Re: gnutls 3.0.14

[Nikos Mavrogiannopoulos]

On 02/24/2012 09:46 PM, Andreas Radke wrote:

> PASS: rsa-encrypt-decrypt
> ==24740== 16,464 bytes in 1 blocks are definitely lost in loss record 21 of 22
> ==24740==    at 0x4C28024: calloc (in 
> /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==24740==    by 0x4E5553A: _mbuffer_alloc (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E56421: _gnutls_io_read_buffered (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E51A13: _gnutls_recv_in_buffers (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E57A43: _gnutls_handshake_io_recv_int (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E5A9D7: _gnutls_recv_handshake (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E5EBBE: _gnutls_recv_server_kx_message (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E5CA79: _gnutls_handshake_client (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4E5CDCF: gnutls_handshake (in 
> /build/src/gnutls-3.0.14/lib/.libs/libgnutls.so.28.6.1)
> ==24740==    by 0x4018CE: doit (in 
> /build/src/gnutls-3.0.14/tests/mini-loss-time)
> ==24740==    by 0x401234: main (in 
> /build/src/gnutls-3.0.14/tests/mini-loss-time)
> ==24740== 
> I guess this shouldn't happen?


Hello Andreas,
 Indeed it shouldn't but it seems to be happening. It think it is
because there are two processes and the parent kills the child, thus
creating a leak for valgrind. I didn't know how to fix that so I've
left it as it is until I get a better idea :) So I think it is
safe to ignore for now.

regards,
Nikos