Make dwarf_vma uint64_t
authorAlan Modra <amodra@gmail.com>
Fri, 12 Aug 2022 08:17:51 +0000 (17:47 +0930)
committerAlan Modra <amodra@gmail.com>
Sat, 13 Aug 2022 04:41:27 +0000 (14:11 +0930)
commit31e5a3a3807e7d5b43845d0d17ddfd1a8692752c
tree3fc68ca03ec7226ba6af6e92b60d79e0bb747d41
parent625d49fce79d6154de07edc00b36335e7e7a0479
Make dwarf_vma uint64_t

This replaces dwarf_vma, dwarf_size_type and dwarf_signed_vma with
uint64_t and int64_t everywhere.  The patch also gets rid of
DWARF_VMA_FMT since we can't use that with uint64_t, and all of the
configure support for deciding the flavour of HOST_WIDEST_INT.
dwarf_vmatoa also disappears, replacing most uses with one of
PRIx64, PRId64 or PRIu64.  Printing of size_t and ptrdiff_t values
now use %z and %t rather than by casting to unsigned long.  Also,
most warning messages that used 0x%lx or similar now use %#lx and a
few that didn't print the 0x hex prefix now also use %#.  The patch
doesn't change normal readelf output, except in odd cases where values
previously might have been truncated.
binutils/config.in
binutils/configure
binutils/configure.ac
binutils/dwarf.c
binutils/dwarf.h
binutils/objdump.c
binutils/readelf.c
binutils/sysdep.h