-----Original Message-----
> The gssq instruction is a Loongson extended instruction, so the gssq patch
> is applied to gdb.
>
> Signed-off-by: Huacai Chen <[email protected]>
> Signed-off-by: Youling Tang <[email protected]>
> ---
> v1 -> v2:
> - Move the patch into the gdb-7.6.patch file.
>
> gdb-7.6.patch | 21 ++++++++++++++++++++-
> 1 file changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/gdb-7.6.patch b/gdb-7.6.patch
> index f64b55f..63c9a14 100644
> --- a/gdb-7.6.patch
> +++ b/gdb-7.6.patch
> @@ -2500,4 +2500,23 @@ diff -up gdb-7.6/opcodes/configure.orig
> gdb-7.6/opcodes/configure
> +struct target_desc *tdesc_aarch64;
> #include "features/aarch64.c"
> #include "features/aarch64-without-fpu.c"
> -
> +
> +--- gdb-7.6/gdb/mips-tdep.c.orig
> ++++ gdb-7.6/gdb/mips-tdep.c
> +@@ -3261,6 +3261,16 @@ restart:
> + /* Irix 6.2 N32 ABI uses sd instructions for saving $gp and $ra. */
> + set_reg_offset (gdbarch, this_cache, reg, sp + low_word);
> + }
> ++ else if (((inst & 0xFFE08020) == 0xeba00020) /* gssq
> reg,reg,offset($sp) */
> ++ && regsize_is_64_bits)
> ++ {
> ++ reg = (inst >> 16) & 0x1F;
> ++ low_word = ((((inst >> 6) & 0x1FF) ^ 0x100) - 0x100) << 4;
> ++ set_reg_offset (gdbarch, this_cache, reg, sp + low_word);
> ++ reg = inst & 0x1F;
> ++ low_word = (((((inst >> 6) & 0x1FF) ^ 0x100) - 0x100) << 4) + 8;
> ++ set_reg_offset (gdbarch, this_cache, reg, sp + low_word);
> ++ }
> + else if (high_word == 0x27be) /* addiu $30,$sp,size */
> + {
> + /* Old gcc frame, r30 is virtual frame pointer. */
> --
> 2.1.0
Note: This can be applied with rebasing, and a build test was OK.
(also with target=PPC64).
$ make warn target=PPC64
...
if [ "x86_64" = "x86_64" ] && [ "PPC64" = "PPC64" ] && [ -f
gdb-7.6-ppc64le-support.patch ]; then \
patch -d gdb-7.6 -p1 -F0 < gdb-7.6-ppc64le-support.patch ; \
fi
...
patching file gdb/mips-tdep.c
Hunk #7 succeeded at 3592 (offset 10 lines).
Hunk #8 succeeded at 3612 (offset 10 lines).
Hunk #9 succeeded at 7636 (offset 10 lines).
Hunk #10 succeeded at 7645 (offset 10 lines).
These are fine.
Thanks,
Kazu
--
Crash-utility mailing list
[email protected]
https://listman.redhat.com/mailman/listinfo/crash-utility
