review (XS) for 6765546: Wrong sscanf used to parse CompilerOracle command >= 32 characters could lead to crash

[Tom Rodriguez]

On Dec 14, 2010, at 4:55 PM, Vladimir Kozlov wrote:

> It should be in the RANGEBASE definition (not in one sscanf) which is used in other places also:
> 
> -#define RANGEBASE "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789$_<>" \
> +#define RANGEBASE "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789$_<>[" \

RANGEBASE is for class and method names.  The signature line adds to RANGEBASE the characters which are allowed as part of signatures.

tom

> 
> Vladimir
> 
> Tom Rodriguez wrote:
>> Sure.  I added that in and tested to make sure array signatures were accepted now.  I've updated the webrev too.  Thanks.
>> tom
>> On Dec 14, 2010, at 3:54 PM, Igor Veresov wrote:
>>> Looks good!
>>> 
>>> May I also suggest piggybacking that on this change:
>>> 
>>> @@ -489,7 +489,7 @@ void CompilerOracle::parse_from_line(cha
>>>    line += bytes_read;
>>>    // there might be a signature following the method.
>>>    // signatures always begin with ( so match that by hand
>>> -    if (1 == sscanf(line, "%*[ \t](%254[);/" RANGEBASE "]%n", sig + 1, &bytes_read)) {
>>> +    if (1 == sscanf(line, "%*[ \t](%254[[);/" RANGEBASE "]%n", sig + 1, &bytes_read)) {
>>> 
>>> This basically makes it accept signatures with arrays.
>>> 
>>> Thanks,
>>> igor
>>> 
>>> On 12/14/10 3:30 PM, Tom Rodriguez wrote:
>>>> http://cr.openjdk.java.net/~never/6765546
>>>> 
>>>> 6765546: Wrong sscanf used to parse CompilerOracle command>= 32 characters could lead to crash
>>>> Reviewed-by:
>>>> 
>>>> The buffer for a sscanf isn't long enough to include the null
>>>> termination and we're missing a check for unknown commands.  Tested
>>>> with various command lines.