Bug 273285

Created attachment 139056 [details]
patch for inbuild drivers

Reassigning to yast2-network maintainer.

it would be nice to have this for alpha4

Thank you for the patch
But it would be useful to make difference between :
- no driver found
- built-in driver

Does hwinfo provide this information? 
Can you attach "hwinfo --netcard", please?

for hwinfo output refer to bug #273135

Stefan, can you attach output from "hwinfo --netcard" ? 
It's not same as netcard part of full hwinfo output and network uses that one (for me attachent in comment #3 is not usefull)

Created attachment 139238 [details]
hwinfo --netcard

here's the hwinfo output. sorry for the other one not being useful. Thanks :)

Steffen, how can I know from hwinfo output, that driver is available (packaged in module or built-in in kernel)?

Simple: if arch == ppc then driver_is_compiled_in.
No idea, though, why that's so. :-/

But more to the point: if you compare with the normal output, like:

25: PCI 8000.0: 0200 Ethernet controller                        
  [Created at pci.288]
  UDI: /org/freedesktop/Hal/devices/pci_14e4_1677
  Unique ID: rBUF.W6GsncWCLSF
  Parent ID: Z7uZ.n6ywvlBHNPC
  SysFS ID: /devices/pci0000:00/0000:00:1c.3/0000:80:00.0
  SysFS BusID: 0000:80:00.0
  Hardware Class: network
  Model: "Hewlett-Packard Company NetXtreme BCM5751 Gigabit Ethernet PCI Express"
  Vendor: pci 0x14e4 "Broadcom"
  Device: pci 0x1677 "NetXtreme BCM5751 Gigabit Ethernet PCI Express"
  SubVendor: pci 0x103c "Hewlett-Packard Company"
  SubDevice: pci 0x3008 
  Revision: 0x01
  Driver: "tg3"
  Driver Modules: "tg3"
  Device File: eth0
  Memory Range: 0xf0400000-0xf040ffff (rw,non-prefetchable)
  IRQ: 19 (1337681 events)
  HW Address: 00:11:0a:9d:48:b6
  Link detected: yes
  Module Alias: "pci:v000014E4d00001677sv0000103Csd00003008bc02sc00i00"
  Driver Info #0:
    Driver Status: tg3 is active
    Driver Activation Cmd: "modprobe tg3"
  Config Status: cfg=no, avail=yes, need=no, active=unknown
  Attached to: #13 (PCI bridge)

'Driver Modules:' is the module name for the driver. This field only exists
if there is a module. The field is (hd_t).driver_module (no 's') in libhd.
Don't know how it's called in yast.

Compared hwinfo from comment #5 and comment #9 is that correct rule (?):

- when "Driver" is not included in "Driver Modules" than it's built-in
- if there is "Module Alias", module is autoloaded by kernel (module.alias)

Is that correct?

Not quite. ;-)

(1)
"driver" and "driver modules" are different strings, even if they look
the same in most cases - that's why they are separated.

Please use the (hd_t).driver_module field (has probably to be added
to the hardware agent first). BTW, hwinfo normally doesn't list this field,
use "hwinfo --debug=-1 --whatever" to see it.

(2)
"module alias" does not imply that there even exists a driver (compiled in
or not). You can look for the existence of 'Driver Info #X'.

fixed in 2.15.44