From: Aditya Kumar Singh aditya.kumar.singh@oss.qualcomm.com
[ Upstream commit dce7aec6b1f74b0a46b901ab8de1f7bd0515f733 ]
With hardware grouping, during reboot, whenever a device is removed, it powers down itself and all its partner devices in the same group. Now this is done by all devices and hence there is multiple power down for devices and hence the following error messages can be seen:
ath12k_pci 0002:01:00.0: failed to set mhi state POWER_OFF(3) in current mhi state (0x0) ath12k_pci 0002:01:00.0: failed to set mhi state: POWER_OFF(3) ath12k_pci 0002:01:00.0: failed to set mhi state DEINIT(1) in current mhi state (0x0) ath12k_pci 0002:01:00.0: failed to set mhi state: DEINIT(1) ath12k_pci 0003:01:00.0: failed to set mhi state POWER_OFF(3) in current mhi state (0x0) ath12k_pci 0003:01:00.0: failed to set mhi state: POWER_OFF(3) ath12k_pci 0003:01:00.0: failed to set mhi state DEINIT(1) in current mhi state (0x0) ath12k_pci 0003:01:00.0: failed to set mhi state: DEINIT(1) ath12k_pci 0004:01:00.0: failed to set mhi state POWER_OFF(3) in current mhi state (0x0) ath12k_pci 0004:01:00.0: failed to set mhi state: POWER_OFF(3)
To prevent this, check if the ATH12K_PCI_FLAG_INIT_DONE flag is already set before powering down. If it is set, it indicates that another partner device has already performed the power down, and this device can skip this step.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1 Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.4.1-00199-QCAHKSWPL_SILICONZ-1 Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
Signed-off-by: Aditya Kumar Singh aditya.kumar.singh@oss.qualcomm.com Reviewed-by: Vasanthakumar Thiagarajan vasanthakumar.thiagarajan@oss.qualcomm.com Link: https://patch.msgid.link/20250408-fix_reboot_issues_with_hw_grouping-v4-3-95... Signed-off-by: Jeff Johnson jeff.johnson@oss.qualcomm.com Signed-off-by: Sasha Levin sashal@kernel.org ---
Based on my analysis of the commit and the surrounding kernel code, here is my assessment: **YES** This commit should be backported to stable kernel trees. ## Detailed Analysis **The Bug Being Fixed:** The commit addresses a specific issue with hardware grouping in ath12k WiFi devices where multiple power-down operations during reboot cause MHI (Modem Host Interface) state transition errors. The error messages show failed attempts to set MHI state to POWER_OFF(3) and DEINIT(1) when the current state is already 0x0, indicating the device has already been powered down. **Code Changes Analysis:** The fix is extremely minimal and surgical - it adds only 3 lines of code: ```c if (!test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags)) return; ``` This check is added at the beginning of `ath12k_pci_power_down()` function at line 1491-1492. The logic is sound: 1. **ATH12K_PCI_FLAG_INIT_DONE** flag is set in `ath12k_pci_start()` (line 1170) when the device is properly initialized 2. **The flag is cleared** in `ath12k_pci_power_down()` (line 1500) after successful power down 3. **The new check** prevents redundant power-down operations by returning early if the flag is already cleared (meaning device was already powered down) **Why This Qualifies for Backporting:** 1. **Clear Bug Fix**: The commit directly addresses user-visible error messages during system reboot - a clear operational issue affecting stability. 2. **Minimal Risk**: The change is extremely contained (3 lines) and only affects the power-down path. It doesn't introduce new functionality or architectural changes. 3. **Hardware-Specific Issue**: This affects specific WiFi hardware (QCN9274, WCN7850) that users may have in stable kernel configurations. 4. **Similar Pattern**: Looking at the similar commits, commit #2 (ath11k IRQ affinity fix) and commit #4 (MHI PCI shutdown callback) were both marked as YES for backporting. They share similar characteristics: - Small, targeted fixes - Address shutdown/reboot issues - Minimal code changes - Clear problem resolution 5. **Race Condition Fix**: This addresses a race condition where multiple devices in a hardware group try to power down simultaneously, which is a classic stability issue suitable for stable trees. 6. **Low Regression Risk**: The change only adds an early return condition based on an existing flag that's already used throughout the codebase for the same purpose - tracking initialization state. **Supporting Evidence from Code:** - The `ATH12K_PCI_FLAG_INIT_DONE` flag is already used consistently throughout the driver to track device state - The MHI error handling code shows these are expected error paths that should be avoided (lines 529, 607 in mhi.c) - The commit includes proper testing on multiple hardware variants This is a textbook example of a stable-tree appropriate fix: small, targeted, fixes a clear user-impacting bug, with minimal regression risk.
drivers/net/wireless/ath/ath12k/pci.c | 3 +++ 1 file changed, 3 insertions(+)
diff --git a/drivers/net/wireless/ath/ath12k/pci.c b/drivers/net/wireless/ath/ath12k/pci.c index 45d537066345a..34855e2d540ca 100644 --- a/drivers/net/wireless/ath/ath12k/pci.c +++ b/drivers/net/wireless/ath/ath12k/pci.c @@ -1301,6 +1301,9 @@ void ath12k_pci_power_down(struct ath12k_base *ab, bool is_suspend) { struct ath12k_pci *ab_pci = ath12k_pci_priv(ab);
+ if (!test_bit(ATH12K_PCI_FLAG_INIT_DONE, &ab_pci->flags)) + return; + /* restore aspm in case firmware bootup fails */ ath12k_pci_aspm_restore(ab_pci);