Commit 592d42ac authored by David S. Miller's avatar David S. Miller

Merge branch 'qed-IOV-cleanups'

Yuval Mintz says:

====================
qed: IOV related clenaups

This patch series targets IOV functionality [on both PF and VF].

Patches #2, #3 and #5 fix flows relating to malicious VFs, either by
upgrading and aligning current safe-guards or by correcing racy flows.

Patches #1 and #8 make some malicious/dysnfunctional VFs logging appear
by default in logs.

The rest of the patches either cleanup the existing code or else correct
some possible [yet fairly insignicant] issues in VF behavior.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parents f9fe1c12 e50728ef
......@@ -178,26 +178,59 @@ static struct qed_vf_info *qed_iov_get_vf_info(struct qed_hwfn *p_hwfn,
return vf;
}
enum qed_iov_validate_q_mode {
QED_IOV_VALIDATE_Q_NA,
QED_IOV_VALIDATE_Q_ENABLE,
QED_IOV_VALIDATE_Q_DISABLE,
};
static bool qed_iov_validate_queue_mode(struct qed_hwfn *p_hwfn,
struct qed_vf_info *p_vf,
u16 qid,
enum qed_iov_validate_q_mode mode,
bool b_is_tx)
{
if (mode == QED_IOV_VALIDATE_Q_NA)
return true;
if ((b_is_tx && p_vf->vf_queues[qid].p_tx_cid) ||
(!b_is_tx && p_vf->vf_queues[qid].p_rx_cid))
return mode == QED_IOV_VALIDATE_Q_ENABLE;
/* In case we haven't found any valid cid, then its disabled */
return mode == QED_IOV_VALIDATE_Q_DISABLE;
}
static bool qed_iov_validate_rxq(struct qed_hwfn *p_hwfn,
struct qed_vf_info *p_vf, u16 rx_qid)
struct qed_vf_info *p_vf,
u16 rx_qid,
enum qed_iov_validate_q_mode mode)
{
if (rx_qid >= p_vf->num_rxqs)
if (rx_qid >= p_vf->num_rxqs) {
DP_VERBOSE(p_hwfn,
QED_MSG_IOV,
"VF[0x%02x] - can't touch Rx queue[%04x]; Only 0x%04x are allocated\n",
p_vf->abs_vf_id, rx_qid, p_vf->num_rxqs);
return rx_qid < p_vf->num_rxqs;
return false;
}
return qed_iov_validate_queue_mode(p_hwfn, p_vf, rx_qid, mode, false);
}
static bool qed_iov_validate_txq(struct qed_hwfn *p_hwfn,
struct qed_vf_info *p_vf, u16 tx_qid)
struct qed_vf_info *p_vf,
u16 tx_qid,
enum qed_iov_validate_q_mode mode)
{
if (tx_qid >= p_vf->num_txqs)
if (tx_qid >= p_vf->num_txqs) {
DP_VERBOSE(p_hwfn,
QED_MSG_IOV,
"VF[0x%02x] - can't touch Tx queue[%04x]; Only 0x%04x are allocated\n",
p_vf->abs_vf_id, tx_qid, p_vf->num_txqs);
return tx_qid < p_vf->num_txqs;
return false;
}
return qed_iov_validate_queue_mode(p_hwfn, p_vf, tx_qid, mode, true);
}
static bool qed_iov_validate_sb(struct qed_hwfn *p_hwfn,
......@@ -217,6 +250,34 @@ static bool qed_iov_validate_sb(struct qed_hwfn *p_hwfn,
return false;
}
static bool qed_iov_validate_active_rxq(struct qed_hwfn *p_hwfn,
struct qed_vf_info *p_vf)
{
u8 i;
for (i = 0; i < p_vf->num_rxqs; i++)
if (qed_iov_validate_queue_mode(p_hwfn, p_vf, i,
QED_IOV_VALIDATE_Q_ENABLE,
false))
return true;
return false;
}
static bool qed_iov_validate_active_txq(struct qed_hwfn *p_hwfn,
struct qed_vf_info *p_vf)
{
u8 i;
for (i = 0; i < p_vf->num_txqs; i++)
if (qed_iov_validate_queue_mode(p_hwfn, p_vf, i,
QED_IOV_VALIDATE_Q_ENABLE,
true))
return true;
return false;
}
static int qed_iov_post_vf_bulletin(struct qed_hwfn *p_hwfn,
int vfid, struct qed_ptt *p_ptt)
{
......@@ -693,6 +754,11 @@ static int qed_iov_enable_vf_access(struct qed_hwfn *p_hwfn,
u32 igu_vf_conf = IGU_VF_CONF_FUNC_EN;
int rc;
/* It's possible VF was previously considered malicious -
* clear the indication even if we're only going to disable VF.
