Release 2.3.0 by Chandrakanth468 · Pull Request #49 · rdkcentral/hotspot

Chandrakanth468 · 2026-04-23T14:25:02Z

Release 2.3.0

Release 2.2.0 2.2.0

… Failure (#32) 1. dhcpsnooper in hotspot component monitors the DHCP exchange process for newly associated hotspot clients. If a client successfully associates with the Access Point (AP) but does not receive a valid IP address assignment from the DHCP server within a predefined period and automatically trigger a de-authentication for that client. 2. Implement a data model( Device.X_COMCAST-COM_GRE.Hotspot.RejectAssociatedClient)for the Rbus event, which will be triggered whenever a client's DHCP transaction fails to assign an IP address within a specified period (i.e., a timeout). --------- Signed-off-by: raghavendra.hegde2@comcast.com

Copilot

Pull request overview

Note

Copilot was unable to run its full agentic suite in this review.

Release 2.3.0 adds Hotspot client deauthentication support on DHCP failure by tracking per-client DHCP state and publishing an RBUS event to OneWifi, along with accompanying unit tests and changelog updates.

Changes:

Added DHCP-state tracking per client (linked list) and timeout-based deauth trigger that publishes an RBUS event.
Extended client bookkeeping to carry vapIndex and updated RSSI update API accordingly.
Added unit tests for client-state helpers, RBUS publish helper, and DHCP timer state logic; updated CHANGELOG.md for 2.3.0.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
source/test/mocks/hotspotfd_internal_mock.h	Exposes new global `client_list` for unit tests.
source/test/HotspotFdTest/HotspotFdTest.cpp	Adds unit tests and shared cleanup helper for DHCP client-state logic and RBUS publishing.
source/hotspotfd/include/dhcpsnooper.h	Introduces DHCP client state structs + APIs, adds `vapIndex`, updates function signatures.
source/hotspotfd/hotspotfd.c	Adds RBUS data element registration + init for the RejectAssociatedClient event.
source/hotspotfd/dhcpsnooper.c	Implements RBUS publish helper, client-state list management, and DHCP timer/deauth logic in packet handler.
source/hotspotfd/cosa_hotspot_dml.c	Updates caller to pass `vapIndex` into RSSI update API.
CHANGELOG.md	Adds 2.3.0 release notes entry.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-04-23T14:29:26Z

+    msg_debug("Command string: %s - %s\n", cmdStr, __FUNCTION__);
+
+    if (cmdStr == NULL || strlen(cmdStr) == 0) {
+        msg_debug("mac address empty and not sending rbus event to onewifi: %d\n", __LINE__);
+        return 0;
+    }
+


cmdStr is logged with %s before it is null-checked; calling publish_to_onewifi(NULL) will dereference a NULL pointer in the log call. Move the log after the NULL/empty check or log safely (e.g., conditional string or %p).

Suggested change

msg_debug("Command string: %s - %s\n", cmdStr, __FUNCTION__);

if (cmdStr == NULL || strlen(cmdStr) == 0) {

msg_debug("mac address empty and not sending rbus event to onewifi: %d\n", __LINE__);

return 0;

}

if (cmdStr == NULL || strlen(cmdStr) == 0) {

msg_debug("mac address empty and not sending rbus event to onewifi: %d\n", __LINE__);

return 0;

}

msg_debug("Command string: %s - %s\n", cmdStr, __FUNCTION__);

Copilot · 2026-04-23T14:29:27Z

+    }
+}
+
+static void constructCommand(char *macaddr, char *cmd)


constructCommand copies into cmd using a fixed length derived from macaddr_with_index (63 bytes) without knowing the capacity of cmd, and it doesn’t guarantee cmd is NUL-terminated. This can overflow the caller’s buffer and/or produce unterminated strings. Update the API to accept cmd length (e.g., constructCommand(..., char* cmd, size_t cmd_len)) and use bounded formatting/copying into cmd with explicit termination.

Copilot · 2026-04-23T14:29:27Z

+        snprintf(macaddr_with_index, sizeof(macaddr_with_index), "%s_%d", macaddr, pNewClient->client.vapIndex);
+        macaddr_with_index[sizeof(macaddr_with_index) - 1] = '\0';
+    }
+    strncpy(cmd, macaddr_with_index, sizeof(macaddr_with_index) - 1);
+    msg_debug("constructCommand :%s\n", cmd);
+    pthread_mutex_unlock(&global_stats_mutex);


constructCommand copies into cmd using a fixed length derived from macaddr_with_index (63 bytes) without knowing the capacity of cmd, and it doesn’t guarantee cmd is NUL-terminated. This can overflow the caller’s buffer and/or produce unterminated strings. Update the API to accept cmd length (e.g., constructCommand(..., char* cmd, size_t cmd_len)) and use bounded formatting/copying into cmd with explicit termination.

