This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
[commit/mips] Wire in convert_register
- From: Andrew Cagney <cagney at gnu dot org>
- To: gdb-patches at sources dot redhat dot com
- Date: Wed, 14 Jan 2004 10:10:46 -0500
- Subject: [commit/mips] Wire in convert_register
Hello,
These functions never got enabled when they were merged in off a branch.
This enables them, and fixes a little bit of bitrot. store.exp's
float tests now work a little bit better.
commited,
Andrew
(PS: I sent this last night but it went awal).
2004-01-12 Andrew Cagney <cagney@redhat.com>
* mips-tdep.c (mips_convert_register_p): Handle both raw and
cooked floating-point registers.
(mips_gdbarch_init): Set convert_register_p, register_to_value,
and value_to_register.
Index: mips-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/mips-tdep.c,v
retrieving revision 1.271
diff -u -r1.271 mips-tdep.c
--- mips-tdep.c 13 Jan 2004 21:38:46 -0000 1.271
+++ mips-tdep.c 14 Jan 2004 02:49:19 -0000
@@ -652,7 +652,8 @@
{
return (TARGET_BYTE_ORDER == BFD_ENDIAN_BIG
&& register_size (current_gdbarch, regnum) == 4
- && (regnum) >= mips_regnum (current_gdbarch)->fp0 && (regnum) < mips_regnum (current_gdbarch)->fp0 + 32
+ && (regnum % NUM_REGS) >= mips_regnum (current_gdbarch)->fp0
+ && (regnum % NUM_REGS) < mips_regnum (current_gdbarch)->fp0 + 32
&& TYPE_CODE(type) == TYPE_CODE_FLT
&& TYPE_LENGTH(type) == 8);
}
@@ -6030,6 +6031,10 @@
set_gdbarch_deprecated_pop_frame (gdbarch, mips_pop_frame);
set_gdbarch_frame_align (gdbarch, mips_frame_align);
set_gdbarch_deprecated_save_dummy_frame_tos (gdbarch, generic_save_dummy_frame_tos);
+
+ set_gdbarch_convert_register_p (gdbarch, mips_convert_register_p);
+ set_gdbarch_register_to_value (gdbarch, mips_register_to_value);
+ set_gdbarch_value_to_register (gdbarch, mips_value_to_register);
set_gdbarch_deprecated_frame_chain (gdbarch, mips_frame_chain);
set_gdbarch_frameless_function_invocation (gdbarch,