Alan Hayward [Tue, 28 May 2019 09:07:54 +0000 (10:07 +0100)]
Suppress SIGTTOU when handling errors
Calls to error () can cause SIGTTOU to send gdb to the background.
For example, on an Arm build:
(gdb) b main
Breakpoint 1 at 0x10774: file /build/gdb/testsuite/../../../src/binutils-gdb/gdb/testsuite/gdb.base/watchpoint.c, line 174.
(gdb) r
Starting program: /build/gdb/testsuite/outputs/gdb.base/watchpoint/watchpoint
Alan Modra [Tue, 28 May 2019 05:36:47 +0000 (15:06 +0930)]
Alpha-linux linker segmentation fault
This patch cures a linker segfault, and "FAIL: Build pr22263-1".
PR 24596
* elf64-alpha.c (elf64_alpha_relocate_section): Don't attempt
to emit R_ALPHA_GOTTPREL in PIEs, for which no space is
allocated in alpha_dynamic_entries_for_reloc.
Alan Modra [Tue, 28 May 2019 02:19:30 +0000 (11:49 +0930)]
m68k linker segmentation faults
This doesn't fix the underlying bug, but an abort is better than a
segfault.
PR 24596
* elf32-m68k.c (elf_m68k_get_got_entry): Don't create a new
entry when MUST_FIND. Abort when MUST_FIND not found.
(elf_m68k_get_bfd2got_entry): Likewise.
(elf_m68k_relocate_section): Remove now useless assert.
Alan Modra [Mon, 27 May 2019 23:11:56 +0000 (08:41 +0930)]
HPPA64 linker segmentation faults
One of the ld tests produces:
failed with: <Segmentation fault>, no expected output
FAIL: Discarded dynamic relocation section
This patch cures the segv. (The test still fails with ld producing
a really messed up output, DT_RELA at address 0!)
PR 24596
* elf64-hppa.c (elf64_hppa_finalize_dynreloc): Get the output bfd
from bfd_link_info, not an output section owner.
(elf64_hppa_finish_dynamic_symbol, elf64_hppa_finalize_opd): Likewise.
(elf_hppa_final_link_relocate): Likewise.
Alan Modra [Mon, 27 May 2019 10:30:11 +0000 (20:00 +0930)]
Obsolete tic30-aout, and linker segmentation faults
See also the FIXME. tic30-aout linker support is so bad (and has been
that way since the initial tic30-aout commit) that I'm obsoleting the
target. This patch fixes numerous linker testsuite segmentation faults.
PR 24596
* aout-tic30.c (MY_bfd_final_link): Don't segfault on missing
create_object_symbols_section, obj_textsec, obj_datasec or
obj_bsssec. Fix other errors in placement.
* config.bfd: Obsolete tic30-aout.
Alan Modra [Mon, 27 May 2019 06:08:55 +0000 (15:38 +0930)]
XCOFF linker segmentation fault
The XCOFF linker temporarily trims the output bfd section list,
without adjusting section_count to suit. This is a little rude, but
the dwarf line number code can easily cope with this situation. So
check for a NULL end of list as well as limiting the saved section
VMAs to the first section_count list entries.
Also fixes
-FAIL: Weak test 3 (main, static) (32-bit)
-FAIL: Weak test 3 (main, static) (64-bit)
PR 24596
* dwarf2.c (save_section_vma, section_vma_same): Check for NULL
end of section list as well as section_count.
* xcofflink.c (xcoff_link_add_symbols): Fix temporarily changed
section list before returning error.
Alan Modra [Mon, 27 May 2019 02:07:51 +0000 (11:37 +0930)]
Fix failure on powerpc 32-bit only targets
Targets that lack ppc64 support were failing the new prefix-reloc
test. This patch adds some test infrastructure to deal with that, and
changes the powerpc gas usage info so that "-a64" is omitted when
unsupported.
I've been meaning to break up the usage message for a long time;
While doing so causes translators some work now, it should make it
easier next time a new powerpc option is added.
* config/tc-ppc.c (is_ppc64_target): New function.
(md_show_usage): Split up usage message. Don't show -a64 when
unsupported.
testsuite/gas/ppc/ppc.exp (supports_ppc64): New.
(prefix-reloc): Only run for ppc64.
Alan Modra [Mon, 27 May 2019 01:07:37 +0000 (10:37 +0930)]
readelf group errors/warnings
I noticed that one of the readelf errors stopped processing of further
group sections. This patch makes readelf continue on to other groups,
like it does with the other errors.
* readelf.c (process_section_groups): Continue processing groups
when sh_entsize exceeds group size.
Alan Modra [Mon, 27 May 2019 00:59:22 +0000 (10:29 +0930)]
Another generic ELF target assertion failure
After fixing the ld-elf/pr22836-1a segmentation fault we run into an
assertion failure due to the generic ELF target not removing empty
SHT_GROUP sections. Avoid that.
* elf.c (bfd_elf_set_group_contents): Exit on zero size section.
Alan Modra [Mon, 27 May 2019 00:08:42 +0000 (09:38 +0930)]
Generic ELF target group signature symbol
Even though the generic ELF target doesn't handle groups correctly,
this helps avoid a segfault in bfd_elf_set_group_contents seen on
d30v-elf, dlx-elf, pj-elf, and xgate-elf when linking the pr22836
testcase.
PR 24596
bfd/
* linker.c (_bfd_generic_link_output_symbols): Heed BSF_KEEP.
ld/
* emultempl/genelf.em (gld${EMULATION_NAME}_after_open): Set
BFS_KEEP on group signature symbol.
* testsuite/ld-aarch64/aarch64-elf.exp: Add new tests.
* testsuite/ld-aarch64/variant_pcs-1.s: New asm for tests.
* testsuite/ld-aarch64/variant_pcs-2.s: New asm for tests.
* testsuite/ld-aarch64/variant_pcs-now.d: New test.
* testsuite/ld-aarch64/variant_pcs-r.d: New test.
* testsuite/ld-aarch64/variant_pcs-shared.d: New test.
* testsuite/ld-aarch64/variant_pcs.ld: New linker script for tests.
