Bug 15003
| Summary: | 3C574 (ep0) reads bogus ethernet address | ||
|---|---|---|---|
| Product: | Base System | Reporter: | gshapiro+freebsd <gshapiro+freebsd> |
| Component: | i386 | Assignee: | Matthew N. Dodd <mdodd> |
| Status: | Closed FIXED | ||
| Severity: | Affects Only Me | ||
| Priority: | Normal | ||
| Version: | 3.3-STABLE | ||
| Hardware: | Any | ||
| OS: | Any | ||
State Changed From-To: open->feedback Is this still a problem in more recent versions of FreeBSD, for example 4.1-RELEASE? Responsible Changed From-To: freebsd-bugs->mdodd Over to ep maintainer. State Changed From-To: feedback->open From the origination --- Date: Tue, 22 Aug 2000 13:04:46 -0700 (PDT) (22:04 CEST) State Changed From-To: open->closed Close my own PR as this isn't an issue any more -- it works in 4.X |
The latest 3.3-STABLE has begun supporting my PCMCIA Ethernet card (3Com Megahertz 3CXE574B) using /sys/i386/isa/if_ep.c with the commit of change 1.78.2.4 with one slight problem. The Ethernet address is read as: Nov 14 13:21:48 monkeyboy /kernel: Card inserted, slot 1 Nov 14 13:21:58 monkeyboy /kernel: ep0: [*UTP*] address 02:01:02:01:02:01 What I find interesting about the bogus Ethernet address is the number 0201 is also the ID number for the new card in the source: case 0x0201: /* 3C574BT */ To make this more interesting, before 3.3-STABLE had this support I was using the patch committed to 4.0-CURRENT in commits 1.86 and 1.87 and the card worked fine, including getting the proper Ethernet address. It appears that the MFC of the changes broken part of it. In the 4.0-CURRENT code, the epb->prod_id == 0x4b57, in the 3.3-STABLE code, it's 0x0201 which shouldn't happen as the product ID is in the card. Although the card works fine beyond the bogus Ethernet address, if more than one of us shows up to a USENIX or IETF with this card, there will be Ethernet address conflicts. Any chance this can be fixed before 3.3-STABLE goes into code slush state in preparation for 3.4?