This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [patch] mips-tdep: info registers
- From: Aleksandar Ristovski <aristovski at qnx dot com>
- To: gdb-patches at sources dot redhat dot com
- Date: Wed, 04 Mar 2009 16:47:34 -0500
- Subject: Re: [patch] mips-tdep: info registers
- References: <gn1iej$o3b$1@ger.gmane.org> <20090223020820.GC26056@adacore.com> <20090223025230.GA11699@caradoc.them.org> <200902230718.n1N7IoBD028396@brahms.sibelius.xs4all.nl> <gnuhhs$kmv$2@ger.gmane.org> <20090223161756.GA19411@caradoc.them.org> <20090223162602.GH26056@adacore.com> <gnumqh$acr$1@ger.gmane.org> <gnupjj$kvk$1@ger.gmane.org> <20090227195607.GJ26056@adacore.com>
Joel Brobecker wrote:
I think there are two parts to this patch:
...
This patch is fine, and approved. Please just run it through the testsuite
before checking in. (can you remember to post a ChangeLog for it as well?)
Ok, I committed this part. I have had weird issues with
running testsuite, I ended up running all the test
one-by-one. (tested on linux).
The mips-tdep part can be treated independently. Although I don't see
any problem with it, and you updated it the same way I would have changed
it, I'm not sure about going against the comment:
/* Map [gdbarch_num_regs .. 2*gdbarch_num_regs) onto the raw registers,
but then don't make the raw register names visible. */
It looks fine to return "0" ... "31" as the names of the raw registers,
but I'd like someone with more experience with the mips target to confirm
it.
...
I made all suggested changes. New patch is attached.
Note: I tested this on our gdb based on gdb 6.8 sources,
using our remote protocol to connect to a MIPS target. The
changes are the same - the code affected by the patch does
not differ.
On the HEAD gdb sources I did a compile but I could not do a
test on the target.
Thanks,
Aleksandar
* mips-tdep.c (mips_register_name): Handle numeric GPR
register numbers.
(mips_print_registers_info): Remove gdb_assert.
Index: gdb/mips-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/mips-tdep.c,v
retrieving revision 1.490
diff -u -p -r1.490 mips-tdep.c
--- gdb/mips-tdep.c 22 Feb 2009 01:02:17 -0000 1.490
+++ gdb/mips-tdep.c 4 Mar 2009 21:39:25 -0000
@@ -440,13 +440,27 @@ mips_register_name (struct gdbarch *gdba
"t8", "t9", "k0", "k1", "gp", "sp", "s8", "ra"
};
+ /* MIPS GPR register numbers, as used by assembler. Order must
+ reflect gdb's regno<->MIPS register number mapping which is
+ currently 1-1. */
+ static char *mips_gpr_numeric_names[] = {
+ "0", "1", "2", "3", "4", "5", "6", "7",
+ "8", "9", "10", "11", "12", "13", "14", "15",
+ "16", "17", "18", "19", "20", "21", "22", "23",
+ "24", "25", "26", "27", "28", "29", "30", "31"
+ };
+
enum mips_abi abi = mips_abi (gdbarch);
- /* Map [gdbarch_num_regs .. 2*gdbarch_num_regs) onto the raw registers,
- but then don't make the raw register names visible. */
+ /* Map [gdbarch_num_regs .. 2*gdbarch_num_regs) onto the raw registers. */
int rawnum = regno % gdbarch_num_regs (gdbarch);
if (regno < gdbarch_num_regs (gdbarch))
- return "";
+ {
+ if (regno >= 0 && regno < 32)
+ return mips_gpr_numeric_names [regno];
+ else
+ return "";
+ }
/* The MIPS integer registers are always mapped from 0 to 31. The
names of the registers (which reflects the conventions regarding
@@ -4601,7 +4615,6 @@ mips_print_registers_info (struct gdbarc
{
if (regnum != -1) /* do one specified register */
{
- gdb_assert (regnum >= gdbarch_num_regs (gdbarch));
if (*(gdbarch_register_name (gdbarch, regnum)) == '\0')
error (_("Not a valid register for the current processor type"));