[PATCH 0/7] Fix issues and false positives in code coverge tool

newer
[PATCH 1/5] fwd: Clarify semantics...

older
[PATCH v8 01/14] dhcpv6: Fix reply...

Jon Maloy

9 Jul 2026 9 Jul '26

12:31 a.m.

This series fixes minor issues found by code coverage tools in the main passt code and in documentation examples. The core code changes addresses missing fd initialisations and explicit invariant checks for unsigned arithmetic. We also fix address buffer handling, type mismatches, uninitialised variables, and resource leaks in the documentation example code. Jon Maloy (7): passt: Initialise listening socket fds to -1 udp_vu: Assert iov_tail_clone() return before assigning to msg_iovlen tap: Assert IPv6 tail size before subtracting header length doc/migration: Use strncpy() for socket path and fix argv access doc/migration: Fix buffer type mismatch in recv() call doc/platform-requirements: Initialise va_list before va_start() doc/platform-requirements: Close leaked sockets in test_close_dup() doc/migration/source.c | 2 +- doc/migration/target.c | 10 ++++------ doc/platform-requirements/common.h | 2 +- doc/platform-requirements/udp-close-dup.c | 3 +++ passt.c | 3 +++ tap.c | 4 +++- udp_vu.c | 7 +++++-- 7 files changed, 20 insertions(+), 11 deletions(-) -- 2.52.0

Show replies by date

Jon Maloy

9 Jul 9 Jul

12:31 a.m.

New subject: [PATCH 2/7] udp_vu: Assert iov_tail_clone() return before assigning to msg_iovlen

iov_tail_clone() returns ssize_t, but its return value was assigned directly to msg.msg_iovlen which is size_t. Add an assert to document the invariant that the clone into a VIRTQUEUE_MAX_SIZE array cannot fail. Signed-off-by: Jon Maloy --- udp_vu.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/udp_vu.c b/udp_vu.c index e4fb1057..5d6f45ec 100644 --- a/udp_vu.c +++ b/udp_vu.c @@ -67,11 +67,14 @@ static ssize_t udp_vu_sock_recv(struct iov_tail *payload, size_t *cnt, int s) struct iovec msg_iov[VIRTQUEUE_MAX_SIZE]; struct msghdr msg = { 0 }; size_t iov_used; + ssize_t iovlen; ssize_t dlen; + iovlen = iov_tail_clone(msg_iov, ARRAY_SIZE(msg_iov), payload); + assert(iovlen >= 0); + msg.msg_iov = msg_iov; - msg.msg_iovlen = iov_tail_clone(msg.msg_iov, ARRAY_SIZE(msg_iov), - payload); + msg.msg_iovlen = iovlen; /* read data from the socket */ dlen = recvmsg(s, &msg, 0); -- 2.52.0

Jon Maloy

12:31 a.m.

New subject: [PATCH 1/7] passt: Initialise listening socket fds to -1

fd_tap_listen, fd_control_listen, and fd_repair_listen are file descriptors and should be initialised to -1 rather than the implicit 0, consistent with the other fd fields in passt_ctx. Signed-off-by: Jon Maloy --- passt.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/passt.c b/passt.c index 65a07d72..4e5b9289 100644 --- a/passt.c +++ b/passt.c @@ -65,6 +65,9 @@ char pkt_buf[PKT_BUF_BYTES] __attribute__ ((aligned(PAGE_SIZE))); struct ctx passt_ctx = { .pidfile_fd = -1, .fd_tap = -1, + .fd_tap_listen = -1, + .fd_control_listen = -1, + .fd_repair_listen = -1, .pasta_netns_fd = -1, .device_state_fd = -1, }; -- 2.52.0

Jon Maloy

12:31 a.m.

