[U-Boot] gcc optimizing option cause printf() and others crash.why?

I use arm-2009q1-203-arm-none-linux-gnueabi-i686-pc-linux-gnu.tar.gz to compile uboot1.3.4, and found that if I set OPTFLAGS to -O1,or O2, uboot will crash, usually in printf(). when i set to O3, it runs to prompt, but after several seconds(or minutes), the code is dead,and the console has no response. Os also has the problem,usually cannot run to main_loop(). How should i resolve this problem?

2010/7/29 yaojin liu lanmanck@gmail.com

hi,all: I encounter strange thing again. :(

it seems the vsprintf will overrun. have someone met before?

To debug it ,I define a function writing character directly to the register: "u_sendstring()"

Here is the code for debugging printf:

void printf(const char *fmt, ...)

{

va_list args;

uint i;

char printbuffer[CONFIG_SYS_PBSIZE];

va_start(args, fmt);

/* For this to work, printbuffer must be larger than

• anything we ever want to print.

*/

u_sendstring("printf2\n");

i = vsprintf(printbuffer, fmt, args);

u_sendstring("printf3\n");

va_end(args);

puts(printbuffer);

}

and here is the output:

DM3xx=>>: DONEprintf2

DM3xx=>>: printf3

DM3xx=>>: U-Boot 2009.03 ( 2010 - 23:14:17)

DM3xx=>>: I2C: ready

DM3xx=>>: DRAM: printf2

DM3xx=>>: printf3

DM3xx=>>: 128printf2

DM3xx=>>: printf3

DM3xx=>>: MB

DM3xx=>>: NAND: board_nand_init end

DM3xx=>>: nand_scan1

DM3xx=>>: nand_scan_ident1

DM3xx=>>: nand_scan_ident2

DM3xx=>>: nand_get_flash_type1

DM3xx=>>: nand_get_flash_type2

DM3xx=>>: nand_get_flash_type4

DM3xx=>>: printf2n BOOTME

DM3xx=>>: BOOTME

DM3xx=>>: BOOTME

DM3xx=>>: BOOTME

/////////////////////// weill, you can see that the first string will output correctly,such as "128" "MB", but when it ran to nand_get_flash_type(), the chip reboot.

static struct nand_flash_dev *nand_get_flash_type(struct mtd_info *mtd, struct nand_chip *chip, int busw, int *maf_id)

{

struct nand_flash_dev *type = NULL;

int i, dev_id, maf_idx;

int tmp_id, tmp_manf;

/* Select the device */

chip->select_chip(mtd, 0);

u_sendstring("nand_get_flash_type1\n");

.....

u_sendstring("nand_get_flash_type4\n");

printk("NAND device: Manufacturer ID:"

     " 0x%02x, Chip ID: 0x%02x (%s %s)\n", *maf_id, dev_id,

     nand_manuf_ids[maf_idx].name, type->name);

u_sendstring("nand_get_flash_type5\n");

return type;

}

printk is just printf. So as i describe above, the code ran to vsprintf and no respond anymore. How could this happen? Is it another hardware bug?