On my DS414, some PCI devices return odd values when probing BAR sizes. An obvious case is all-ones response, the Linux driver (drivers/pci/probe.c) catches those explicitly and a comment explains that either bit 0 or bit 1 must be clear (depending on MEM or IO type). Other BARs return e.g. 0xfff0000f or 0xfff00004 and thus manage to break size calculation due to the "middle" zeroes. Mitigate that copying more or less what Linux does and do a "find least bit set".
Signed-off-by: Phil Sutter phil@nwl.cc --- drivers/pci/pci_auto.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/drivers/pci/pci_auto.c b/drivers/pci/pci_auto.c index 3f46b7697d7ca..ea202b0e0959e 100644 --- a/drivers/pci/pci_auto.c +++ b/drivers/pci/pci_auto.c @@ -47,16 +47,17 @@ void dm_pciauto_setup_device(struct udevice *dev, int bars_num, dm_pci_write_config32(dev, bar, 0xffffffff); dm_pci_read_config32(dev, bar, &bar_response);
- /* If BAR is not implemented go to the next BAR */ - if (!bar_response) + /* If BAR is not implemented (or invalid) go to the next BAR */ + if (!bar_response || bar_response == 0xffffffff) continue;
found_mem64 = 0;
/* Check the BAR type and set our address mask */ if (bar_response & PCI_BASE_ADDRESS_SPACE) { - bar_size = ((~(bar_response & PCI_BASE_ADDRESS_IO_MASK)) - & 0xffff) + 1; + bar_size = bar_response & PCI_BASE_ADDRESS_IO_MASK; + bar_size &= ~(bar_size - 1); + if (!enum_only) bar_res = io;