August 2009 - U-Boot

[U-Boot] ext2fs.c/ext2fs_mount() fails when inode_size = 256
by Bob Furber 12 Aug '09

12 Aug '09

Our SBC have been happily booting uClinux from ext3 partitioned SD cards prepared on a Fedora Linux-2.6.15 PC. But, when we prepared our 2GB SD cards on a brand new Ubuntu Linux-2.6.27 PC, out SBCs were no longer able to find /boot/linux.bin. Linux/dumpe2fs showed that the bootable [Fedora] 2GB SD cards had inode_size = 128, whereas the delinquent [Ubuntu] SD cards had inode_size = 256. Adding a "-I 128" switch to mkfs.ext3 in the script that prepares an loads the SD cards solved the problem. That is, our SBCs can now boot from Ubuntu prepared SD cards. However, this brings up the point that, somewhere along the line, the default inode_size for SD cards is no longer 128 bytes. It could be double, quadruple or more. There is some suspicious code in ext2fs.c/ext2fs_mount(): inodes_per_block = EXT2_BLOCK_SIZE (data) / 128; However, when this was replaced by inodes_per_block = EXT2_BLOCK_SIZE (data) / INODE_SIZE(data); ..the 2GB/128 byte inode SD cards booted and the 2GB/256 byte inode SD cards didn't. When reading a 2GB/256 byte inode card, ext2fs_mount()/ext2fs_read_inode(data, 2, data->inode) returns data->inode filled with zeros. Either there is a flaw in data or data is not being interpreted correctly. Regrettably, the root problem eludes me. Any thoughts or comments would be appreciated. Bob Furber

2 1

[U-Boot] modem pass-through
by David Collier 12 Aug '09

12 Aug '09

I was reading the notes about openmoko and saw that they had implemented a serial pass-through to allow the on-board GSM modem to be driven via an external serial port. Did that feature ever make it's way into the standard u-boot code??? David Collier www.dexdyne.com

2 1

[U-Boot] reset environment
by Dexdyne Postmaster 12 Aug '09

12 Aug '09

In development, it's a bit of a pain that installing a new u-boot image with different settings for bootargs and bootcmd doesn't actually change anything, because the environment in flash is still valid. Worse, there doesn't seem to be a command to say "over-write that with your new defaults" If that is indeed so, then can I propose such a command should be added? I'll try to knock out the patch if no-one more skilled wants to leap into action! David Collier www.dexdyne.com

3 3

[U-Boot] taking the time pressure off board init
by David Collier 12 Aug '09

12 Aug '09

I'm working on an AVR32 with a funny system of warm boots caused by watchdog which don't fully reset the watchdog or real-time-counter hardware. I think the system allows RTC interrupts to be generated during or after the warm boot which can cause havoc with the existing code. It certainly means you have to rush to the watchdog and disable it within a second, or you'll loop. but I have a modem which I need/want to handle in board-init, and it takes 1-second pulses to get it going.... I have found it reasonable to add an cpu_early_init leading into an clock_early_init. That means that board_early_init can run without any time pressure on it because WDT hasn't yet been stomped on. The early clock init can disable the WDT, then kill off RTC interrupts. Any good reason why I shouldn't upload a patch to cover this? Do I do it here, or on AVR32-specific u-boot list? David Collier www.dexdyne.com

2 1

[U-Boot] NAND issues
by alfred steele 12 Aug '09

12 Aug '09

Hi All, I have been using the "nand erase.e" command with favourable results till now. I was enticed by the "CONFIG_JFFS2_NAND" option in the platform config. When i turn this config option on and use the "NAND. write.jffs2 ", and use the resulting image, i get abnormal results including a kernel crash. What is the difference between a "nand write.jffs2" and the traditional nand write.e" command? I haven't been able to dig up on any relevant documention. I haven't grepped through the "common" source for the NAND subsystem. Are the commands any different , which of the commands is deprecated? i was guessing "nand write.e" understands bad blocks whereas nand write.jffs2" understands bad blocks as well as jffs2 internals like cleanmarker etc. Which one should i be using? Thanks, Alfred

2 1

[U-Boot] Write back flash content to SD/MMC Card
by A. Geisreiter 12 Aug '09

12 Aug '09

Hello, I will write back the complete content of my flash memory to a binary File on SD/MMC Card. How can I do this? Is there any support to create a file on SD/MMC card? Thanks, Andreas

2 1

[U-Boot] CFI Flash write speed
by A. Geisreiter 12 Aug '09

12 Aug '09

Hello, I have add CFI support to my project. I use Intel strata flash P30 and PXA270 CPU. The flash is connected with 16bit to the PXA data bus. If I write to the flash I have transfer rates of roughly 75kB/sec. I think this is very slow. How can I increase the transfer rate? Thanks, Andreas

2 1

Re: [U-Boot] [PATCH 2/2] MTD:NAND: ADD new ECC mode NAND_ECC_HW_OOB_FIRST
by Scott Wood 12 Aug '09

12 Aug '09

On Mon, Aug 10, 2009 at 01:27:56PM -0400, s-paulraj(a)dal.design.ti.com wrote: > /** > + * nand_read_page_hwecc_oob_first - [REPLACABLE] hw ecc, read oob first > + * @mtd: mtd info structure > + * @chip: nand chip info structure > + * @buf: buffer to store read data > + * > + * Hardware ECC for large page chips, require OOB to be read first. That statement may be true on some controllers, but not all. > + * For this ECC mode, the write_page method is re-used from ECC_HW. > + * These methods read/write ECC from the OOB area, unlike the > + * ECC_HW_SYNDROME support with multiple ECC steps, follows the > + * "infix ECC" scheme and reads/writes ECC from the data area, by > + * overwriting the NAND manufacturer bad block markings. > + */ > +static int nand_read_page_hwecc_oob_first(struct mtd_info *mtd, > + struct nand_chip *chip, uint8_t *buf, int page) > +{ > + int i, eccsize = chip->ecc.size; > + int eccbytes = chip->ecc.bytes; > + int eccsteps = chip->ecc.steps; > + uint8_t *p = buf; > + uint8_t *ecc_code = chip->buffers->ecccode; > + uint32_t *eccpos = chip->ecc.layout->eccpos; > + uint8_t *ecc_calc = chip->buffers->ecccalc; > + > + /* Read the OOB area first */ > + chip->cmdfunc(mtd, NAND_CMD_READOOB, 0, page); Note that a READ0 command will have already been issued at this point. I guess it looks to the NAND chip as a zero-byte read, but still things are getting quite ugly. The "generic" interface is one big layering violation that assumes a certain type of very simple controller, and we're accumulating hacks to deal with less straightforward controllers. :-( -Scott

2 2

[U-Boot] test mail server
by Kumar Gala 12 Aug '09

12 Aug '09

please ignore. It seems like the list server or something isn't quite happy and sending this to see if it comes back. - k

1 0

[U-Boot] Urgent
by Mike listowel 12 Aug '09

12 Aug '09

Hello Good Day This is Listowel, With regards to your Company i am sending this email Regards to order some(CORDLESS DRILL) ,i will like to know the type and sizes you have in stock and get me the sales price of one so that i will tell you the quantity i will be ordering, and also if you accept credit card as a form of payment . God Bless . NAME.LISTOWEL Mike Listowel Listowel And Wife Co Ltd. 101 31st Ave West Number 5 Dickinson,ND 58505 ********************************************************************************

1 0