May 2014 - U-Boot

[U-Boot] [PATCH] net: phy/vitesse: Add support for VSC8664 phy module
by Chunhe Lan 20 May '14

20 May '14

This patch adds support for VSC8664 PHY module which can be found on Freescale's T4240RDB boards. Signed-off-by: Chunhe Lan <Chunhe.Lan(a)freescale.com> --- drivers/net/phy/vitesse.c | 46 +++++++++++++++++++++++++++++++++++++++++++++ 1 files changed, 46 insertions(+), 0 deletions(-) diff --git a/drivers/net/phy/vitesse.c b/drivers/net/phy/vitesse.c index c555979..b04d1e0 100644 --- a/drivers/net/phy/vitesse.c +++ b/drivers/net/phy/vitesse.c @@ -58,6 +58,14 @@ #define MIIM_VSC8514_18G_QSGMII 0x80e0 #define MIIM_VSC8514_18G_CMDSTAT 0x8000 +/* Vitesse VSC8664 Control/Status Register */ +#define MIIM_VSC8664_SERDES_AND_SIGDET 0x13 +#define MIIM_VSC8664_ADDITIONAL_DEV 0x16 +#define MIIM_VSC8664_EPHY_CON 0x17 +#define MIIM_VSC8664_LED_CON 0x1E + +#define PHY_EXT_PAGE_ACCESS_EXTENDED 0x0001 + /* CIS8201 */ static int vitesse_config(struct phy_device *phydev) { @@ -244,6 +252,33 @@ static int vsc8514_config(struct phy_device *phydev) return 0; } +static int vsc8664_config(struct phy_device *phydev) +{ + u32 val; + + /* Enable MAC interface auto-negotiation */ + phy_write(phydev, MDIO_DEVAD_NONE, PHY_EXT_PAGE_ACCESS, 0); + val = phy_read(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_EPHY_CON); + val |= (1 << 13); + phy_write(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_EPHY_CON, val); + + phy_write(phydev, MDIO_DEVAD_NONE, PHY_EXT_PAGE_ACCESS, + PHY_EXT_PAGE_ACCESS_EXTENDED); + val = phy_read(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_SERDES_AND_SIGDET); + val |= (1 << 11); + phy_write(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_SERDES_AND_SIGDET, val); + phy_write(phydev, MDIO_DEVAD_NONE, PHY_EXT_PAGE_ACCESS, 0); + + /* Enable LED blink */ + val = phy_read(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_LED_CON); + val &= ~(1 << 2); + phy_write(phydev, MDIO_DEVAD_NONE, MIIM_VSC8664_LED_CON, val); + + genphy_config_aneg(phydev); + + return 0; +} + static struct phy_driver VSC8211_driver = { .name = "Vitesse VSC8211", .uid = 0xfc4b0, @@ -334,6 +369,16 @@ static struct phy_driver VSC8662_driver = { .shutdown = &genphy_shutdown, }; +static struct phy_driver VSC8664_driver = { + .name = "Vitesse VSC8664", + .uid = 0x70660, + .mask = 0xffff0, + .features = PHY_GBIT_FEATURES, + .config = &vsc8664_config, + .startup = &vitesse_startup, + .shutdown = &genphy_shutdown, +}; + /* Vitesse bought Cicada, so we'll put these here */ static struct phy_driver cis8201_driver = { .name = "CIS8201", @@ -366,6 +411,7 @@ int phy_vitesse_init(void) phy_register(&VSC8574_driver); phy_register(&VSC8514_driver); phy_register(&VSC8662_driver); + phy_register(&VSC8664_driver); phy_register(&cis8201_driver); phy_register(&cis8204_driver); -- 1.7.6.5

2 1

[U-Boot] [PATCH] Prevent a buffer overflow in mkimage when signing with SHA256
by Michael van der Westhuizen 20 May '14