New subject: [PATCH 3/7] tap: Assert IPv6 tail size before subtracting header length

In tap6_handler(), iov_tail_size(&data) - sizeof(*ip6h) is used to compute the expected payload length. The subtraction is safe because IOV_PEEK_HEADER() already verified that the tail contains at least sizeof(*ip6h) bytes, but add an assert to make the invariant explicit. Signed-off-by: Jon Maloy --- tap.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tap.c b/tap.c index 6d93c7ce..d0d5148e 100644 --- a/tap.c +++ b/tap.c @@ -986,7 +986,9 @@ resume: if (!ip6h) continue; - check = iov_tail_size(&data) - sizeof(*ip6h); + check = iov_tail_size(&data); + assert(check >= sizeof(*ip6h)); + check -= sizeof(*ip6h); saddr = &ip6h->saddr; daddr = &ip6h->daddr; -- 2.52.0

Jon Maloy

12:32 a.m.

New subject: [PATCH 4/7] doc/migration: Use strncpy() for socket path and fix argv access

Replace strcpy() with strncpy() when copying the helper socket path into sun_path, preventing a potential buffer overflow from a long argument. In target.c, move the argc check before accessing argv[4], so we don't dereference past the end of argv. Signed-off-by: Jon Maloy --- doc/migration/source.c | 2 +- doc/migration/target.c | 8 +++----- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/doc/migration/source.c b/doc/migration/source.c index d44ebf1f..77ff8fda 100644 --- a/doc/migration/source.c +++ b/doc/migration/source.c @@ -47,7 +47,7 @@ int main(int argc, char **argv) return -1; } - strcpy(a_helper.sun_path, argv[4]); + strncpy(a_helper.sun_path, argv[4], sizeof(a_helper.sun_path) - 1); getaddrinfo(argv[1], argv[2], &hints, &r); /* Connect socket to server and send some data */ diff --git a/doc/migration/target.c b/doc/migration/target.c index f7d31083..0eb1f6ac 100644 --- a/doc/migration/target.c +++ b/doc/migration/target.c @@ -38,11 +38,6 @@ int main(int argc, char **argv) struct cmsghdr *cmsg = CMSG_FIRSTHDR(&msg); struct addrinfo *r; - (void)argc; - - strcpy(a_helper.sun_path, argv[4]); - getaddrinfo(argv[1], argv[2], &hints, &r); - if (argc != 7) { fprintf(stderr, "%s DST_ADDR DST_PORT SRC_PORT HELPER_PATH SSEQ RSEQ\n", @@ -50,6 +45,9 @@ int main(int argc, char **argv) return -1; } + strncpy(a_helper.sun_path, argv[4], sizeof(a_helper.sun_path) - 1); + getaddrinfo(argv[1], argv[2], &hints, &r); + /* Prepare socket, bind to source port */ s = socket(r->ai_family, SOCK_STREAM, IPPROTO_TCP); setsockopt(s, SOL_SOCKET, SO_REUSEADDR, &((int){ 1 }), sizeof(int)); -- 2.52.0

Jon Maloy

12:32 a.m.

New subject: [PATCH 5/7] doc/migration: Fix buffer type mismatch in recv() call

The compound literal passed to recv() was int (4 bytes) but only 1 byte is received. Use int8_t to match the length, consistent with the other recv() calls in this file and in source.c. Signed-off-by: Jon Maloy --- doc/migration/target.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/migration/target.c b/doc/migration/target.c index 0eb1f6ac..78cfd564 100644 --- a/doc/migration/target.c +++ b/doc/migration/target.c @@ -69,7 +69,7 @@ int main(int argc, char **argv) /* Send command to helper: turn repair mode on, wait for reply */ cmd = TCP_REPAIR_ON; sendmsg(s_helper, &msg, 0); - recv(s_helper, &((int){ 0 }), 1, 0); + recv(s_helper, &((int8_t){ 0 }), 1, 0); /* Set sending sequence */ seq = TCP_SEND_QUEUE; -- 2.52.0

Jon Maloy

12:32 a.m.