Szabolcs Nagy [Thu, 25 Apr 2019 14:07:10 +0000 (15:07 +0100)]
aarch64: override default elf .set handling in gas
Allow st_other values such as STO_AARCH64_VARIANT_PCS to be set for alias
symbols independently. This is needed for ifunc symbols which are
aliased to the resolver using .set and don't expect resolver attributes
to override the ifunc symbol attributes. This means .variant_pcs must be
added explicitly to aliases.
gas/ChangeLog:
* config/tc-aarch64.c (aarch64_elf_copy_symbol_attributes): Define.
* config/tc-aarch64.h (aarch64_elf_copy_symbol_attributes): Declare.
(OBJ_COPY_SYMBOL_ATTRIBUTES): Define.
* testsuite/gas/aarch64/symbol-variant_pcs-3.d: New test.
* testsuite/gas/aarch64/symbol-variant_pcs-3.s: New test.
Szabolcs Nagy [Thu, 25 Apr 2019 14:06:53 +0000 (15:06 +0100)]
aarch64: handle .variant_pcs directive in gas
In ELF objects the specified symbol is marked with STO_AARCH64_VARIANT_PCS.
gas/ChangeLog:
* config/tc-aarch64.c (s_variant_pcs): New function.
* doc/c-aarch64.texi: Document .variant_pcs.
* testsuite/gas/aarch64/symbol-variant_pcs-1.d: New test.
* testsuite/gas/aarch64/symbol-variant_pcs-1.s: New test.
* testsuite/gas/aarch64/symbol-variant_pcs-2.d: New test.
* testsuite/gas/aarch64/symbol-variant_pcs-2.s: New test.
Szabolcs Nagy [Thu, 25 Apr 2019 12:46:01 +0000 (13:46 +0100)]
aarch64: add STO_AARCH64_VARIANT_PCS and DT_AARCH64_VARIANT_PCS
The bottom 2 bits of st_other are used for visibility, the top 6 bits are
de facto reserved for processor specific use. This patch defines a
bits to mark function symbols that follow a variant procedure call standard
with different register usage convention.
A dynamic tag is also defined that marks modules with R_<CLS>_JUMP_SLOT
relocations referencing symbols marked with STO_AARCH64_VARIANT_PCS.
This can be used by dynamic linkers that support lazy binding to decide
what registers need to be preserved during symbol resolution.
Alan Modra [Tue, 30 Apr 2019 07:03:25 +0000 (16:33 +0930)]
PowerPC GOT_PCREL34 optimisation
bfd/
* elf64-ppc.c (ppc64_elf_check_relocs): Set has_gotrel for
R_PPC64_GOT_PCREL34.
(xlate_pcrel_opt): New function.
(ppc64_elf_edit_toc): Handle R_PPC64_GOT_PCREL34.
(ppc64_elf_relocate_section): Edit GOT indirect to GOT relative
for R_PPC64_GOT_PCREL34. Implement R_PPC64_PCREL_OPT optimisation.
ld/
* testsuite/ld-powerpc/pcrelopt.s,
* testsuite/ld-powerpc/pcrelopt.d,
* testsuite/ld-powerpc/pcrelopt.sec: New test.
* testsuite/ld-powerpc/powerpc.exp: Run it.
Peter Bergner [Tue, 15 May 2018 21:48:14 +0000 (16:48 -0500)]
PowerPC add initial -mfuture instruction support
This patch adds initial 64-bit insn assembler/disassembler support.
The only instruction added is "pnop" along with the automatic aligning
of prefix instruction so they do not cross 64-byte boundaries.
include/
* dis-asm.h (WIDE_OUTPUT): Define.
* opcode/ppc.h (prefix_opcodes, prefix_num_opcodes): Declare.
(PPC_OPCODE_POWERXX, PPC_GET_PREFIX, PPC_GET_SUFFIX),
(PPC_PREFIX_P, PPC_PREFIX_SEG): Define.
opcodes/
* ppc-dis.c (ppc_opts): Add "future" entry.
(PREFIX_OPCD_SEGS): Define.
(prefix_opcd_indices): New array.
(disassemble_init_powerpc): Initialize prefix_opcd_indices.
(lookup_prefix): New function.
(print_insn_powerpc): Handle 64-bit prefix instructions.
* ppc-opc.c (PREFIX_OP, PREFIX_FORM, SUFFIX_MASK, PREFIX_MASK),
(PMRR, POWERXX): Define.
(prefix_opcodes): New instruction table.
(prefix_num_opcodes): New constant.
binutils/
* objdump.c (disassemble_bytes): Set WIDE_OUTPUT in flags.
gas/
* config/tc-ppc.c (ppc_setup_opcodes): Handle prefix_opcodes.
(struct insn_label_list): New.
(insn_labels, free_insn_labels): New variables.
(ppc_record_label, ppc_clear_labels, ppc_start_line_hook): New funcs.
(ppc_frob_label, ppc_new_dot_label): Move functions earlier in file
and call ppc_record_label.
(md_assemble): Handle 64-bit prefix instructions. Align labels
that are on the same line as a prefix instruction.
* config/tc-ppc.h (tc_frob_label, ppc_frob_label): Move to
later in the file.
(md_start_line_hook): Define.
(ppc_start_line_hook): Declare.
* testsuite/gas/ppc/prefix-align.d,
* testsuite/gas/ppc/prefix-align.s: New test.
* testsuite/gas/ppc/ppc.exp: Run new test.
Jose E. Marchesi [Thu, 23 May 2019 21:17:39 +0000 (23:17 +0200)]
bfd: fix build with --enable-targets=all in 32-bit hosts
This patch avoids for bpf_elf64_le_vec to be referenced in targmatch.h
when building a BFD without BFD64, resulting in an undefined symbol.
This was a regression introduced along with the BPF target.
bfd/ChangeLog:
2019-05-23 Jose E. Marchesi <jose.marchesi@oracle.com>
* config.bfd (targ_cpu): Process bpf-*-none only if BFD64.
* configure.ac: Set target_size=64 for bpf_elf64_le_vec and
bpf_elf64_be_vec.
* configure: Regenerate.
