[PATCH AUTOSEL 6.19-6.6] wifi: ath12k: fix preferred hardware mode calculation

[PATCH AUTOSEL 6.19-6.6] wifi: ath12k: fix preferred hardware mode calculation

[Sasha Levin]

From: Baochen Qiang <baochen.qiang@oss.qualcomm.com>

[ Upstream commit 7f852de0003219c431a6f2ffd951fd82a4673660 ]

For single pdev device like WCN7850/QCC2072, preferred_hw_mode is
initialized to WMI_HOST_HW_MODE_SINGLE. Later when firmware sends
supported modes to host, each mode is compared with the initial one
and if the priority of the new mode is higher, update the parameter
and store mode capability.

For WCN7850, this does not result in issue, as one of the supported
mode indeed has a higher priority. However the only available mode of
QCC2072 at this stage is WMI_HOST_HW_MODE_SINGLE, which fails the
comparison, hence mode capability is not stored. Subsequently driver
initialization fails.

Fix it by accepting a mode with the same priority.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.1.c5-00302-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.115823.3

Signed-off-by: Baochen Qiang <baochen.qiang@oss.qualcomm.com>
Reviewed-by: Vasanthakumar Thiagarajan <vasanthakumar.thiagarajan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260112-ath12k-support-qcc2072-v2-4-fc8ce1e43969@oss.qualcomm.com
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---

LLM Generated explanations, may be completely bogus:

The code context confirms that `ath12k_hw_mode_pri_map` is a priority
mapping array where lower values mean higher priority. The
`WMI_HOST_HW_MODE_SINGLE` is in the array, and the fix ensures that when
a mode with equal priority is found, it's still accepted.

### Conclusion

This is a textbook stable backport candidate:

1. **Fixes a real bug**: Complete driver initialization failure on
   QCC2072 hardware
2. **Obviously correct**: The logic error is clear — when the only
   available mode has the same priority as the initial value, strict `<`
   prevents it from being stored
3. **Minimal change**: Single character change (`<` to `<=`)
4. **No new features**: Just fixes the comparison logic
5. **Low risk**: For already-working hardware (WCN7850), the higher-
   priority mode still wins; the `<=` only affects the equal-priority
   case
6. **Tested and reviewed**: By Qualcomm engineers and the subsystem
   maintainer
7. **High user impact**: Without this, an entire class of hardware
   doesn't work

**YES**

 drivers/net/wireless/ath/ath12k/wmi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 3ce5fcb0e4600..12f4d378f50d4 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -4545,7 +4545,7 @@ static int ath12k_wmi_hw_mode_caps(struct ath12k_base *soc,
 
 		pref = soc->wmi_ab.preferred_hw_mode;
 
-		if (ath12k_hw_mode_pri_map[mode] < ath12k_hw_mode_pri_map[pref]) {
+		if (ath12k_hw_mode_pri_map[mode] <= ath12k_hw_mode_pri_map[pref]) {
 			svc_rdy_ext->pref_hw_mode_caps = *hw_mode_caps;
 			soc->wmi_ab.preferred_hw_mode = mode;
 		}
-- 
2.51.0

[PATCH AUTOSEL 6.19-6.18] wifi: ath12k: fix mac phy capability parsing

[Sasha Levin]

From: Baochen Qiang <baochen.qiang@oss.qualcomm.com>

[ Upstream commit b5151c9b6e3a347416a4b4b55fc00195526d8771 ]

Currently ath12k_pull_mac_phy_cap_svc_ready_ext() assumes only one band
supported in each phy, hence it skips 5 GHz band if 2 GHz band support
is detected. This does not work for device which gets only one phy but
has both bands supported, such as QCC2072.

Change to check each band individually to fix this issue.

Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.1.c5-00302-QCAHMTSWPL_V1.0_V2.0_SILICONZ-1.115823.3

Signed-off-by: Baochen Qiang <baochen.qiang@oss.qualcomm.com>
Reviewed-by: Vasanthakumar Thiagarajan <vasanthakumar.thiagarajan@oss.qualcomm.com>
Link: https://patch.msgid.link/20260112-ath12k-support-qcc2072-v2-6-fc8ce1e43969@oss.qualcomm.com
Signed-off-by: Jeff Johnson <jeff.johnson@oss.qualcomm.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---

LLM Generated explanations, may be completely bogus:

### Dependency Concerns

The investigation reveals an important dependency issue:

- The constants were **renamed** from
  `WMI_HOST_WLAN_2G_CAP`/`WMI_HOST_WLAN_5G_CAP` to
  `WMI_HOST_WLAN_2GHZ_CAP`/`WMI_HOST_WLAN_5GHZ_CAP` in March 2025
  (commit `6a88093f79ea0`). If backporting to a stable tree that
  predates this rename, the constant names would need adjustment.
- The `nss_ratio` fields were added in June 2025 - older stable trees
  may not have these.
- The ath12k driver itself was introduced in November 2022, so stable
  trees from 6.1.y onward should have the driver.

The core logic change (if/else if → independent if checks) is self-
contained and the fix concept applies regardless of these naming
differences, though minor backport adjustments to constant names may be
needed for older stable trees.

### Risk vs. Benefit

- **Benefit:** Fixes a real functional bug where dual-band single-PHY
  WiFi devices (like QCC2072) cannot use 5 GHz band at all. This is a
  significant hardware functionality issue.
- **Risk:** Very low. The logic change is straightforward and doesn't
  affect single-band devices. The only behavioral change for dual-band
  devices is that both bands are now properly initialized instead of
  only one.