*/
vf->b_malicious = false;
if (vf->to_disable)
return 0;
......@@ -705,9 +771,6 @@ static int qed_iov_enable_vf_access(struct qed_hwfn *p_hwfn,
qed_iov_vf_igu_reset(p_hwfn, p_ptt, vf);
/* It's possible VF was previously considered malicious */
vf->b_malicious = false;
rc = qed_mcp_config_vf_msix(p_hwfn, p_ptt, vf->abs_vf_id, vf->num_sbs);
if (rc)
return rc;
......@@ -1134,13 +1197,17 @@ static void qed_iov_send_response(struct qed_hwfn *p_hwfn,
(sizeof(union pfvf_tlvs) - sizeof(u64)) / 4,
&params);
qed_dmae_host2host(p_hwfn, p_ptt, mbx->reply_phys,
mbx->req_virt->first_tlv.reply_address,
sizeof(u64) / 4, &params);
/* Once PF copies the rc to the VF, the latter can continue
* and send an additional message. So we have to make sure the
* channel would be re-set to ready prior to that.
*/
REG_WR(p_hwfn,
GTT_BAR0_MAP_REG_USDM_RAM +
USTORM_VF_PF_CHANNEL_READY_OFFSET(eng_vf_id), 1);
qed_dmae_host2host(p_hwfn, p_ptt, mbx->reply_phys,
mbx->req_virt->first_tlv.reply_address,
sizeof(u64) / 4, &params);
}
static u16 qed_iov_vport_to_tlv(struct qed_hwfn *p_hwfn,
......@@ -1749,6 +1816,8 @@ static void qed_iov_vf_mbx_start_vport(struct qed_hwfn *p_hwfn,
vf->state = VF_ENABLED;
start = &mbx->req_virt->start_vport;
qed_iov_enable_vf_traffic(p_hwfn, p_ptt, vf);
/* Initialize Status block in CAU */
for (sb_id = 0; sb_id < vf->num_sbs; sb_id++) {
if (!start->sb_addr[sb_id]) {
......@@ -1762,7 +1831,6 @@ static void qed_iov_vf_mbx_start_vport(struct qed_hwfn *p_hwfn,
start->sb_addr[sb_id],
vf->igu_sbs[sb_id], vf->abs_vf_id, 1);
}
qed_iov_enable_vf_traffic(p_hwfn, p_ptt, vf);
vf->mtu = start->mtu;
vf->shadow_config.inner_vlan_removal = start->inner_vlan_removal;
......@@ -1819,6 +1887,16 @@ static void qed_iov_vf_mbx_stop_vport(struct qed_hwfn *p_hwfn,
vf->vport_instance--;
vf->spoof_chk = false;
if ((qed_iov_validate_active_rxq(p_hwfn, vf)) ||
(qed_iov_validate_active_txq(p_hwfn, vf))) {
vf->b_malicious = true;
DP_NOTICE(p_hwfn,
"VF [%02x] - considered malicious; Unable to stop RX/TX queuess\n",
vf->abs_vf_id);
status = PFVF_STATUS_MALICIOUS;
goto out;
}
rc = qed_sp_vport_stop(p_hwfn, vf->opaque_fid, vf->vport_id);
if (rc) {
DP_ERR(p_hwfn, "qed_iov_vf_mbx_stop_vport returned error %d\n",
......@@ -1830,6 +1908,7 @@ static void qed_iov_vf_mbx_stop_vport(struct qed_hwfn *p_hwfn,
vf->configured_features = 0;
memset(&vf->shadow_config, 0, sizeof(vf->shadow_config));
out:
qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_VPORT_TEARDOWN,
sizeof(struct pfvf_def_resp_tlv), status);
}
......@@ -1886,7 +1965,8 @@ static void qed_iov_vf_mbx_start_rxq(struct qed_hwfn *p_hwfn,
req = &mbx->req_virt->start_rxq;
if (!qed_iov_validate_rxq(p_hwfn, vf, req->rx_qid) ||
if (!qed_iov_validate_rxq(p_hwfn, vf, req->rx_qid,