Jose E. Marchesi [Thu, 23 May 2019 17:06:17 +0000 (19:06 +0200)]
binutils: add support for eBPF
This patch adds support for ELF64 eBPF to readelf, and fixes a `nm'
test to run properly in bpf-*-* targets.
binutils/ChangeLog:
2019-05-23 Jose E. Marchesi <jose.marchesi@oracle.com>
* readelf.c: Include elf/bpf.h.
(guess_is_rela): Hanle EM_BPF.
(dump_relocations): Likewise.
(is_32bit_abs_reloc): Likewise.
* testsuite/binutils-all/nm.exp: Add bpf-*-* to the list of
ELF targets.
Alan Modra [Thu, 23 May 2019 00:52:56 +0000 (10:22 +0930)]
Re: Have the linker report an error if the same script is used twice
git commit 6ec6968b1b2 results in
... error: linker script file '/usr/local/lib64/libgcc_s.so' appears multiple times
collect2: error: ld returned 1 exit status
FAIL: bootstrap
This patch changes things so that an error is given only when a -T
script or the default script is invoked more than once. I'm still a
little nervous that we match script file names, not the entire path.
PR 24576
* ldfile.c (enum script_open_style): New.
(struct script_name_list): New.
(ldfile_open_command_file_1): Take a script_open_style param
rather than booleans. Adjust callers. Only fail when -T or
default -T script is invoked twice.
(ldfile_try_open_bfd): Revert last change.
Tom Tromey [Wed, 22 May 2019 19:41:28 +0000 (15:41 -0400)]
Constify target_ops::follow_exec
I noticed that target_ops::follow_exec took a "char *" parameter,
where "const char *" would be more correct. This patch changes this
(and related functions) to be constified.
Tested by rebuilding.
gdb/ChangeLog
2019-05-22 Tom Tromey <tromey@adacore.com>
Nick Clifton [Wed, 22 May 2019 14:58:57 +0000 (15:58 +0100)]
Have the linker report an error if the same script is used twice.
PR 24576
* ld/ldfile.c: (ldfile_open_command_file_1): Add new parameter -
is_script. If true check that the file has not already been
parsed as a linker script.
(ldfile_open_script_file): New function.
(ldfile_try_open_bfd): Use the new function in place of
ldfile_open_command_line.
* ldmain.c (main): Likewise.
* lexsup.c (parse_args): Use the new function for opening linker
scripts with the -T option.
* ldfile.h (ldfile_open_script_file): Add prototype.
Alan Hayward [Wed, 22 May 2019 13:02:17 +0000 (14:02 +0100)]
AArch64: Treat pauth ops as nops on non-pauth systems
Running an address signed binary through GDB on a non pauth system
gives the following error:
Call Frame Instruction op 45 in vendor extension space is not handled on this architecture.
Instead GDB should ignore the op, treating it as a nop.
Add test case for pauth binaries, regardless of whether the target
supports it.
gdb/ChangeLog:
* aarch64-tdep.c (aarch64_execute_dwarf_cfa_vendor_op): Treat
DW_CFA_AARCH64_negate_ra_state as nop on non pauth targets.
gdb/testsuite/ChangeLog:
* gdb.arch/aarch64-pauth.c: New test.
* gdb.arch/aarch64-pauth.exp: New file.
Simon Cook [Wed, 22 May 2019 11:59:58 +0000 (12:59 +0100)]
gdb/riscv: Improve flen length determination
This solves an assertion failure when a remote provides a target
description which only refers to floating point registers by their
hardware name (e.g. f0), rather than their ABI name (e.g. ft0). GDB
assumed that should the floating point register feature be presented,
it would contain a register called ft0.
The floating point length is now instead determined by searching for
the same register, but looking for any of its aliases.
gdb/ChangeLog:
* riscv-tdep.c (riscv_gdbarch_init): Support determining flen from
target descriptions using exclusively floating point register name
aliases.
Alan Modra [Wed, 22 May 2019 08:22:59 +0000 (17:52 +0930)]
bfdtest1 segfaults on hppa-hp-hpux10
The archive element cache needs tidying when closing an archive element.
This patch fixes these failures:
-FAIL: ar long file names (bfdtest1)
-FAIL: ar thin archive (bfdtest1)
-FAIL: ar thin archive with nested archive (bfdtest1)
John Darrington [Wed, 22 May 2019 05:16:14 +0000 (07:16 +0200)]
S12Z: GAS: New option --mdollar-hex.
This option (also implied by --traditional) causes '$' to introduce
literal hexadecimal constants, rather than the modern convention '0x'.
gas/
* config/tc-s12z.c (s12z_strtol): New function. (md_show_usage): Update.
(md_parse_option): new case OPTION_DOLLAR_HEX. (s12z_init_after_args):
(<global>): Use s12z_strtol instead of strtol.
* doc/c-s12z.texi (S12Z Options): Document new option -mdollar-hex.
* testsuite/gas/s12z/dollar-hex.d: New file.
* testsuite/gas/s12z/dollar-hex.s: New file.
* testsuite/gas/s12z/s12z.exp: Add them.
gdb/fortran: Use floatformats_ia64_quad for fortran 16-byte floats
GDB was changed such that the Fortran's 16-byte float format was
obtained by calling gdbarch_floatformat_for_type instead of just using
gdbarch_long_double_format as it was before.
The problem with this default_floatformat_for_type can return NULL in
some cases, and the code introduced in 34d11c682f didn't consider
this.
This commit introduces several alternative strategies for finding a
suitable 16-byte floating point type. First GDB calls
gdbarch_floatformat_for_type (this was what 34d11c682f added), if this
returns null GDB will use gdbarch_long_double_format if it is the
correct size (this was the format used before 34d11c682f). Finally,
if neither of the above provides a suitable type then GDB will create
a new dummy type.
This final dummy type is unlikely to provide an correct debug
experience as far as examining the 16-byte floats, but it should
prevent GDB crashing.
gdb/ChangeLog:
PR gdb/18644:
* f-lang.c (build_fortran_types): Handle the case where
gdbarch_floatformat_for_type returns a nullptr.
