August 2010 - U-Boot

[U-Boot] [PATCH 5/7] powerpc/p1021: Add more P1021 defines
by Haiying Wang 16 Aug '10

16 Aug '10

Signed-off-by: Haiying Wang <Haiying.Wang(a)freescale.com> --- arch/powerpc/include/asm/immap_85xx.h | 6 ++++++ arch/powerpc/include/asm/immap_qe.h | 7 ++++++- 2 files changed, 12 insertions(+), 1 deletions(-) diff --git a/arch/powerpc/include/asm/immap_85xx.h b/arch/powerpc/include/asm/immap_85xx.h index c1382c8..5eeb93a 100644 --- a/arch/powerpc/include/asm/immap_85xx.h +++ b/arch/powerpc/include/asm/immap_85xx.h @@ -1962,6 +1962,12 @@ typedef struct ccsr_gur { u8 res10b[76]; par_io_t qe_par_io[7]; u8 res10c[1600]; +#elif defined(CONFIG_P1021) + u8 res10b1[12]; + u32 iovselsr; + u8 res10b2[60]; + par_io_t qe_par_io[3]; + u8 res10c[1496]; #else u8 res10b[1868]; #endif diff --git a/arch/powerpc/include/asm/immap_qe.h b/arch/powerpc/include/asm/immap_qe.h index 531cfc8..a523876 100644 --- a/arch/powerpc/include/asm/immap_qe.h +++ b/arch/powerpc/include/asm/immap_qe.h @@ -588,6 +588,9 @@ typedef struct qe_immap { #elif defined(CONFIG_MPC8569) u8 muram[0x20000]; /* 0x1_0000 - 0x3_0000 Multi-user RAM */ u8 res17[0x10000]; /* 0x3_0000 - 0x4_0000 */ +#elif defined(CONFIG_P1021) + u8 muram[0x06000]; /* 0x1_0000 - 0x1_6000 Multi-user RAM */ + u8 res17[0x1a000]; /* 0x1_6000 - 0x3_0000 */ #else u8 muram[0xC000]; /* 0x110000 - 0x11C000 Multi-user RAM */ u8 res17[0x24000]; /* 0x11C000 - 0x140000 */ @@ -601,13 +604,15 @@ extern qe_map_t *qe_immr; #define QE_MURAM_SIZE 0x10000UL #elif defined(CONFIG_MPC8569) #define QE_MURAM_SIZE 0x20000UL +#elif defined(CONFIG_P1021) +#define QE_MURAM_SIZE 0x6000UL #elif defined(CONFIG_MPC8360) #define QE_MURAM_SIZE 0xc000UL #elif defined(CONFIG_MPC832x) #define QE_MURAM_SIZE 0x4000UL #endif -#if defined(CONFIG_MPC8323) +#if defined(CONFIG_MPC8323) || defined(CONFIG_P1021) #define MAX_QE_RISC 1 #define QE_NUM_OF_SNUM 28 #elif defined(CONFIG_MPC8569) -- 1.7.0

1 0

[U-Boot] Transfer of US$10.5 Million
by Sun Moyo 16 Aug '10

16 Aug '10

Please open the attachment letter for more details and get back to me

1 0

[U-Boot] [PATCH V2 1/2] USB-CDC: correct wrong alignment in ether.c
by Stefano Babic 15 Aug '10

15 Aug '10

The buffer for the status request must be word aligned because it is accessed with 32 bit pointer in the eth_status_complete function. Signed-off-by: Stefano Babic <sbabic(a)denx.de> --- drivers/usb/gadget/ether.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 313f15f..0b4ed18 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -684,7 +684,7 @@ static struct usb_gadget_strings stringtab = { /*============================================================================*/ static u8 control_req[USB_BUFSIZ]; -static u8 status_req[STATUS_BYTECOUNT]; +static u8 status_req[STATUS_BYTECOUNT] __attribute__ ((aligned(32))); -- 1.6.3.3

3 8

[U-Boot] [PATCH V2 2/2] USB-CDC: called handle_interrupts inside usb_eth_send
by Stefano Babic 15 Aug '10

