Alan Modra [Mon, 21 Feb 2005 10:30:49 +0000 (10:30 +0000)]
* lib/ld-lib.exp (run_dump_test): Don't require a dump program if
#warning given. Rearrange to allow $program to remain unset.
Don't allow gas errors. Append objcopy_as_link output to that
from the linker before testing against expected output. Fail the
test if warning not found when expected. Conversely fail the
test if ld errors or warnings given when not expected.
Alan Modra [Mon, 21 Feb 2005 08:18:49 +0000 (08:18 +0000)]
* ld-elf/warn1.d: Specify -Ttext.
* ld-scripts/align.exp (align2a, align2b: Don't run on aout targets.
* ld-scripts/align2.t: Discard all but .text and .data.
* ld-scripts/align2a.d: Accept non-readonly for coff.
* ld-scripts/align2b.d: Likewise.
* lib/ld-lib.exp (is_aout_format): New function.
Replace string variables with a filename completer with
add_setshow_optional_filename_cmd.
* solib.c, symfile.c: Update.
* cli/cli-setshow.c: Do not tidle expand the file.
Mark Kettenis [Sun, 20 Feb 2005 01:19:24 +0000 (01:19 +0000)]
* value.h (value_contents_equal): New prototype.
* value.c (value_contents_equal): New function.
* varobj.c: Include "exceptions.h" and "gdb_assert.h". Don't
include <math.h>.
(varobj_set_value): Initialize error to zero.
(varobj_update): Rename error2 to error and initialize it to zero.
Slightly change the wording of some comments.
(my_value_equal): Reimplement using TRY_CATCH and
value_contents_equal.
Thiemo Seufer [Fri, 18 Feb 2005 22:12:53 +0000 (22:12 +0000)]
/gas/ChangeLog
* config/tc-mips.c (macro_build_ldst_constoffset): Fail on $at
uses after .set noat.
(load_address): Likewise.
(macro): Likewise. Don't try to avoid $at use by sacrificing
the target register before it is stored, it won't work.
/gas/testsuite/ChangeLog
* gas/mips/noat-1.s, gas/mips/noat-1.d, gas/mips/noat-2.s,
gas/mips/noat2.l, gas/mips/noat-3.s, gas/mips/noat-3.l,
gas/mips/noat-4.s, gas/mips/noat-4.l, gas/mips/noat-5.s,
gas/mips/noat-5.l, gas/mips/noat-6.s, gas/mips/noat-6.l,
gas/mips/noat-7.s, gas/mips/noat-7.l: New files, testcases for
.set noat in macro expansions.
* gas/mips/mips.exp: Run new testcases.
* gas/mips/rol-hw.d, gas/mips/rol-hw.l, gas/mips/rol.d,
gas/mips/rol.l, gas/mips/rol.s, gas/mips/rol64-hw.d,
gas/mips/rol64-hw.l, gas/mips/rol64.d, gas/mips/rol64.l,
gas/mips/rol64.s, gas/mips/uld2-eb.d, gas/mips/uld2-el.d,
gas/mips/uld2.l, gas/mips/uld2.s, gas/mips/ulh2-eb.d,
gas/mips/ulh2-el.d, gas/mips/ulh2.l, gas/mips/ulh2.s,
gas/mips/ulw2-eb-ilocks.d, gas/mips/ulw2-eb.d,
gas/mips/ulw2-el-ilocks.d, gas/mips/ulw2-el.d, gas/mips/ulw2.l,
gas/mips/ulw2.s: Don't try to test .set noat.
Alan Modra [Fri, 18 Feb 2005 01:55:16 +0000 (01:55 +0000)]
* ld-scripts/align2a.d: Don't assume anything about section
padding or alignment, or max page size. Allow 64-bit addresses.
* ld-scripts/align2b.d: Likewise.
Zack Weinberg [Thu, 17 Feb 2005 20:57:01 +0000 (20:57 +0000)]
* ldexp.c (assigning_to_dot): New global flag.
