There's this little bit of ecoff support in the h8300 port that I ran
into that I think we could remove. Grepping around BFD, I don't think
ECOFF is supported for h8300.
And if not, I'm failing to see why we'd support ECOFF debug info in
ELF for h8300 (.mdebug). It kind of seems like someone just installed
the set_gdbarch_ecoff_reg_to_regnum hook at the same time the dwarf2
and stabs hooks were installed.
I'm removing this so that greps for ecoff stuff only lead to MIPS, as
ISTR that Linux/MIPS support still relies on some subset of ECOFF for
something.
gdb/ChangeLog:
2018-07-11 Pedro Alves <palves@redhat.com>
* h8300-tdep.c (h8300_gdbarch_init): Remove
set_gdbarch_ecoff_reg_to_regnum calls.
This patch fixes a failure that happens when a structure has a static
member whose type is the same as itself. From the bug report:
Example code:
struct A
{
static A Empty;
int a;
};
int main(void) {
A a;
return 0;
}
Output:
(gdb) ptype/o A
/* offset | size */ type = struct A {
static struct A {
static struct A {
static struct A {
static struct A {
static struct A {
static struct A {
... # infinite loop
The problem here is that GDB is not taking into account the fact that
static members inside a class/struct are not stored in the
class/struct, and therefore they should not be accounted for during
the display of the offsets/sizes. The fix is simple: we just check if
the field we're dealing with (on
c-typeprint.c:c_type_print_base_struct_union) is static, and if it is
then we don't iterate over it.
This patch also adds a new test for this case, and doesn't introduce
any regressions. I believe it is important enough to be included in
the 8.2 branch.
OK?
gdb/ChangeLog:
2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com>
PR c++/23373
* c-typeprint.c (c_type_print_base_struct_union): Don't print
offsets/sizes for static members of a class/struct.
gdb/testsuite/ChangeLog:
2018-07-11 Sergio Durigan Junior <sergiodj@redhat.com>
PR c++/23373
* gdb.base/ptype-offsets.cc (struct static_member): New
struct.
(main) <stmember>: New variable.
* gdb.base/ptype-offsets.exp: Add test for printing a struct
with a static member in it.
Running gdbserver under Valgrind I get:
==26925== Conditional jump or move depends on uninitialised value(s)
==26925== at 0x473E7F: i387_cache_to_xsave(regcache*, void*) (i387-fp.c:579)
==26925== by 0x46E3ED: x86_fill_xstateregset(regcache*, void*) (linux-x86-low.c:418)
==26925== by 0x45E747: regsets_store_inferior_registers(regsets_info*, regcache*) (linux-low.c:5456)
==26925== by 0x45EEF8: linux_store_registers(regcache*, int) (linux-low.c:5731)
==26925== by 0x426441: regcache_invalidate_thread(thread_info*) (regcache.c:89)
==26925== by 0x45CCAF: linux_resume_one_lwp_throw(lwp_info*, int, int, siginfo_t*) (linux-low.c:4447)
==26925== by 0x45CE2A: linux_resume_one_lwp(lwp_info*, int, int, siginfo_t*) (linux-low.c:4519)
==26925== by 0x45E17C: proceed_one_lwp(thread_info*, lwp_info*) (linux-low.c:5216)
==26925== by 0x45DC81: linux_resume_one_thread(thread_info*, bool) (linux-low.c:5031)
==26925== by 0x45DD34: linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}::operator()(thread_info*) const (linux-low.c:5095)
==26925== by 0x462907: void for_each_thread<linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}>(linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}) (gdbthread.h:150)
==26925== by 0x45DE62: linux_resume(thread_resume*, unsigned long) (linux-low.c:5093)
==26925==
==26925== Conditional jump or move depends on uninitialised value(s)
==26925== at 0x473EBD: i387_cache_to_xsave(regcache*, void*) (i387-fp.c:586)
==26925== by 0x46E3ED: x86_fill_xstateregset(regcache*, void*) (linux-x86-low.c:418)
==26925== by 0x45E747: regsets_store_inferior_registers(regsets_info*, regcache*) (linux-low.c:5456)
==26925== by 0x45EEF8: linux_store_registers(regcache*, int) (linux-low.c:5731)
==26925== by 0x426441: regcache_invalidate_thread(thread_info*) (regcache.c:89)
==26925== by 0x45CCAF: linux_resume_one_lwp_throw(lwp_info*, int, int, siginfo_t*) (linux-low.c:4447)
==26925== by 0x45CE2A: linux_resume_one_lwp(lwp_info*, int, int, siginfo_t*) (linux-low.c:4519)
==26925== by 0x45E17C: proceed_one_lwp(thread_info*, lwp_info*) (linux-low.c:5216)
==26925== by 0x45DC81: linux_resume_one_thread(thread_info*, bool) (linux-low.c:5031)
==26925== by 0x45DD34: linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}::operator()(thread_info*) const (linux-low.c:5095)
==26925== by 0x462907: void for_each_thread<linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}>(linux_resume(thread_resume*, unsigned long)::{lambda(thread_info*)#2}) (gdbthread.h:150)
==26925== by 0x45DE62: linux_resume(thread_resume*, unsigned long) (linux-low.c:5093)
The problem is a type/width mismatch in code like this, in
gdbserver/i387-fp.c:
/* Some registers are 16-bit. */
collect_register_by_name (regcache, "fctrl", &val);
fp->fctrl = val;
In the above code:
#1 - 'val' is a 64-bit unsigned long.