### Conclusion

This is a clear bug fix in a wireless driver that prevents hardware from
working correctly. The fix is:
- Small and contained (single function, single file)
- Obviously correct (the if/else if logic was clearly wrong for dual-
  band PHYs)
- Tested and reviewed
- Low risk of regression
- Fixes a real user-visible issue (5 GHz band not working)

The only minor concern is that backporting may require adjusting
constant names for older stable trees, but this is a trivial mechanical
adjustment.

**YES**

 drivers/net/wireless/ath/ath12k/wmi.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/net/wireless/ath/ath12k/wmi.c b/drivers/net/wireless/ath/ath12k/wmi.c
index 12f4d378f50d4..1613492b38350 100644
--- a/drivers/net/wireless/ath/ath12k/wmi.c
+++ b/drivers/net/wireless/ath/ath12k/wmi.c
@@ -496,6 +496,7 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 	struct ath12k_band_cap *cap_band;
 	struct ath12k_pdev_cap *pdev_cap = &pdev->cap;
 	struct ath12k_fw_pdev *fw_pdev;
+	u32 supported_bands;
 	u32 phy_map;
 	u32 hw_idx, phy_idx = 0;
 	int i;
@@ -519,14 +520,19 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 		return -EINVAL;
 
 	mac_caps = wmi_mac_phy_caps + phy_idx;
+	supported_bands = le32_to_cpu(mac_caps->supported_bands);
+
+	if (!(supported_bands & WMI_HOST_WLAN_2GHZ_CAP) &&
+	    !(supported_bands & WMI_HOST_WLAN_5GHZ_CAP))
+		return -EINVAL;
 
 	pdev->pdev_id = ath12k_wmi_mac_phy_get_pdev_id(mac_caps);
 	pdev->hw_link_id = ath12k_wmi_mac_phy_get_hw_link_id(mac_caps);
-	pdev_cap->supported_bands |= le32_to_cpu(mac_caps->supported_bands);
+	pdev_cap->supported_bands |= supported_bands;
 	pdev_cap->ampdu_density = le32_to_cpu(mac_caps->ampdu_density);
 
 	fw_pdev = &ab->fw_pdev[ab->fw_pdev_count];
-	fw_pdev->supported_bands = le32_to_cpu(mac_caps->supported_bands);
+	fw_pdev->supported_bands = supported_bands;
 	fw_pdev->pdev_id = ath12k_wmi_mac_phy_get_pdev_id(mac_caps);
 	fw_pdev->phy_id = le32_to_cpu(mac_caps->phy_id);
 	ab->fw_pdev_count++;
@@ -535,10 +541,12 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 	 * band to band for a single radio, need to see how this should be
 	 * handled.
 	 */
-	if (le32_to_cpu(mac_caps->supported_bands) & WMI_HOST_WLAN_2GHZ_CAP) {
+	if (supported_bands & WMI_HOST_WLAN_2GHZ_CAP) {
 		pdev_cap->tx_chain_mask = le32_to_cpu(mac_caps->tx_chain_mask_2g);
 		pdev_cap->rx_chain_mask = le32_to_cpu(mac_caps->rx_chain_mask_2g);
-	} else if (le32_to_cpu(mac_caps->supported_bands) & WMI_HOST_WLAN_5GHZ_CAP) {
+	}
+
+	if (supported_bands & WMI_HOST_WLAN_5GHZ_CAP) {
 		pdev_cap->vht_cap = le32_to_cpu(mac_caps->vht_cap_info_5g);
 		pdev_cap->vht_mcs = le32_to_cpu(mac_caps->vht_supp_mcs_5g);
 		pdev_cap->he_mcs = le32_to_cpu(mac_caps->he_supp_mcs_5g);
@@ -548,8 +556,6 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 			WMI_NSS_RATIO_EN_DIS_GET(mac_caps->nss_ratio);
 		pdev_cap->nss_ratio_info =
 			WMI_NSS_RATIO_INFO_GET(mac_caps->nss_ratio);
-	} else {
-		return -EINVAL;
 	}
 
 	/* tx/rx chainmask reported from fw depends on the actual hw chains used,
@@ -565,7 +571,7 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 	pdev_cap->rx_chain_mask_shift =
 			find_first_bit((unsigned long *)&pdev_cap->rx_chain_mask, 32);
 
-	if (le32_to_cpu(mac_caps->supported_bands) & WMI_HOST_WLAN_2GHZ_CAP) {
+	if (supported_bands & WMI_HOST_WLAN_2GHZ_CAP) {
 		cap_band = &pdev_cap->band[NL80211_BAND_2GHZ];
 		cap_band->phy_id = le32_to_cpu(mac_caps->phy_id);
 		cap_band->max_bw_supported = le32_to_cpu(mac_caps->max_bw_supported_2g);
@@ -585,7 +591,7 @@ ath12k_pull_mac_phy_cap_svc_ready_ext(struct ath12k_wmi_pdev *wmi_handle,
 				le32_to_cpu(mac_caps->he_ppet2g.ppet16_ppet8_ru3_ru0[i]);
 	}
 
-	if (le32_to_cpu(mac_caps->supported_bands) & WMI_HOST_WLAN_5GHZ_CAP) {
+	if (supported_bands & WMI_HOST_WLAN_5GHZ_CAP) {
 		cap_band = &pdev_cap->band[NL80211_BAND_5GHZ];
 		cap_band->phy_id = le32_to_cpu(mac_caps->phy_id);
 		cap_band->max_bw_supported =
-- 
2.51.0