| Message ID | 20230424092420.12794-1-quic_rajkbhag@quicinc.com |
|---|---|
| State | New |
| Headers | show |
| Series | wifi: ath11k: update proper pdev/vdev id for testmode command | expand |
Raj Kumar Bhagat <quic_rajkbhag@quicinc.com> writes: > From: Venkateswara Naralasetty <quic_vnaralas@quicinc.com> > > User can extend test mode command support to set vdev and pdev params > for debug purpose at run time by sending vdev/pdev param set command > through the testmode command interface. > > Fill the proper pdev/vdev id in driver since, pdev/vdev id details may > not be available at user space. It will make sure that the proper > vdev/pdev ids are sent to firmware. > > Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1 > > Depends-On: > [v3,3/4] wifi: ath11k: factory test mode support > > Signed-off-by: Venkateswara Naralasetty <quic_vnaralas@quicinc.com> > Signed-off-by: Raj Kumar Bhagat <quic_rajkbhag@quicinc.com> > --- Please add 'Depends-on' after the '---' line. This is for future patches, no need to resend because of this.
On 4/26/2023 12:59 PM, Kalle Valo wrote: > Raj Kumar Bhagat <quic_rajkbhag@quicinc.com> writes: > >> From: Venkateswara Naralasetty <quic_vnaralas@quicinc.com> >> >> User can extend test mode command support to set vdev and pdev params >> for debug purpose at run time by sending vdev/pdev param set command >> through the testmode command interface. >> >> Fill the proper pdev/vdev id in driver since, pdev/vdev id details may >> not be available at user space. It will make sure that the proper >> vdev/pdev ids are sent to firmware. >> >> Tested-on: QCN9074 hw1.0 PCI WLAN.HK.2.7.0.1-01744-QCAHKSWPL_SILICONZ-1 >> >> Depends-On: >> [v3,3/4] wifi: ath11k: factory test mode support >> >> Signed-off-by: Venkateswara Naralasetty <quic_vnaralas@quicinc.com> >> Signed-off-by: Raj Kumar Bhagat <quic_rajkbhag@quicinc.com> >> --- > > Please add 'Depends-on' after the '---' line. This is for future > patches, no need to resend because of this. > Thanks for pointing this. Will take care for future patches. Thanks, Raj
diff --git a/drivers/net/wireless/ath/ath11k/testmode.c b/drivers/net/wireless/ath/ath11k/testmode.c index bebd9c1776ba..e1edbc532eb6 100644 --- a/drivers/net/wireless/ath/ath11k/testmode.c +++ b/drivers/net/wireless/ath/ath11k/testmode.c @@ -250,13 +250,16 @@ static int ath11k_tm_cmd_testmode_start(struct ath11k *ar, struct nlattr *tb[]) return ret; } -static int ath11k_tm_cmd_wmi(struct ath11k *ar, struct nlattr *tb[]) +static int ath11k_tm_cmd_wmi(struct ath11k *ar, struct nlattr *tb[], + struct ieee80211_vif *vif) { struct ath11k_pdev_wmi *wmi = ar->wmi; struct sk_buff *skb; + struct ath11k_vif *arvif; u32 cmd_id, buf_len; - int ret; + int ret, tag; void *buf; + u32 *ptr; mutex_lock(&ar->conf_mutex); @@ -280,6 +283,34 @@ static int ath11k_tm_cmd_wmi(struct ath11k *ar, struct nlattr *tb[]) cmd_id = nla_get_u32(tb[ATH11K_TM_ATTR_WMI_CMDID]); + /* Make sure that the buffer length is long enough to + * hold TLV and pdev/vdev id. + */ + if (buf_len < sizeof(struct wmi_tlv) + sizeof(u32)) { + ret = -EINVAL; + goto out; + } + + ptr = buf; + tag = FIELD_GET(WMI_TLV_TAG, *ptr); + + /* pdev/vdev id start after TLV header */ + ptr++; + + if (tag == WMI_TAG_PDEV_SET_PARAM_CMD) + *ptr = ar->pdev->pdev_id; + + if (ar->ab->fw_mode != ATH11K_FIRMWARE_MODE_FTM && + (tag == WMI_TAG_VDEV_SET_PARAM_CMD || tag == WMI_TAG_UNIT_TEST_CMD)) { + if (vif) { + arvif = (struct ath11k_vif *)vif->drv_priv; + *ptr = arvif->vdev_id; + } else { + ret = -EINVAL; + goto out; + } + } + ath11k_dbg(ar->ab, ATH11K_DBG_TESTMODE, "testmode cmd wmi cmd_id %d buf length %d\n", cmd_id, buf_len); @@ -411,7 +442,7 @@ int ath11k_tm_cmd(struct ieee80211_hw *hw, struct ieee80211_vif *vif, case ATH11K_TM_CMD_TESTMODE_START: return ath11k_tm_cmd_testmode_start(ar, tb); case ATH11K_TM_CMD_WMI: - return ath11k_tm_cmd_wmi(ar, tb); + return ath11k_tm_cmd_wmi(ar, tb, vif); case ATH11K_TM_CMD_WMI_FTM: set_bit(ATH11K_FLAG_FTM_SEGMENTED, &ab->dev_flags); return ath11k_tm_cmd_process_ftm(ar, tb);