(fold_name): If assigning_to_dot is true, object immediately to
an undefined symbol.
(exp_fold_tree): Set and clear assigning_to_dot around the
recursive call to exp_fold_tree to process the right-hand side
of an assignment to the location counter.
testsuite:
* ld-scripts/align.exp: Rename existing "ALIGN" test to "align1".
Add dump tests "align2a", "align2b", "align2c".
* ld-scripts/align2.t, ld-scripts/align2a.s, ld-scripts/align2a.d
* ld-scripts/align2b.s, ld-scripts/align2b.d
* ld-scripts/align2c.s, ld-scripts/align2c.d: New files.
Alexandre Oliva [Thu, 17 Feb 2005 17:11:01 +0000 (17:11 +0000)]
bfd/ChangeLog:
* elf32-frv.c (elf32_frv_relocate_section): Remove warning from
uninitialized check_segment[1] in TLSMOFF case.
Reported by Alan Modra.
(elf32_frv_relocate_section): Improve errors and warnings.
ld/testsuite/ChangeLog:
* ld-frv/fdpic-static-6.d: Update.
* ld-frv/fdpic*.d: Remove explicit -mfdpic from #as. Update
spelling of errors and warnings.
Alan Modra [Thu, 17 Feb 2005 12:59:59 +0000 (12:59 +0000)]
* deflex.l (YY_NO_UNPUT): Define.
* rclex.l (YY_NO_UNPUT): Define.
* rcparse.y (null_unichar): New static var.
(res_null_text): Use it rather than attempting to init from wchar_t.
* windres.c: Include assert.h and time.h before getopt.h.
Include config.h and unistd.h too.
* config/tc-ia64.c (ia64_parse_name): Don't advance 'name' when
parsing inN, locN, outN. Set 'idx' to offset register number starts
at. Don't handle numbers with leading zeroes or beyond 95. Remove
pointless cast.
Thiemo Seufer [Wed, 16 Feb 2005 15:49:13 +0000 (15:49 +0000)]
* config/tc-mips.c (load_address): Fix formatting.
(macro): Don't use AT if .set noat is in effect. Fix formatting.
Catch macros which are unexpandable without AT. Remove duplicate
zeroing of used_at.
(macro2): Remove duplicate zeroing of used_at.
Mark Mitchell [Wed, 16 Feb 2005 07:49:26 +0000 (07:49 +0000)]
* configure.in (AC_CHECK_FUNCS): Add glob.
* configure: Regenerated.
* emultempl/elf32.em (<glob.h>): Do not include if HAVE_GLOB is
not defined.
(gld${EMULATION_NAME}_parse_ld_so_conf_include): Do not use glob
if HAVE_GLOB is not defined.
Alan Modra [Wed, 16 Feb 2005 02:20:13 +0000 (02:20 +0000)]
bfd/
* elf64-ppc.c (ppc64_elf_edit_toc): Skip toc if size is zero.
Skip toc reloc adjust if reloc_count is zero.
ld/
* emultempl/ppc64elf.em (ppc_create_output_section_statements): Make
an error creating fake bfd fatal.
(ppc_before_allocation): Continue on after errors.
(gld${EMULATION_NAME}_finish): Likewise.
* elf32-mips.c (elf_mips16_howto_table_rel): New array for MIPS16
reloc howtos. Add R_MIPS16_HI16 and R_MIPS16_LO16 relocs and
R_MIPS16_GOT16 and R_MIPS16_CALL16 placeholders.
(elf_mips16_jump_howto): Move into elf_mips16_howto_table_rel.
(elf_mips16_gprel_howto): Likewise. Redefine src_mask and
dst_mask.
(mips16_gprel_reloc): Remove bit shuffling; call
_bfd_mips16_elf_reloc_unshuffle(), _bfd_mips_elf_gprel16_with_gp()
and _bfd_mips16_elf_reloc_shuffle() instead.
(mips16_reloc_map): New reloc map for MIPS16 relocs.
