By the way, I confirm the new pal code binary fixes both this GDB issue and the IPI warning message problem from my other thread.
Thanks, Brad -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steve Reinhardt Sent: Wednesday, June 04, 2008 1:53 PM To: M5 users mailing list Subject: Re: [m5-users] Config File for Linux 2.6.22 Looks like it was a problem with using the wrong PAL. Turns out that tsb_osfpal isn't in the m5_system_2.0b3 tarball, and the one in zizzer:/z/m5/system/stable is broken (out of date). We've got the one from /dist/m5/system now and it seems to be working. I'll put a link up on the wiki for the working tsb_osfpal. Steve On Wed, Jun 4, 2008 at 7:01 AM, nathan binkert <[EMAIL PROTECTED]> wrote: > The very low phisical address 0x10134 indicates that this code is > either in the console, or in the PAL. The reason you're hitting the > fencepost is probably because gdb can't make any sense out of what the > stack should be because the console and the PAL don't exactly follow > normal ABIs. What stage of the system are you in? I can imagine you > being in the console code if either your CPUs haven't spun up yet > because you're still booting, or if they're idle. You can nm the > console and PAL binaries to figure out what's up. You can also load > the symbols for those into GDB in addition to (or instead of) the > linux kernel. If this error is only on CPUs > the fourth, then it's > probably because you're using the wrong PAL. > > Nate > _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
