valgrind: Adjust suppression for MSG_TRUNC with NULL buffer

valgrind complains if we pass a NULL buffer to recv(), even if we use
MSG_TRUNC, in which case it's actually safe.  For a long time we've had
a valgrind suppression for this.  It singles out the recv() in
tcp_sock_consume(), the only place we use MSG_TRUNC.

However, tcp_sock_consume() only has a single caller, which makes it a
prime candidate for inlining.  If inlined, it won't appear on the stack and
valgrind won't match the correct suppression.

It appears that certain compiler versions (for example gcc-13.2.1 in
Fedora 39) will inline this function even with the -O0 we use for valgrind
builds.  This breaks the suppression leading to a spurious failure in the
tests.

There's not really any way to adjust the suppression itself without making
it overly broad (we don't want to match other recv() calls).  So, as a hack
explicitly prevent inlining of this function when we're making a valgrind
build.  To accomplish this add an explicit -DVALGRIND when making such a
build.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Stefano Brivio <sbrivio@redhat.com>
This commit is contained in:
David Gibson 2023-11-16 20:15:58 +11:00 committed by Stefano Brivio
parent 457ff122e3
commit f7724647b1
3 changed files with 11 additions and 3 deletions

View file

@ -128,7 +128,7 @@ qrap: $(QRAP_SRCS) passt.h
valgrind: EXTRA_SYSCALLS += rt_sigprocmask rt_sigtimedwait rt_sigaction \ valgrind: EXTRA_SYSCALLS += rt_sigprocmask rt_sigtimedwait rt_sigaction \
getpid gettid kill clock_gettime mmap \ getpid gettid kill clock_gettime mmap \
munmap open unlink gettimeofday futex munmap open unlink gettimeofday futex
valgrind: FLAGS:=-g -O0 $(filter-out -O%,$(FLAGS)) valgrind: FLAGS:=-g -O0 $(filter-out -O%,$(FLAGS)) -DVALGRIND
valgrind: all valgrind: all
.PHONY: clean .PHONY: clean

9
tcp.c
View file

@ -2097,6 +2097,15 @@ static void tcp_conn_from_tap(struct ctx *c,
* *
* Return: 0 on success, negative error code from recv() on failure * Return: 0 on success, negative error code from recv() on failure
*/ */
#ifdef VALGRIND
/* valgrind doesn't realise that passing a NULL buffer to recv() is ok if using
* MSG_TRUNC. We have a suppression for this in the tests, but it relies on
* valgrind being able to see the tcp_sock_consume() stack frame, which it won't
* if this gets inlined. This has a single caller making it a likely inlining
* candidate, and certain compiler versions will do so even at -O0.
*/
__attribute__((noinline))
#endif /* VALGRIND */
static int tcp_sock_consume(const struct tcp_tap_conn *conn, uint32_t ack_seq) static int tcp_sock_consume(const struct tcp_tap_conn *conn, uint32_t ack_seq)
{ {
/* Simply ignore out-of-order ACKs: we already consumed the data we /* Simply ignore out-of-order ACKs: we already consumed the data we

View file

@ -3,7 +3,6 @@
passt_recv_MSG_TRUNC_into_NULL_buffer passt_recv_MSG_TRUNC_into_NULL_buffer
Memcheck:Param Memcheck:Param
socketcall.recvfrom(buf) socketcall.recvfrom(buf)
fun:recv
... ...
fun:tcp_sock_consume* fun:tcp_sock_consume
} }