Re: [Qemu-devel] [PATCH 1/8] qdict: fix unbounded stack for qdict_array_entries

[Markus Armbruster]

Cc: Kevin, because he added the array in question.

Peter Xu <address@hidden> writes:

> Suggested-by: Paolo Bonzini <address@hidden>
> CC: Luiz Capitulino <address@hidden>
> Signed-off-by: Peter Xu <address@hidden>
> ---
>  qobject/qdict.c | 15 +++++++++------
>  1 file changed, 9 insertions(+), 6 deletions(-)
>
> diff --git a/qobject/qdict.c b/qobject/qdict.c
> index 9833bd0..eb602a7 100644
> --- a/qobject/qdict.c
> +++ b/qobject/qdict.c
> @@ -704,17 +704,19 @@ int qdict_array_entries(QDict *src, const char 
> *subqdict)
>      for (i = 0; i < INT_MAX; i++) {
>          QObject *subqobj;
>          int subqdict_entries;
> -        size_t slen = 32 + subqdict_len;
> -        char indexstr[slen], prefix[slen];
> +#define __SLEN_MAX (128)
> +        char indexstr[__SLEN_MAX], prefix[__SLEN_MAX];
>          size_t snprintf_ret;
>  
> -        snprintf_ret = snprintf(indexstr, slen, "%s%u", subqdict, i);
> -        assert(snprintf_ret < slen);
> +        assert(__SLEN_MAX >= 32 + subqdict_len);
> +
> +        snprintf_ret = snprintf(indexstr, __SLEN_MAX, "%s%u", subqdict, i);
> +        assert(snprintf_ret < __SLEN_MAX);
>  
>          subqobj = qdict_get(src, indexstr);
>  
> -        snprintf_ret = snprintf(prefix, slen, "%s%u.", subqdict, i);
> -        assert(snprintf_ret < slen);
> +        snprintf_ret = snprintf(prefix, __SLEN_MAX, "%s%u.", subqdict, i);
> +        assert(snprintf_ret < __SLEN_MAX);
>  
>          subqdict_entries = qdict_count_prefixed_entries(src, prefix);
>          if (subqdict_entries < 0) {
> @@ -745,6 +747,7 @@ int qdict_array_entries(QDict *src, const char *subqdict)
>      }
>  
>      return i;
> +#undef __SLEN_MAX
>  }
>  
>  /**

Same arguments as for PATCH 2, except here an argument on the maximum
length of subqdict would probably be easier.

Unrelated to your patch: I think we've pushed QDict use father than
sensible.  Encoding multiple keys in a string so you can use a flat
associative array as your catch-all data structure is appropriate in
AWK, but in C?  Not so much...