieces > - Drivers: This is essentially a few new drivers and fixes in already exiasting > drivers > - Enhancement: This series enhances the already existing spear support > - spear13xx: This is a totally new suppoort for spear13xx architecture Thanks for all these patches. This is a massive patch-set. I'll try to review at least some of them later this week. Please be patient. But this amount of patches does raise a question that I discussed a time ago. If you continue to work on upstream U-Boot, which we greatly appreciate, then its perhaps better to create/add an SPEAr U-Boot custodian responsible for the SPEAr related patches. Reviewing, collecting patches and pushing them to Alberts ARM repository. This would take a bit of work from Albert, since without such an custodian all those SPEAr patches would fall into his area. What do you guys think? Albert, what are your thoughts on this? Vipin, would you "volunteer" to eventually become this SPEAr U-Boot custodian? Or someone else? Just my thought, comments welcome! Thanks, Stefan DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: office(a)denx.de

2 2

[U-Boot] [RFC, PATCH] mmc: prepare SDHCI_HOST_CONTROL for SDMA operation
by Rommel G Custodio 08 Nov '12

08 Nov '12

modification for commit 804c7f422169212e92530e1ddaf74bf1ca9ebfa1 The original patch contained a bug. Using a char with sdhci_readl/sdhci_writel operation. The adjacent bytes to SDHCI_HOST_CONTROL (specially SDHCI_POWER_CONTROL) get mangled and can result in the controller entering an unstable state. This patch also moves the process inside sdhci_init(). Signed-off-by: Rommel G Custodio <sessyargc(a)gmail.com> --- drivers/mmc/sdhci.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/mmc/sdhci.c b/drivers/mmc/sdhci.c index 7845f87..fae2169 100644 --- a/drivers/mmc/sdhci.c +++ b/drivers/mmc/sdhci.c @@ -82,13 +82,6 @@ static int sdhci_transfer_data(struct sdhci_host *host, struct mmc_data *data, unsigned int start_addr) { unsigned int stat, rdy, mask, timeout, block = 0; -#ifdef CONFIG_MMC_SDMA - unsigned char ctrl; - ctrl = sdhci_readl(host, SDHCI_HOST_CONTROL); - ctrl &= ~SDHCI_CTRL_DMA_MASK; - ctrl |= SDHCI_CTRL_SDMA; - sdhci_writel(host, ctrl, SDHCI_HOST_CONTROL); -#endif timeout = 1000000; rdy = SDHCI_INT_SPACE_AVAIL | SDHCI_INT_DATA_AVAIL; @@ -386,6 +379,14 @@ int sdhci_init(struct mmc *mmc) { struct sdhci_host *host = (struct sdhci_host *)mmc->priv; +#ifdef CONFIG_MMC_SDMA + unsigned int ctrl; + ctrl = sdhci_readl(host, SDHCI_HOST_CONTROL); + ctrl &= ~SDHCI_CTRL_DMA_MASK; + ctrl |= SDHCI_CTRL_SDMA; + sdhci_writel(host, ctrl, SDHCI_HOST_CONTROL); +#endif + if ((host->quirks & SDHCI_QUIRK_32BIT_DMA_ADDR) && !aligned_buffer) { aligned_buffer = memalign(8, 512*1024); if (!aligned_buffer) { -- 1.8.0

2 1

[U-Boot] [PATCH 1/3] common: add ifdefs around bouncebuf.c body
by Stephen Warren 08 Nov '12

08 Nov '12

From: Stephen Warren <swarren(a)nvidia.com> If a U-Boot config file enables CONFIG_BOUNCE_BUFFER only for the main U-Boot build and not for the SPL, then config.mk will contain CONFIG_BOUNCE_BUFFER=y, so common/Makefile will build bouncebuf.c for both the SPL and main U-Boot, but config.h won't set CONFIG_BOUNCE_BUFFER for the SPL, so bouncebuf.h will provide static inline functions, which will conflict with the compiled bouncebuf.c. Solve this by guarding the body of bouncebuf.c with the ifdef to avoid conflicts. Signed-off-by: Stephen Warren <swarren(a)nvidia.com> --- This series is based on u-boot/master. I've CC'd the MMC and Tegra maintainers since they'll presumably need to ack the changes in order for these patches to all be applied in one place. Marek seems to be the main MXS MMC maintainer as far as I can tell. common/bouncebuf.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/common/bouncebuf.c b/common/bouncebuf.c index 4f827f8..ffd3c90 100644 --- a/common/bouncebuf.c +++ b/common/bouncebuf.c @@ -27,6 +27,7 @@ #include <errno.h> #include <bouncebuf.h> +#ifdef CONFIG_BOUNCE_BUFFER static int addr_aligned(void *data, size_t len) { const ulong align_mask = ARCH_DMA_MINALIGN - 1; @@ -90,3 +91,4 @@ int bounce_buffer_stop(void **data, size_t len, void **backup, uint8_t flags) return 0; } +#endif -- 1.7.0.4

