Well, that fixes it. The driver works out-of-the-box with just some minor makefile modifications. So, we've got a problem somewhere in the module handling. Either the symbol wasn't being relocated properly, or it wasn't being allocated properly, or something. I'm not an expert in this region of the kernel, but my guess is that we're going to see this more and more often, so someone with a clue should take a look at this. I'm more than willing to help, as I seem to be the only person with a 100% reproducable situation. But I really have no idea even where to begin looking... my expertise ends right about at objdump, and even then I'm not certain how some of that data should look for loadable modules. Matt -- Matthew D. Dharm Senior Software Designer Momentum Computer Inc. 1815 Aston Ave. Suite 107 (760) 431-8663 X-115 Carlsbad, CA 92008-7310 Momentum Works For You www.momenco.com > -----Original Message----- > From: owner-linux-mips@oss.sgi.com > [mailto:owner-linux-mips@oss.sgi.com]On Behalf Of Kevin Paul Herbert > Sent: Monday, February 25, 2002 10:36 AM > To: Matthew Dharm; Linux-MIPS > Subject: Re: Is this a toolchain bug? > > > At 5:57 PM -0800 2/22/02, Matthew Dharm wrote: > > > >The toolchain I'm using is the one from oss.sgi.com by H.J. Liu > >(toolchain-20011020-1). Because of the way the e1000 > driver Makefile > >works, I'm actually compiling it using the native compiler > on-target. > I just tossed out most of intel's makefile so that I could build it > as part of our cross-build (hosted on RedHat i386), > building it as a > module. When I was using the 3.0 driver, I hacked it to be > built with > the kernel (not a module at all). It was just a few minutes of > makefile hacking, and may save you some grief. > > There is a driver V4.0.7? I *just* picked up 3.5 a little while > ago... can you point me at where on intel's website you got your > driver? > > Thanks, > Kevin > -- >