20 May '14

Due to the FIT_MAX_HASH_LEN constant not having been updated to support SHA256 signatures one will always see a buffer overflow in fit_image_process_hash when signing images that use this larger hash. This is exposed by vboot_test.sh. Signed-off-by: Michael van der Westhuizen <michael(a)smart-africa.com> --- include/image.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/image.h b/include/image.h index b278778..fdaab5e 100644 --- a/include/image.h +++ b/include/image.h @@ -702,7 +702,8 @@ int bootz_setup(ulong image, ulong *start, ulong *end); #define FIT_FDT_PROP "fdt" #define FIT_DEFAULT_PROP "default" -#define FIT_MAX_HASH_LEN 20 /* max(crc32_len(4), sha1_len(20)) */ +/* max(crc32_len(4), sha1_len(20), sha256_len(32)) */ +#define FIT_MAX_HASH_LEN 32 /* cmdline argument format parsing */ int fit_parse_conf(const char *spec, ulong addr_curr, -- 2.0.0.rc0

1 0

[U-Boot] build u-boot tools for win32
by Hannes Petermaier 20 May '14

20 May '14

Hi, did anyone a successfull build of the u-boot tools for win32 ? i try: make HOSTCC=i686-w64-mingw32-gcc tools and get: scripts/basic/fixdep.c:108:22: fatal error: sys/mman.h: Datei oder Verzeichnis nicht gefunden "Datei oder Verzeichnis nicht gefunden" means 'file not found' many thanks and best regards, hannes

1 0

Re: [U-Boot] [RFC] POWER framework v3 - wish list
by Marek Vasut 20 May '14

20 May '14

On Friday, May 09, 2014 at 08:58:02 AM, Przemyslaw Marczak wrote: > Hello, [...] > struct power_ops_key_power { > int (*key_state) (int *state); > }; This could be a key input device. > struct power_ops_rtc { > int (*sec) (int set_get, int *val); > int (*min) (int set_get, int *val); > int (*hour) (int set_get, int *val); > int (*day) (int set_get, int *val); > int (*month) (int set_get, int *val); > int (*year) (int set_get, int *val); > }; RTC device. > struct power_ops_motor { > int (*configure) (void); > int (*enable) (int time, int gain); > }; > > struct power_ops_led_flash { > int (*configure) (void); > int (*enable) (void); > int (*disable) (void); > }; LED device. It seems like you're trying to assemble a huge framework while avoiding the already-present frameworks. No? Best regards, Marek Vasut