#2 - "fctrl" is 32-bit in the register cache, thus half of 'val' is
left uninitialized by collect_register_by_name, which works with
an untyped raw buffer output (i.e., void*).
#3 - fp->fctrl is an unsigned short (16-bit). For some such
registers we're masking off the uninitialized bits with 0xffff,
but not in all cases.
We end up in such a fragile situation because
collect_registers_by_name works with an untyped output buffer pointer,
making it easy to pass a pointer to a variable of the wrong size.
Fix this by using regcache_raw_get_unsigned instead (actually a new
regcache_raw_get_unsigned_by_name wrapper), which always returns a
zero-extended ULONGEST register value. It ends up simplifying the
i387-tdep.c code a bit, even.
gdb/gdbserver/ChangeLog:
2018-07-11 Pedro Alves <palves@redhat.com>
* i387-fp.c (i387_cache_to_fsave, cache_to_fxsave)
(i387_cache_to_xsave): Use regcache_raw_get_unsigned_by_name
instead of collect_register_by_name.
* regcache.c (regcache_raw_get_unsigned_by_name): New.
* regcache.h (regcache_raw_get_unsigned_by_name): New.
See:
https://developer.arm.com/support/arm-security-updates/speculative-processor-vulnerability/download-the-whitepaper
opcodes * arm-dis.c (arm_opcodes): Add ssbb and pssbb and move
csdb together with them.
(thumb32_opcodes): Likewise.
gas * config/tc-arm.c (insns): Add new ssbb and pssbb instructions.
* testsuite/gas/arm/csdb.s: Add new tests for ssbb and pssbb.
* testsuite/gas/arm/csdb.d: Likewise
* testsuite/gas/arm/thumb2_it_bad.s: Likewise.
* testsuite/gas/arm/thumb2_it_bad.l: Likewise.
* testsuite/gas/arm/barrier.d: Update with ssbb.
* testsuite/gas/arm/barrier-thumb.d: Likewise.
Correct an issue with commit 0f20cc3522e7 ("TLS support for MIPS"),
<https://sourceware.org/ml/binutils/2005-02/msg00607.html>, where a
condition used to determine whether to use a dynamic symbol for GD, LD
and IE TLS dynamic relocations against a symbol that has been defined
locally has been incorrectly reversed.
It's executables rather than dynamic shared objects where no symbol is
required, because such symbols cannot be preempted and therefore their
values (thread pointer offsets) are fixed at the static link time as is
the associated module ID of the main executable, so the original
condition should have been `shared' instead of `!shared'. This wrong
condition was then later converted from `!shared' to `!bfd_link_pic',
with commit 0e1862bb401f ("Add output_type to bfd_link_info").
Use the correct `bfd_link_dll' condition then, and adjust code for the
dynamic symbol index possibly being -1 as with symbols that have been
forced local, removing unnecessary dynamic relocations from dynamic
regular executables. PIE executables are unaffected as the existing
condition excluded them by chance due to the conversion mentioned above.
Adjust test cases accordingly.
bfd/
* elfxx-mips.c (mips_tls_got_relocs): Use `bfd_link_dll' rather
than `!bfd_link_pic' in determining the dynamic symbol index.
Avoid the index of -1.
(mips_elf_initialize_tls_slots): Likewise. Flatten code by
moving `dyn' to the beginning of the function block.
ld/
* testsuite/ld-mips-elf/tlsdyn-o32.d: Update test for dynamic
relocation removal.
