1d509aa625f8 ("infrun: step through indirect branch thunks")
missed adding x86-tdep.o to the list of object file included in an amd64
or i386 build. The problem is not seen with --enable-targets=all
because that file is included in ALL_TARGET_OBS.
H.J. Lu [Sun, 15 Apr 2018 15:38:23 +0000 (08:38 -0700)]
x86: Allow 32-bit registers for tpause and umwait
Since only the first 32 bits of input operand are used for tpause and
umwait, the REX.W bit is skipped. Both 32-bit registers and 64-bit
registers are allowed.
gas/
* testsuite/gas/i386/x86-64-waitpkg.s: Add 32-bit registers
tests for tpause and umwait.
* testsuite/gas/i386/x86-64-waitpkg-intel.d: Updated.
* testsuite/gas/i386/x86-64-waitpkg.d: Likewise.
opcodes/
* i386-dis.c (prefix_table): Replace Em with Edq on tpause and
umwait.
* i386-opc.tbl: Allow 32-bit registers for tpause and umwait in
64-bit mode.
* i386-tbl.h: Regenerated.
Alan Modra [Sat, 14 Apr 2018 06:53:56 +0000 (16:23 +0930)]
powerpc common-page-size
max-page-size only matters for demand paged executables or shared
libraries, and the ideal size is the largest value used by your
operating system. Values larger than necessary just waste file space
and memory. common-page-size also affects file and memory size,
trading a possible small increase in file size for a decrease in
memory size when the operating system is using a common-page-size
page. With a powerpc max-page-size of 64k and common-page-size of 4k
many executables will use no more memory pages when the system page
size is 4k than an executable linked with -z max-page-size=0x1000,
yet will still run on a system using 64k pages. However, when running
on a system using 64k pages relro protection will not be completely
effective.
Due to the relro problem, powerpc binutils has been using a default
common-page-size of 64k since 2014-12-18 (git commit 04c6a44c7),
leading to complaints about increased file and memory sizes. People
not using relro do have a valid reason to complain..
So this patch introduces an extra back-end value to use as the default
for common-page-size when generating relro executables, and enables
the support for powerpc. Non relro executables will now be generated
with a default common-page-size of 4k.
bfd/
* elf-bfd.h (struct elf_backend_data): Add relropagesize.
* elfxx-target.h (ELF_RELROPAGESIZE): Provide default and
sanity test.
(elfNN_bed): Init relropagesize.
* bfd.c (bfd_emul_get_commonpagesize): Add boolean param to
select relropagesize.
* elf32-ppc.c (ELF_COMMONPAGESIZE): Define as 0x1000.
(ELF_RELROPAGESIZE): Define as ELF_MAXPAGESIZE.
(ELF_MINPAGESIZE): Don't define.
* elf64-ppc.c (ELF_COMMONPAGESIZE): Define as 0x1000.
(ELF_RELROPAGESIZE): Define as ELF_MAXPAGESIZE.
* bfd-in2.h: Regenerate.
ld/
* ldmain.c (main): Move config.maxpagesize and
config.commonpagesize initialization to..
* ldemul.c (after_parse_default): ..here.
* testsuite/ld-powerpc/ppc476-shared.d: Pass -z common-page-size.
* testsuite/ld-powerpc/ppc476-shared2.d: Likewise.
Alan Modra [Thu, 12 Apr 2018 10:40:36 +0000 (20:10 +0930)]
powerpc max-page-size vs __QNXTARGET__
The default max-page-size on ppc32 has been 64k since 1995-02-15 (git
commit bcbe2c71). There was a change committed 2003-07-12 to chose a
4k page if __QNXTARGET__ is defined, but that particular commit was
from an earlier posted patch
https://sourceware.org/ml/binutils/2003-07/msg00211.html that only
made the change effective for arm, rather than the later one
https://sourceware.org/ml/binutils/2003-07/msg00220.html that also
changed powerpc and sh..
Since the __QNXTARGET__ #ifdef in elf32-ppc.c is ineffective unless
the user defines it in his or her CFLAGS, I'm removing that code.
* elf32-ppc.c (ELF_MAXPAGESIZE, ELF_COMMONPAGESIZE): Don't depend
on __QNXTARGET__ define.
Alan Modra [Sat, 14 Apr 2018 01:21:23 +0000 (10:51 +0930)]
powerpc-lynxos and powerpc-windiss fixes
Enabling shared lib tests showed the powerpc-lynxos target is broken,
and has been for a long time. The breakage happened in a 2005-05-07
patch of mine, git commit 3b36f7e62, I think. There have been no bug
reports I recall so it seems the target is dead. powerpc-windiss is
similarly broken.
This patch fixes the breakage, and puts the targets on the obsolete
list.
bfd/
* config.bfd: Add powerpc-*-lynxos* and powerpc-*-windiss*
to obsolete list.
ld/
* emulparams/elf32ppcwindiss.sh: Rewrite to use elf32ppc.sh.
* emulparams/ppclynx.sh: Likewise.
Andreas Arnez [Fri, 13 Apr 2018 17:26:05 +0000 (19:26 +0200)]
Show line numbers in output for "info var/func/type"
The GDB commands "info variables", "info functions", and "info types" show
the appropriate list of definitions matching the given pattern. They also
group them by source files. But no line numbers within these source files
are shown.