2 1

Re: [U-Boot] Unable to boot the new FIT image
by Simon Glass 20 May '14

20 May '14

Hi, On May 19, 2014 1:32 PM, "KJ H. Kiran" <kj.h.kiran(a)us.abb.com> wrote: > > Hi Simon and Jagan, > > Thanks for the help. I tried your suggestions but still I could not able to boot up my device. > Here is what I tried: > 1) Applied the patch from Simon's email ( http://patchwork.ozlabs.org/patch/339609/) I think you need at least the next patch in that series. > 2) Compiled the u-boot with my public key signed device tree of U-boot > harsha@harshakiran_kasha:/abb/Experiments/u-boot$ sudo make ARCH=arm CROSS_COMPILE=/abb/compilers/gcc-linaro-arm-linux-gnueabihf-4.7-2013.04-20130415_linux/bin/arm-linux-gnueabihf- -j8 DEV_TREE_DTB=./am335x-evmsk-pubkey.dtb > > Please find attached log in the txt file attached. I got the MLO, u-boot.img, u-boot.bin, u-boot-dtb.bin binaries. > > 3) Copied the MLO, u-boot.bin and u-boot-dtb.bin and the FIT image into the SD card. This was the response: > > U-Boot SPL 2014.07-rc1 (May 19 2014 - 13:17:49) > reading u-boot.img > spl_load_image_fat: error reading image u-boot.img, err - -1 > ### ERROR ### Please RESET the board ### > > 4) Second time , I copied MLO, u-boot.img, u-boot-dtb.bin and FIT image. This was the response: > U-Boot SPL 2014.07-rc1 (May 19 2014 - 13:17:49) > reading u-boot.img > reading u-boot.img > No valid FDT found - please append one to U-Boot binary, use u-boot-dtb.bin or define CONFIG_OF_EMBED. For sandbox, use -d <file > .dtb> > ### ERROR ### Please RESET the board ### > > 5) Since I added the CONFIG's to support the FIT, I was thinking even the u-boot.img expects a fdt binary. That’s the reason even the u-boot.img didn’t work.! > > 6) I commented out the CONFIGs from the am335x_evm.h file and compiled again. This time I loaded the MLO , u-boot.img, u-boot.dtb.bin and FIT just to check and follow jagan's suggestions.. > > I was able to successfully load MLO and U-boot and came to the u-boot# prompt. > .... > U-Boot# fatload mmc 0 ${TEXTBASE} u-boot-dtb.bin > reading zImage > ** Unable to read file zImage ** > U-Boot# fatload mmc0 > ** No device specified ** > U-Boot# fatload mmc0 u-boot-dtb.bin > ** Bad device specification mmc0 u-boot-dtb.bin ** > U-Boot# fatload mmc 0 ${TEXTBASE} u-boot-fdt.bin > reading zImage > ** Unable to read file zImage ** > U-Boot# > U-Boot# > U-Boot# > > Please let me know if I am still missing anything. > > Thanks, > Harsha kiran > > > > > > -----Original Message----- > From: sjg(a)google.com [mailto:sjg@google.com] On Behalf Of Simon Glass > Sent: Monday, May 19, 2014 10:43 AM > To: Jagan Teki > Cc: KJ H. Kiran; u-boot(a)lists.denx.de; trini(a)ti.com > Subject: Re: [U-Boot] Unable to boot the new FIT image > > +Tom > > On 19 May 2014 07:39, Jagan Teki <jagannadh.teki(a)gmail.com> wrote: > > > > On Mon, May 19, 2014 at 10:06 AM, KJ H. Kiran <kj.h.kiran(a)us.abb.com> wrote: > > > > > > Hi , > > > > > > > > > I am trying to make a eval board from TI to boot it with the new FIT image format. > > > > > > > > > I have followed the below procedure but iam getting the "No valid FDT found - please append one to U-Boot binary, use u-boot-dtb.bin or define CONFIG_OF_EMBED." error message. Cannot find out what is the exact problem. > > > > > > > This shouldn't be an issue with FIT - issue with booting u-boot fdt > > bin > > > > Just for FIT work usage - try to boot with u-boot.bin (fdt less u-boot > > bin) So-that you can end up with normal u-boot booting. > > > > On u-boot prompt and load u-boot-fdt.bin with TEXTBASE then load > > u-boot> fatload mmc 0 ${TEXTBASE} u-boot-fdt.bin go ${TEXTBASE} > > > > ------ fdt u-boot boot prompt> > > u-boot> fatload mmc 0 ${LOADADDR} kernel_fdt.itb bootm ${LOADADDR} > > ..... > > > > Please send output log for this. > > I hit this problem myself and posted this patch and the following series: > > http://patchwork.ozlabs.org/patch/339609/ > > I heard from one person that it didn't work for them, but I haven't got back to it yet. Still, it might help you. There was a problem with the hash pushing out the end of the image. > > > > > > > > > > > U-Boot Source: > > > > > > The latest source code from denx. > > > > > > patman: Suppress duplicate signoffs only for real patches on > > > 05/17/2014 > > > > > > > > > Procedure: > > > > > > Followed the exact same process mentioned in > > > > > > http://www.denx.de/wiki/pub/U-Boot/MiniSummitELCE2013/U-Boot_verifie > > > d_RSA_boot_flow_on_arm_target.pdf > > > > > > > > > I created a device tree and was able to generate a u-boot-dtb.bin file. > > > > > > > > > SD CARD: > > > > > > 1) copied the MLO in /boot > > > > > > 2) copied the u-boot-dtb.bin in /boot > > > > > > 3) coped the kernel_fdt.itb file > > > > > > 4) uEnv has > > > > > > console=ttyO0,115200n8 > > > fitfdt=/boot/kernel_fdt.itb > > > setenv loadaddr 0x82000000; run mmcargs; ext2load mmc ${mmcdev}:2 > > > ${loadaddr} ${fitfdt}; > > > > > > > > > I loaded the u-boot-dtb.bin file but it still complains ! Not much > > > help form the forums/internet > > > > > > > > > Please let me know if i missed anything here. > > > > > > thanks! > > -- > > Jagan. > > > > Regards, > Simon