Faraz Shahbazker [Sat, 18 May 2019 08:19:03 +0000 (01:19 -0700)]
MIPS/LD: Fix memory fault linking non-PIC object in to shared library
bfd/
* elfxx-mips.c (_bfd_mips_elf_check_relocs): Add NULL pointer
checks. Search the RELA table for n64 relocations.
ld/
* testsuite/ld-mips-elf/pic-reloc-5.d: New test.
* testsuite/ld-mips-elf/pic-reloc-6.d: New test.
* testsuite/ld-mips-elf/pic-reloc-7.d: New test.
* testsuite/ld-mips-elf/pic-reloc-5.s: New test source.
* testsuite/ld-mips-elf/pic-reloc-6.s: New test source.
* testsuite/ld-mips-elf/pic-reloc-7.s: New test source.
* testsuite/ld-mips-elf/mips-elf.exp: Run the new tests.
Sudakshina Das [Tue, 21 May 2019 17:20:48 +0000 (18:20 +0100)]
[binutils, ARM] <spec_reg> changes for VMRS and VMSR instructions
This patch makes changes to the <spec_reg> operand for VMRS and VMSR
instructions as per the Armv8.1-M Mainline.
New <spec_reg> options to support are:
0b0010: FPSCR_nzcvqc, access to FPSCR condition and saturation flags.
0b1100: VPR, privileged only access to the VPR register.
0b1101: P0, access to VPR.P0 predicate fields
0b1110: FPCXT_NS, enables saving and restoring of Non-secure floating
point context.
0b1111: FPCXT_S, enables saving and restoring of Secure floating point
context
*** gas/ChangeLog ***
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* config/tc-arm.c (parse_operands): Update case OP_RVC to
parse p0 and P0.
(do_vmrs): Add checks for valid operands with respect to
cpu and fpu options.
(do_vmsr): Likewise.
(reg_names): New reg_names for FPSCR_nzcvqc, VPR, FPCXT_NS
and FPCXT_S.
* testsuite/gas/arm/armv8_1-m-spec-reg.d: New.
* testsuite/gas/arm/armv8_1-m-spec-reg.s: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad1.d: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad2.d: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad3.d: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad1.l: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad2.l: New.
* testsuite/gas/arm/armv8_1-m-spec-reg-bad3.l: New.
* testsuite/gas/arm/vfp1xD.d: Updated to allow new valid values.
* testsuite/gas/arm/vfp1xD_t2.d: Likewise.
*** opcodes/ChangeLog ***
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* arm-dis.c (coprocessor_opcodes): New instructions for VMRS
and VMSR with the new operands.
Sudakshina Das [Tue, 21 May 2019 17:15:13 +0000 (18:15 +0100)]
[binutils, Arm] Add support for conditional instructions in Armv8.1-M Mainline
This patch adds the following instructions which are part of the
Armv8.1-M Mainline:
CINC
CINV
CNEG
CSINC
CSINV
CSNEG
CSET
CSETM
CSEL
gas/ChangeLog:
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* config/tc-arm.c (TOGGLE_BIT): New.
(T16_32_TAB): New entries for cinc, cinv, cneg, csinc,
csinv, csneg, cset, csetm and csel.
(operand_parse_code): New OP_RR_ZR.
(parse_operand): Handle case for OP_RR_ZR.
(do_t_cond): New.
(insns): New instructions for cinc, cinv, cneg, csinc,
csinv, csneg, cset, csetm, csel.
* testsuite/gas/arm/armv8_1-m-cond-bad.d: New test.
* testsuite/gas/arm/armv8_1-m-cond-bad.l: New test.
* testsuite/gas/arm/armv8_1-m-cond-bad.s: New test.
* testsuite/gas/arm/armv8_1-m-cond.d: New test.
* testsuite/gas/arm/armv8_1-m-cond.s: New test.
opcodes/ChangeLog:
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* arm-dis.c (enum mve_instructions): New enum
for csinc, csinv, csneg, csel, cset, csetm, cinv, cinv
and cneg.
(mve_opcodes): New instructions as above.
(is_mve_encoding_conflict): Add cases for csinc, csinv,
csneg and csel.
(print_insn_mve): Accept new %<bitfield>c and %<bitfield>C.
Sudakshina Das [Tue, 21 May 2019 17:11:08 +0000 (18:11 +0100)]
[binutils, Arm] Add support for shift instructions in MVE
This patch adds the following instructions which are part of
Armv8.1-M MVE:
ASRL (imm)
ASRL (reg)
LSLL (imm)
LSLL (reg)
LSRL
SQRSHRL
SRQSHR
SQSHLL
SQSHL
SRSHRL
SRSHR
UQRSHLL
UQRSHL
UQSHLL
UQSHL
URSHLL
URSHL
*** gas/ChangeLog ***
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* config/tc-arm.c (operand_parse_code): New entries for
OP_RRnpcsp_I32 (register or integer operands).
(do_mve_scalar_shift): New.
(insns): New instructions for asrl, lsll, lsrl, sqrshrl, sqrshr, sqshl
sqshll, srshr, srshrl, uqrshll, uqrshl, uqshll, uqshl, urshrl and urshr.
* testsuite/gas/arm/mve-shift.d: New.
* testsuite/gas/arm/mve-shift.s: New.
* testsuite/gas/arm/mve-shift-bad.d: New.
* testsuite/gas/arm/mve-shift-bad.s: New.
* testsuite/gas/arm/mve-shift-bad.l: New.
*** opcodes/ChangeLog ***
2019-05-21 Sudakshina Das <sudi.das@arm.com>
* arm-dis.c (emun mve_instructions): Updated for new instructions.
(mve_opcodes): New instructions for asrl, lsll, lsrl, sqrshrl,
sqrshr, sqshl, sqshll, srshr, srshrl, uqrshll, uqrshl, uqshll,
uqshl, urshrl and urshr.
(is_mve_okay_in_it): Add new instructions to TRUE list.
(is_mve_unpredictable): Add cases for UNPRED_R13 and UNPRED_R15.
(print_insn_mve): Updated to accept new %j,
%<bitfield>m and %<bitfield>n patterns.