New subject: [PATCH 7/7] doc/platform-requirements: Close leaked sockets in test_close_dup()

s1 and send_s were not closed before returning, leaking file descriptors across the loop iterations in main(). Signed-off-by: Jon Maloy --- doc/platform-requirements/udp-close-dup.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/platform-requirements/udp-close-dup.c b/doc/platform-requirements/udp-close-dup.c index 99060fcb..abb719a5 100644 --- a/doc/platform-requirements/udp-close-dup.c +++ b/doc/platform-requirements/udp-close-dup.c @@ -87,6 +87,9 @@ static void test_close_dup(enum dup_method method) token = random(); send_token(send_s, token); recv_token(s1, token); + + close(s1); + close(send_s); } int main(int argc, char *argv[]) -- 2.52.0

Jon Maloy

12:32 a.m.

New subject: [PATCH 6/7] doc/platform-requirements: Initialise va_list before va_start()

va_start() initialises the va_list, but zero-initialise it at declaration as well to make it clear the variable is not used uninitialised. Signed-off-by: Jon Maloy --- doc/platform-requirements/common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/platform-requirements/common.h b/doc/platform-requirements/common.h index e85fc2b5..815e8271 100644 --- a/doc/platform-requirements/common.h +++ b/doc/platform-requirements/common.h @@ -18,7 +18,7 @@ __attribute__((format(printf, 1, 2), noreturn)) static inline void die(const char *fmt, ...) { - va_list ap; + va_list ap = { 0 }; va_start(ap, fmt); (void)vfprintf(stderr, fmt, ap); -- 2.52.0

David Gibson

4:13 a.m.

New subject: [PATCH 3/7] tap: Assert IPv6 tail size before subtracting header length

On Wed, Jul 08, 2026 at 06:31:59PM -0400, Jon Maloy wrote:

...

In tap6_handler(), iov_tail_size(&data) - sizeof(*ip6h) is used to compute the expected payload length. The subtraction is safe because IOV_PEEK_HEADER() already verified that the tail contains at least sizeof(*ip6h) bytes, but add an assert to make the invariant explicit.

Again, I think "assumption" would be clearer. I think of "invariant" as something that applies at every iteration of a loop, or to a data structure at all times, rather than just to a particular variable at a particular point

...

Signed-off-by: Jon Maloy

Reviewed-by: David Gibson

...

--- tap.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tap.c b/tap.c index 6d93c7ce..d0d5148e 100644 --- a/tap.c +++ b/tap.c @@ -986,7 +986,9 @@ resume: if (!ip6h) continue;

- check = iov_tail_size(&data) - sizeof(*ip6h); + check = iov_tail_size(&data); + assert(check >= sizeof(*ip6h)); + check -= sizeof(*ip6h);

saddr = &ip6h->saddr; daddr = &ip6h->daddr; -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:20 a.m.

New subject: [PATCH 4/7] doc/migration: Use strncpy() for socket path and fix argv access

On Wed, Jul 08, 2026 at 06:32:00PM -0400, Jon Maloy wrote:

...

Replace strcpy() with strncpy() when copying the helper socket path into sun_path, preventing a potential buffer overflow from a long argument.

In target.c, move the argc check before accessing argv[4], so we don't dereference past the end of argv.

This fixes real bugs in the examples, but.. strncpy() does not guarantee the target is \0-terminated. In this case we will end up \0-terminated, because we limit the strncpy() to exclude the last byte of the target buffer, which is initialized to all 0s. That's rather more riskly subtle than I'd like. In the passt code proper we usually use snprintf() rather than strncpy() to move an arbitrary length string into a fixed sized buffer like (including sun_path in several cases). It's strictly speaking more heavyweight than we need, but the meaning is pretty clear and it guarantees \0 termination. I'd suggest using that instead.

...

Signed-off-by: Jon Maloy --- doc/migration/source.c | 2 +- doc/migration/target.c | 8 +++----- 2 files changed, 4 insertions(+), 6 deletions(-)

diff --git a/doc/migration/source.c b/doc/migration/source.c index d44ebf1f..77ff8fda 100644 --- a/doc/migration/source.c +++ b/doc/migration/source.c @@ -47,7 +47,7 @@ int main(int argc, char **argv) return -1; }

- strcpy(a_helper.sun_path, argv[4]); + strncpy(a_helper.sun_path, argv[4], sizeof(a_helper.sun_path) - 1); getaddrinfo(argv[1], argv[2], &hints, &r);

/* Connect socket to server and send some data */ diff --git a/doc/migration/target.c b/doc/migration/target.c index f7d31083..0eb1f6ac 100644 --- a/doc/migration/target.c +++ b/doc/migration/target.c @@ -38,11 +38,6 @@ int main(int argc, char **argv) struct cmsghdr *cmsg = CMSG_FIRSTHDR(&msg); struct addrinfo *r;

- (void)argc; - - strcpy(a_helper.sun_path, argv[4]); - getaddrinfo(argv[1], argv[2], &hints, &r); - if (argc != 7) { fprintf(stderr, "%s DST_ADDR DST_PORT SRC_PORT HELPER_PATH SSEQ RSEQ\n", @@ -50,6 +45,9 @@ int main(int argc, char **argv) return -1; }

+ strncpy(a_helper.sun_path, argv[4], sizeof(a_helper.sun_path) - 1); + getaddrinfo(argv[1], argv[2], &hints, &r); + /* Prepare socket, bind to source port */ s = socket(r->ai_family, SOCK_STREAM, IPPROTO_TCP); setsockopt(s, SOL_SOCKET, SO_REUSEADDR, &((int){ 1 }), sizeof(int)); -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:09 a.m.

On Wed, Jul 08, 2026 at 06:31:56PM -0400, Jon Maloy wrote:

...

This series fixes minor issues found by code coverage tools in the main passt code and in documentation examples.

Nit: If you're talking about the tool I think you are, it's not really a "code coverage" tool. Static anlaysis, sure, but "coverage" implies it's about measuring the amount of the codebase tested in some way.

...

The core code changes addresses missing fd initialisations and explicit invariant checks for unsigned arithmetic.

We also fix address buffer handling, type mismatches, uninitialised variables, and resource leaks in the documentation example code.

Jon Maloy (7): passt: Initialise listening socket fds to -1 udp_vu: Assert iov_tail_clone() return before assigning to msg_iovlen tap: Assert IPv6 tail size before subtracting header length doc/migration: Use strncpy() for socket path and fix argv access doc/migration: Fix buffer type mismatch in recv() call doc/platform-requirements: Initialise va_list before va_start() doc/platform-requirements: Close leaked sockets in test_close_dup()

doc/migration/source.c | 2 +- doc/migration/target.c | 10 ++++------ doc/platform-requirements/common.h | 2 +- doc/platform-requirements/udp-close-dup.c | 3 +++ passt.c | 3 +++ tap.c | 4 +++- udp_vu.c | 7 +++++-- 7 files changed, 20 insertions(+), 11 deletions(-)

-- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:09 a.m.

New subject: [PATCH 1/7] passt: Initialise listening socket fds to -1

On Wed, Jul 08, 2026 at 06:31:57PM -0400, Jon Maloy wrote:

...

fd_tap_listen, fd_control_listen, and fd_repair_listen are file descriptors and should be initialised to -1 rather than the implicit 0, consistent with the other fd fields in passt_ctx.

Signed-off-by: Jon Maloy

Reviewed-by: David Gibson

...

--- passt.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/passt.c b/passt.c index 65a07d72..4e5b9289 100644 --- a/passt.c +++ b/passt.c @@ -65,6 +65,9 @@ char pkt_buf[PKT_BUF_BYTES] __attribute__ ((aligned(PAGE_SIZE))); struct ctx passt_ctx = { .pidfile_fd = -1, .fd_tap = -1, + .fd_tap_listen = -1, + .fd_control_listen = -1, + .fd_repair_listen = -1, .pasta_netns_fd = -1, .device_state_fd = -1, }; -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:11 a.m.