Copilot · 2026-04-23T14:29:27Z

+    clock_gettime(CLOCK_MONOTONIC, &now);
+    return (now.tv_sec - start.tv_sec) * 1000 + (now.tv_nsec - start.tv_nsec) / 1000000;


This elapsed-time calculation is incorrect when now.tv_nsec < start.tv_nsec (common across second boundaries), producing a negative millisecond component and inaccurate totals. Normalize the timespec delta (borrow 1s when nsec diff is negative) or compute using a single nanosecond total with proper signed arithmetic before converting to ms.

Suggested change

clock_gettime(CLOCK_MONOTONIC, &now);

return (now.tv_sec - start.tv_sec) * 1000 + (now.tv_nsec - start.tv_nsec) / 1000000;

time_t sec_diff;

long nsec_diff;

clock_gettime(CLOCK_MONOTONIC, &now);

sec_diff = now.tv_sec - start.tv_sec;

nsec_diff = now.tv_nsec - start.tv_nsec;

if (nsec_diff < 0) {

sec_diff--;

nsec_diff += 1000000000L;

}

return sec_diff * 1000 + nsec_diff / 1000000;

Copilot · 2026-04-23T14:29:27Z

+                msg_debug("All DHCP states completed with ACK IP 172.20.20.20 in %ld ms. Timer stopped.\n", elapsed_time);
+                state->timer_running = false;
+                state->dhcp_discover = state->dhcp_request = state->dhcp_offer = state->dhcp_ack = false;
+                remove_client_state(mac_str);


remove_client_state(mac_str) frees the list node backing state, leaving state as a dangling pointer for the remainder of snoop_packetHandler. Even if it’s not currently used afterward, this is fragile and can easily become a use-after-free with future edits. Set state = NULL immediately after removal (and avoid any subsequent reads), or defer removal until the end of the handler.

Suggested change

remove_client_state(mac_str);

remove_client_state(mac_str);

state = NULL;

Copilot · 2026-04-23T14:29:28Z

+    event.data = data;
+    event.type = RBUS_EVENT_GENERAL;
+
+    rc = rbusEvent_Publish(disassoc_rbus_handle, &event);


disassoc_rbus_handle is used unconditionally here, but RBUS initialization in hotspotDisassocClientRbusInit() can fail (RBUS not available / open failure / reg failure). Publishing with an uninitialized/invalid handle can fail unpredictably (and may crash depending on RBUS implementation). Add a guard in publish_to_onewifi (e.g., check handle validity and/or a “rbus ready” flag set only on successful init) and return a failure code early if RBUS isn’t initialized.

Copilot · 2026-04-23T14:29:28Z

+// Helper to clean up the client_list linked list between tests
+static void cleanup_client_list() {
+    client_node_t *cur = client_list;
+    while (cur) {
+        client_node_t *next = cur->next;
+        free(cur);
+        cur = next;
+    }
+    client_list = NULL;
+}


Many tests call cleanup_client_list() manually at the end. If any test hits an ASSERT_* and returns early, cleanup won’t run, leaking global state into subsequent tests and causing order-dependent failures. Prefer moving this cleanup into the fixture’s TearDown() (or using a scoped cleanup guard) so it always executes, and remove per-test cleanup calls.

bunnam988 and others added 3 commits April 1, 2026 07:42

Merge tag '2.2.0' into develop

1c7c790

Release 2.2.0 2.2.0

Add changelog for release 2.3.0

b9ea302

Copilot AI review requested due to automatic review settings April 23, 2026 14:25

Chandrakanth468 requested review from a team as code owners April 23, 2026 14:25

Chandrakanth468 closed this Apr 23, 2026

Chandrakanth468 deleted the release/2.3.0 branch April 23, 2026 14:25

Copilot AI reviewed Apr 23, 2026

View reviewed changes

github-actions Bot locked and limited conversation to collaborators Apr 23, 2026

Copilot started reviewing on behalf of Chandrakanth468 April 23, 2026 14:32 View session

Copilot started reviewing on behalf of Chandrakanth468 April 23, 2026 14:33 View session

Copilot started reviewing on behalf of Chandrakanth468 April 23, 2026 14:34 View session

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Release 2.3.0#49

Release 2.3.0#49
Chandrakanth468 wants to merge 3 commits intomainfrom
release/2.3.0

Chandrakanth468 commented Apr 23, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Copilot AI Apr 23, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

		clock_gettime(CLOCK_MONOTONIC, &now);
		return (now.tv_sec - start.tv_sec) * 1000 + (now.tv_nsec - start.tv_nsec) / 1000000;

-    clock_gettime(CLOCK_MONOTONIC, &now);
-    return (now.tv_sec - start.tv_sec) * 1000 + (now.tv_nsec - start.tv_nsec) / 1000000;
+    time_t sec_diff;
+    long nsec_diff;
+    clock_gettime(CLOCK_MONOTONIC, &now);
+    sec_diff = now.tv_sec - start.tv_sec;
+    nsec_diff = now.tv_nsec - start.tv_nsec;
+    if (nsec_diff < 0) {
+        sec_diff--;
+        nsec_diff += 1000000000L;
+    }
+    return sec_diff * 1000 + nsec_diff / 1000000;

	remove_client_state(mac_str);
	remove_client_state(mac_str);
	state = NULL;

Conversation

Chandrakanth468 commented Apr 23, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants