On Friday 18 March 2005 03:10, Jean Delvare wrote:
> Hi Tony, hi all,
> I think I found the problem. If count is less than in_row1, bad things
> will happen with the code above. Shame on me for not noticing when I
> wrote it in the first place.
>
> Suggested fix would be:
>
> --- linux-2.6.11.4/drivers/i2c/chips/eeprom.c.orig 2005-03-13
> 10:00:01.000000000 +0100 +++
> linux-2.6.11.4/drivers/i2c/chips/eeprom.c 2005-03-17 19:54:07.000000000
> +0100 @@ -130,7 +130,8 @@
>
> /* Hide Vaio security settings to regular users (16 first bytes) */
> if (data->nature == VAIO && off < 16 && !capable(CAP_SYS_ADMIN)) {
> - int in_row1 = 16 - off;
> + size_t in_row1 = 16 - off;
> + in_row1 = min(in_row1, count);
> memset(buf, 0, in_row1);
> if (count - in_row1 > 0)
> memcpy(buf + in_row1, &data->data[16], count - in_row1);
>
>
> Please test and report.
>
Tested on nvidiafb. No more hang, thanks.
I can test on savagefb too, if you want, but I expect the same result.
Tony