New subject: [PATCH 2/7] udp_vu: Assert iov_tail_clone() return before assigning to msg_iovlen

On Wed, Jul 08, 2026 at 06:31:58PM -0400, Jon Maloy wrote:

...

iov_tail_clone() returns ssize_t, but its return value was assigned directly to msg.msg_iovlen which is size_t. Add an assert to document the invariant that the clone into a VIRTQUEUE_MAX_SIZE array cannot

Nit: I think s/invariant/assumption/ would be clearer.

...

fail.

Signed-off-by: Jon Maloy

Reviewed-by: David Gibson

...

--- udp_vu.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/udp_vu.c b/udp_vu.c index e4fb1057..5d6f45ec 100644 --- a/udp_vu.c +++ b/udp_vu.c @@ -67,11 +67,14 @@ static ssize_t udp_vu_sock_recv(struct iov_tail *payload, size_t *cnt, int s) struct iovec msg_iov[VIRTQUEUE_MAX_SIZE]; struct msghdr msg = { 0 }; size_t iov_used; + ssize_t iovlen; ssize_t dlen;

+ iovlen = iov_tail_clone(msg_iov, ARRAY_SIZE(msg_iov), payload); + assert(iovlen >= 0); + msg.msg_iov = msg_iov; - msg.msg_iovlen = iov_tail_clone(msg.msg_iov, ARRAY_SIZE(msg_iov), - payload); + msg.msg_iovlen = iovlen;

/* read data from the socket */ dlen = recvmsg(s, &msg, 0); -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:41 a.m.

New subject: [PATCH 7/7] doc/platform-requirements: Close leaked sockets in test_close_dup()

On Wed, Jul 08, 2026 at 06:32:03PM -0400, Jon Maloy wrote:

...

s1 and send_s were not closed before returning, leaking file descriptors across the loop iterations in main().

Signed-off-by: Jon Maloy

Insignificant as a bug, since the whole program is over in milliseconds. But might as well make the checkers happy. Reviewed-by: David Gibson

...

--- doc/platform-requirements/udp-close-dup.c | 3 +++ 1 file changed, 3 insertions(+)

diff --git a/doc/platform-requirements/udp-close-dup.c b/doc/platform-requirements/udp-close-dup.c index 99060fcb..abb719a5 100644 --- a/doc/platform-requirements/udp-close-dup.c +++ b/doc/platform-requirements/udp-close-dup.c @@ -87,6 +87,9 @@ static void test_close_dup(enum dup_method method) token = random(); send_token(send_s, token); recv_token(s1, token); + + close(s1); + close(send_s); }

int main(int argc, char *argv[]) -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:22 a.m.

New subject: [PATCH 5/7] doc/migration: Fix buffer type mismatch in recv() call

On Wed, Jul 08, 2026 at 06:32:01PM -0400, Jon Maloy wrote:

...

The compound literal passed to recv() was int (4 bytes) but only 1 byte is received. Use int8_t to match the length, consistent with the other recv() calls in this file and in source.c.

Signed-off-by: Jon Maloy

Oops. Reviewed-by: David Gibson Although, arguably to act as better documentation it should actually check the value we get back from passt-repair.

...

--- doc/migration/target.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/migration/target.c b/doc/migration/target.c index 0eb1f6ac..78cfd564 100644 --- a/doc/migration/target.c +++ b/doc/migration/target.c @@ -69,7 +69,7 @@ int main(int argc, char **argv) /* Send command to helper: turn repair mode on, wait for reply */ cmd = TCP_REPAIR_ON; sendmsg(s_helper, &msg, 0); - recv(s_helper, &((int){ 0 }), 1, 0); + recv(s_helper, &((int8_t){ 0 }), 1, 0);

/* Set sending sequence */ seq = TCP_SEND_QUEUE; -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

David Gibson

4:40 a.m.

