samsung-sm8650/android_kernel_samsung_sm8650-modules
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

qcacmn: Cleanup assert in mgmt rx reo - Part 4

Browse Source 
Cleanup asserts in management Rx reorder module.

Change-Id: I37f13dc04356d20fe3fcce43ef8a0d48c8854e31
CRs-Fixed: 3593930
This commit is contained in:
Edayilliam Jayadev
2023-08-21 21:04:15 +05:30
committed by Rahul Choudhary
parent 845c861380
commit b1439341ca
1 changed files with 52 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

71
umac/cmn_services/mgmt_txrx/core/src/wlan_mgmt_txrx_rx_reo.c
View File

@@ -3074,8 +3074,11 @@ mgmt_rx_reo_release_egress_list_entries(struct mgmt_rx_reo_context *reo_context,
goto exit_unlock_frame_release_lock;
exit_unlock_egress_list_lock:
qdf_assert_always(qdf_list_size(&reo_egress_list->list) <=
reo_egress_list->max_list_size);
if (qdf_list_size(&reo_egress_list->list) >
reo_egress_list->max_list_size)
mgmt_rx_reo_err("Egress list overflow size =%u, max = %u",
qdf_list_size(&reo_egress_list->list),
reo_egress_list->max_list_size);
qdf_spin_unlock_bh(&reo_egress_list->list_lock);
exit_unlock_frame_release_lock:
qdf_spin_unlock(&reo_context->frame_release_lock);
@@ -3378,52 +3381,75 @@ mgmt_rx_reo_move_entries_ingress_to_egress_list
/* Check if ingress list has at least one frame ready to be delivered */
if (num_frames_ready_to_deliver) {
qdf_list_t temp_list_frames_ready_to_deliver;
qdf_list_t temp_list_frames_to_deliver;
qdf_list_create(&temp_list_frames_ready_to_deliver,
qdf_list_create(&temp_list_frames_to_deliver,
INGRESS_TO_EGRESS_MOVEMENT_TEMP_LIST_MAX_SIZE);
status = qdf_list_split(&temp_list_frames_ready_to_deliver,
status = qdf_list_split(&temp_list_frames_to_deliver,
&reo_ingress_list->list,
&latest_frame_ready_to_deliver->node);
qdf_assert_always(QDF_IS_STATUS_SUCCESS(status));
if (QDF_IS_STATUS_ERROR(status)) {
mgmt_rx_reo_err("Failed to split list");
qdf_list_destroy(&temp_list_frames_to_deliver);
goto exit_unlock_ingress_list;
}
qdf_assert_always(num_frames_ready_to_deliver ==
qdf_list_size(&temp_list_frames_ready_to_deliver));
if (num_frames_ready_to_deliver !=
qdf_list_size(&temp_list_frames_to_deliver)) {
uint32_t list_size;
list_size = qdf_list_size(&temp_list_frames_to_deliver);
mgmt_rx_reo_err("Mismatch in frames ready %u and %u",
num_frames_ready_to_deliver,
list_size);
status = QDF_STATUS_E_INVAL;
qdf_list_destroy(&temp_list_frames_to_deliver);
goto exit_unlock_ingress_list;
}
qdf_spin_lock_bh(&reo_egress_list->list_lock);
status = qdf_list_join(&reo_egress_list->list,
&temp_list_frames_ready_to_deliver);
qdf_assert_always(QDF_IS_STATUS_SUCCESS(status));
&temp_list_frames_to_deliver);
if (QDF_IS_STATUS_ERROR(status)) {
mgmt_rx_reo_err("Failed to join lists");
qdf_list_destroy(&temp_list_frames_to_deliver);
goto exit_unlock_egress_and_ingress_list;
}
if (mgmt_rx_reo_list_overflowed(reo_egress_list)) {
status =
mgmt_rx_reo_handle_egress_overflow(reo_egress_list);
if (QDF_IS_STATUS_ERROR(status)) {
mgmt_rx_reo_err("Failed to handle overflow");
qdf_assert_always(0);
qdf_list_destroy(&temp_list_frames_to_deliver);
goto exit_unlock_egress_and_ingress_list;
}
}
qdf_assert_always(qdf_list_size(&reo_ingress_list->list) <=
reo_ingress_list->max_list_size);
status = mgmt_rx_reo_check_sanity_lists(reo_egress_list,
reo_ingress_list);
if (QDF_IS_STATUS_ERROR(status)) {
mgmt_rx_reo_err("Sanity check of reo lists failed");
qdf_assert_always(0);
qdf_list_destroy(&temp_list_frames_to_deliver);
goto exit_unlock_egress_and_ingress_list;
}
qdf_spin_unlock_bh(&reo_egress_list->list_lock);
qdf_list_destroy(&temp_list_frames_ready_to_deliver);
qdf_list_destroy(&temp_list_frames_to_deliver);
}
status = QDF_STATUS_SUCCESS;
goto exit_unlock_ingress_list;
exit_unlock_egress_and_ingress_list:
qdf_spin_unlock_bh(&reo_egress_list->list_lock);
exit_unlock_ingress_list:
qdf_spin_unlock_bh(&reo_ingress_list->list_lock);
return QDF_STATUS_SUCCESS;
return status;
}
/**
@@ -3679,8 +3705,15 @@ mgmt_rx_reo_update_wait_count(
{
uint8_t link_id;
qdf_assert_always(wait_count_old_frame);
qdf_assert_always(wait_count_new_frame);
if (!wait_count_old_frame) {
mgmt_rx_reo_err("Pointer to old frame wait count is null");
return QDF_STATUS_E_NULL_VALUE;
}
if (!wait_count_new_frame) {
mgmt_rx_reo_err("Pointer to new frame wait count is null");
return QDF_STATUS_E_NULL_VALUE;
}
for (link_id = 0; link_id < MAX_MLO_LINKS; link_id++) {
if (wait_count_old_frame->per_link_count[link_id]) {