Faraz Shahbazker [Tue, 14 May 2019 00:19:37 +0000 (17:19 -0700)]
MIPS/gas: Reject $0 as source register for DAUI instruction
The MIPS64R6 TRM requires that the source register for DAUI
not be r0.
[1] "MIPS Architecture for Programmers Volume II-A: The MIPS64
Instruction Set Manual", Imagination Technologies Ltd., Document
Number: MD00087, Revision 6.06, December 15, 2016, Section 3.2
"Alphabetical List of Instructions", pp. 67-68.
gas/
* testsuite/gas/mips/r6-branch-constraints.s: Rename to ...
* testsuite/gas/mips/r6-reg-constraints.s: this and add test
case for DAUI.
* testsuite/gas/mips/r6-branch-constraints.l: Rename to ...
* testsuite/gas/mips/r6-reg-constraints.l: this and add test
for DAUI.
* testsuite/gas/mips/mips.exp: Rename test from
r6-branch-constraints to r6-reg-constraints.
opcodes/
* mips-opc.c (mips_builtin_opcodes): Change source register
constraint for DAUI.
Matthew Fortune [Tue, 14 May 2019 00:03:19 +0000 (17:03 -0700)]
[MIPS] Add generation of PLT entries with compact jumps for MIPS R6
Add a new option to get the linker to emit PLTs that use compact
branches instead of delay slot branches.
bfd/
* elfxx-mips.c (LA25_BC): New macro.
(mips_elf_link_hash_table)<compact_branches>: New field.
(STUB_JALRC): New macro.
(mipsr6_o32_exec_plt0_entry_compact): New array.
(mipsr6_n32_exec_plt0_entry_compact): Likewise.
(mipsr6_n64_exec_plt0_entry_compact): Likewise.
(mipsr6_exec_plt_entry_compact): Likewise.
(mips_elf_create_la25_stub): Use BC instead of J for stubs
when compact_branches is true.
(_bfd_mips_elf_finish_dynamic_symbol): Choose the compact
PLT for MIPSR6 with compact_branches. Do not reorder the
compact branches PLT. Switch the lazy stub for MIPSR6
with compact_branches to use JALRC.
(mips_finish_exec_plt): Choose the compact PLT0 for MIPSR6
when compact_branches is true.
(_bfd_mips_elf_compact_branches): New function.
* elfxx-mips.h (_bfd_mips_elf_compact_branches): New prototype.
ld/
* emultempl/mipself.em (compact_branches): New static variable.
(mips_create_output_section_statements): Call
_bfd_mips_elf_compact_branches.
(PARSE_AND_LIST_PROLOGUE): Add OPTION_COMPACT_BRANCHES and
OPTION_NO_COMPACT_BRANCHES.
(PARSE_AND_LIST_LONGOPTS): Add compact-branches,
no-compact-branches.
(PARSE_AND_LIST_OPTIONS): Add --compact-branches,
--no-compact-branches.
(PARSE_AND_LIST_ARGS_CASES): Handle the above.
* ld.texinfo: Document --compact-branches, --no-compact-branches.
* testsuite/ld-mips-elf/pic-and-nonpic-1-r6.dd: New test.
* testsuite/ld-mips-elf/pic-and-nonpic-1-r6.nd: New test.
* testsuite/ld-mips-elf/pic-and-nonpic-3a-r6.dd: New test.
* testsuite/ld-mips-elf/pic-and-nonpic-3a-r6.gd: New test.
* testsuite/ld-mips-elf/pic-and-nonpic-1a-r6.s: New test source.
* testsuite/ld-mips-elf/pic-and-nonpic-3a-r6.s: New test source.
* testsuite/ld-mips-elf/mips-elf.exp: Run the new tests.
Tom de Vries [Tue, 21 May 2019 14:32:41 +0000 (16:32 +0200)]
[gdb/cli] Fix use of uninitialized variable in complete_command
When building gdb on ubuntu 16.04 with gcc 5.4.0, and running the gdb
testsuite we run into:
...
FAIL: gdb.linespec/explicit.exp: complete after -line: \
cmd complete "b -line argument " (timeout)
...
The failure is reproducible outside the testsuite like this:
...
$ gdb -q build/gdb/testsuite/outputs/gdb.linespec/explicit/explicit \
-ex "complete b -line argument"
Reading symbols from \
build/gdb/testsuite/outputs/gdb.linespec/explicit/explicit...
terminate called after throwing an instance of 'std::length_error'
what(): basic_string::_M_create
Aborted (core dumped)
...
The problem is here in complete_command:
...
completion_result result = complete (arg, &word, "e_char);
std::string arg_prefix (arg, word - arg);
if (result.number_matches != 0)
...
The problem is that the word variable is not initialized when
result.number_matches == 0, but the variable is still used in the arg_prefix
initialization.
Fix this by guarding the arg_prefix initialization with the
'result.number_matches != 0' test.
Build and tested on x86_64-linux.
gdb/ChangeLog:
2019-05-21 Tom de Vries <tdevries@suse.de>
PR cli/24587
* cli/cli-cmds.c (complete_command): Fix use of unitialized variable.
Andre Vieira [Tue, 21 May 2019 13:51:43 +0000 (14:51 +0100)]
[GAS, Arm] PR24559: Fix pseudo load-operations for Armv8-M Baseline
gas/ChangeLog:
2019-05-21 Andre Vieira <andre.simoesdiasvieira@arm.com>
PR 24559
* config/tc-arm.c (move_or_literal_pool): Set size_req to 0
for MOVW replacement.
* testsuite/gas/arm/load-pseudo.s: New test input.
* testsuite/gas/arm/m0-load-pseudo.d: New test.
* testsuite/gas/arm/m23-load-pseudo.d: New test.
* testsuite/gas/arm/m33-load-pseudo.d: New test.
Tamar Christina [Tue, 21 May 2019 12:04:08 +0000 (13:04 +0100)]
AArch64: Implement choice between Cortex-A53 erratum workarounds. (PR ld/24373)
The Cortex-A53 erratum currently has two ways it can resolve the erratum when
using the flag --fix-cortex-a53-843419:
1) If the address is within the range of an ADR instruction it rewrites the ADRP
into an ADR, and those doesn't need the use of a veneer.