QED_IOV_VALIDATE_Q_DISABLE) ||
!qed_iov_validate_sb(p_hwfn, vf, req->hw_sb))
goto out;
......@@ -2000,7 +2080,8 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn,
memset(&params, 0, sizeof(params));
req = &mbx->req_virt->start_txq;
if (!qed_iov_validate_txq(p_hwfn, vf, req->tx_qid) ||
if (!qed_iov_validate_txq(p_hwfn, vf, req->tx_qid,
QED_IOV_VALIDATE_Q_DISABLE) ||
!qed_iov_validate_sb(p_hwfn, vf, req->hw_sb))
goto out;
......@@ -2037,57 +2118,53 @@ static void qed_iov_vf_mbx_start_txq(struct qed_hwfn *p_hwfn,
static int qed_iov_vf_stop_rxqs(struct qed_hwfn *p_hwfn,
struct qed_vf_info *vf,
u16 rxq_id, u8 num_rxqs, bool cqe_completion)
u16 rxq_id, bool cqe_completion)
{
struct qed_vf_q_info *p_queue;
int rc = 0;
int qid;
if (rxq_id + num_rxqs > ARRAY_SIZE(vf->vf_queues))
if (!qed_iov_validate_rxq(p_hwfn, vf, rxq_id,
QED_IOV_VALIDATE_Q_ENABLE)) {
DP_VERBOSE(p_hwfn,
QED_MSG_IOV,
"VF[%d] Tried Closing Rx 0x%04x which is inactive\n",
vf->relative_vf_id, rxq_id);
return -EINVAL;
}
for (qid = rxq_id; qid < rxq_id + num_rxqs; qid++) {
p_queue = &vf->vf_queues[qid];
p_queue = &vf->vf_queues[rxq_id];
if (!p_queue->p_rx_cid)
continue;
rc = qed_eth_rx_queue_stop(p_hwfn,
p_queue->p_rx_cid,
false, cqe_completion);
if (rc)
return rc;
rc = qed_eth_rx_queue_stop(p_hwfn,
p_queue->p_rx_cid,
false, cqe_completion);
if (rc)
return rc;
vf->vf_queues[qid].p_rx_cid = NULL;
vf->num_active_rxqs--;
}
p_queue->p_rx_cid = NULL;
vf->num_active_rxqs--;
return rc;
return 0;
}
static int qed_iov_vf_stop_txqs(struct qed_hwfn *p_hwfn,
struct qed_vf_info *vf, u16 txq_id, u8 num_txqs)
struct qed_vf_info *vf, u16 txq_id)
{
int rc = 0;
struct qed_vf_q_info *p_queue;
int qid;
int rc = 0;
if (txq_id + num_txqs > ARRAY_SIZE(vf->vf_queues))
if (!qed_iov_validate_txq(p_hwfn, vf, txq_id,
QED_IOV_VALIDATE_Q_ENABLE))
return -EINVAL;
for (qid = txq_id; qid < txq_id + num_txqs; qid++) {
p_queue = &vf->vf_queues[qid];
if (!p_queue->p_tx_cid)
continue;
p_queue = &vf->vf_queues[txq_id];
rc = qed_eth_tx_queue_stop(p_hwfn, p_queue->p_tx_cid);
if (rc)
return rc;
rc = qed_eth_tx_queue_stop(p_hwfn, p_queue->p_tx_cid);
if (rc)
return rc;
p_queue->p_tx_cid = NULL;
}
p_queue->p_tx_cid = NULL;
return rc;
return 0;
}
static void qed_iov_vf_mbx_stop_rxqs(struct qed_hwfn *p_hwfn,
......@@ -2096,20 +2173,28 @@ static void qed_iov_vf_mbx_stop_rxqs(struct qed_hwfn *p_hwfn,
{
u16 length = sizeof(struct pfvf_def_resp_tlv);
struct qed_iov_vf_mbx *mbx = &vf->vf_mbx;
u8 status = PFVF_STATUS_SUCCESS;
u8 status = PFVF_STATUS_FAILURE;
struct vfpf_stop_rxqs_tlv *req;
int rc;
/* We give the option of starting from qid != 0, in this case we
* need to make sure that qid + num_qs doesn't exceed the actual
* amount of queues that exist.