New subject: [PATCH 6/7] doc/platform-requirements: Initialise va_list before va_start()

On Wed, Jul 08, 2026 at 06:32:02PM -0400, Jon Maloy wrote:

...

va_start() initialises the va_list, but zero-initialise it at declaration as well to make it clear the variable is not used uninitialised.

I really dislike this approach to suppressing uninitialised variable warnings. A decent tool should be able to tell that va_start() initialises ap, and pre-initialising means that tool will no longer warning if we accidentally deleted the va_start(). Is there really no other way for the warning in question?

...

Signed-off-by: Jon Maloy --- doc/platform-requirements/common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/platform-requirements/common.h b/doc/platform-requirements/common.h index e85fc2b5..815e8271 100644 --- a/doc/platform-requirements/common.h +++ b/doc/platform-requirements/common.h @@ -18,7 +18,7 @@ __attribute__((format(printf, 1, 2), noreturn)) static inline void die(const char *fmt, ...) { - va_list ap; + va_list ap = { 0 };

va_start(ap, fmt); (void)vfprintf(stderr, fmt, ap); -- 2.52.0

-- David Gibson (he or they) | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you, not the other way | around. http://www.ozlabs.org/~dgibson

Jon Maloy

6:49 p.m.

New subject: [PATCH 6/7] doc/platform-requirements: Initialise va_list before va_start()

On 2026-07-08 22:40, David Gibson wrote:

...

On Wed, Jul 08, 2026 at 06:32:02PM -0400, Jon Maloy wrote:

...
va_start() initialises the va_list, but zero-initialise it at declaration as well to make it clear the variable is not used uninitialised.

I really dislike this approach to suppressing uninitialised variable warnings. A decent tool should be able to tell that va_start() initialises ap, and pre-initialising means that tool will no longer warning if we accidentally deleted the va_start().

Is there really no other way for the warning in question?

TBH I dislike it too. I still decided to post it just to see the feedback. I cannot see any other way to get rid of this warning, but maybe somebody else has? /jon

...

...
Signed-off-by: Jon Maloy --- doc/platform-requirements/common.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/platform-requirements/common.h b/doc/platform-requirements/common.h index e85fc2b5..815e8271 100644 --- a/doc/platform-requirements/common.h +++ b/doc/platform-requirements/common.h @@ -18,7 +18,7 @@ __attribute__((format(printf, 1, 2), noreturn)) static inline void die(const char *fmt, ...) { - va_list ap; + va_list ap = { 0 };

va_start(ap, fmt); (void)vfprintf(stderr, fmt, ap); -- 2.52.0

Stefano Brivio

6:59 p.m.

New subject: [PATCH 6/7] doc/platform-requirements: Initialise va_list before va_start()

On Thu, 9 Jul 2026 12:49:10 -0400 Jon Maloy wrote:

...

On 2026-07-08 22:40, David Gibson wrote:

...
On Wed, Jul 08, 2026 at 06:32:02PM -0400, Jon Maloy wrote:

...
va_start() initialises the va_list, but zero-initialise it at declaration as well to make it clear the variable is not used uninitialised.

I really dislike this approach to suppressing uninitialised variable warnings. A decent tool should be able to tell that va_start() initialises ap, and pre-initialising means that tool will no longer warning if we accidentally deleted the va_start().

Is there really no other way for the warning in question?

TBH I dislike it too. I still decided to post it just to see the feedback. I cannot see any other way to get rid of this warning, but maybe somebody else has?

I don't think we need to get rid of any warnings here as we don't run static checkers on the code in platform-requirements/ anyway, it's just there as documentation, proof of concept, or test code. -- Stefano

Age (days ago)

Last active (days ago)

List overview

Download

17 comments

3 participants

participants (3)

David Gibson
Jon Maloy
Stefano Brivio

[PATCH 0/7] Fix issues and false positives in code coverge tool

tags

participants (3)