August 2017 - U-Boot

[U-Boot] [PATCH] rockchip: puma-rk3399: remove redudance code
by Kever Yang 06 Aug '17

06 Aug '17

Some code are duplicate, remove one of them Signed-off-by: Kever Yang <kever.yang(a)rock-chips.com> --- board/theobroma-systems/puma_rk3399/puma-rk3399.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/board/theobroma-systems/puma_rk3399/puma-rk3399.c b/board/theobroma-systems/puma_rk3399/puma-rk3399.c index 36e9cd7..5375247 100644 --- a/board/theobroma-systems/puma_rk3399/puma-rk3399.c +++ b/board/theobroma-systems/puma_rk3399/puma-rk3399.c @@ -9,22 +9,16 @@ #include <ram.h> #include <dm/pinctrl.h> #include <dm/uclass-internal.h> -#include <misc.h> #include <asm/setup.h> #include <asm/arch/periph.h> #include <power/regulator.h> #include <u-boot/sha256.h> -#define RK3399_CPUID_OFF 0x7 -#define RK3399_CPUID_LEN 0x10 - DECLARE_GLOBAL_DATA_PTR; #define RK3399_CPUID_OFF 0x7 #define RK3399_CPUID_LEN 0x10 -DECLARE_GLOBAL_DATA_PTR; - int board_init(void) { struct udevice *pinctrl, *regulator; -- 1.9.1

2 1

[U-Boot] [PATCH 0/2] Fix remaining errors and warning on u-boot-usb tree
by patrice.chotard＠st.com 06 Aug '17

06 Aug '17

From: Patrice Chotard <patrice.chotard(a)st.com> This series fixes remaining compilation errors for poplar platform and warning for dra7xx_evm / dra7xx_hs_evm platforms Patrice Chotard (2): usb: host: xhci-dxc3: fix compilation warnings clk: fix compilation errors for poplar platform drivers/clk/clk-uclass.c | 47 ++++++++++++++++++++++---------------------- drivers/usb/host/xhci-dwc3.c | 4 ++-- include/clk.h | 35 ++++++++++++++++++++------------- 3 files changed, 47 insertions(+), 39 deletions(-) -- 1.9.1

3 5

[U-Boot] [PATCH v2] rockchip: rk3288: Add reset reason detection
by Wadim Egorov 06 Aug '17

06 Aug '17

Sometimes it's helpful to know the reset reason caused in the SoC. Add reset reason detection for the RK3288 SoC. This will set an environemt variable which represents the reset reason. Signed-off-by: Wadim Egorov <w.egorov(a)phytec.de> Acked-by: Philipp Tomsich <philipp.tomsich(a)theobroma-systems.com> --- Changes in v2: - Added Acked-by: Philipp Tomsich <philipp.tomsich(a)theobroma-systems.com> - Clear cru_glb_rst_st - POR if cru_glb_rst_st is 0 - RST if 1st or 2nd bit is set Can be tested with mw 0xff7601b0 0xfffffdb9 mw 0xff7601b4 0xffffeca8 --- arch/arm/mach-rockchip/rk3288-board.c | 35 +++++++++++++++++++++++++++++++++++ 1 file changed, 35 insertions(+) diff --git a/arch/arm/mach-rockchip/rk3288-board.c b/arch/arm/mach-rockchip/rk3288-board.c index 613967c..da1995c 100644 --- a/arch/arm/mach-rockchip/rk3288-board.c +++ b/arch/arm/mach-rockchip/rk3288-board.c @@ -11,6 +11,7 @@ #include <syscon.h> #include <asm/io.h> #include <asm/arch/clock.h> +#include <asm/arch/cru_rk3288.h> #include <asm/arch/periph.h> #include <asm/arch/pmu_rk3288.h> #include <asm/arch/qos_rk3288.h> @@ -70,10 +71,44 @@ int rk3288_qos_init(void) return 0; } +static void rk3288_detect_reset_reason(void) +{ + struct rk3288_cru *cru = rockchip_get_cru(); + if (IS_ERR(cru)) + return; + + switch (cru->cru_glb_rst_st) { + case 0: + setenv("reset_reason", "POR"); + break; + case (1 << 0): + case (1 << 1): + setenv("reset_reason", "RST"); + break; + case (1 << 2): + case (1 << 3): + setenv("reset_reason", "THERMAL"); + break; + case (1 << 4): + case (1 << 5): + setenv("reset_reason", "WDOG"); + break; + default: + setenv("reset_reason", "unknown reset"); + } + + /* + * Clear cru_glb_rst_st, so we can determine the last reset cause + * for following resets. + */ + rk_clrreg(&cru->cru_glb_rst_st, 0x3f); +} + int board_late_init(void) { setup_boot_mode(); rk3288_qos_init(); + rk3288_detect_reset_reason(); return rk_board_late_init(); } -- 1.9.1