/* There has never been an official driver that used this interface
* for stopping multiple queues, and it is now considered deprecated.
* Validate this isn't used here.
*/
req = &mbx->req_virt->stop_rxqs;
rc = qed_iov_vf_stop_rxqs(p_hwfn, vf, req->rx_qid,
req->num_rxqs, req->cqe_completion);
if (rc)
status = PFVF_STATUS_FAILURE;
if (req->num_rxqs != 1) {
DP_VERBOSE(p_hwfn, QED_MSG_IOV,
"Odd; VF[%d] tried stopping multiple Rx queues\n",
vf->relative_vf_id);
status = PFVF_STATUS_NOT_SUPPORTED;
goto out;
}
rc = qed_iov_vf_stop_rxqs(p_hwfn, vf, req->rx_qid,
req->cqe_completion);
if (!rc)
status = PFVF_STATUS_SUCCESS;
out:
qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_STOP_RXQS,
length, status);
}
......@@ -2120,19 +2205,27 @@ static void qed_iov_vf_mbx_stop_txqs(struct qed_hwfn *p_hwfn,
{
u16 length = sizeof(struct pfvf_def_resp_tlv);
struct qed_iov_vf_mbx *mbx = &vf->vf_mbx;
u8 status = PFVF_STATUS_SUCCESS;
u8 status = PFVF_STATUS_FAILURE;
struct vfpf_stop_txqs_tlv *req;
int rc;
/* We give the option of starting from qid != 0, in this case we
* need to make sure that qid + num_qs doesn't exceed the actual
* amount of queues that exist.
/* There has never been an official driver that used this interface
* for stopping multiple queues, and it is now considered deprecated.
* Validate this isn't used here.
*/
req = &mbx->req_virt->stop_txqs;
rc = qed_iov_vf_stop_txqs(p_hwfn, vf, req->tx_qid, req->num_txqs);
if (rc)
status = PFVF_STATUS_FAILURE;
if (req->num_txqs != 1) {
DP_VERBOSE(p_hwfn, QED_MSG_IOV,
"Odd; VF[%d] tried stopping multiple Tx queues\n",
vf->relative_vf_id);
status = PFVF_STATUS_NOT_SUPPORTED;
goto out;
}
rc = qed_iov_vf_stop_txqs(p_hwfn, vf, req->tx_qid);
if (!rc)
status = PFVF_STATUS_SUCCESS;
out:
qed_iov_prepare_resp(p_hwfn, p_ptt, vf, CHANNEL_TLV_STOP_TXQS,
length, status);
}
......@@ -2157,22 +2250,17 @@ static void qed_iov_vf_mbx_update_rxqs(struct qed_hwfn *p_hwfn,
complete_event_flg = !!(req->flags & VFPF_RXQ_UPD_COMPLETE_EVENT_FLAG);
/* Validate inputs */
if (req->num_rxqs + req->rx_qid > QED_MAX_VF_CHAINS_PER_PF ||
!qed_iov_validate_rxq(p_hwfn, vf, req->rx_qid)) {
DP_INFO(p_hwfn, "VF[%d]: Incorrect Rxqs [%04x, %02x]\n",
vf->relative_vf_id, req->rx_qid, req->num_rxqs);
goto out;
}
for (i = 0; i < req->num_rxqs; i++) {
qid = req->rx_qid + i;
if (!vf->vf_queues[qid].p_rx_cid) {
DP_INFO(p_hwfn,
"VF[%d] rx_qid = %d isn`t active!\n",
vf->relative_vf_id, qid);
for (i = req->rx_qid; i < req->rx_qid + req->num_rxqs; i++)
if (!qed_iov_validate_rxq(p_hwfn, vf, i,
QED_IOV_VALIDATE_Q_ENABLE)) {
DP_INFO(p_hwfn, "VF[%d]: Incorrect Rxqs [%04x, %02x]\n",
vf->relative_vf_id, req->rx_qid, req->num_rxqs);
goto out;
}
/* Prepare the handlers */
for (i = 0; i < req->num_rxqs; i++) {
qid = req->rx_qid + i;