1 0

Re: [U-Boot] Unable to boot the new FIT image
by KJ H. Kiran 20 May '14

20 May '14

Hi Simon and Jagan, Thanks for the help. I tried your suggestions but still I could not able to boot up my device. Here is what I tried: 1) Applied the patch from Simon's email (.....http://patchwork.ozlabs.org/patch/339609/) 2) Compiled the u-boot with my public key signed device tree of U-boot harsha@harshakiran_kasha:/abb/Experiments/u-boot$ sudo make ARCH=arm CROSS_COMPILE=/abb/compilers/gcc-linaro-arm-linux-gnueabihf-4.7-2013.04-20130415_linux/bin/arm-linux-gnueabihf- -j8 DEV_TREE_DTB=./am335x-evmsk-pubkey.dtb Please find attached log in the txt file attached. I got the MLO, u-boot.img, u-boot.bin, u-boot-dtb.bin binaries. 3) Copied the MLO, u-boot.bin and u-boot-dtb.bin and the FIT image into the SD card. This was the response: U-Boot SPL 2014.07-rc1 (May 19 2014 - 13:17:49) reading u-boot.img spl_load_image_fat: error reading image u-boot.img, err - -1 ### ERROR ### Please RESET the board ### 4) Second time , I copied MLO, u-boot.img, u-boot-dtb.bin and FIT image. This was the response: U-Boot SPL 2014.07-rc1 (May 19 2014 - 13:17:49) reading u-boot.img reading u-boot.img No valid FDT found - please append one to U-Boot binary, use u-boot-dtb.bin or define CONFIG_OF_EMBED. For sandbox, use -d <file .dtb> ### ERROR ### Please RESET the board ### 5) Since I added the CONFIG's to support the FIT, I was thinking even the u-boot.img expects a fdt binary. That’s the reason even the u-boot.img didn’t work.! 6) I commented out the CONFIGs from the am335x_evm.h file and compiled again. This time I loaded the MLO , u-boot.img, u-boot.dtb.bin and FIT just to check and follow jagan's suggestions.. I was able to successfully load MLO and U-boot and came to the u-boot# prompt. .... U-Boot# fatload mmc 0 ${TEXTBASE} u-boot-dtb.bin reading zImage ** Unable to read file zImage ** U-Boot# fatload mmc0 ** No device specified ** U-Boot# fatload mmc0 u-boot-dtb.bin ** Bad device specification mmc0 u-boot-dtb.bin ** U-Boot# fatload mmc 0 ${TEXTBASE} u-boot-fdt.bin reading zImage ** Unable to read file zImage ** U-Boot# U-Boot# U-Boot# Please let me know if I am still missing anything. Thanks, Harsha kiran -----Original Message----- From: sjg(a)google.com [mailto:sjg@google.com] On Behalf Of Simon Glass Sent: Monday, May 19, 2014 10:43 AM To: Jagan Teki Cc: KJ H. Kiran; u-boot(a)lists.denx.de; trini(a)ti.com Subject: Re: [U-Boot] Unable to boot the new FIT image +Tom On 19 May 2014 07:39, Jagan Teki <jagannadh.teki(a)gmail.com> wrote: > > On Mon, May 19, 2014 at 10:06 AM, KJ H. Kiran <kj.h.kiran(a)us.abb.com> wrote: > > > > Hi , > > > > > > I am trying to make a eval board from TI to boot it with the new FIT image format. > > > > > > I have followed the below procedure but iam getting the "No valid FDT found - please append one to U-Boot binary, use u-boot-dtb.bin or define CONFIG_OF_EMBED." error message. Cannot find out what is the exact problem. > > > > This shouldn't be an issue with FIT - issue with booting u-boot fdt > bin > > Just for FIT work usage - try to boot with u-boot.bin (fdt less u-boot > bin) So-that you can end up with normal u-boot booting. > > On u-boot prompt and load u-boot-fdt.bin with TEXTBASE then load > u-boot> fatload mmc 0 ${TEXTBASE} u-boot-fdt.bin go ${TEXTBASE} > > ------ fdt u-boot boot prompt> > u-boot> fatload mmc 0 ${LOADADDR} kernel_fdt.itb bootm ${LOADADDR} > ..... > > Please send output log for this. I hit this problem myself and posted this patch and the following series: http://patchwork.ozlabs.org/patch/339609/ I heard from one person that it didn't work for them, but I haven't got back to it yet. Still, it might help you. There was a problem with the hash pushing out the end of the image. > > > > > > U-Boot Source: > > > > The latest source code from denx. > > > > patman: Suppress duplicate signoffs only for real patches on > > 05/17/2014 > > > > > > Procedure: > > > > Followed the exact same process mentioned in > > > > http://www.denx.de/wiki/pub/U-Boot/MiniSummitELCE2013/U-Boot_verifie > > d_RSA_boot_flow_on_arm_target.pdf > > > > > > I created a device tree and was able to generate a u-boot-dtb.bin file. > > > > > > SD CARD: > > > > 1) copied the MLO in /boot > > > > 2) copied the u-boot-dtb.bin in /boot > > > > 3) coped the kernel_fdt.itb file > > > > 4) uEnv has > > > > console=ttyO0,115200n8 > > fitfdt=/boot/kernel_fdt.itb > > setenv loadaddr 0x82000000; run mmcargs; ext2load mmc ${mmcdev}:2 > > ${loadaddr} ${fitfdt}; > > > > > > I loaded the u-boot-dtb.bin file but it still complains ! Not much > > help form the forums/internet > > > > > > Please let me know if i missed anything here. > > > thanks! > -- > Jagan. Regards, Simon

1 0

[U-Boot] [PATCH v2] powerpc/t4rdb: Add alternate serdes protocols to align with A-007186
by Chunhe Lan 20 May '14

20 May '14

A-007186: SerDes PLL is calibrated at reset. It is possible for jitter to increase and cause the PLL to unlock when the temperature delta from the time the PLL is calibrated exceeds +56C/-66C when using X VDD of 1.35 V (or +70C/-80C when using XnVDD of 1.5 V). No issues are seen with LC VCO. The protocols only using Ring VCOs are impacted. Workaround: For all 1.25/2.5/5 GHz protocols, use LC VCO instead of Ring VCO, this need to use alternate serdes protocols. Alternate option has the same functionality as the original option; the only difference being LC VCO rather than Ring VCO. Signed-off-by: Chunhe Lan <Chunhe.Lan(a)freescale.com> --- board/freescale/t4rdb/eth.c | 2 +- board/freescale/t4rdb/t4_rcw.cfg | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/board/freescale/t4rdb/eth.c b/board/freescale/t4rdb/eth.c index d220475..142c6a8 100644 --- a/board/freescale/t4rdb/eth.c +++ b/board/freescale/t4rdb/eth.c @@ -67,7 +67,7 @@ int board_eth_init(bd_t *bis) /* Register the 10G MDIO bus */ fm_memac_mdio_init(bis, &tgec_mdio_info); - if (srds_prtcl_s1 == 28) { + if ((srds_prtcl_s1 == 28) || (srds_prtcl_s1 == 27)) { /* SGMII */ fm_info_set_phy_address(FM1_DTSEC1, SGMII_PHY_ADDR1); fm_info_set_phy_address(FM1_DTSEC2, SGMII_PHY_ADDR2); diff --git a/board/freescale/t4rdb/t4_rcw.cfg b/board/freescale/t4rdb/t4_rcw.cfg index 13408bd..51e9539 100644 --- a/board/freescale/t4rdb/t4_rcw.cfg +++ b/board/freescale/t4rdb/t4_rcw.cfg @@ -1,7 +1,7 @@ #PBL preamble and RCW header aa55aa55 010e0100 -#serdes protocol 28_56_2_10 +#serdes protocol 27_56_1_9 16070019 18101916 00000000 00000000 -70701050 00448c00 6c020000 f5000000 +6c700848 00448c00 0c020000 f5000000 00000000 ee0000ee 00000000 000287fc 00000000 50000000 00000000 00000028 -- 1.7.6.5

1 0

[U-Boot] [PATCH] dfu: mmc: return data from raw partition reads
by Stephen Warren 20 May '14