2) If the address is not within range, it adds a branch to a veneer which will
execute the final bit of the erratum workaround and branch back to the call
site.
When we do this we always generate the veneers and we always align the size of
the text section to 4KB. This is because we only know which workaround we can
use after all linking has finished and all addresses are known. This means even
though the veneers are not used, we still generate the section and we still
change the size of the input section.
This is problematic for small memory devices as this would require the user to
take about a ~4KB hit in memory even though it's not even used.
Since there's no real way to restart the linking process from the final write
phase this patch solves the issue by allowing the user more control over which
erratum workaround gets used.
Concretely this changes the option --fix-cortex-a53-843419 to take optional
arguments --fix-cortex-a53-843419[=full|adr|adrp]
- full (default): Use both ADRP and ADR workaround. This is equivalent to not
specifying any options and is the default behavior before this
patch.
- adr: Only use the ADR workaround, this will not cause any increase in binary
size but linking will fail if the referenced address is out of range of
an ADR instruction.
- adrp: Use only the ADRP workaround, this will never rewrite your ADRP.
In the cases where the user knows how big their binaries are the `adr` option
would prevent the unneeded overhead.
bfd/ChangeLog:
PR ld/24373
* bfd-in.h (enum erratum_84319_opts): New
(bfd_elf64_aarch64_set_options, bfd_elf32_aarch64_set_options): Change
int to enum erratum_84319_opts.
* bfd-in2.h: Regenerate.
* elfnn-aarch64.c (struct elf_aarch64_link_hash_table): Change
fix_erratum_843419 to use new enum, remove fix_erratum_843419_adr.
(_bfd_aarch64_add_stub_entry_after): Conditionally create erratum stub.
(aarch64_size_one_stub): Conditionally size erratum 843419 stubs.
(_bfd_aarch64_resize_stubs): Amend comment.
(elfNN_aarch64_size_stubs): Don't generate stubs when no workaround
requested.
(bfd_elfNN_aarch64_set_options): Use new fix_erratum_843419 enum.
(_bfd_aarch64_erratum_843419_branch_to_stub): Implement selection of
erratum workaround.
(clear_erratum_843419_entry): Update erratum conditional.
ld/ChangeLog:
PR ld/24373
* emultempl/aarch64elf.em (PARSE_AND_LIST_LONGOPTS): Add optional args
to flags.
* NEWS: Add changes to flag.
(PARSE_AND_LIST_OPTIONS): Update help descriptions.
(PARSE_AND_LIST_ARGS_CASES): Add new options to parser.
* testsuite/ld-aarch64/aarch64-elf.exp: Add new run_dump_tests.
* testsuite/ld-aarch64/erratum843419-adr.d: New test.
* testsuite/ld-aarch64/erratum843419-adrp.d: New test.
* testsuite/ld-aarch64/erratum843419-far-adr.d: New test.
* testsuite/ld-aarch64/erratum843419-far-full.d: New test.
* testsuite/ld-aarch64/erratum843419-far.s: New test.
* testsuite/ld-aarch64/erratum843419-full.d: New test.
* testsuite/ld-aarch64/erratum843419-near.s: New test.
* testsuite/ld-aarch64/erratum843419-no-args.d: New test.
Tamar Christina [Tue, 21 May 2019 10:03:45 +0000 (11:03 +0100)]
AArch64: Add SVE DWARF registers
The SVE DRAWF register names are missing from binutils, this may cause objdump
and readelf to ignore certain DRAWF output as the registers are unknown (most
notably CIEs).
This patch adds the registers in accordance to the "DWARF for ARM(r) 64-bit
Architecture (AARch64) with SVE support" documentation [1].
* dwarf.c (dwarf_regnames_aarch64): Add SVE registers.
* testsuite/binutils-all/aarch64/sve-dwarf-registers.d: New test.
* testsuite/binutils-all/aarch64/sve-dwarf-registers.s: New test.
John Darrington [Tue, 21 May 2019 08:11:40 +0000 (10:11 +0200)]
GAS: Replace macro LITERAL_PREFIXDOLLAR_HEX with a runtime value.
In an upcoming commit, I need to be able to set the prefix used
to introduce hexadecimal literal constants using a command line
flag. This is not currently possible, because the switch which
determines this (LITERAL_PREFIXDOLLAR_HEX) is a macro set at
build time.
This change substitutes it for a variable to be set at start up.
gas/ChangeLog:
* expr.c (literal_prefix_dollar_hex): New variable.
(operand)[case '$']: Use the new variable instead of the old macro.
Also, move this instance of "case '$'" next to the other one, and
enable it only in the complementary proprocessor case.
* expr.h (literal_prefix_dollar_hex): Declare it.
* config/tc-epiphany.c (md_begin): Assign literal_prefix_dollar_hex.
* config/tc-ip2k.c: ditto
* config/tc-mt.c: ditto
* config/tc-epiphany.h (LITERAL_PREFIXDOLLAR_HEX): Remove macro definition.
* config/tc-ip2k.h: ditto
* config/tc-mt.h: ditto
Fix PR 24571 - Relaxation does not shorten jmp or call to target at pc-relative range boundary
The range check done to transform an absolute call/jump to a pc-relative one is
off-by-one, and that causes this shortening optimization to be missed if the
branch target is right at the range boundary.
In the non-shrinkable case, the range is what is mentioned in the ISA - -4094
bytes in the backward direction, and 4096 bytes in the positive direction.
In the shrinkable case, the forward jump range increases by two bytes (deleted
because of the shortening from call/jmp to rcall/rjmp), and therefore, the
range is -4094 in the reverse, and 4098 in the positive direction.
Fix the ranges for !shrinkable and shrinkable cases, and add a test caes to
ensure jumps to max forward and backward ranges get relaxed to rjmp.
Fix PR 24564 - link fails for some rcalls/rjmps with wraparound
The current code to compute relative distance in the wrap around case does not
handle the edge case of the target (after adjusting for implicit PC increment)
being exactly half of the wrap around distance. This patch fixes that and adds a
testcase.
