Subject: Re: where is cpu_exec_ecoff_probe()? (Re: CVS commit: syssrc)
To: None <email@example.com>
From: Jonathan Stone <jonathan@DSG.Stanford.EDU>
Date: 11/23/2000 00:02:56
>> I guess that NetBSD/mips have not used native ecoff binaries
>> in its history (NetBSD/alpha did, BTW)
I beleive the first kernel I bootstrapped was runinng ECOFF binaries,
but compiled wiht the Ultirx 4.a mips cc. That aside, I think we went
from 4.4bsd-style a.out binaries to ELF.
>Doing the same
>for NetBSD would make it impossible to execute either NetBSD
>or Ultrix ecoff binaries, since the first matching execsw entry
>is used for binary.
This begs the question, *why* suport ecoff binaries? We can't do
ECOFF/mips shared-libs, as there's no standard for ECOFF shared-libs
on MIPs and (last I looked) no support for it in the GNU
toolchain. ECOFF on mips has been obsolete in Unix-style systems
since IRIX 4.0.5 was supplanted by IRIX 5.
Supporting other little-endian ECOFF mips systems may sound sensible,
but the only one I can think of is NT. "nuff said.
Ted Lemon and I made a concerted effort to get rid of ECOFF on pmax;
Simon Burge mostly finished it (the only vesitges are for network
booting on old PROms). It seemed to all of us that getting rid of
ECOFF on mips was a worthy goal.
Now, perhaps there are reasons to try and support native ECOFF; but
could someone explain what they are? Or, if that's not the goal, what