(bfd_elf32_bfd_reloc_type_lookup): Use mips16_reloc_map for MIPS16
relocs.
(mips_elf32_rtype_to_howto): Fetch MIPS16 howtos from
elf_mips16_howto_table_rel.
* elf64-mips.c (mips16_elf64_howto_table_rel): New array for
MIPS16 REL reloc howtos. Add R_MIPS16_HI16 and R_MIPS16_LO16
relocs and R_MIPS16_GOT16 and R_MIPS16_CALL16 placeholders.
(elf_mips16_jump_howto): Move into mips16_elf64_howto_table_rel.
(elf_mips16_gprel_howto): Likewise. Redefine src_mask and
dst_mask.
(mips16_elf64_howto_table_rela): New array for MIPS16 RELA
reloc howtos. Add R_MIPS16_26, R_MIPS16_GPREL, R_MIPS16_HI16 and
R_MIPS16_LO16 relocs and R_MIPS16_GOT16 and R_MIPS16_CALL16
placeholders.
(mips16_gprel_reloc): Remove bit shuffling; call
_bfd_mips16_elf_reloc_unshuffle(), _bfd_mips_elf_gprel16_with_gp()
and _bfd_mips16_elf_reloc_shuffle() instead.
(mips16_reloc_map): New reloc map for MIPS16 relocs.
(bfd_elf64_bfd_reloc_type_lookup): Use mips16_reloc_map for MIPS16
relocs.
(mips_elf64_rtype_to_howto): Fetch MIPS16 howtos from
mips16_elf64_howto_table_rela or mips16_elf64_howto_table_rel.
* elfn32-mips.c (elf_mips16_howto_table_rel): New array for MIPS16
REL reloc howtos. Add R_MIPS16_HI16 and R_MIPS16_LO16 relocs and
R_MIPS16_GOT16 and R_MIPS16_CALL16 placeholders.
(elf_mips16_jump_howto): Move into elf_mips16_howto_table_rel.
(elf_mips16_gprel_howto): Likewise. Redefine src_mask and
dst_mask.
(mips16_gprel_reloc): Remove bit shuffling; call
_bfd_mips16_elf_reloc_unshuffle(), _bfd_mips_elf_gprel16_with_gp()
and _bfd_mips16_elf_reloc_shuffle() instead.
(mips16_reloc_map): New reloc map for MIPS16 relocs.
(bfd_elf32_bfd_reloc_type_lookup): Use mips16_reloc_map for MIPS16
relocs.
(mips_elf_n32_rtype_to_howto): Fetch MIPS16 howtos from
elf_mips16_howto_table_rela or elf_mips16_howto_table_rel.
* elfxx-mips.c (_bfd_mips16_elf_reloc_unshuffle): New function to
handle bit shuffling for MIPS16 relocs.
(_bfd_mips16_elf_reloc_shuffle): Likewise.
(_bfd_mips_elf_lo16_reloc): Use _bfd_mips16_elf_reloc_unshuffle()
and _bfd_mips16_elf_reloc_shuffle().
(_bfd_mips_elf_generic_reloc): Likewise.
(mips_elf_calculate_relocation): Likewise. Handle R_MIPS16_HI16
and R_MIPS16_LO16.
(mips_elf_obtain_contents): Remove bit shuffling.
(mips_elf_perform_relocation): Likewise; call
_bfd_mips16_elf_reloc_unshuffle() and _bfd_mips16_elf_reloc_shuffle()
instead.
(_bfd_mips_elf_relocate_section): Likewise. Handle R_MIPS16_HI16
and R_MIPS16_LO16.
* elfxx-mips.h (_bfd_mips16_elf_reloc_unshuffle): Declare.
(_bfd_mips16_elf_reloc_shuffle): Likewise.
* reloc.c (BFD_RELOC_MIPS16_HI16): New reloc.
(BFD_RELOC_MIPS16_HI16_S): Likewise.