* testsuite/ld-mips-elf/tlsdyn-o32.got: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-1.d: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-1.got: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-2.d: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-2.got: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-3.d: Likewise.
* testsuite/ld-mips-elf/tlsdyn-o32-3.got: Likewise.
Correct a commit 0e1862bb401f ("Add output_type to bfd_link_info") issue
and use `bfd_link_dll' rather than `bfd_link_pic' in determining whether
to fully resolve GD, LD and IE TLS relocations referring to symbols
locally defined rather than deferring them to the load time by means of
dynamic relocations.
Such symbols cannot be preempted in PIE executables, which are
necessarily PIC, and therefore their values (thread pointer offsets) are
fixed at the static link time as is the associated module ID of the main
executable.
Given the `tlsbin-o32.s' and `tlsdyn-o32.s' sources from our test suite
this removes the absolute TLS relocations from the static:
DYNAMIC RELOCATION RECORDS
OFFSET TYPE VALUE
00000000 R_MIPS_NONE *ABS*
1000002c R_MIPS_TLS_TPREL32 *ABS*
10000030 R_MIPS_TLS_DTPMOD32 *ABS*
10000038 R_MIPS_TLS_DTPMOD32 *ABS*
and the dynamic:
DYNAMIC RELOCATION RECORDS
OFFSET TYPE VALUE
00000000 R_MIPS_NONE *ABS*
1000002c R_MIPS_TLS_TPREL32 *ABS*
10000038 R_MIPS_TLS_DTPMOD32 *ABS*
10000044 R_MIPS_TLS_DTPMOD32 *ABS*
10000030 R_MIPS_TLS_DTPMOD32 tlsvar_gd
10000034 R_MIPS_TLS_DTPREL32 tlsvar_gd
10000040 R_MIPS_TLS_TPREL32 tlsvar_ie
PIE executable respectively, as reported by `objdump -R', and fills the
corresponding GOT slots with the values expected, as recorded with the
test cases added. The new output from `objdump -R' is:
DYNAMIC RELOCATION RECORDS (none)
and:
DYNAMIC RELOCATION RECORDS
OFFSET TYPE VALUE
00000000 R_MIPS_NONE *ABS*
10000030 R_MIPS_TLS_DTPMOD32 tlsvar_gd
10000034 R_MIPS_TLS_DTPREL32 tlsvar_gd
10000040 R_MIPS_TLS_TPREL32 tlsvar_ie
for the static and the dynamic executable respectively.
2018-07-11 Maciej W. Rozycki <macro@mips.com>
Rich Felker <bugdal@aerifal.cx>
bfd/
PR ld/22570
* elfxx-mips.c (mips_tls_got_relocs): Use `bfd_link_dll' rather
than `bfd_link_pic' to determine whether dynamic relocations are
to be produced.
(mips_elf_initialize_tls_slots): Likewise.
ld/
PR ld/22570
* testsuite/ld-mips-elf/tlsbin-pie-o32.d: New test.
* testsuite/ld-mips-elf/tlsbin-pie-o32.got: New test.
* testsuite/ld-mips-elf/tlsdyn-pie-o32.d: New test.
* testsuite/ld-mips-elf/tlsdyn-pie-o32.got: New test.
* testsuite/ld-mips-elf/mips-elf.exp: Run the new tests.
I just stumbled over this with 2.29.1 while building a cross-toolchain, on Cygwin64, but it's still the same for 2.30. m.arena has size_t on Cygwin64 and thus errors out due to -Werror=format.
gold * main.cc: Print m.arena as long long.
tdesc_register_size returns number of bits, not bytes.
Rename to make it clearer.
Also, fixed bug in aarch64_get_tdesc_vq which assumed bytes.
gdb/
* target-descriptions.c (tdesc_register_bitsize): Rename.
* target-descriptions.h (tdesc_register_bitsize): Likewise.
* rs6000-tdep.c (rs6000_gdbarch_init): Use new name.
* aarch64-tdep.c (aarch64_get_tdesc_vq): Convert size.
Architecturally, MONITOR's and MONITORX'es memory operand is a 16- or
32-bit register outside of 64-bit mode, and a 64- or 32-bit register
inside 64-bit mode. The other register operands, including all of them
for MWAIT and MWAITX, are uniformly 32-bit, irrespective of mode. Retain
the original 64-bit MONITOR{,X} templates for compatibility only, and
fold the MWAIT{,X} ones.
