Re: [PATCH] Reduce stack used by lib/hexdump.c

[Andrew Morton]

On Thu, 29 Nov 2007 15:28:42 -0800
Joe Perches <joe@xxxxxxxxxxx> wrote:

> On Thu, 2007-11-29 at 22:07 +0100, Jan Engelhardt wrote:
> > I'd add GFP_ATOMIC here. Who knows whether tomorrow, the oops dumper
> > or warn_on will use print_hex_dump.
> 
> Signed-off-by: Joe Perches <joe@xxxxxxxxxxx>
> 
> diff --git a/lib/hexdump.c b/lib/hexdump.c
> index 70e23fb..be94934 100644
> --- a/lib/hexdump.c
> +++ b/lib/hexdump.c
> @@ -140,13 +140,20 @@ EXPORT_SYMBOL(hex_dump_to_buffer);
>   * Example output using %DUMP_PREFIX_ADDRESS and 4-byte mode:
>   * ffffffff88089af0: 73727170 77767574 7b7a7978 7f7e7d7c  pqrstuvwxyz{|}~.
>   */
> +
> +#define HEX_LINE_SIZE 200
> +
>  void print_hex_dump(const char *level, const char *prefix_str, int prefix_type,
>  			int rowsize, int groupsize,
>  			const void *buf, size_t len, bool ascii)
>  {
>  	const u8 *ptr = buf;
>  	int i, linelen, remaining = len;
> -	unsigned char linebuf[200];
> +	unsigned char *linebuf;
> +
> +	linebuf = kmalloc(HEX_LINE_SIZE, GFP_ATOMIC);
> +	if (!linebuf) {
> +		WARN_ON(1);
> +		return;
> +	}

No, I think print_hex_dump() is too low-level to be doing allocations. 
For example, one could easily choose to call print_hex_dump() at oops time,
and then what happens if we oops in kmalloc() (as we often do...)?

You could trim linebuf[] to 80 chars or so.  Extra points for making it
very clear when someone tries to exceed that - strcpy(linebuf, "stop being
stupid").

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/