Commit a4e772898f8b ("PCI: Add missing bridge lock to pci_bus_lock()")
delegates the bridge device's pci_dev_trylock() to pci_bus_trylock()
in pci_slot_trylock(), but it leaves a redundant pci_dev_unlock() when
pci_bus_trylock() fails.

Remove the redundant bridge-device pci_dev_unlock() in pci_slot_trylock(),
since that lock is no longer taken there.

Fixes: a4e772898f8b ("PCI: Add missing bridge lock to pci_bus_lock()")
Cc: stable@vger.kernel.org
Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
---
 drivers/pci/pci.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 13dbb405dc31..75a98819db6f 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -5347,7 +5347,6 @@ static int pci_slot_trylock(struct pci_slot *slot)
 			continue;
 		if (dev->subordinate) {
 			if (!pci_bus_trylock(dev->subordinate)) {
-				pci_dev_unlock(dev);
 				goto unlock;
 			}
 		} else if (!pci_dev_trylock(dev))
-- 
2.20.1

On Thu, 11 Dec 2025, Jinhui Guo wrote:

> Commit a4e772898f8b ("PCI: Add missing bridge lock to pci_bus_lock()")
> delegates the bridge device's pci_dev_trylock() to pci_bus_trylock()
> in pci_slot_trylock(), but it leaves a redundant pci_dev_unlock() when
> pci_bus_trylock() fails.
> 
> Remove the redundant bridge-device pci_dev_unlock() in pci_slot_trylock(),
> since that lock is no longer taken there.

Doesn't it cause issues if trying to unlock something that wasn't locked 
so saying its "redundant" seem a bit an understatement?

> Fixes: a4e772898f8b ("PCI: Add missing bridge lock to pci_bus_lock()")
> Cc: stable@vger.kernel.org
> Signed-off-by: Jinhui Guo <guojinhui.liam@bytedance.com>
> ---
>  drivers/pci/pci.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
> index 13dbb405dc31..75a98819db6f 100644
> --- a/drivers/pci/pci.c
> +++ b/drivers/pci/pci.c
> @@ -5347,7 +5347,6 @@ static int pci_slot_trylock(struct pci_slot *slot)
>  			continue;
>  		if (dev->subordinate) {
>  			if (!pci_bus_trylock(dev->subordinate)) {
> -				pci_dev_unlock(dev);
>  				goto unlock;
>  			}
>  		} else if (!pci_dev_trylock(dev))
> 

-- 
 i.

On Thu, Dec 11, 2025 20:13:59 +0200, Ilpo Järvinen wrote:
> > Commit a4e772898f8b ("PCI: Add missing bridge lock to pci_bus_lock()")
> > delegates the bridge device's pci_dev_trylock() to pci_bus_trylock()
> > in pci_slot_trylock(), but it leaves a redundant pci_dev_unlock() when
> > pci_bus_trylock() fails.
> > 
> > Remove the redundant bridge-device pci_dev_unlock() in pci_slot_trylock(),
> > since that lock is no longer taken there.
> 
> Doesn't it cause issues if trying to unlock something that wasn't locked 
> so saying its "redundant" seem a bit an understatement?

Hi, Ilpo Järvinen

Thanks for your time. The commit message was a bit brief, so I've sent v2
https://lore.kernel.org/all/20251212133737.2367-1-guojinhui.liam@bytedance.com/
with more details.

Best Regards,
Jinhui