June 2016 - U-Boot

[U-Boot] [PATCH 2 00/22] DM: Cmd: GPIO/LED/STM32/CLK: provide command-line support for device-tree configured gpios and leds
by Benjamin Tietz 01 Aug '16

01 Aug '16

This series begins to provide device-tree support on stm32 devices, starting with a the stack of a simple clock-driver (at least enabling/disabling peripheral clocks), the gpio driver and leds. As the current led command-line interface isn't aware of any device-tree configured led, the command gets rewritten and extended for device-tree LEDs on the way. These changes are architecture indipendent. To accomplish these changes the led-uclass driver had to be extended, too. Changes in v2: * more verbose commit messages * correct clock calculation in stm32_clk * minor adjustments --- Benjamin Tietz (22): stm32: gpio: fix otype access stm32: gpio_direction_output: make sure, output is set to push-pull stm32: gpio_get_value: always return 0 or 1 stm32f429-discovery: config: enable status leds Cmd: led: provide a selector in kconfig DTS: stm32f429: provide device-tree files (from linux kernel) clock-uclass: allow disabling a peripheral clock STM32: clock: provide dts-accessible clock driver DTS: STM32f429: add gpio-banks STM32: gpio: group SOC-specific code to one ifdef/elif construct GPIO: STM32: make DTS-aware STM32F429-discovery: led: disable board-specific code, if DM is selected GPIO/LED: make more robust, if STATUS_LED isn't selected Cmd: LED: rewrite to prepare non-static access DTS: STM32F429-disco: add board leds and enable rcc LED: add function to retrieve a device's label LED: provide function to count and get all (DM-)LEDs cmd: LED: be aware of DTS-configured leds LED: provide functionality to get led status LED: GPIO: provide get_on() op LED: provide toggling interface Cmd: LED: make DM-leds toggle arch/arm/dts/Makefile | 2 arch/arm/dts/armv7-m.dtsi | 24 ++ arch/arm/dts/stm32429i-eval.dts | 75 ++++++ arch/arm/dts/stm32f429-disco.dts | 97 ++++++++ arch/arm/dts/stm32f429.dtsi | 282 +++++++++++++++++++++++ board/st/stm32f429-discovery/Makefile | 3 cmd/Kconfig | 4 cmd/led.c | 401 ++++++++++++++++++++++++--------- drivers/clk/Kconfig | 4 drivers/clk/Makefile | 1 drivers/clk/clk-uclass.c | 10 + drivers/clk/clk_stm32.c | 112 +++++++++ drivers/gpio/stm32_gpio.c | 202 ++++++++++++++--- drivers/led/led-uclass.c | 83 +++++++ drivers/led/led_gpio.c | 11 + drivers/misc/gpio_led.c | 4 drivers/misc/status_led.c | 2 include/clk.h | 18 + include/configs/stm32f429-discovery.h | 14 + include/led.h | 65 +++++ include/status_led.h | 4 21 files changed, 1277 insertions(+), 141 deletions(-) create mode 100644 arch/arm/dts/armv7-m.dtsi create mode 100644 arch/arm/dts/stm32429i-eval.dts create mode 100644 arch/arm/dts/stm32f429-disco.dts create mode 100644 arch/arm/dts/stm32f429.dtsi create mode 100644 drivers/clk/clk_stm32.c -- best regards Benjamin Tietz

5 35

[U-Boot] [PATCH 00/27] dm: mmc: Add driver-model support for MMC operations
by Simon Glass 01 Aug '16