The checking against reg16_inoutportreg can be had with a simple test of
a bit, and the value setting from inoutportreg can be replaced by using
the actual register's reg_type field.
Note that the so far redundant 2nd instance of OPERAND_TYPE_INOUTPORTREG
is left in place, for its use in type_names[].
The check_prefix label was bogus from the beginning: The special
checking is supposed to happen for PadLock insns only; no
3-opcode-byte insn should go this path.
With its name it should equal to the largest prior enumerator, which is
not very helpful as that will change every once in a while. Rename it to
OTNum instead to make name and value match, and correct use sites to no
longer use one too large a value.
git commit 624940689a wrongly changed fill.d and fill16.d notarget
entries to xfail, so revert that change and instead replace notarget
with skip. The two tests are complementary.
This patch also modifies the pr20022 test to use .data, avoiding
xfails on targets that complain about dynamic relocations in .text.
This shows a bug in metag which discards section _foo in pr2022.so.
* testsuite/ld-elf/hash.d: Comment on mips xfail.
* testsuite/ld-gc/pr20022b.s: Use .data not .text.
* testsuite/ld-gc/pr20022.d: Remove xfails.
* testsuite/ld-scripts/fill.d: Revert most of last patch, but use
skip rather than notarget.
* testsuite/ld-scripts/fill16.d: Likewise.
This changes moribund_locations to use a std::vector rather than VEC.
I also noticed that moribund_locations is only used in breakpoint.c,
so now it is static as well.
It might be possible to make this code a bit simpler by using a
ref_ptr in moribund_locations; however, I have not done this.
Tested by the buildbot.
gdb/ChangeLog
2018-07-10 Tom Tromey <tom@tromey.com>
* breakpoint.c (moribund_locations): Now static and a
std::vector.
(breakpoint_init_inferior, moribund_breakpoint_here_p)
(build_bpstat_chain, update_global_location_list)
(breakpoint_retire_moribund): Update.
* breakpoint.h (bp_location_p): Remove typedef. Don't declare
VEC.
Only run MIPS/LD export class call relocation tests with shared library
targets, avoiding:
.../ld/ld-new: -shared not supported
linker errors and removing:
FAIL: MIPS export class call relocation test (o32)
regressions with `mipsel-ps2-elf' and `mips64el-ps2-elf' targets.
ld/
* testsuite/ld-mips-elf/mips-elf.exp: Limit export class call
relocation tests to shared library targets.
Remove `-Wshadow' compilation errors:
cc1: warnings being treated as errors
.../bfd/elflink.c: In function 'bfd_elf_final_link':
.../bfd/elflink.c:11722: error: declaration of 'remove' shadows a global declaration
/usr/include/stdio.h:154: error: shadowed declaration is here
which for versions of GCC before 4.8 prevent support for ELF targets
from being built. See also GCC PR c/53066.
bfd/
* elflink.c (bfd_elf_final_link): Rename `remove' local variable
to `remove_section'.
elf32-xgate.c contains many functions that are only stubs and
elf32-xgate.h contains unused declarations. While this might be
reasonable for the initial commit of a port with subsequent work
fleshing out the stubs, xgate has only had two minor target specific
patches since the initial commit over six years ago. The rest of the
changes have been general maintenance work applied to all ELF targets,
and some of this work could have been avoided if the stubs hadn't been
there. So this patch removes all the stubs.
I've kept the functionality of the old elf32_xgate_add_symbol_hook,
implemented in elf32_xgate_backend_symbol_processing. Presumably,
that's to set the symbol st_target_internal flag for use in
elf32-m68hc1x.c:elf32_m68hc11_relocate_section.
The empty elf32_xgate_relocate_section meant that xgate had no linker.
Or at least, no linker relocation processing. Deleting the
elf_backend_relocate_section define means the target will now use the
generic linker reloc processing. How good that is will depend on the
accuracy of the reloc howtos..
I haven't updated the ld testsuite to xfail tests expected to fail
for generic elf targets.
bfd/
* elf32-xgate.h: Delete.
* elf32-xgate.c: Delete unnecessary forward declarations, add two
that are now needed.
(xgate_elf_bfd_link_hash_table_create)
(xgate_elf_bfd_link_hash_table_free)
(xgate_elf_set_mach_from_flags, struct xgate_scan_param)
(stub_hash_newfunc, elf32_xgate_add_symbol_hook)
(elf32_xgate_setup_section_lists, elf32_xgate_size_stubs)
(elf32_xgate_build_stubs, elf32_xgate_check_relocs)
(elf32_xgate_relocate_section, _bfd_xgate_elf_set_private_flags)
(elf32_xgate_post_process_headers): Delete.