3 2

[U-Boot] [PATCH v8 0/6] Extend xhci-dwc3
by patrice.chotard＠st.com 06 Aug '17

06 Aug '17

From: Patrice Chotard <patrice.chotard(a)st.com> This series extend xhci-dwc3.c drivers by : _ converting it to DM model, _ adding dual role mode support from DT _ adding new generic_phy_valid() method in PHY framework _ adding support of generic PHY framework v8: _ Add CONFIG_DM_USB flag in path 1 to fix compilation issue for platforms which uses xhci-dwc3 driver and doesn't support yet driver model _ add PHY stub functions in patch 5 to fix compilation issue for platforms which uses xhci-dwc3 driver and doesn't support generic PHY framework _ fix bisection issue with patch 1 reported by Łukasz Majewski, move struct xhci_dwc3_platdata in patch 6 to fix it. v7: _ add Reviewed-by in patch 1 _ declare bool generic_phy_valid(struct phy *phy) as static inline in patch 4 v6: _ remove useless struct xhci_dwc3 in patch 1 v5: _ rebase on top of dm/master in order to use the last livetree update _ replace dev_get_addr() by devfdt_get_addr() in patch 1 v4: _ set phy->dev to NULL in case of generic_phy_get_by_index() v3: _ introduce generic_phy_valid() method _ add Reviewed-by v2: _ use dev_get_addr() in PATCH 1 and removed useless piece of code Patrice Chotard (6): usb: host: xhci-dwc3: Convert driver to DM usb: host: xhci-dwc3: Add dual role mode support from DT drivers: phy: Set phy->dev to NULL when generic_phy_get_by_index() fails drivers: phy: add generic_phy_valid() method dm: phy: add PHY stub function usb: host: xhci-dwc3: Add generic PHY support drivers/phy/phy-uclass.c | 1 + drivers/usb/host/xhci-dwc3.c | 89 ++++++++++++++++++++++++++++++++++++++++++++ include/generic-phy.h | 53 ++++++++++++++++++++++++++ 3 files changed, 143 insertions(+) -- 1.9.1

3 10

[U-Boot] [PATCH v2 0/3] efi_loader: OpenProtocol, CloseProtocol
by Heinrich Schuchardt 05 Aug '17

05 Aug '17

The implementation of OpenProtocol was incomplete up to now. CloseProtocol and OpenProtocolInformation were not implemented. The first patch prints a debug message with the protocol GUID in OpenProtocol. The second patch adds a table of efi_open_protocol_info_entry to keep track of locks to each protocol entry of each handle. This table is maintained in OpenProtocol and CloseProtocol. The third patch implements OpenProtocolInformation which returns said table. --- v2 Fixed typo in 1st patch which is resubmitted. Added 2nd and 3rd patch. --- Heinrich Schuchardt (3): efi_loader: write protocol GUID in OpenProtocol efi_loader: open_info in OpenProtocol, CloseProtocol efi_loader: implement OpenProtocolInformation include/efi_loader.h | 15 +++ lib/efi_loader/efi_boottime.c | 257 ++++++++++++++++++++++++++++++++++++++++-- 2 files changed, 262 insertions(+), 10 deletions(-) -- 2.11.0

2 7

[U-Boot] Booting Wandboard through USB
by Vincent Prince 05 Aug '17

05 Aug '17

following: - https://lists.denx.de/pipermail/u-boot/2015-May/215573.html - https://lists.denx.de/pipermail/u-boot/2015-June/215606.html Hi everyone, I'd like to know if something is going on regarding this topic ? I managed to flash a yocto BSP with SPL with USB+serial with Fabio help, but I found it a bit painful, and hard to automate. - Go into serial console to reboot, interrupt uboot, and type bmode usb. - Quit serial console - Launch imx_usb_loader, - Send u-boot.bin with ymodem (teraterm on windows, kermit on linux) - Go back to serial console, interrupt u-boot - Mount Usb OTG (ums 1 mmc 0) - Flash wic.gz image - Reset u-boot. What is a bit annoying is the use of ymodem to load u-boot.bin. Anybody has a cleaner solution ? Thanks

4 4

[U-Boot] [PATCH 00/41] Kconfig: Convert remaining commands to Kconfig
by Simon Glass 05 Aug '17

05 Aug '17

4 87

[U-Boot] [PATCH] dm: scsi: fix callto function with wrong signature
by Philipp Tomsich 05 Aug '17

