Fixed some bugs in handling vxworks960 -- not reading registers right,

trying to use FP regs (not supported by target), botching IP.

diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index fbac9ba9e787687770bcc31d8156a802e8ca596a..460affc5823ddb221577d3db36ffae44aff86d2a 100644 (file)
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,3 +1,23 @@
+Thu Jul  9 18:44:26 1992  Ken Raeburn  (raeburn@cygnus.com)
+
+       * remote-vx.c (vx_kill): Don't look for arguments; they aren't
+       being passed.
+       (vx_read_register): Declare inferior_fp_regs.  Delete code for
+       copying "inferior_registers" around; the values are already in var
+       "registers".  Use correct size for register set.
+
+       * remote-vx.c (vx_write_register): Don't try writing FP regs; the
+       target doesn't support it.
+       (vx_read_register): Don't try reading them either.
+
+       * tm-i960.h (IP_REGNUM): Move to end of non-FP list, since the
+       VxWorks back end reads them in one contiguous set except for this
+       one.
+       (register_names): Change "pc" to "pcw" to avoid confusion printing
+       "$pc".
+       * tm-vx960.h (PC_REGNUM): Use RIP_REGNUM, since that's where we
+       find the PC under VxWorks.
+
 Wed Jul  8 21:34:30 1992  Fred Fish  (fnf@cygnus.com)
 
        * dwarfread.c (alloc_utype, decode_subscr_data):  Call alloc_type