[PATCH v6 10/29] pci: Adjust dm_pci_read_bar32() to return errors correctly

[Simon Glass]

At present if reading a BAR returns 0xffffffff then the value is masked
and a different value is returned. This makes it harder to detect the
problem when debugging.

Update the function to avoid masking in this case.

Signed-off-by: Simon Glass <sjg at chromium.org>
Reviewed-by: Bin Meng <bmeng.cn at gmail.com>
Reviewed-by: Wolfgang Wallner <wolfgang.wallner at br-automation.com>
---

Changes in v6:
- Rework commit message to avoid mention of missing device

Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 drivers/pci/pci-uclass.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/pci-uclass.c b/drivers/pci/pci-uclass.c
index 213381da6bd..7f46e901fb2 100644
--- a/drivers/pci/pci-uclass.c
+++ b/drivers/pci/pci-uclass.c
@@ -1213,7 +1213,14 @@ u32 dm_pci_read_bar32(const struct udevice *dev, int barnum)
 
 	bar = PCI_BASE_ADDRESS_0 + barnum * 4;
 	dm_pci_read_config32(dev, bar, &addr);
-	if (addr & PCI_BASE_ADDRESS_SPACE_IO)
+
+	/*
+	 * If we get an invalid address, return this so that comparisons with
+	 * FDT_ADDR_T_NONE work correctly
+	 */
+	if (addr == 0xffffffff)
+		return addr;
+	else if (addr & PCI_BASE_ADDRESS_SPACE_IO)
 		return addr & PCI_BASE_ADDRESS_IO_MASK;
 	else
 		return addr & PCI_BASE_ADDRESS_MEM_MASK;
-- 
2.26.0.292.g33ef6b2f38-goog