15 Aug '10

The patch removes an endless loop in the usb_eth_send if the tx_complete is not called before going in the loop. The driver interrupt routine is called allowing the driver to check if the TX is completed. Signed-off-by: Stefano Babic <sbabic(a)denx.de> --- Changes from V1: - Comments by Remy Bohmer: moved usb_ep_queue inside timeout to not break at91 code drivers/usb/gadget/ether.c | 12 ++++++++++-- 1 files changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c index 0b4ed18..f082c03 100644 --- a/drivers/usb/gadget/ether.c +++ b/drivers/usb/gadget/ether.c @@ -1808,6 +1808,8 @@ static int usb_eth_send(struct eth_device* netdev, volatile void* packet, int le struct usb_request *req = NULL; struct eth_dev *dev = &l_ethdev; + unsigned long ts; + unsigned long timeout = USB_CONNECT_TIMEOUT; dprintf("%s:...\n",__func__); req = dev->tx_req; @@ -1833,13 +1835,19 @@ static int usb_eth_send(struct eth_device* netdev, volatile void* packet, int le #endif dev->tx_qlen=1; + ts = get_timer(0); + packet_sent=0; retval = usb_ep_queue (dev->in_ep, req, GFP_ATOMIC); if (!retval) - dprintf("%s: packet queued\n",__func__); + debug("%s: packet queued\n",__func__); while(!packet_sent) { - packet_sent=0; + if (get_timer(ts) > timeout) { + printf ("timeout sending packets to usb ethernet\n"); + return -1; + } + usb_gadget_handle_interrupts(); } return 0; -- 1.6.3.3

3 3

[U-Boot] [PATCH] Add mvsata support for the Sheevaplug devices
by GEAK 14 Aug '10

14 Aug '10

Add mvsata support to SHEEVAPLUG Signed-off-by: Gérald Kerma <geraker(a)gmail.com> --- include/configs/sheevaplug.h | 29 +++++++++++++++++++++++++++++ 1 files changed, 29 insertions(+), 0 deletions(-) diff --git a/include/configs/sheevaplug.h b/include/configs/sheevaplug.h index c5de86e..6a80c15 100644 --- a/include/configs/sheevaplug.h +++ b/include/configs/sheevaplug.h @@ -98,6 +98,7 @@ #define CONFIG_CMD_NAND #define CONFIG_CMD_PING #define CONFIG_CMD_USB +#define CONFIG_CMD_IDE /* * NAND configuration @@ -196,6 +197,34 @@ #endif /* CONFIG_CMD_USB */ /* + * IDe Support on SATA port0 + */ +#ifdef CONFIG_CMD_IDE +#define __io +#define CONFIG_CMD_EXT2 +#define CONFIG_MVSATA_IDE +#define CONFIG_IDE_PREINIT +#define CONFIG_MVSATA_IDE_USE_PORT0 +/* Needs byte-swapping for ATA data register */ +#define CONFIG_IDE_SWAP_IO +/* Data, registers and alternate blocks are at the same offset */ +#define CONFIG_SYS_ATA_DATA_OFFSET (0x0100) +#define CONFIG_SYS_ATA_REG_OFFSET (0x0100) +#define CONFIG_SYS_ATA_ALT_OFFSET (0x0100) +/* Each 8-bit ATA register is aligned to a 4-bytes address */ +#define CONFIG_SYS_ATA_STRIDE 4 +/* Controller supports 48-bits LBA addressing */ +#define CONFIG_LBA48 +/* CONFIG_CMD_IDE requires some #defines for ATA registers */ +#define CONFIG_SYS_IDE_MAXBUS 1 +#define CONFIG_SYS_IDE_MAXDEVICE 1 +/* ATA registers base is at SATA controller base */ +#define CONFIG_SYS_ATA_BASE_ADDR KW_SATA_BASE +/* ATA bus 0 is Kirkwood port 1 on sheevaplug */ +#define CONFIG_SYS_ATA_IDE0_OFFSET KW_SATA_PORT1_OFFSET +#endif /* CONFIG_CMD_IDE */ + +/* * File system */ #define CONFIG_CMD_EXT2 --

5 4

[U-Boot] commit 64419e47 causes data aborts in md on orion5x internal registers
by Albert ARIBAUD 14 Aug '10