05 Aug '17

buildman fails for the qemu-x86 with the following error: drivers/ata/ahci-pci.c: In function 'ahci_pci_probe': drivers/ata/ahci-pci.c:21:9: error: too few arguments to function 'ahci_probe_scsi' return ahci_probe_scsi(dev); ^ In file included from drivers/ata/ahci-pci.c:8:0: include/ahci.h:224:5: note: declared here int ahci_probe_scsi(struct udevice *ahci_dev, ulong base); This changes the function call to call ahci_probe_scsi_pci (which matches the signature and probes for a controller via on a PCI bus). Signed-off-by: Philipp Tomsich <philipp.tomsich(a)theobroma-systems.com> --- drivers/ata/ahci-pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/ahci-pci.c b/drivers/ata/ahci-pci.c index f46fad8..5a45edc 100644 --- a/drivers/ata/ahci-pci.c +++ b/drivers/ata/ahci-pci.c @@ -18,7 +18,7 @@ static int ahci_pci_bind(struct udevice *dev) static int ahci_pci_probe(struct udevice *dev) { - return ahci_probe_scsi(dev); + return ahci_probe_scsi_pci(dev); } static const struct udevice_id ahci_pci_ids[] = { -- 2.1.4

2 1

[U-Boot] Issues about u-boot tools (mkimage and dumpimage)
by Sun,Mingshen 05 Aug '17

05 Aug '17

Hi all, I am working on a project related to secure boot. We are using NXP’s iMX7d development board. I am using `make –j32 V=1` to verbosely print build commands. The last command for making u-boot is `./tools/mkimage -n board/freescale/mx7dsabresd/imximage.cfg.cfgtmp -T imximage -e 0x87800000 -d u-boot-dtb.bin u-boot-dtb.imx`. Then it will print out the image header imformation like this: ``` Image Type: Freescale IMX Boot Image Image Ver: 2 (i.MX53/6/7 compatible) Mode: DCD Data Size: 561152 Bytes = 548.00 KiB = 0.54 MiB Load Address: 877ff420 Entry Point: 87800000 HAB Blocks: 877ff400 00000000 00086c00 DCD Blocks: 00910000 0000002c 000001b4 ``` The last two lines about “HAB” and “DCD” are special header fields for this type of board because of secure boot (i.e., high assurance boot). I’d like to later examine the values of HAB Blocks. I found `dumpimage` tools can print the header of built image. However, it will report segmentation fault. After spending some time to understand the code. I found it’s caused by `imximage_check_params()` function which requires a non-null `imagename` in `prams`. I tried to fix this issue. However, the dumpimage tool still cannot dump “HAB Blocks” information in the header. The reason is that the `print_header()` functions of `imximage` type needs a configuration file which located in "board/freescale/mx7dsabresd/imximage.cfg.cfgtmp" for my board. Again, I tried to fix it, but failed. Because the `print_header()` function only accept one parameter which is header and this is a unified function among all types of board. I cannot change the definition of it. In a nutshell, there are two issues with the code related with imximage type: 1. The mkimage tool *misuses* “-n” (i.e., image name) flag to pass an image configuration file path. 2. The dumpimage tool cannot dump the “HAB” header information due to the *lack* of the imx image configuration file. Here is my temporary workaround patch for imx board: ``` diff --git a/tools/dumpimage.c b/tools/dumpimage.c index 75a5d47..ad67f22 100644 --- a/tools/dumpimage.c +++ b/tools/dumpimage.c @@ -15,6 +15,7 @@ static void usage(void); /* parameters initialized by core will be used by the image type code */ static struct image_tool_params params = { .type = IH_TYPE_KERNEL, + .imagename = "board/freescale/mx7dsabresd/imximage.cfg.cfgtmp", }; /* diff --git a/tools/imximage.c b/tools/imximage.c index eb7e682..e495345 100644 --- a/tools/imximage.c +++ b/tools/imximage.c @@ -810,6 +810,21 @@ static void imximage_print_header(const void *ptr) struct imx_header *imx_hdr = (struct imx_header *) ptr; uint32_t version = detect_imximage_version(imx_hdr); + memset(&imximage_header, 0, sizeof(imximage_header)); + /* + * In order to not change the old imx cfg file + * by adding VERSION command into it, here need + * set up function ptr group to V1 by default. + */ + imximage_version = IMXIMAGE_V1; + /* Be able to detect if the cfg file has no BOOT_FROM tag */ + imximage_ivt_offset = FLASH_OFFSET_UNDEFINED; + imximage_csf_size = 0; + set_hdr_func(); + + /* Parse dcd configuration file */ + parse_cfg_file(&imximage_header, "board/freescale/mx7dsabresd/imximage.cfg.cfgtmp"); + switch (version) { case IMXIMAGE_V1: print_hdr_v1(imx_hdr); ``` Is there any suggestion for me to contribute patches to the mainstream code in order to permanently fix above issues. Thank you. Best, Mingshen

1 0

[U-Boot] [PATCH] common, image-sig: [BUG?] if no valid signature node found, do not boot signed FIT image
by Heiko Schocher 05 Aug '17