01 Aug '16

At present MMC does not use driver model for its operations. It uses its own structure and passes a struct mmc instead of a struct udevice. This series addresses this by adding driver-model operations for MMC. The conversion process is also started, with patches for rockchip, zynq and qualcomm. The zynq patches are a starting point only and need more work. Simon Glass (27): dm: mmc: dwmmc: Add comments to the dwmmc setup functions rockchip: Use 'select' instead of defaults in Kconfig mmc: Add function declarations for mmc_bread() and mmc_switch_part() dm: mmc: Move CONFIG_BLK code into the mmc uclass dm: mmc: Move non-CONFIG_BLK code into mmc_legacy.c mmc: Move MMC boot code into its own file dm: mmc: rockchip: Support only CONFIG_BLK mmc: Move tracing code into separate functions rockchip: Disable CONFIG_SDHCI dm: mmc: Add a way to use driver model for MMC operations dm: mmc: dwmmc: Support CONFIG_DM_MMC_OPS dm: mmc: rockchip: Enable CONFIG_DM_MMC_OPS for all boards rockchip: Add MAINTAINER files for kylin_rk3036, evb_rk3036 dm: sandbox: Convert to use CONFIG_CMD_MMC_OPS dm: mmc: sdhci: Refactor configuration setup to support DM dm: mmc: sdhci: Support CONFIG_BLK and CONFIG_DM_MMC_OPS dm: mmc: msm_sdhci: Support CONFIG_BLK and CONFIG_DM_MMC_OPS dm: mmc: Move dragonboard410c to use CONFIG_BLK and CONFIG_DM_MMC_OPS dm: mmc: msmsdhic: Drop old MMC code dm: spl: mmc: Support CONFIG_BLK in SPL MMC dm: dfu: mmc: Support CONFIG_BLK in DFU for MMC net: phy: marvell: Add a missing errno.h header zynq: Increase the early malloc() size dm: zynq: usb: Convert to CONFIG_DM_USB dm: mmc: zynq: Convert zynq to use driver model for MMC dm: mmc: Enable DM_MMC_OPS by default with DM_MMC dm: blk: Enable CONFIG_BLK if DM_MMC is enabled arch/Kconfig | 1 + arch/arm/Kconfig | 18 ++ arch/arm/cpu/armv8/zynqmp/Kconfig | 4 + arch/arm/mach-rockchip/Kconfig | 27 --- arch/arm/mach-zynq/Kconfig | 3 + board/evb_rk3036/evb_rk3036/MAINTAINERS | 6 + board/kylin/kylin_rk3036/MAINTAINERS | 6 + common/spl/spl_mmc.c | 6 +- configs/dragonboard410c_defconfig | 2 + configs/sandbox_defconfig | 4 +- drivers/block/Kconfig | 1 + drivers/dfu/dfu_mmc.c | 11 +- drivers/mmc/Kconfig | 12 +- drivers/mmc/Makefile | 3 + drivers/mmc/dw_mmc.c | 33 +++ drivers/mmc/mmc-uclass.c | 146 +++++++++++++ drivers/mmc/mmc.c | 371 +++++--------------------------- drivers/mmc/mmc_boot.c | 131 +++++++++++ drivers/mmc/mmc_legacy.c | 91 ++++++++ drivers/mmc/mmc_private.h | 47 ++++ drivers/mmc/msm_sdhci.c | 38 +++- drivers/mmc/rockchip_dw_mmc.c | 14 +- drivers/mmc/sandbox_mmc.c | 17 +- drivers/mmc/sdhci.c | 147 ++++++++----- drivers/mmc/zynq_sdhci.c | 39 +++- drivers/net/phy/marvell.c | 1 + drivers/usb/host/ehci-zynq.c | 102 +++++---- include/configs/rk3036_common.h | 1 - include/configs/rk3288_common.h | 1 - include/dwmmc.h | 73 +++++++ include/mmc.h | 66 +++++- include/sdhci.h | 80 +++++++ 32 files changed, 1006 insertions(+), 496 deletions(-) create mode 100644 drivers/mmc/mmc_boot.c -- 2.8.0.rc3.226.g39d4020

6 57

[U-Boot] [PATCH] imx: ventana: re-enable late board info display
by Tim Harvey 28 Jul '16

28 Jul '16

3b1f681131149b5f62602f582a7e60b0185a2a49 caused a regression that removes board info dispaly for Gateworks Ventana boards because it made the invalid assumption that CONFIG_DISPLAY_BOARDINFO_LATE was the same thing as CONFIG_DISPLAY_BOARDINFO. Ventana needs to call show_board_info in late init because we need to have the i2c eeprom based model info. Re-define CONFIG_DISPLAY_BOARDINFO_LATE to allow that to happen. Cc: Peter Robinson <pbrobinson(a)gmail.com> Signed-off-by: Tim Harvey <tharvey(a)gateworks.com> --- include/configs/gw_ventana.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/configs/gw_ventana.h b/include/configs/gw_ventana.h index 4acfca6..343f3c5 100644 --- a/include/configs/gw_ventana.h +++ b/include/configs/gw_ventana.h @@ -33,6 +33,8 @@ #include "imx6_spl.h" /* common IMX6 SPL configuration */ #include "mx6_common.h" #undef CONFIG_SPL_EXT_SUPPORT +#undef CONFIG_DISPLAY_BOARDINFO +#define CONFIG_DISPLAY_BOARDINFO_LATE #define CONFIG_MACH_TYPE 4520 /* Gateworks Ventana Platform */ -- 1.9.1

2 1

[U-Boot] [PATCH 0/2] imx: ventana: add PCIe dt fixups
by Tim Harvey 28 Jul '16

28 Jul '16

There are a couple of device-tree fixups that need to be applied to the PCIe node in the devicetree which is a set of nested nodes based on bus topology. This series adds the ability to track the PCI devices enumerated and build a nested topology which is supported on newer kernels. Tim Harvey (2): imx: ventana: add dt fixup for GW16082 irq mapping imx: ventana: add dt fixup for eth1 mac-address board/gateworks/gw_ventana/gw_ventana.c | 263 ++++++++++++++++++++++++++++++++ 1 file changed, 263 insertions(+) -- 1.9.1

2 4

[U-Boot] [PATCH 0/2] imx: ventana: allow pci to be disabled via env
by Tim Harvey 28 Jul '16

28 Jul '16

This series adds the ability for PCI to be disabled via an env var then disables it by default for the Gateworks Ventana boards because of an adverse affect enabling PCI in the bootloader has on the kernel. The user may want to remove the env var that disables it if they require features that need PCI enabled in the bootloader. Tim Harvey (2): pci: allow disabling of pci init/enum via env imx: ventana: default pci to disabled drivers/pci/pci.c | 4 ++++ include/configs/gw_ventana.h | 1 + 2 files changed, 5 insertions(+) -- 1.9.1

2 4

[U-Boot] [PATCH 1/4] mmc: fsl_esdhc: don't set XFERTYP_RSPTYP_48_BUSY for CMD with busy response
by Yangbo Lu 28 Jul '16

28 Jul '16

For CMD with busy response, the eSDHC driver would poll DAT0 until CMD completion rather than polling IRQSTAT. So, don't set XFERTYP_RSPTYP_48_BUSY to avoid interrupts (DTOE or TC) in IRQSTAT. Signed-off-by: Yangbo Lu <yangbo.lu(a)nxp.com> --- drivers/mmc/fsl_esdhc.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/mmc/fsl_esdhc.c b/drivers/mmc/fsl_esdhc.c index 3acf9e8..b06dd69 100644 --- a/drivers/mmc/fsl_esdhc.c +++ b/drivers/mmc/fsl_esdhc.c @@ -126,8 +126,16 @@ static uint esdhc_xfertyp(struct mmc_cmd *cmd, struct mmc_data *data) xfertyp |= XFERTYP_CICEN; if (cmd->resp_type & MMC_RSP_136) xfertyp |= XFERTYP_RSPTYP_136; - else if (cmd->resp_type & MMC_RSP_BUSY) - xfertyp |= XFERTYP_RSPTYP_48_BUSY; + /* + * For CMD with busy response, the eSDHC driver would poll DAT0 + * until CMD completion rather than polling IRQSTAT. So, don't + * set XFERTYP_RSPTYP_48_BUSY to avoid interrupts (DTOE or TC) + * in IRQSTAT. + * + * Remove: + * else if (cmd->resp_type & MMC_RSP_BUSY) + * xfertyp |= XFERTYP_RSPTYP_48_BUSY; + */ else if (cmd->resp_type & MMC_RSP_PRESENT) xfertyp |= XFERTYP_RSPTYP_48; -- 2.1.0.27.g96db324