14 Aug '10

(adding Mike for info as he is the submitter of 64419e47) Hi all, on current origin/master, doing an md on internal registers area (0xf1xxxxxx) causes a data abort on orion5x-based edminiv2. A git bisect between origin/master ('bad') and the last edminiv2 commit ecaf3af2977e76484e0231d3113be3f4bbf2b8d6 ('good', known to work) points to commit 64419e47518bbba059c80b77558f93ad4804145c (print_buffer: optimize & shrink). The data abort is in the aliasing of uip, usp and ucp to linebuf, which is not 4-bytes aligned. Aligning it fixes the issue. I've just posted a fix to the u-boot list: <http://article.gmane.org/gmane.comp.boot-loaders.u-boot/82812> Rogan, please test it and see if it fixes your issue as well (can't swear that it will). Mike, please test it and see if it does not regress on your side (can't see why it should, but maybe the __aligned__ attribute is going to choke your build toolchain?) Amicalement, -- Albert.

2 1

[U-Boot] [PATCH 00/7] ARMV7: OMAP3: Board updates for Beagle and Overo
by Steve Sakoman 13 Aug '10

13 Aug '10

This series does a bit of OMAP housekeeping as well as adding board revision detection for Beagle and Overo. The first patch moves syslib.c from omap3 to omap-common where it can be shared by both OMAP3 and OMAP4 boards. The second patch creates a convenience function to set the voltage regulators in the TWL4030 companion chip. It also modifies existing code to use the new function. The third patch enables the input drivers on the Overo's MMC_CLK pins for channels 1 and 3 (channel 2 was already enabled, missing the other 2 was an oversight) The fourth patch adds an environment variable to allow one to easily set the processor clock rate via a linux boot arg. The fifth patch adds a board revision detection function for Overo. The sixth patch adds enhanced revision detection for Beagle and support for the C4 revision (it uses features introduced by patches 2 and 4) The seventh patch initializes the OMAP GPMC and interrupt GPIO for the second ethernet chip on Overo boards. These patches were tested on Beagle and Overo as well as the OMAP4 Panda to make sure there were no regressions on OMAP4 boards. I also did a MAKEALL ARMV7 to verify that there were no build errors on any ARMV7 boards. Steve Sakoman (7): ARMV7: OMAP: Move syslib.c to omap-common since it can be shared by OMAP3 and OMAP4 ARMV7: OMAP: add convenience function to set TWL4030 regulator voltages ARMV7: OMAP: Enable input driver on Overo's MMC1_CLK and MMC3_CLK pinmux setup ARMV7: OMAP: Add mpurate boot arg for Overo and Beagle ARMV7: OMAP: Add board revision detection for Overo ARMV7: OMAP: Add detection and support for Beagle C4 revision ARMV7: OMAP: Configure Overo's second network chip

1 7

[U-Boot] [PATCH 0/8] USB: gadget: fix porting bugs
by Vitaly Kuzmichev 13 Aug '10

13 Aug '10

1) Restuct Makefile to unable usage old and new stacks simultaneously 2) Add lost 'qmult' definition for High Speed devices 3) Take debug printout macros back and make them more useful 4) Fix in_ep, out_ep, ep0 and status_ep confusions 5) Replace 'strcpy' by 'strlcpy' 6) Fix possible oops on stat_req->buf initialization 7) 2 updates backported from linux-2.6.git David Brownell (1): USB: gadget: ethernet error path potential oops fix Julia Lawall (1): USB: gadget: change simple_strtol to simple_strtoul Vitaly Kuzmichev (6): USB-CDC: Restuct USB gadget Makefile USB-CDC: Add lost 'qmult' definition USB-CDC: Linux-like debug printout USB-CDC: Correct freeing usb requests USB-CDC: Replace 'strcpy' by 'strlcpy' USB-CDC: Correct stat_req initialization drivers/usb/gadget/Makefile | 9 +++- drivers/usb/gadget/epautoconf.c | 2 +- drivers/usb/gadget/ether.c | 95 ++++++++++++++++++++++----------------- 3 files changed, 61 insertions(+), 45 deletions(-)

