cvs commit: src/sys/dev/em if_em.c if_em.h if_em_hw.c if_em_hw.h if_em_osdep.h

[Hiroki Sato]

"Jack Vogel" <jfvogel at gmail.com> wrote
  in <2a41acea0610280019r15d1e40bgbec37d9e0f72633e at mail.gmail.com>:

jf> On 10/28/06, Hiroki Sato <hrs at freebsd.org> wrote:
jf> > Jack F Vogel <jfv at FreeBSD.org> wrote
jf> >   in <200610280137.k9S1bFq2089275 at repoman.freebsd.org>:
jf> >
jf> > jf> jfv         2006-10-28 01:37:14 UTC
jf> > jf>
jf> > jf>   FreeBSD src repository
jf> > jf>
jf> > jf>   Modified files:        (Branch: RELENG_6)
jf> > jf>     sys/dev/em           if_em.c if_em.h if_em_hw.c if_em_hw.h
jf> > jf>                          if_em_osdep.h
jf> > jf>   Log:
jf> > jf>   Merge of Intel 6.2.9 em driver code.
jf> > jf>   Approved by: re, scottl, jhb, pdeuskar
jf> > jf>
jf> > jf>   Revision   Changes    Path
jf> > jf>   1.65.2.19  +731 -589  src/sys/dev/em/if_em.c
jf> > jf>   1.32.2.5   +97 -71    src/sys/dev/em/if_em.h
jf> > jf>   1.16.2.4   +574 -531  src/sys/dev/em/if_em_hw.c
jf> > jf>   1.15.2.5   +96 -148   src/sys/dev/em/if_em_hw.h
jf> > jf>   1.14.2.3   +46 -52    src/sys/dev/em/if_em_osdep.h
jf> >
jf> > Just wanted to make sure, but is the following change in if_em.c
jf> > really intentional?  This means that the new version no longer
jf> > supports 82542...
jf> >
jf> > Index: if_em.c
jf> > @@ -116,8 +117,6 @@
jf> >         { 0x8086, E1000_DEV_ID_82541GI_LF,      PCI_ANY_ID, PCI_ANY_ID, 0},
jf> >         { 0x8086, E1000_DEV_ID_82541GI_MOBILE,  PCI_ANY_ID, PCI_ANY_ID, 0},
jf> >
jf> > -       { 0x8086, E1000_DEV_ID_82542,           PCI_ANY_ID, PCI_ANY_ID, 0},
jf> > -
jf> >         { 0x8086, E1000_DEV_ID_82543GC_FIBER,   PCI_ANY_ID, PCI_ANY_ID, 0},
jf> >         { 0x8086, E1000_DEV_ID_82543GC_COPPER,  PCI_ANY_ID, PCI_ANY_ID, 0},
jf>
jf> Yes that was intentional, its an ID that the Intel source has not
jf> had for some time. When I put it back in our source a while back
jf> due to a merge our test group came to me and said these adapters
jf> dont even work with the driver, so clearly no one is using them :)
jf>
jf> I asked about dropping the ID to a set of developers and got the OK to drop
jf> it.
jf>
jf> If someone actually speaks up about having hardware that was working and
jf> now is broken I'll take it all back and we can put the ID back in, is that good
jf> enough? :)

 So here is the report (from nyan@):

 em0 at pci1:1:0:   class=0x020000 card=0x10008086 chip=0x10008086 rev=0x03 hdr=0x00
    vendor   = 'Intel Corporation'
    device   = '82542 Gigabit Ethernet Controller'
    class    = network
    subclass = ethernet

 It is not recognized by the new driver at boot time while worked fine
 with the old one.  I think we should not remove the existing
 hardware support (especially in -STABLE) if possible.

--
| Hiroki SATO
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/cvs-src/attachments/20061028/0811266d/attachment.pgp