3 19

[U-Boot] [PATCH] USB: add arrow key support to usb_kbd
by Allen Martin 08 Nov '12

08 Nov '12

Check for scancodes for arrow keys and map them to ^F/^B, ^N/^P. Control characters are used instead of ANSI sequence because the queueing code in usb_kbd doesn't handle the data increase when one keypress generates 3 keycodes. The real fix is to convert this driver to use the input subsystem and queue, but this allows arrow keys to work until this driver is converted. Signed-off-by: Allen Martin <amartin(a)nvidia.com> --- common/usb_kbd.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/common/usb_kbd.c b/common/usb_kbd.c index 24467ce..4efbcfe 100644 --- a/common/usb_kbd.c +++ b/common/usb_kbd.c @@ -94,6 +94,15 @@ static const unsigned char usb_kbd_num_keypad[] = { }; /* + * map arrow keys to ^F/^B ^N/^P, can't really use the proper + * ANSI sequence for arrow keys because the queuing code breaks + * when a single keypress expands to 3 queue elements + */ +static const unsigned char usb_kbd_arrow[] = { + 0x6, 0x2, 0xe, 0x10 +}; + +/* * NOTE: It's important for the NUM, CAPS, SCROLL-lock bits to be in this * order. See usb_kbd_setled() function! */ @@ -224,6 +233,10 @@ static int usb_kbd_translate(struct usb_kbd_pdata *data, unsigned char scancode, keycode = usb_kbd_numkey[scancode - 0x1e]; } + /* Arrow keys */ + if ((scancode >= 0x4f) && (scancode <= 0x52)) + keycode = usb_kbd_arrow[scancode - 0x4f]; + /* Numeric keypad */ if ((scancode >= 0x54) && (scancode <= 0x67)) keycode = usb_kbd_num_keypad[scancode - 0x54]; -- 1.7.10.4

4 12

[U-Boot] [PATCH] powerpc/mpc8xxx: Fix compiling error caused in DDR driver
by York Sun 08 Nov '12

08 Nov '12

mpc86xx platforms should use CONFIG_SYS_MPC86xx_DDR2_ADDR in utils.c if applicable. Signed-off-by: York Sun <yorksun(a)freescale.com> --- arch/powerpc/cpu/mpc8xxx/ddr/util.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/powerpc/cpu/mpc8xxx/ddr/util.c b/arch/powerpc/cpu/mpc8xxx/ddr/util.c index 940ffff..9a07d9d 100644 --- a/arch/powerpc/cpu/mpc8xxx/ddr/util.c +++ b/arch/powerpc/cpu/mpc8xxx/ddr/util.c @@ -152,7 +152,11 @@ void board_add_ram_info(int use_default) #if CONFIG_NUM_DDR_CONTROLLERS >= 2 if (!(sdram_cfg & SDRAM_CFG_MEM_EN)) { +#ifdef CONFIG_MPC86xx + ddr = (void __iomem *)CONFIG_SYS_MPC86xx_DDR2_ADDR; +#else ddr = (void __iomem *)CONFIG_SYS_MPC85xx_DDR2_ADDR; +#endif sdram_cfg = in_be32(&ddr->sdram_cfg); } #endif -- 1.7.9.5

2 2

[U-Boot] [PATCH] patman: Issue empty change logs for unchanged patches
by Simon Glass 08 Nov '12

08 Nov '12

Often a particular patch may change only for some versions of a series. For versions where there is no change, issue a change log indicating that (for example 'Changes in v4: None'). For such lines, don't add a blank line afterwards, to conserve space. Use list.insert() instead of list = [item] + list. Signed-off-by: Simon Glass <sjg(a)chromium.org> --- tools/patman/series.py | 21 +++++++++++++-------- 1 files changed, 13 insertions(+), 8 deletions(-) diff --git a/tools/patman/series.py b/tools/patman/series.py index a283a2d..d2971f4 100644 --- a/tools/patman/series.py +++ b/tools/patman/series.py @@ -145,10 +145,11 @@ class Series(dict): Return: The change log as a list of strings, one per line - Changes in v2: + Changes in v4: - Jog the dial back closer to the widget - Changes in v1: + Changes in v3: None + Changes in v2: - Fix the widget - Jog the dial @@ -162,12 +163,16 @@ class Series(dict): if commit and this_commit != commit: continue out.append(text) - if out: - out = ['Changes in v%d:' % change] + out - if need_blank: - out = [''] + out - final += out - need_blank = True + line = 'Changes in v%d:' % change + have_changes = len(out) > 0 + if have_changes: + out.insert(0, line) + else: + out = [line + ' None'] + if need_blank: + out.insert(0, '') + final += out + need_blank = have_changes if self.changes: final.append('') return final -- 1.7.7.3

3 2