Hi, I'm back for a bit of hacking... The notes are all ok so that was barking up the wrong tree, the culprint here seems to be somwehere when reading or dumping the fpregset data (which really should have been obvious!). I looked at the actual core dump a bit, and it seems that the .reg2 (the ones that hold floating point data) sections are a bit screwed. So it seems that the actual code that dumps those registers is screwed. The sections are noted, but they haven't been written to the core dump.
Sections: Idx Name Size VMA LMA File off Algn 0 note0 0000190c 00000000 00000000 00001000 2**0 1 .reg/28315 0000004c 00000000 00000000 00002554 2**2 2 .reg 0000004c 00000000 00000000 00002554 2**2 3 .reg2/28315 00000000 00000000 00000000 00000000 2**2 4 .reg2 00000000 00000000 00000000 00000000 2**2 5 .reg/93851 0000004c 00000000 00000000 00002744 2**2 6 .reg2/93851 00000000 00000000 00000000 00000000 2**2 The size of .reg (which should be the size of gregset_t, 76) is correct, but the size of .reg2 is bogus, it should be 0x17c (the size of fpregset_t). Anyway, this is where I'm stuck and don't know how to proceed (probobly because of the sleep deprivation). Tips of any kind would be apperciated. Good night. _______________________________________________ Bug-hurd mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-hurd