The range for a forward relative jump call is 4096 bytes ((2 * 2047) + (2 bytes
for the implicit PC increment)). If the target of the jump is at a distance of
4098 bytes, it is out of range for a forward jump - however, a backward jump can
still reach that address if pmem-wrap-around is 8192.
Assume address 0 has rjmp to address 4098. With a wrap around of 8192 and
*without* adjusting for the implicit PC increment of 2 bytes, rjmp .-4096 will
jump to address 4096 (wrap around at 8192 and decreasing addresses from then
on). Adjusting 2 bytes for the implicit PC increment, the actual target is 4098.
avr_relative_distance_considering_wrap_around though, does the wrap around only
if the passed in distance is less than half of the wrap around distance. In this
case, it is exactly equal to half (original distance 4098, adjusted distance of
4096 and wraparound of 8192), and the bypassed wrap around causes the reloc
overflow error.
Fix by wrapping around even if adjusted distance is equal to half of wrap around
distance.
[MIPS] PR gas/14798: Limit IRIX5 specific default typing to IRIX targets
On IRIX 5, every global symbol that is not explicitly labelled as
being a function is assumed to be an object. There is no reason
why IRIX behaviour should extend to all MIPS targets, so limit this
to only IRIX targets.
gas/
PR 14798
* config/tc-mips.c (s_mips_globl): Only treat symbols that are
not explicitly labelled as BSF_OBJECTs for IRIX targets.
* testsuite/gas/mips/pr14798.s: New test source.
* testsuite/gas/mips/pr14798-irix.d: New test.
* testsuite/gas/mips/pr14798.d: Likewise.
* testsuite/gas/mips/mips.exp: Run the new tests.
John Darrington [Mon, 20 May 2019 17:53:30 +0000 (19:53 +0200)]
GAS: Replace macro LITERAL_PREFIXDOLLAR_HEX with a runtime value.
In an upcoming commit, I need to be able to set the prefix used
to introduce hexadecimal literal constants using a command line
flag. This is not currently possible, because the switch which
determines this (LITERAL_PREFIXDOLLAR_HEX) is a macro set at
build time.
This change substitutes it for a variable to be set at start up.
gas/ChangeLog:
* expr.c (literal_prefix_dollar_hex): New variable.
(operand)[case '$']: Use the new variable instead of the old macro.
* expr.h (literal_prefix_dollar_hex): Declare it.
* config/tc-epiphany.c (md_begin): Assign literal_prefix_dollar_hex.
* config/tc-ip2k.c: ditto
* config/tc-mt.c: ditto
* config/tc-epiphany.h (LITERAL_PREFIXDOLLAR_HEX): Remove macro definition.
* config/tc-ip2k.h: ditto
* config/tc-mt.h: ditto
Nick Clifton [Mon, 20 May 2019 16:17:24 +0000 (17:17 +0100)]
Correct the alpha sorting of the short options in the usage description of the gprof program.
PR 24569
* gprof.c (usage): Restore alphabetical sorting to short options
list. Add -r, -R and -t short options to the list.
(main): Add comment about -g and -G possibly being deprecated.
* gprof.texi: Update usage example in line with changes above.
Andrew Burgess [Fri, 3 May 2019 14:23:55 +0000 (15:23 +0100)]
gdb/fortran: Use floatformats_ia64_quad for fortran 16-byte floats
PR gdb/18644 is caused by GDB using the wrong floating point format
for gfortran's 16-byte floating point type, including when the 16-byte
float is used as the component of a 32-byte complex type.
This commit addresses the issue in two places, first in i386-tdep.c,
there is already some code to force the use of floatformats_ia64_quad
for specific named types, this is extended to include the type names
that gfortran uses for its 16-byte floats.
Second, the builtin 16-byte float type (in f-lang.c) is changed so it
no longer uses gdbarch_long_double_format. On i386 this type is not
16-bytes, but is smaller, this is not what gfortran is expecting.
Instead we now use gdbarch_floatformat_for_type and ask for a
16-byte (128 bit) type using the common gfortran type name. This is
then spotted in i386-tdep.c (thanks to the first change above) and we
again get floatformats_ia64_quad returned.
This patch was tested on X86-64/GNU-Linux using '--target_board=unix'
and '--target_board=unix/-m32', and resolves all of the known failures
associated with PR gdb/18644. I've also added the test case from the
original bug report.
gdb/ChangeLog:
PR gdb/18644:
* f-lang.c (build_fortran_types): Use floatformats_ia64_quad for
16-byte floats.
* i386-tdep.c (i386_floatformat_for_type): Use
floatformats_ia64_quad for the 16-byte floating point component
within a fortran 32-byte complex number.
gdb/testsuite/ChangeLog:
PR gdb/18644
* gdb.fortran/complex.exp: Remove setup_kfail calls.
* gdb.fortran/printing-types.exp: Add new test.
* gdb.fortran/printing-types.f90: Add 16-byte real variable for
testing.
* gdb.fortran/type-kinds.exp (test_cast_1_to_type_kind): Remove
setup_kfail call.
Andrew Burgess [Fri, 17 May 2019 21:30:34 +0000 (22:30 +0100)]
gdb: Add constructor to struct cu_partial_die_info
Adds a constructor to 'struct cu_partial_die_info' and disables the
default constructor, preventing partially initialised instances from
being created.
Update 'find_partial_die' to return a const struct.
Users of 'find_partial_die' are updated to take account of the above
two changes.
There should be no user visible changes after this commit.
gdb/ChangeLog:
* dwarf2read.c (struct cu_partial_die_info): Add constructor,
delete default constructor.
(find_partial_die): Update to return const struct.
(partial_die_parent_scope): Move variable declaration into scope
of its use and change its type to auto.
(guess_partial_die_structure_name): Likewise.
(partial_die_info::fixup): Likewise.
Alan Hayward [Tue, 30 Apr 2019 15:00:29 +0000 (16:00 +0100)]
testsuite: Add replay logging to GDBSERVER_DEBUG
Add "replay" to the list of GDBSERVER_DEBUG options. This will
cause a gdbserver.replay file to be written to the test output
directory.
At the same time switch this to a comma separated list in order
to easily handle all possible options.