4 12

[U-Boot] [PATCH v2 0/4] armv8/fsl-layerscape: add dwc3 gadget driver support
by Rajat Srivastava 27 Jul '16

27 Jul '16

From: Rajesh Bhagat <rajesh.bhagat(a)nxp.com> Adds support for dwc3 gadget driver for ARM v8 platform of layerscape series. Currently this patch has been tested for LS1043A Rajat Srivastava (3): usb: dwc3: Add helper functions to enable snooping and burst settings armv8/fsl-layerscape: add dwc3 gadget driver support armv8/fsl-layerscape: enable dwc3 gadget driver support Rajesh Bhagat (1): usb: ums: support multiple controllers using controller_index arch/arm/cpu/armv8/fsl-layerscape/soc.c | 93 ++++++++++++++++++++++ .../include/asm/arch-fsl-layerscape/immap_lsch2.h | 6 ++ cmd/usb_mass_storage.c | 2 +- drivers/usb/dwc3/core.c | 45 +++++++++++ drivers/usb/dwc3/core.h | 7 ++ drivers/usb/gadget/f_mass_storage.c | 11 ++- include/configs/ls1043aqds.h | 15 ++++ include/usb_mass_storage.h | 2 +- 8 files changed, 177 insertions(+), 4 deletions(-) -- 2.6.2.198.g614a2ac

6 21

[U-Boot] [PATCH 1/2] board: ls1043ardb: move sec_init to board_init
by Aneesh Bansal 27 Jul '16

27 Jul '16

sec_init() which was earlier called in misc_init_r() is now done in board_init() before PPA init as SEC block will be used during PPA image validation. Signed-off-by: Aneesh Bansal <aneesh.bansal(a)nxp.com> --- The patchset is dependent on http://patchwork.ozlabs.org/patch/571339/ board/freescale/ls1043ardb/ls1043ardb.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) diff --git a/board/freescale/ls1043ardb/ls1043ardb.c b/board/freescale/ls1043ardb/ls1043ardb.c index eff09aa..aa2c962 100644 --- a/board/freescale/ls1043ardb/ls1043ardb.c +++ b/board/freescale/ls1043ardb/ls1043ardb.c @@ -117,6 +117,23 @@ int board_init(void) enable_layerscape_ns_access(); #endif +#ifdef CONFIG_SECURE_BOOT + /* In case of Secure Boot, the IBR configures the SMMU + * to allow only Secure transactions. + * SMMU must be reset in bypass mode. + * Set the ClientPD bit and Clear the USFCFG Bit + */ + u32 val; + val = (in_le32(SMMU_SCR0) | SCR0_CLIENTPD_MASK) & ~(SCR0_USFCFG_MASK); + out_le32(SMMU_SCR0, val); + val = (in_le32(SMMU_NSCR0) | SCR0_CLIENTPD_MASK) & ~(SCR0_USFCFG_MASK); + out_le32(SMMU_NSCR0, val); +#endif + +#ifdef CONFIG_FSL_CAAM + sec_init(); +#endif + #ifdef CONFIG_FSL_LS_PPA ppa_init_pre(&ppa_entry); @@ -136,21 +153,6 @@ int config_board_mux(void) int misc_init_r(void) { config_board_mux(); -#ifdef CONFIG_SECURE_BOOT - /* In case of Secure Boot, the IBR configures the SMMU - * to allow only Secure transactions. - * SMMU must be reset in bypass mode. - * Set the ClientPD bit and Clear the USFCFG Bit - */ - u32 val; - val = (in_le32(SMMU_SCR0) | SCR0_CLIENTPD_MASK) & ~(SCR0_USFCFG_MASK); - out_le32(SMMU_SCR0, val); - val = (in_le32(SMMU_NSCR0) | SCR0_CLIENTPD_MASK) & ~(SCR0_USFCFG_MASK); - out_le32(SMMU_NSCR0, val); -#endif -#ifdef CONFIG_FSL_CAAM - return sec_init(); -#endif return 0; } #endif -- 1.8.1.4

