Regards,
Tigran.
On Mon, 15 Nov 1999, Eleonora Autore wrote:
> > > binfmt-0000 happens to me always when I to exec a zero length file. (2.2.x)
> > > I guess this could be classified as a kernel bug but no idea why you are
> > > getting it - unless your ramdisk fs is corrupted.
> > > No idea what it is trying to execute?
> >
> > if ramdisk was corrupted then 2.2.x wouldn't work either, would it? But
> > it does. So, it does look like a bug in 2.3.28 (most likely binfmt or
> > kmod) so I am going to debug it as soon as I get a spare minute.
>
> it seems taht in 2.3.28 the piece of memory prepared by setup containing
> command_line[256] is corrupted, e.g. I get my command_line chopped down
> to only 80-90 bytes (e.g. I had "console=ttyS1,9600 console=tty0" at the
> end and sometimes I get only serial console output). Also I put some
> printk() in setup_arch() of arch/i386/kernel/setup.c like this:
>
> #ifdef CONFIG_BLK_DEV_RAM
> rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
> rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
> rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
> printk(KERN_ERR "rd_start=%d, rd_prompt=%d, rd_doload=%d\n",
> rd_image_start, rd_prompt, rd_doload);
> #endif
>
> and I get wrong values of rd_start/rd_prompt/rd_doload (all 0's) even
> when I specify load_ramdisk=1 prompt_ramdisk=1.
>
> I don't have time to look at it closer now but I will certainly work on
> it in the evening.
>
> Regards,
> Tigran.
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/