On 23 March 2016 at 06:41, Ben Pfaff wrote:
> On Mon, Mar 07, 2016 at 11:31:02AM -0800, Joe Stringer wrote:
>> Address pointed by header_ptr might be free'd due to realloc
>> happened in ofpbuf_put_hex(). Reported by valgrind in the test
>> 379: check TCP flags expression in OXM and NXM.
>>
>> Inv
On Mon, Mar 07, 2016 at 11:31:02AM -0800, Joe Stringer wrote:
> Address pointed by header_ptr might be free'd due to realloc
> happened in ofpbuf_put_hex(). Reported by valgrind in the test
> 379: check TCP flags expression in OXM and NXM.
>
> Invalid write of size 4
> nx_match_from_string_raw
Hi Joe,
I've tested this patch (with modification to ofpbuf_put to force using
newly allocated address) and it works fine. Thanks!
Regards,
William
On Mon, Mar 7, 2016 at 11:31 AM, Joe Stringer wrote:
> Address pointed by header_ptr might be free'd due to realloc
> happened in ofpbuf_put_hex()
On 4 March 2016 at 17:35, William Tu wrote:
> Address pointed by header_ptr might be free'd due to realloc
> happened at ofpbuf_put_uninit() and ofpbuf_put_hex(). Reported
> by valgrind 379: check TCP flags expression in OXM and NXM.
>
> Invalid write of size 4
> nx_match_from_string_raw (nx-m
Hi Jarno,
Thanks for the feedback. I forgot to mention that this issue is found by
changing the ofpbuf code to make each put reallocate the memory. I patched
the code with:
--- a/lib/ofpbuf.c
+++ b/lib/ofpbuf.c
@@ -383,6 +383,7 @@ ofpbuf_put_zeros(struct ofpbuf *b, size_t size)
void *
ofpbuf_pu
It might be super slow, but how about running the test suite with valgrind and
ofpbuf code changed so that each put reallocates the memory? That way we would
not have to be lucky about the timing/placement of reallocations to find these
bugs?
Jarno
> On Mar 4, 2016, at 5:35 PM, William Tu w