The replay log is created by GDB, but has been added to
GDBSERVER_DEBUG as it is only required for gdbserver tests. To
enable it, the gdb_debug_init is overridden to allow the additional
checking, before calling the original function.
gdb/testsuite/ChangeLog:
* README (Testsuite Parameters): Add replay logging to
GDBSERVER_DEBUG.
(gdbserver,debug): Refer to GDBSERVER_DEBUG.
* lib/gdbserver-support.exp (gdbserver_start): Treat gdbserverdebug
as a comma separated list.
(gdb_debug_init): Override procedure.
Alan Hayward [Fri, 17 May 2019 15:38:02 +0000 (16:38 +0100)]
testsuite: Create .cmd files for gdb and gdbserver
When spawning gdb or gdbserver create a .cmd file in the test output
directory containing the full command line, ensuring the current gdb
instance is appended to the files so that they can be quickly matched
to the corresponding gdb.in file.
gdb/testsuite/ChangeLog:
* lib/gdb.exp (default_gdb_spawn): Call gdb_write_cmd_file.
(gdb_write_cmd_file): New procedure.
* lib/gdbserver-support.exp (gdbserver_start): Call
gdbserver_write_cmd_file.
(gdbserver_write_cmd_file): New proedure.
Alan Hayward [Fri, 17 May 2019 15:30:09 +0000 (16:30 +0100)]
testsuite: Record all gdb input to gdb.in
When debugging testsuite failures, it can be awkward parsing gdb.log to
obtain all the commands run in order to manually re-run the test.
This patch adds the functionality to save all gdb commands to the file gdb.in
when the testsuite is run. The file is saved in the directory for the test and
if gdb is restarted then .1, .2, .3 etc is added to the filename.
Once a test has been run, the .in file can be used to re-run the test in the
following way:
The code works by intercepting send_gdb. I've added a TYPE to ensure that any
commands that would destroy the playback are kept from the log (for example the
Y from an answer to a y/n question).
Adds library function standard_output_file_with_gdb_instance to open a file
postfixed with count of the gdb instance. Ensure this count is reset when a new
.exp script is run.
I've re-run a random selection of .in files to check they do not error. Logs with
commands such as "attach <pid>" will not directly work when re-run.
gdb/testsuite/ChangeLog:
* lib/gdb.exp (gdb_unload): Mark Y as an answer.
(delete_breakpoints): Likewise.
(gdb_run_cmd): Likewise.
(gdb_start_cmd): Likewise.
(gdb_starti_cmd): Likewise.
(gdb_internal_error_resync): Likewise.
(gdb_test_multiple): Likewise.
(gdb_reinitialize_dir): Likewise.
(default_gdb_exit): Likewise.
(gdb_file_cmd): Mark kill as optional.
(default_gdb_start): Call gdb_stdin_log_init.
(send_gdb): Call gdb_stdin_log_write.
(rerun_to_main): Mark Y as an answer.
(gdb_stdin_log_init): New function.
(gdb_stdin_log_write): Likewise.
Alan Hayward [Fri, 17 May 2019 13:35:23 +0000 (14:35 +0100)]
testsuite: Add option to capture GDB debug
Add both board option and environment variable which enables gdb
debug via a comma separated list and sends it to the file gdb.debug,
located in the output directory for the current test. Document this.
Add support for the environment variable in the Makefile.
The testsuite can be run with gdb debug enabled in the following way:
make check GDB_DEBUG="infrun,target,remote"
A Test with multiple invocations of GDB will all append debug to the
same log file.
Alan Hayward [Fri, 17 May 2019 13:15:01 +0000 (14:15 +0100)]
Add debug redirect option
Currently, when logging is enabled, output will be sent to both a
logfile and standard terminal output. The redirect option sends output
only to the logfile. This includes all debug output.
Add the option to redirect debug output seperately to normal
output, using the cli command:
set logging debugredirect on
By setting this and enabling logging, all output and debug will
be sent to the logfile. The user will still see all output but
no debug output.
This causes a change in behaviour for anyone currently using
logging redirect, as now only output will be redirected. Users
will have to issue the additional command above to also redirect
debug.
Alan Hayward [Fri, 17 May 2019 13:07:04 +0000 (14:07 +0100)]
Change file close behavior for tee_file
Instead of using two bools to decide if the files should close when tee_file
is closed, make file one stay open and file two close. This simplifies the
use cases for it.
Inline the make_logging_output into the calling functions (the logic here
looks ugly in order to simplify a later change).
Expand ui-redirect.exp to cover the changes, similar to mi-logging.exp.
Alan Modra [Fri, 17 May 2019 09:39:42 +0000 (19:09 +0930)]
PR24567, assertion failure in ldlang.c:6868 when compiling with -flto
As the existing comment said: "a common ought to be overridden by a
def in a -flto object". This patch makes the code actually do that,
rather than allowing a normal object file common to override a -flto
defined symbol.
PR 24567
* plugin.c (plugin_notice): Do not let a common symbol override
a non-common definition in IR.
Jan Vrany [Fri, 17 May 2019 09:58:23 +0000 (10:58 +0100)]
MI: Add new command -complete
There is a CLI command 'complete' intended to use with emacs. Such a command
would also be useful for MI frontends, when separate CLI and MI channels cannot
be used. For example, on Windows (because of lack of PTYs) or when GDB is used
through SSH session.
This commit adds a new '-complete' MI command.
gdb/Changelog:
2019-01-28 Jan Vrany <jan.vrany@fit.cvut.cz>
* mi/mi-cmds.h (mi_cmd_complete): New function.
* mi/mi-main.c (mi_cmd_complete): Likewise.
* mi/mi-cmds.c: Define new MI command -complete.
* NEWS: Mention new -complete command.
gdb/doc/ChangeLog:
2019-01-28 Jan Vrany <jan.vrany@fit.cvut.cz>
* gdb.texinfo (Miscellaneous GDB/MI Commands): Document new
MI command -complete.
gdb/testsuite/ChangeLog:
2019-01-28 Jan Vrany <jan.vrany@fit.cvut.cz>
* gdb.mi/mi-complete.exp: New file.
* gdb.mi/mi-complete.cc: Likewise.
projects / thirdparty / binutils-gdb.git / log