(elf32_xgate_backend_symbol_processing): New function.
(xgate_elf_ignore_reloc, xgate_elf_special_reloc)
(_bfd_xgate_elf_print_private_bfd_data): Make static.
(ELF_TARGET_ID, elf_info_to_howto, elf_backend_check_relocs)
(elf_backend_relocate_section, elf_backend_object_p)
(elf_backend_final_write_processing, elf_backend_can_gc_sections)
(elf_backend_post_process_headers, elf_backend_add_symbol_hook)
(bfd_elf32_bfd_link_hash_table_create)
(bfd_elf32_bfd_set_private_flags)
(xgate_stub_hash_lookup): Don't define.
(elf_backend_symbol_processing): Define.
* elf-bfd.h (elf_target_id): Delete XGATE_ELF_DATA.
ld/
* emulparams/xgateelf.sh (TEMPLATE_NAME) Set to generic.
(EXTRA_EM_FILE): Set to genelf.
Expand the gdb.base/watchpoint-reuse-slot.exp test to be run twice,
once with hardware watchpoints disabled (this is new) and then with
hardware watchpoints enabled (the old way).
Running with hardware watchpoints enabled is skipped if the board file
says that hardware watchpoints are not supported on this target.
gdb/testsuite/ChangeLog:
* gdb.base/watchpoint-reuse-slot.exp: Test with hardware
watchpoints enabled and disabled.
An assertion when setting up arguments for an inferior call checks the
size of the argument against xlen. However, if xlen and flen are
different sizes, and the argument is being placed into a floating
pointer register then we should be comparing against flen not xlen.
This issue shows up as an assertion failure when running on an rv32g
target with a binary compiled using the rv32f abi and making an
inferior call involving large floating point arguments, for example
the test gdb.base/infcall-nested-structs.exp.
gdb/ChangeLog:
* riscv-tdep.c (riscv_is_fp_regno_p): New function.
(riscv_register_reggroup_p): Use new function, remove unneeded
parenthesis.
(riscv_push_dummy_call): Extend assert to compare against xlen or
flen based on register type.
In debug printing, use TYPE_SAFE_NAME instead of replicating the
functionality using TYPE_NAME and a null check.
gdb/ChangeLog:
* riscv-tdep.c (riscv_print_arg_location): Use TYPE_SAFE_NAME.
The current default dumps for n64 GOT relocation tests only work for
IRIX targets really, while the Linux dumps are suitable for non-Linux
targets as well. Make the Linux dumps the default then and handle the
IRIX ones as a special case.
Define an `irix' variable as a shorthand for `[istarget *-*-irix*]' and
use it for these n64 GOT relocation tests as well as export class call
relocation tests.
ld/
* testsuite/ld-mips-elf/elf-rel-got-n64.d: Rename to...
* testsuite/ld-mips-elf/elf-rel-got-n64-irix.d: ... this.
* testsuite/ld-mips-elf/elf-rel-got-n64-linux.d: Rename to...
* testsuite/ld-mips-elf/elf-rel-got-n64.d: ... this.
* testsuite/ld-mips-elf/elf-rel-xgot-n64.d: Rename to...
* testsuite/ld-mips-elf/elf-rel-xgot-n64-irix.d: ... this.
* testsuite/ld-mips-elf/elf-rel-xgot-n64-linux.dd: Rename to...
* testsuite/ld-mips-elf/elf-rel-xgot-n64.d: ... this.
* testsuite/ld-mips-elf/mips-elf.exp: Update accordingly.
Define `irix' as a shorthand for `[istarget *-*-irix*]' and use
it throughout.
Testing for the GCC version 5 or later isn't right, since C++ 11 support
wasn't enabled by default until later. This patch tests the C++ standard
support directly instead of inferring it from the GCC version.
gold/
* incremental.cc (Sized_incremental_binary::setup_readers): Use
emplace_back for C++ 11 or later.
