This is the mail archive of the ecos-discuss@sources.redhat.com mailing list for the eCos project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

pci1 test queries !!


I tried to run test pci1 on i386 host and this test detected many devices on
PCI bus, which includes 3COM NIC cards also. This NIC is located on
motherboard only. But this test does not detect Intel Ether Pro NIC located
on PCI expansion slots. But I am able to reach network through Ether Pro NIC
very well. Does it mean test pci1 does not detect devices on Expansion
slots?
The same test while running on ARM Integrator board detects PCI-PCI bridge
and Fast Ethernet LAN Controller but does not detect Ether Pro NIC in PCI
Expansion slot but here I am not able to reach network :) Any idea?

Another question is given a device id (devid) which headers should I look
for mapping to device? Say I do see devid as 0x6000 while debugging, so
where I should look for knowing which device is this.

And finally, some queries regarding line 123 in file
packages/io/pci/v2_0/src/pci.c from function cyg_pci_get_device_info
 for (i = 0; i < dev_info->num_bars; i++)
        cyg_pcihw_read_config_uint32(bus, devfn,
                                     CYG_PCI_CFG_BAR_BASE + 4*i,
                                     &dev_info->base_address[i]);

and at line 169 in same file and same function
bar = dev_info->base_address[i];
     // No reason to scan beyond first inactive BAR.
if (bar == 0) {
  dev_info->num_bars = i;
  break;    /****** control reaches here *****/
}

>From file if_82559.c function cyg_pci_get_device_info is called. In this
function ( I have given code snippet above) code at line number 123 is not
working properly for me. I am getting dev_info->base_address[i] as 0. So
later at line number 169 loop breaks and dev_info->base_map[] members remain
zero (as function break). Later in file if_82559.c in function code tries to
read/write at that address considering as PCI address and device get timed
out there.
Can somebody conclude the problem from above scenario?

Regards
Mohanlal

-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]