05 Aug '17

fit_image_verify_required_sigs() must return != 0, on error. When fit_image_verify_required_sigs() does not find a signature node, it returns 0, which leads in booting a signed FIT image. Fix this! Signed-off-by: Heiko Schocher <hs(a)denx.de> --- Found on an imx28 based board, with key dtb appended to u-boot.bin. Booting signed FIT image without an valid key dtb appended to u-boot.bin shows: Using FEC1 device TFTP from server 192.168.1.1; our IP address is 192.168.20.103 Filename '/tftpboot/xxx/20170509/signed.fit'. Load address: 0x42000000 Loading: ################################################################# [...] ########### 3.3 MiB/s done Bytes transferred = 12560801 (bfa9a1 hex) Using 'conf@1' configuration Verifying Hash Integrity ... OK Trying 'kernel@1' kernel subimage Description: Linux kernel Created: 2017-06-08 9:10:14 UTC Type: Kernel Image Compression: uncompressed Data Start: 0x420000c0 Data Size: 4078928 Bytes = 3.9 MiB Architecture: ARM OS: Linux Load Address: 0x40008000 Entry Point: 0x40008000 Hash algo: sha256 Hash value: 6d1dce3e08133ac4d34c0e07ce266f5cffc6f5a2713619c9ff76ca4b04df4a5b Sign algo: sha256,rsa4096:dev Sign value: xxx Timestamp: 2017-06-08 9:10:15 UTC Verifying Hash Integrity ... sha256+ sha256,rsa4096:dev- OK ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Using 'conf@1' configuration Trying 'ramdisk@1' ramdisk subimage Description: miro ramdisk Created: 2017-06-08 9:10:14 UTC Type: RAMDisk Image Compression: gzip compressed Data Start: 0x423e92b0 Data Size: 8457506 Bytes = 8.1 MiB Architecture: ARM OS: Linux Load Address: 0x00000000 Entry Point: 0x00000000 Hash algo: sha256 Hash value: da60884efa4373e7003940a56271c326f159ff74356ded28d8ebe108af807cda Sign algo: sha256,rsa4096:dev Sign value: xxx Timestamp: 2017-06-08 9:10:15 UTC Verifying Hash Integrity ... sha256+ sha256,rsa4096:dev- OK ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Using 'conf@1' configuration Trying 'fdt@1' fdt subimage Description: miro device tree Created: 2017-06-08 9:10:14 UTC Type: Flat Device Tree Compression: uncompressed Data Start: 0x423e41b4 Data Size: 19852 Bytes = 19.4 KiB Architecture: ARM Hash algo: sha256 Hash value: 9b39c3ab6227bb8f0bcebc0bb64439248a6670dfe873bb1c6536764e0dc1623c Sign algo: sha256,rsa4096:dev Sign value: xxx Timestamp: 2017-06-08 9:10:15 UTC Verifying Hash Integrity ... sha256+ sha256,rsa4096:dev- OK ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Booting using the fdt blob at 0x423e41b4 Loading Kernel Image ... OK Loading Ramdisk to 47303000, end 47b13d22 ... OK Loading Device Tree to 472fb000, end 47302d8b ... OK Starting kernel ... [ 0.000000] Booting Linux on physical CPU 0x0 [ 0.000000] Linux version 4.11.0 (hs(a)hercules.denx.de) (gcc version 6.2.0 (GCC) ) #1 Tue May 16 07:31:30 CEST 2017 [...] And linux boots ... which is bad, as we have no valid key appended to u-boot.bin ... With this patch it shows: Using 'conf@1' configuration Verifying Hash Integrity ... OK Trying 'kernel@1' kernel subimage Description: Linux kernel Created: 2017-06-08 9:34:28 UTC Type: Kernel Image Compression: uncompressed Data Start: 0x420000c0 Data Size: 4078928 Bytes = 3.9 MiB Architecture: ARM OS: Linux Load Address: 0x40008000 Entry Point: 0x40008000 Hash algo: sha256 Hash value: 6d1dce3e08133ac4d34c0e07ce266f5cffc6f5a2713619c9ff76ca4b04df4a5b Sign algo: sha256,rsa4096:dev Sign value: xxx Timestamp: 2017-06-08 9:34:29 UTC Verifying Hash Integrity ... error! Unable to verify required signature for '' hash node in 'kernel@1' image node Bad Data Hash ERROR: can't get kernel image! => common/image-sig.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/common/image-sig.c b/common/image-sig.c index 455f2b9..646fb08 100644 --- a/common/image-sig.c +++ b/common/image-sig.c @@ -265,7 +265,7 @@ int fit_image_verify_required_sigs(const void *fit, int image_noffset, if (sig_node < 0) { debug("%s: No signature node found: %s\n", __func__, fdt_strerror(sig_node)); - return 0; + return 1; } fdt_for_each_subnode(noffset, sig_blob, sig_node) { -- 2.7.4

3 4