Factor in the presence of a `.gnu.attributes' section at the end of the
respective dumps:
Disassembly of section .gnu.attributes:
0000000000000000 <.gnu.attributes>:
0: 41000000 bc0f 4 <__dso_displacement+0x4>
4: 0f676e75 jal d9db9d4 <__dso_displacement+0xd9db9d4>
8: 00010000 sll zero,at,0x0
c: 00070401 0x70401
and also ignore any trailing output with the IRIX variant of n64 GOT
relocation tests, like already done with the Linux and the embedded
ones, removing:
FAIL: MIPS ELF got reloc n64
FAIL: MIPS ELF xgot reloc n64
LD testsuite regressions with the `mips-sgi-irix6' target.
ld/
* testsuite/ld-mips-elf/elf-rel-got-n64.d: Add `#pass' at the
end.
* testsuite/ld-mips-elf/elf-rel-xgot-n64.d: Likewise.
Correct a MIPS/BFD linker issue with dynamic symbol and corresponding
GOT entry values being redirected to lazy binding stubs where the stubs
section has been discarded by assigning to the `/DISCARD/' output
section in the linker script used. The issue manifests itself by the
values entered being relative to the absolute section, which is what any
discarded sections are internally assigned in the linker.
For the `stub-dynsym-2.s' piece of code included as a test case with
this change this issue results in the dynamic symbol table and the GOT
looking like:
Symbol table '.dynsym' contains 3 entries:
Num: Value Size Type Bind Vis Ndx Name
0: 00000000 0 NOTYPE LOCAL DEFAULT UND
1: 00000010 0 FUNC GLOBAL DEFAULT UND bar
2: 00000000 0 FUNC GLOBAL DEFAULT UND foo
Primary GOT:
Canonical gp value: 00097ff0
Reserved entries:
Address Access Initial Purpose
00090000 -32752(gp) 00000000 Lazy resolver
00090004 -32748(gp) 80000000 Module pointer (GNU extension)
Global entries:
Address Access Initial Sym.Val. Type Ndx Name
00090008 -32744(gp) 00000010 00000010 FUNC UND bar
0009000c -32740(gp) 00000000 00000000 FUNC UND foo
if assembled to regular MIPS code, or:
Symbol table '.dynsym' contains 3 entries:
Num: Value Size Type Bind Vis Ndx Name
0: 00000000 0 NOTYPE LOCAL DEFAULT UND
1: 0000000d 0 FUNC GLOBAL DEFAULT UND bar
2: 00000001 0 FUNC GLOBAL DEFAULT UND foo
Primary GOT:
Canonical gp value: 00097ff0
Reserved entries:
Address Access Initial Purpose
00090000 -32752(gp) 00000000 Lazy resolver
00090004 -32748(gp) 80000000 Module pointer (GNU extension)
Global entries:
Address Access Initial Sym.Val. Type Ndx Name
00090008 -32744(gp) 0000000d 0000000d FUNC UND bar
0009000c -32740(gp) 00000001 00000001 FUNC UND foo
if assembled to microMIPS code. Symbol values and GOT entries record
the offset into the inexistent stubs section and the ISA bit rather than
zero, which would be the case if a lazy binding stub was not used for
other reasons, such as the value of the symbol being taken for a purpose
other than making a function call (e.g. an R_MIPS_GOT16 relocation).
Correct the issue by refraining from redirecting symbols to lazy binding
stubs if the stubs section is going to be discarded.
bfd/
* elfxx-mips.c (_bfd_mips_elf_adjust_dynamic_symbol): Don't set
`->needs_lazy_stub' if the stubs output section is the absolute
section.
ld/
* testsuite/ld-mips-elf/stub-dynsym-2.dd: New test.
* testsuite/ld-mips-elf/stub-dynsym-2.gd: New test.
* testsuite/ld-mips-elf/stub-dynsym-2.sd: New test.
* testsuite/ld-mips-elf/stub-dynsym-discard-2.gd: New test.
* testsuite/ld-mips-elf/stub-dynsym-discard-2.sd: New test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-2.dd: New test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-2.gd: New test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-2.sd: New test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-insn32-2.dd: New
test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-insn32-2.gd: New
test.
* testsuite/ld-mips-elf/stub-dynsym-micromips-insn32-2.sd: New
test.
* testsuite/ld-mips-elf/stub-dynsym-2.ld: New test linker
script.
* testsuite/ld-mips-elf/stub-dynsym-discard-2.ld: New test
linker script.
* testsuite/ld-mips-elf/mips-elf.exp: Run the new tests.
Switch to use add_setshow_zuinteger_unlimited_cmd for some of the
control variables in remote.c. The variables
hardware-watchpoint-limit, hardware-breakpoint-limit, and
hardware-watchpoint-length-limit are all changed. For example, a user
will now see this:
(gdb) show remote hardware-breakpoint-limit
The maximum number of target hardware breakpoints is unlimited.