20 May '14

From: Stephen Warren <swarren(a)nvidia.com> Nothing in dfu_mmc.c ever sets dfu->r_left. Consequently, reads from MMC DFU altsettings would return 0 bytes. Fix this by modifying dfu_read_medium_mmc() to set *len in the same way that dfu_nand.c's dfu_read_medium_nand() does. Signed-off-by: Stephen Warren <swarren(a)nvidia.com> --- Even though this mirrors how dfu_nand.c works, it feels odd to do this every time a HW read is performed, since I believe this can happen multiple times in a single DFU read transaction. Surely *len (and hence dfu->r_left) should be set up once when the altsetting is initialized, likely by calling a separate function in struct dfu_entity? P.S. Could you please add a "dfu" entry into doc/git-mailrc? --- drivers/dfu/dfu_mmc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/dfu/dfu_mmc.c b/drivers/dfu/dfu_mmc.c index 63cc876612c9..3addcef68bde 100644 --- a/drivers/dfu/dfu_mmc.c +++ b/drivers/dfu/dfu_mmc.c @@ -203,6 +203,7 @@ int dfu_read_medium_mmc(struct dfu_entity *dfu, u64 offset, void *buf, switch (dfu->layout) { case DFU_RAW_ADDR: + *len = dfu->data.mmc.lba_size * dfu->data.mmc.lba_blk_size; ret = mmc_block_op(DFU_OP_READ, dfu, offset, buf, len); break; case DFU_FS_FAT: -- 1.8.1.5

1 1

[U-Boot] [PATCH v2 0/7] MIPS: Malta generic board conversion
by Paul Burton 19 May '14

19 May '14

This series adds generic board support for the MIPS architecture, and then converts the MIPS Malta development board to use it. Changes since v1: - Call init_func_ram/initdram to preserve the existing behaviour for boards, as requested by Daniel. - Call timer_init, which I'd missed as pointed out by Daniel. Note that I've only submitted the whole series again because patch numbers changed; the content of patches 1-3,6 is unchanged from v1. Paul Burton (7): MIPS: stub interrupt_init function MIPS: move mips_io_port_base out of board.c MIPS: define __init_end in u-boot.lds board_f: call init_func_ram on MIPS board_f: call timer_init on MIPS MIPS: allow use of generic board MIPS: Malta: convert to generic board arch/mips/config.mk | 2 ++ arch/mips/cpu/mips32/interrupts.c | 5 +++++ arch/mips/cpu/mips64/interrupts.c | 5 +++++ arch/mips/cpu/u-boot.lds | 1 + arch/mips/include/asm/u-boot.h | 9 +++++++++ arch/mips/lib/Makefile | 4 ++++ arch/mips/lib/board.c | 6 ------ arch/mips/lib/io.c | 12 ++++++++++++ common/board_f.c | 6 +++--- doc/README.generic-board | 1 + include/configs/malta.h | 3 +++ 11 files changed, 45 insertions(+), 9 deletions(-) create mode 100644 arch/mips/lib/io.c -- 1.8.5.3

3 9

[U-Boot] Unable to boot the new FIT image
by KJ H. Kiran 19 May '14

19 May '14

Hi , I am trying to make a eval board from TI to boot it with the new FIT image format. I have followed the below procedure but iam getting the "No valid FDT found - please append one to U-Boot binary, use u-boot-dtb.bin or define CONFIG_OF_EMBED." error message. Cannot find out what is the exact problem. U-Boot Source: The latest source code from denx. patman: Suppress duplicate signoffs only for real patches on 05/17/2014 Procedure: Followed the exact same process mentioned in http://www.denx.de/wiki/pub/U-Boot/MiniSummitELCE2013/U-Boot_verified_RSA_b… I created a device tree and was able to generate a u-boot-dtb.bin file. SD CARD: 1) copied the MLO in /boot 2) copied the u-boot-dtb.bin in /boot 3) coped the kernel_fdt.itb file 4) uEnv has console=ttyO0,115200n8 fitfdt=/boot/kernel_fdt.itb setenv loadaddr 0x82000000; run mmcargs; ext2load mmc ${mmcdev}:2 ${loadaddr} ${fitfdt}; I loaded the u-boot-dtb.bin file but it still complains ! Not much help form the forums/internet Please let me know if i missed anything here. Thanks, Harsha

3 2