Remove gdbarch_bits_big_endian
authorTom Tromey <tromey@adacore.com>
Mon, 25 Nov 2019 19:31:02 +0000 (12:31 -0700)
committerTom Tromey <tromey@adacore.com>
Wed, 4 Dec 2019 16:31:18 +0000 (09:31 -0700)
commitd5a22e77b56529eade787419ea80a7e9e7d2aa6f
tree9efbd056575604aae8ac03e072aac8e2439ea3cd
parent7ab4a236ce85f1e0950e88e267e679a066b77bed
Remove gdbarch_bits_big_endian

From what I can tell, set_gdbarch_bits_big_endian has never been used.
That is, all architectures since its introduction have simply used the
default, which is simply check the architecture's byte-endianness.

Because this interferes with the scalar_storage_order code, this patch
removes this gdbarch setting entirely.  In some places,
type_byte_order is used rather than the plain gdbarch.

gdb/ChangeLog
2019-12-04  Tom Tromey  <tromey@adacore.com>

* ada-lang.c (decode_constrained_packed_array)
(ada_value_assign, value_assign_to_component): Update.
* dwarf2loc.c (rw_pieced_value, access_memory)
(dwarf2_compile_expr_to_ax): Update.
* dwarf2read.c (dwarf2_add_field): Update.
* eval.c (evaluate_subexp_standard): Update.
* gdbarch.c, gdbarch.h: Rebuild.
* gdbarch.sh (bits_big_endian): Remove.
* gdbtypes.h (union field_location): Update comment.
* target-descriptions.c (make_gdb_type): Update.
* valarith.c (value_bit_index): Update.
* value.c (struct value) <bitpos>: Update comment.
(unpack_bits_as_long, modify_field): Update.
* value.h (value_bitpos): Update comment.

Change-Id: I379b5e0c408ec8742f7a6c6b721108e73ed1b018
13 files changed:
gdb/ChangeLog
gdb/ada-lang.c
gdb/dwarf2loc.c
gdb/dwarf2read.c
gdb/eval.c
gdb/gdbarch.c
gdb/gdbarch.h
gdb/gdbarch.sh
gdb/gdbtypes.h
gdb/target-descriptions.c
gdb/valarith.c
gdb/value.c
gdb/value.h