Instead of this:
(gdb) show remote hardware-breakpoint-limit
The maximum number of target hardware breakpoints is -1.
And can do this:
(gdb) set remote hardware-breakpoint-limit unlimited
However, previously any negative value implied "unlimited", now only
-1, or the text "unlimited" can be used for unlimited. Any other
negative value will give an error about invalid input. This is a
small change in the user interface, but, hopefully, this will not
cause too many problems.
I've also added show functions for these three variables to allow for
internationalisation.
gdb/ChangeLog:
* remote.c (show_hardware_watchpoint_limit): New function.
(show_hardware_watchpoint_length_limit): New function.
(show_hardware_breakpoint_limit): New function.
(_initialize_remote): Use add_setshow_zuinteger_unlimited_cmd
where appropriate, update help text.
gdb/doc/ChangeLog:
* gdb.texinfo (Remote Configuration): Update descriptions for
set/show of hardware-watchpoint-limit, hardware-breakpoint-limit,
and hardware-watchpoint-length-limit variables.
Bring the LD implementation of `run_dump_test' in line with its binutils
and GAS counterparts and report the UNSUPPORTED status for tests the run
of which has been prevented by means of one or more of `target',
`alltargets' and `notarget' tags. Define `skip', `anyskip' and `noskip'
tags, which do not report anything for tests that are not run.
The rationale behind this is that we want to have unsupported tests
reported to ensure that they have actually been attempted and have not
been accidentally suppressed. Then tests which have target-specific
variants that cannot be expressed with a single .d file can make use of
the newly added tags to silently suppress the uninteresting variants.
ld/
* testsuite/lib/ld-lib.exp (run_dump_test): Call `unsupported'
if the target being tested has been excluded by means of one or
more of `target', `alltargets' and `notarget' tags. Add support
for `skip', `anyskip' and `noskip' tags.
Bring the LD implementation of `run_dump_test' in line with its binutils
and GAS counterparts and use the intended test name, from the `name' tag
if set, for test status reporting as soon as it is available.
ld/
* testsuite/lib/ld-lib.exp (run_dump_test): Move the setting of
`testname' earlier on and use it for test status reporting.
Remove `-Wshadow' compilation errors:
cc1: warnings being treated as errors
.../opcodes/s12z-dis.c: In function 'lea_reg_xys_opr':
.../opcodes/s12z-dis.c:814: error: declaration of 'reg' shadows a global declaration
.../opcodes/s12z-dis.c:783: error: shadowed declaration is here
.../opcodes/s12z-dis.c: In function 'lea_reg_xys':
.../opcodes/s12z-dis.c:843: error: declaration of 'reg' shadows a global declaration
.../opcodes/s12z-dis.c:783: error: shadowed declaration is here
.../opcodes/s12z-dis.c: In function 'print_insn_loop_primitive':
.../opcodes/s12z-dis.c:2206: error: declaration of 'reg' shadows a global declaration
.../opcodes/s12z-dis.c:783: error: shadowed declaration is here
which for versions of GCC before 4.8 prevent support for S12Z targets
from being built. See also GCC PR c/53066.
opcodes/
* s12z-dis.c (lea_reg_xys_opr): Rename `reg' local variable to
`reg_xys'.
(lea_reg_xys): Likewise.
(print_insn_loop_primitive): Rename `reg' local variable to
`reg_dxy'.
In the gdb.base/examine-backward.exp test script, we check to see if
address zero is readable, and then read memory first forward from
address zero, and then backward from address zero.
The problem is, being able to read address zero does not guarantee
that you'll be able to read from the other end of the address space,
and the test probably shouldn't assume that is the case.
This patch updates the test script so that even if address zero is
known non-readable, we still run the tests, the tests in question are
mostly about, can GDB calculate the correct address to read from, we
can confirm this even if the final read ultimately fails. We also no
longer assume that if address zero is readable, then the other end of
the address space will be readable.
One further change is that, when we examined the memory at address
zero, the regexp used to match the address expected that the zero
address would have two '0' digits as the least significant digits. As
GDB strips leading zeros from addresses this was causing the test to
fail. I've reduced the zero address to a single 0 digit.
gdb/testsuite/ChangeLog:
* gdb.base/examine-backward.exp: Still run tests around address
0x0, even if address 0x0 is not readable. Update the pattern for
matching address 0x0 in expected output.