handlers[i] = vf->vf_queues[qid].p_rx_cid;
}
......@@ -2388,7 +2476,8 @@ qed_iov_vp_update_rss_param(struct qed_hwfn *p_hwfn,
for (i = 0; i < table_size; i++) {
q_idx = p_rss_tlv->rss_ind_table[i];
if (!qed_iov_validate_rxq(p_hwfn, vf, q_idx)) {
if (!qed_iov_validate_rxq(p_hwfn, vf, q_idx,
QED_IOV_VALIDATE_Q_ENABLE)) {
DP_VERBOSE(p_hwfn,
QED_MSG_IOV,
"VF[%d]: Omitting RSS due to wrong queue %04x\n",
......@@ -2397,15 +2486,6 @@ qed_iov_vp_update_rss_param(struct qed_hwfn *p_hwfn,
goto out;
}
if (!vf->vf_queues[q_idx].p_rx_cid) {
DP_VERBOSE(p_hwfn,
QED_MSG_IOV,
"VF[%d]: Omitting RSS due to inactive queue %08x\n",
vf->relative_vf_id, q_idx);
b_reject = true;
goto out;
}
p_rss->rss_ind_table[i] = vf->vf_queues[q_idx].p_rx_cid;
}
......@@ -3058,9 +3138,10 @@ qed_iov_vf_flr_cleanup(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
return rc;
}
int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *p_disabled_vfs)
bool qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *p_disabled_vfs)
{
u16 i, found = 0;
bool found = false;
u16 i;
DP_VERBOSE(p_hwfn, QED_MSG_IOV, "Marking FLR-ed VFs\n");
for (i = 0; i < (VF_MAX_STATIC / 32); i++)
......@@ -3070,7 +3151,7 @@ int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *p_disabled_vfs)
if (!p_hwfn->cdev->p_iov_info) {
DP_NOTICE(p_hwfn, "VF flr but no IOV\n");
return 0;
return false;
}
/* Mark VFs */
......@@ -3099,7 +3180,7 @@ int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *p_disabled_vfs)
* VF flr until ACKs, we're safe.
*/
p_flr[rel_vf_id / 64] |= 1ULL << (rel_vf_id % 64);
found = 1;
found = true;
}
}
......@@ -3305,11 +3386,17 @@ static void qed_sriov_vfpf_malicious(struct qed_hwfn *p_hwfn,
if (!p_vf)
return;
DP_INFO(p_hwfn,
"VF [%d] - Malicious behavior [%02x]\n",
p_vf->abs_vf_id, p_data->err_id);
if (!p_vf->b_malicious) {
DP_NOTICE(p_hwfn,
"VF [%d] - Malicious behavior [%02x]\n",
p_vf->abs_vf_id, p_data->err_id);
p_vf->b_malicious = true;
p_vf->b_malicious = true;
} else {
DP_INFO(p_hwfn,
"VF [%d] - Malicious behavior [%02x]\n",
p_vf->abs_vf_id, p_data->err_id);
}
}
int qed_sriov_eqe_event(struct qed_hwfn *p_hwfn,
......@@ -3858,6 +3945,7 @@ static int qed_get_vf_config(struct qed_dev *cdev,
void qed_inform_vf_link_state(struct qed_hwfn *hwfn)
{
struct qed_hwfn *lead_hwfn = QED_LEADING_HWFN(hwfn->cdev);
struct qed_mcp_link_capabilities caps;
struct qed_mcp_link_params params;
struct qed_mcp_link_state link;
......@@ -3874,9 +3962,15 @@ void qed_inform_vf_link_state(struct qed_hwfn *hwfn)
if (!vf_info)
continue;
memcpy(&params, qed_mcp_get_link_params(hwfn), sizeof(params));
memcpy(&link, qed_mcp_get_link_state(hwfn), sizeof(link));
memcpy(&caps, qed_mcp_get_link_capabilities(hwfn),
/* Only hwfn0 is actually interested in the link speed.