7 51

Re: [U-Boot] TFTP timeout issue while downloading the linux kernel in openpxa vpac270 u-boot source code
by Marek Vasut 13 Aug '10

13 Aug '10

Dne Pá 13. srpna 2010 12:27:12 StephenPaulraj Chinnadurai-ERS,HCLTech napsal(a): > Dear Marek, > YOU DROPPED THE MAILING LIST AGAIN !! HOW MANY TIMES DO I HAVE TO TELL YOU TO KEEP IT CCED ?! > Even if I assign 0x1234 to MSC2 I am getting the same timeout problem. Because 0x1234 is 16 bit bus ... dont you use 32bit bus to that card ? > > Can you suggest me some other solution? > > Is this problem because of static memory configuration? > Check the bus with a scope I guess, that should give you some ideas. > Where should I concentrate for the solution, whether in net/tftp.c file or > in drivers/net/smc91111.c file? You can enable debuging output in the smc91111 driver. > > Kindly guide me to come out the problem. > > Thanks and Regards > Stephen Paulraj C > > HCL Technologies Ltd. > D-12, Sidco Industrial Estate, > Ambattur, Chennai -58 > Tel: +91 44 42004800 Extn: 2319 > Mob:+91 9962583934 > www.hcl.in > > -----Original Message----- > From: u-boot-bounces(a)lists.denx.de [mailto:u-boot-bounces@lists.denx.de] On > Behalf Of Marek Vasut Sent: Thursday, August 12, 2010 11:05 PM > To: StephenPaulraj Chinnadurai-ERS,HCLTech > Cc: u-boot(a)lists.denx.de > Subject: Re: [U-Boot] TFTP timeout issue while downloading the linux kernel > in openpxa vpac270 u-boot source code > > Dne Čt 12. srpna 2010 19:11:29 StephenPaulraj Chinnadurai-ERS,HCLTech napsal(a): > > Hello Marek, > > > > I am using a customize board which has PXA270 controller. SMCS > > LAN91C111Ii-NU is connected to the static memory 4 (nCS4). > > I see, so it's not vpac270 bug. > > > I have defined following macro value in the include/configs/<myboard.h> > > file. > > > > #define CONFIG_SYS_GPDR2_VAL 0x70DF3FFF > > GPIO 80 is configured as output > > > > #define CONFIG_SYS_GPSR2_VAL 0x10010000 /*nCS4*/ > > GPIO 80 is set > > You don't really need to configure these if you then set it to AF2. > > > #define CONFIG_SYS_GAFR2_U_VAL 0x1504A176 > > Alternate function 2 is selected for GPIO 80 > > > > #define CONFIG_SYS_MSC2_VAL 0x0000B884 > > Zylonite 300 uses value 0x3c for the SMC91111 in the corresponding part of > MSCx register in original Marvell sources, Zylonite320 uses 0x123c ... > maybe you should review the MSCx configuration here. For pxa27x, I guess > it might be something like 0x123c too. > > From the log below, it really seems the ethernet chip is choking. > > Cheers > > > RT4 = 4 Variable Latency I/O (VLIO) > > > > ROM4 Bus Width > > 0 = 32 Bits > > > > RDF4 = 8 > > > > RDN4 = 8 > > > > RRR4 = 3 > > > > RBUFF4 = 1 Faster device > > > > #define CONFIG_SMC91111 > > #define CONFIG_SMC91111_BASE 0x10000300 /*nCS4*/ > > #define CONFIG_SMC_USE_32_BIT > > #undef CONFIG_SMC_91111_EXT_PHY > > > > > > Ethernet controller is working fine. I can able to ping the machines > > connected to the network. > > > > tftpboot command is also working fine, I can able to download the > > u-boot.bin to SDRAM. > > > > But I have issue in downloading bigger files (zImage which is 2.5 MB) to > > the SDRAM. I am getting the following timeout messages and the download > > is not getting completed. > > > > Loading: *########T T T T T T T T T T T T T T T T T T T T > > Retry count exceeded; starting again > > > > Please help me to come out of this issue. > > > > Kindly find the complete log below: > > > > __ > > > > > > U-Boot 2010.06-rc1 (Aug 13 2010 - 