Neither XM_CDEPS nor NAT_CLIBS are defined anywhere, so remove the
uses.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (CDEPS): Don't mention XM_CDEPS.
(CLIBS): Don't mention NAT_CLIBS.
Nothing defines XM_ADD_FILES, TM_ADD_FILES, or NAT_ADD_FILES any more,
so consequently ADD_FILES and ADD_DEPS are no longer needed. So, this
removes them.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (ADD_FILES, ADD_DEPS): Remove.
(LIBGDB_OBS, clean mostlyclean): Update.
(gdb$(EXEEXT), insight$(EXEEXT)): Update.
This minimizes the "make" output from the yacc and lex rules,
following the same technique as the rest of the Makefile.
The lex rule had a special case to deal with the situation where flex
is not available. I don't think this is needed, so I removed it. If
flex is truly unavailable, the person building gdb can simply "touch"
the output file.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (%.c: %.y): Use ECHO_YACC.
(%.c: %.l): Use ECHO_LEX. Just fail if flex not available.
* silent-rules.mk (ECHO_YACC, ECHO_LEX): New variables.
exec.c ws handled specially in COMMON_OBS, but there doesn't seem to
be a reason for this. This changes the Makefile to treat exec.c as an
ordinary source file.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (ALLDEPFILES): Remove exec.c.
(COMMON_OBS): Remove exec.o.
(COMMON_SFILES): Add exec.c.
I don't think anyone uses lint any more, so this removes the support
for it from the Makefile.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (LINT, LINTFLAGS, LINTFILES, lint): Remove.
This introduces a stamp file for version.c, preventing unnecessary
version.o rebuilds.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (clean mostlyclean): Remove stamp-version.
(version.c): Depend on stamp-version.
(stamp-version): New rule, from version.c rule.
This introduces a stamp file for init.c. This prevents constant
rebuilds of init.o, by arranging for init.c to only be modified when
its contents change. (FWIW this is a standard idiom in use by
Automake and by gdb itself for many years.)
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (init.c): Depend on stamp-init.
(stamp-init): New rule, from init.c rule.
(clean mostlyclean): Remove stamp-init.
This simplifies the INIT_FILES variable. COMMON_OBS includes
CONFIG_OBS and SUBDIR_GCC_COMPILE_OBS, so there's no need to reference
CONFIG_OBS or SUBDIR_GCC_COMPILE_SRCS there. Once this is done, it it
clear that duplicates can't occur, so remove the duplicate-removing
code as well.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (INIT_FILES): Remove CONFIG_SRCS,
SUBDIR_GCC_COMPILE_SRCS.
The init.c build rule has a few sed expressions that aren't necessary
any more. This removes them.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (init.c): Remove some unused sed rules.
The TSOBS variable doesn't seem to serve a useful purpose in
gdb/Makefile.in, so remove it.
gdb/ChangeLog
2018-07-09 Tom Tromey <tom@tromey.com>
* Makefile.in (TSOBS): Remove.
(INIT_FILES): Update.
(LIBGDB_OBS): Update.
(COMMON_SFILES): Add inflow.c.
(SFILES): Remove inflow.c.
Use changequote to match "i[3-7]86-*-linux-*", instead of
"i3-786-*-linux-*".
PR ld/23388
* configure.ac: Use changequote for "i[3-7]86-*-linux-*".
* configure: Regenerated.
With config.sub now properly returning m68hc12-unknown-elf rather than
m68hc12-unknown-none, more ELF tests run. This patch enables
STB_GNU_UNIQUE processing fixing some testsuite failures that probably
no m68hc12 user cares about, and removes some XPASSes.
bfd/
* elf32-m68hc1x.c (elf32_m68hc11_post_process_headers): Call
_bfd_elf_post_process_headers.
ld/
* testsuite/ld-discard/extern.d: Remove m68hc12 xfail.
* testsuite/ld-discard/start.d: Likewise.
* testsuite/ld-discard/static.d: Likewise.
In case params.thumb_entry_symbol has its default NULL value, using it
to print a warning in gld${EMULATION_NAME}_finish results in a crash.
Use h->root.string instead which either points to
params.thumb_entry_symbol name or to entry_symbol name.
2018-07-09 Christophe Lyon <christophe.lyon@linaro.org>
* emultempl/armelf.em (gld${EMULATION_NAME}_finish): Use the right
symbol name in case of warning.
Change-Id: Iec61a833c0ad538b1440bf326ba67834c314dd63