* But since only it would receive an MFW indication of link,
* need to take configuration from it - otherwise things like
* rate limiting for hwfn1 VF would not work.
*/
memcpy(&params, qed_mcp_get_link_params(lead_hwfn),
sizeof(params));
memcpy(&link, qed_mcp_get_link_state(lead_hwfn), sizeof(link));
memcpy(&caps, qed_mcp_get_link_capabilities(lead_hwfn),
sizeof(caps));
/* Modify link according to the VF's configured link state */
......
......@@ -348,9 +348,9 @@ int qed_sriov_eqe_event(struct qed_hwfn *p_hwfn,
* @param p_hwfn
* @param disabled_vfs - bitmask of all VFs on path that were FLRed
*
* @return 1 iff one of the PF's vfs got FLRed. 0 otherwise.
* @return true iff one of the PF's vfs got FLRed. false otherwise.
*/
int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *disabled_vfs);
bool qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn, u32 *disabled_vfs);
/**
* @brief Search extended TLVs in request/reply buffer.
......@@ -407,10 +407,10 @@ static inline int qed_sriov_eqe_event(struct qed_hwfn *p_hwfn,
return -EINVAL;
}
static inline int qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn,
u32 *disabled_vfs)
static inline bool qed_iov_mark_vf_flr(struct qed_hwfn *p_hwfn,
u32 *disabled_vfs)
{
return 0;
return false;
}
static inline void qed_iov_wq_stop(struct qed_dev *cdev, bool schedule_first)
......
......@@ -134,14 +134,20 @@ static int qed_send_msg2pf(struct qed_hwfn *p_hwfn, u8 *done, u32 resp_size)
}
if (!*done) {
DP_VERBOSE(p_hwfn, QED_MSG_IOV,
"VF <-- PF Timeout [Type %d]\n",
p_req->first_tlv.tl.type);
DP_NOTICE(p_hwfn,
"VF <-- PF Timeout [Type %d]\n",
p_req->first_tlv.tl.type);
rc = -EBUSY;
} else {
DP_VERBOSE(p_hwfn, QED_MSG_IOV,
"PF response: %d [Type %d]\n",
*done, p_req->first_tlv.tl.type);
if ((*done != PFVF_STATUS_SUCCESS) &&
(*done != PFVF_STATUS_NO_RESOURCE))
DP_NOTICE(p_hwfn,
"PF response: %d [Type %d]\n",
*done, p_req->first_tlv.tl.type);
else
DP_VERBOSE(p_hwfn, QED_MSG_IOV,
"PF response: %d [Type %d]\n",
*done, p_req->first_tlv.tl.type);
}
return rc;
......
......@@ -275,6 +275,8 @@ struct vfpf_stop_rxqs_tlv {
struct vfpf_first_tlv first_tlv;
u16 rx_qid;
/* this field is deprecated and should *always* be set to '1' */
u8 num_rxqs;
u8 cqe_completion;
u8 padding[4];
......@@ -285,6 +287,8 @@ struct vfpf_stop_txqs_tlv {
struct vfpf_first_tlv first_tlv;
u16 tx_qid;
/* this field is deprecated and should *always* be set to '1' */
u8 num_txqs;
u8 padding[5];
};
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment