cvs commit: src/sys/pci if_dc.c if_pcn.c if_rl.c if_sf.cif_sis.c if_sk.c if_ste.c if_ti.c if_tl.c if_vr.c if_wb.c if_xl.c

[Nate Lawson]

On Mon, 31 Mar 2003, John Baldwin wrote:
> On 31-Mar-2003 Nate Lawson wrote:
> > On Mon, 31 Mar 2003, Nate Lawson wrote:
> >>   Modified files:
> >>     sys/pci              if_dc.c if_pcn.c if_rl.c if_sf.c if_sis.c 
> >>                          if_sk.c if_ste.c if_ti.c if_tl.c if_vr.c 
> >>                          if_wb.c if_xl.c 
> >>   Log:
> >>   Clean up locking and resource management for pci/if_*
> > 
> > This is a rather large patch so if you begin to experience problems
> > attaching or detaching (say on module load/unload) a previously-working
> > card, please let me know.
> 
> Already found a bogon in tl(4) when it conflicted with some local
> changes.  Might want to double check that you don't call detach()
> for very early failures when you have no resources to teardown or
> release but just do a direct return in the other drivers.

There is only one other driver that does anything before the mtx_init
(ti(4), ti_mbuf_sanity case) and I handled that correctly.  The KASSERT in
detach should find any case that detach is called before mtx_init.  Thank
you for finding this and fixing it.

-Nate