09:11:31) > > > > > > DRAM: 128 MiB > > > > Flash: 64 MiB > > > > In: serial > > > > Out: serial > > > > Err: serial > > > > Net: SMC91111-0 > > > > Hit any key to stop autoboot: 2 ___ 0 > > > > MDT-Boot> print > > > > bootargs=console=tty0 console=ttyS1, root=/dev/sda1 rootdelay=15 > > > > bootcmd=if mmc init && fatload mmc 0 0xa0000000 uImage; then bootm > > 0xa0000000; fi; if usb reset && fatload usb 0 0xa0000000 uImage; then > > bootm 0xa0000000; fi; bootm 0x150000; > > > > bootdelay=2 > > > > baudrate=115200 > > > > ethaddr=00:00:70:26:1E:01 > > > > netmask=255.255.255.0 > > > > ethact=SMC91111-0 > > > > ipaddr=10.100.144.42 > > > > serverip=10.100.144.245 > > > > gatewayip=10.100.144.255 > > > > stdin=serial > > > > stdout=serial > > > > stderr=serial > > > > > > Environment size: 445/32764 bytes > > > > MDT-Boot> ping 10.100.144.245 > > > > SMC91111: MAC 00:00:70:26:1e:01 > > > > Using SMC91111-0 device > > > > host 10.100.144.245 is alive > > > > MDT-Boot> tftp a3000000 u-boot.bin > > > > SMC91111: MAC 00:00:70:26:1e:01 > > > > Using SMC91111-0 device > > > > TFTP from server 10.100.144.245; our IP address is 10.100.144.42 > > > > Filename 'u-boot.bin'. > > > > Load address: 0xa3000000 > > > > Loading: *_############ > > > > done > > > > Bytes transferred = 335280 (51db0 hex) > > > > MDT-Boot> tftp a3000000 zImage > > > > SMC91111: MAC 00:00:70:26:1e:01 > > > > Using SMC91111-0 device > > > > TFTP from server 10.100.144.245; our IP address is 10.100.144.42 > > > > Filename 'zImage'. > > > > Load address: 0xa3000000 > > > > Loading: *_####################T T T T T T T T T T T T T T T T T T T T > > > > Retry count exceeded; starting again > > > > SMC91111: MAC 00:00:70:26:1e:01 > > > > Using SMC91111-0 device > > > > TFTP from server 10.100.144.245; our IP address is 10.100.144.42 > > > > Filename 'zImage'. > > > > Load address: 0xa3000000 > > > > Loading: *_#######T ##T #T T T T T T T T T T T T T T T T T T > > > > Retry count exceeded; starting again > > > > SMC91111: MAC 00:00:70:26:1e:01 > > > > Using SMC91111-0 device > > > > TFTP from server 10.100.144.245; our IP address is 10.100.144.42 > > > > Filename 'zImage'. > > > > Load address: 0xa3000000 > > > > Loading: *_#######################T T T T T T T T T T T T T T T T T T T T > > > > Retry count exceeded; starting again > > > > > > Thanks and Regards > > Stephen Paulraj C > > _______________________________________________ > U-Boot mailing list > U-Boot(a)lists.denx.de > http://lists.denx.de/mailman/listinfo/u-boot

1 0

[U-Boot] [PATCH] env_nand: return error when no device is found
by Mike Frysinger 13 Aug '10

13 Aug '10

Currently, if there is an error probing the NAND chip and the env is based in NAND, the readenv() function will use a NULL function pointer and thus jump to address 0. Here I just check for a non-zero value of blocksize as that shouldn't be zero when a valid device is found, but perhaps there is a better way for someone familiar with the NAND internals to suggest. Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- common/env_nand.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/common/env_nand.c b/common/env_nand.c index a5e1038..d38bcca 100644 --- a/common/env_nand.c +++ b/common/env_nand.c @@ -266,6 +266,8 @@ int readenv (size_t offset, u_char * buf) u_char *char_ptr; blocksize = nand_info[0].erasesize; + if (!blocksize) + return 1; len = min(blocksize, CONFIG_ENV_SIZE); while (amount_loaded < CONFIG_ENV_SIZE && offset < end) { -- 1.7.2

3 2