Hi Rainer,

All you would have to do is add support for the card and test it in your 
setup. After that you can submit the code to be included in Solaris. 
Before your changes are incorportaed we will do an internal code review 
and testing.

Feel free to contact me if you have more questions.


Rainer Orth wrote:

>Back during S10 Platinum Beta, I raised this RFE
>       http://bugs.opensolaris.org/bugdatabase/view_bug.do?bug_id=4958870
>to have bge(7D) support the Broadcom NetXtreme 5901 card which seems to be
>popular in some laptops (like my IBM ThinkPad R40e :-).  Having used
>Broadcom's own bcme(7D) driver for a long time, I've recently tried bge(7D)
>as of snv_33, and it seemed to just work initially, except for the expected
>Mar 17 20:13:03 llaima bge: [ID 801725 kern.warning] WARNING: bge0: Device 
>'pci14e4,170d' not recognized (5901?)
>Mar 17 20:13:03 llaima bge: [ID 801725 kern.info] NOTICE: bge0: 
>ddi_intr_get_supported_types() returned: 1
>Mar 17 20:13:03 llaima bge: [ID 801725 kern.info] NOTICE: bge0: bge_add_intrs: 
>interrupt type 0x1
>Mar 17 20:13:03 llaima bge: [ID 801725 kern.info] NOTICE: bge0: Using FIXED 
>interrupt type
>Mar 17 20:13:03 llaima mac: [ID 543131 kern.info] NOTICE: bge0/0 registered
>Unfortunately, I'm just now using the driver for a large package and patch
>installation (Studio 11, actually), and the link goes up and down
>regularly, so there is obviously more work to do ;-(  I initially had a
>look at some BSD driver, and apart from mentioning the pci id, there seemed
>to be no special support for it.  I vaguely recall things were different
>when I looked at Broadcom's sources to their Linux bcme driver, but I
>cannot find those right now.
>Anyway, in order to make this card fully supported, I'd like to know what
>it would take to have a change for that integrated.  As I see it, this
>would primarily mean listing the card in bge-known-subsystems in
>usr/src/uts/common/io/bge/bge.conf and handling the pci id in
>usr/src/pkgdefs/SUNWbge/post{install, remove}, and of course investigating
>and fixing the problems noticed above.
>Apart from the actual code changes, what sort of testing would be required
>to guarantee that the driver fully works for that card?
>I hope this is the right list for such questions, but couldn't come up with
>a better idea to ask those questions (maybe apart from the testing
>       Rainer
>Rainer Orth, Faculty of Technology, Bielefeld University
>request-sponsor mailing list
>request-sponsor at opensolaris.org

Reply via email to