(BFD_RELOC_MIPS16_LO16): Likewise.
* bfd-in2.h: Regenerate.
* libbfd.h: Regenerate.
* config/tc-mips.c (reloc_needs_lo_p): Handle
BFD_RELOC_MIPS16_HI16_S.
(fixup_has_matching_lo_p): Handle BFD_RELOC_MIPS16_LO16.
(append_insn): Add BFD_RELOC_MIPS16_GPREL, BFD_RELOC_MIPS16_HI16_S
and BFD_RELOC_MIPS16_LO16 to relocs to suppress overflow
complaints on.
(mips16_ip): Resolve BFD_RELOC_MIPS16_HI16_S,
BFD_RELOC_MIPS16_HI16 and BFD_RELOC_MIPS16_LO16 for constants.
Call my_getSmallExpression() to parse percent operators.
(percent_op_match, mips_percent_op): Separate definitions.
(mips16_percent_op): Define percent operators for the MIPS16 mode.
(parse_relocation): Handle the MIPS16 mode using
mips16_percent_op.
(md_apply_fix3): Handle BFD_RELOC_MIPS16_HI16,
BFD_RELOC_MIPS16_HI16_S and BFD_RELOC_MIPS16_LO16.
* gas/mips/mips16-hilo.d: New test for the R_MIPS16_HI16 and
R_MIPS16_LO16 relocs.
* gas/mips/mips16-hilo-n32.d: Likewise, for the n32 ABI.
* gas/mips/mips16-hilo.s: Source for the new tests.
* gas/mips/mips.exp: Run the new tests.
* ld-mips-elf/mips16-hilo.d: New test for the R_MIPS16_HI16 and
R_MIPS16_LO16 relocs.
* ld-mips-elf/mips16-hilo-n32.d: Likewise, for the n32 ABI.
* ld-mips-elf/mips16-hilo.s: Auxiliary source for the new tests.
* ld-mips-elf/mips-elf.exp: Run the new tests.
* config/tc-ia64.c (md_apply_fix3): Call ia64_gen_real_reloc_type
instead of explicitly dealing with the translation; exclude
relocations that are already pcrel, however.
* config/tc-ia64.c: Include limits.h (if available).
(gr_values[0]): Set path to INT_MAX.
(dot_reg_val): Don't allow changing value of r0. Limit range of
general registers at r127.
(specify_resource): Default resource index is -1. Don't set resource
index (in case IA64_RS_RSE) without setting the specific flag.
(note_register_values): Check operand is O_constant before tracking
input value of moves. Add tracking for dep.z with constant inputs.
(print_dependency): Resource index of specific resource may be zero.
(check_dependencies): Likewise.
* gas/ia64/dv-raw-err.s: Don't use r0 or f0 as output operand.
* gas/ia64/dv-waw-err.s: Likewise.
* gas/ia64/reg-err.[ls]: New.
* gas/ia64/ia64.exp: Run new test.
* config/tc-ia64.c (ia64_gen_real_reloc_type): Define and initialize
new variables type, suffix, and width. Handle
BFD_RELOC_IA64_DIR(32|64)[LM]SB in FUNC_LT_FPTR_RELATIVE case.
Handle BFD_RELOC_IA64_DIR64[LM]SB in FUNC_TP_RELATIVE case. Add
FUNC_DTP_MODULE case. Handle BFD_RELOC_IA64_DIR32[LM]SB in
FUNC_DTP_RELATIVE case. Return incoming relocation type if
BFD_RELOC_IA64_IPLT[LM]SB in FUNC_IPLT_RELOC case. Generate warning
if unable to translate relocation type, using the new variables.
* elfxx-ia64.c (elfNN_ia64_relax_section): Allow relax
backward branch in the same section.
(elfNN_ia64_relocate_section): Inform users that the input
section is too big to relax br instruction when overflow
happens to R_IA64_PCREL21B, R_IA64_PCREL21BI, R_IA64_PCREL21M
and R_IA64_PCREL21F.