2 2

[U-Boot] [RFC PATCH] cmd: gpt: add - partition size parsing
by Michael Trimarchi 27 Jul '16

27 Jul '16

This patch try to parse name=userdata,size=-,uuid=${uuid_gpt_userdata}; gpt mmc write 0 $partitions gpt mmc verify 0 $partitions Signed-off-by: Michael Trimarchi <michael(a)amarulasolutions.com> --- cmd/gpt.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/cmd/gpt.c b/cmd/gpt.c index 8ffaef3..3d9706b 100644 --- a/cmd/gpt.c +++ b/cmd/gpt.c @@ -181,6 +181,7 @@ static int set_gpt_info(struct blk_desc *dev_desc, disk_partition_t *parts; int errno = 0; uint64_t size_ll, start_ll; + lbaint_t offset = 0; debug("%s: lba num: 0x%x %d\n", __func__, (unsigned int)dev_desc->lba, (unsigned int)dev_desc->lba); @@ -296,8 +297,14 @@ static int set_gpt_info(struct blk_desc *dev_desc, } if (extract_env(val, &p)) p = val; - size_ll = ustrtoull(p, &p, 0); - parts[i].size = lldiv(size_ll, dev_desc->blksz); + if ((strcmp(p, "-") == 0)) { + /* remove first usable lba and last block */ + parts[i].size = dev_desc->lba - 34 - 1 - offset; + } else { + size_ll = ustrtoull(p, &p, 0); + parts[i].size = lldiv(size_ll, dev_desc->blksz); + } + free(val); /* start address */ @@ -310,6 +317,8 @@ static int set_gpt_info(struct blk_desc *dev_desc, free(val); } + offset += parts[i].size + parts[i].start; + /* bootable */ if (found_key(tok, "bootable")) parts[i].bootable = 1; -- 2.8.3

4 4

[U-Boot] [PATCH] spi: zynq_spi: Fix infinite looping while xfer
by Lad, Prabhakar 27 Jul '16

27 Jul '16

During spi transfer, for example: sspi 1:1.0 8 ff the rx_len values will be: rx_len = 0 rx_len = 4294967295 This caused a busy looping during xfer, this patch fixes it by adding a check while reading the rx fifo Signed-off-by: Lad, Prabhakar <Prabhakar.Lad(a)racelogic.co.uk> Cc: Michal Simek <michal.simek(a)xilinx.com> Cc: Siva Durga Prasad Paladugu <sivadur(a)xilinx.com> Cc: Jagan Teki <jteki(a)openedev.com> --- drivers/spi/zynq_spi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/spi/zynq_spi.c b/drivers/spi/zynq_spi.c index 09ae1be..dd3de27 100644 --- a/drivers/spi/zynq_spi.c +++ b/drivers/spi/zynq_spi.c @@ -230,7 +230,7 @@ static int zynq_spi_xfer(struct udevice *dev, unsigned int bitlen, /* Read the data from RX FIFO */ status = readl(&regs->isr); - while (status & ZYNQ_SPI_IXR_RXNEMPTY_MASK) { + while ((status & ZYNQ_SPI_IXR_RXNEMPTY_MASK) && rx_len) { buf = readl(&regs->rxdr); if (rx_buf) *rx_buf++ = buf; -- 2.1.4

2 7