The line number information is particularly useful to the user when a
simple "grep" doesn't readily point to a definition. This is often the
case when the definition involves a macro, occurs within a namespace, or
when the identifier appears very frequently in the source file.
This patch enriches the printout of these commands by the line numbers and
adjusts affected test cases to the changed output where necessary. The
new output looks like this:
(gdb) i variables
All defined variables:
File foo.c:
3: const char * const foo;
1: int x;
The line number is followed by a colon and a tab character, which is then
followed by the symbol definition. If no line number is available, the
tab is printed out anyhow, so definitions line up.
gdb/ChangeLog:
* symtab.c (print_symbol_info): Precede the symbol definition by
the line number when available.
* NEWS: Advertise this enhancement.
gdb/doc/ChangeLog:
* gdb.texinfo (Symbols): Mention the fact that "info
variables/functions/types" show source files and line numbers.
gdb/testsuite/ChangeLog:
* gdb.ada/info_types.exp: Adjust expected output to the line
numbers now printed by "info var/func/type".
* gdb.base/completion.exp: Likewise.
* gdb.base/included.exp: Likewise.
* gdb.cp/cp-relocate.exp: Likewise.
* gdb.cp/cplusfuncs.exp: Likewise.
* gdb.cp/namespace.exp: Likewise.
* gdb.dwarf2/dw2-case-insensitive.exp: Likewise.
Markus Metzger [Fri, 2 Feb 2018 11:29:48 +0000 (12:29 +0100)]
btrace: set/show record btrace cpu
Add new set/show commands to set the processor that is used for enabling
errata workarounds when decoding branch trace.
The general format is "<vendor>:<identifier>" but we also allow two
special values "auto" and "none".
The default is "auto", which is the current behaviour of having GDB
determine the processor on which the trace was recorded.
If that cpu is not known to the trace decoder, e.g. when using an old
decoder on a new system, decode may fail with "unknown cpu". In most
cases it should suffice to 'downgrade' decode to assume an older cpu.
Unfortunately, we can't do this automatically.
The other special value, "none", disables errata workarounds.
gdb/
* NEWS (New options): announce set/show record btrace cpu.
* btrace.c: Include record-btrace.h.
(btrace_compute_ftrace_pt): Skip enabling errata workarounds if
the vendor is unknown.
(btrace_compute_ftrace_1): Add cpu parameter. Update callers.
Maybe overwrite the btrace configuration's cpu.
(btrace_compute_ftrace): Add cpu parameter. Update callers.
(btrace_fetch): Add cpu parameter. Update callers.
(btrace_maint_update_pt_packets): Call record_btrace_get_cpu.
Maybe overwrite the btrace configuration's cpu. Skip enabling
errata workarounds if the vendor is unknown.
* python/py-record-btrace.c: Include record-btrace.h.
(recpy_bt_begin, recpy_bt_end, recpy_bt_instruction_history)
(recpy_bt_function_call_history): Call record_btrace_get_cpu.
* record-btrace.c (record_btrace_cpu_state_kind): New.
(record_btrace_cpu): New.
(set_record_btrace_cpu_cmdlist): New.
(record_btrace_get_cpu): New.
(require_btrace_thread, record_btrace_info)
(record_btrace_resume_thread): Call record_btrace_get_cpu.
(cmd_set_record_btrace_cpu_none): New.
(cmd_set_record_btrace_cpu_auto): New.
(cmd_set_record_btrace_cpu): New.
(cmd_show_record_btrace_cpu): New.
(_initialize_record_btrace): Initialize set/show record btrace cpu
commands.
* record-btrace.h (record_btrace_get_cpu): New.
testsuite/
* gdb.btrace/cpu.exp: New.
doc/
* gdb.texinfo: Document set/show record btrace cpu.
keep behaves as before
thunk jumps through a thunk
thunk-external jumps through an external thunk
thunk-inline jumps through an inlined thunk
For thunk and thunk-external, GDB would, on a call to the thunk, step into
the thunk and then resume to its caller assuming that this is an
undebuggable function. On a return thunk, GDB would stop inside the
thunk.
Make GDB step through such thunks instead.
Before:
Temporary breakpoint 1, main ()
at gdb.base/step-indirect-call-thunk.c:37
37 x = apply (inc, 41);
(gdb) s
apply (op=0x80483e6 <inc>, x=41)
at gdb.base/step-indirect-call-thunk.c:29
29 return op (x);
(gdb)
30 }
After:
Temporary breakpoint 1, main ()
at gdb.base/step-indirect-call-thunk.c:37
37 x = apply (inc, 41);
(gdb) s
apply (op=0x80483e6 <inc>, x=41)
at gdb.base/step-indirect-call-thunk.c:29
29 return op (x);
(gdb)
inc (x=41) at gdb.base/step-indirect-call-thunk.c:23
23 return x + 1;
This is independent of the step-mode. In order to step into the thunk,
you would need to use stepi.
When stepping over an indirect call thunk, GDB would first step through
the thunk, then recognize that it stepped into a sub-routine and resume to
the caller (of the thunk). Not sure whether this is worth optimizing.
Thunk detection is implemented via gdbarch. I implemented the methods for
IA. Other architectures may run into unexpected fails.
The tests assume a fixed number of instruction steps to reach a thunk.
This depends on the compiler as well as the architecture. They may need
adjustments when we add support for more architectures. Or we can simply
drop those tests that cover being able to step into thunks using
instruction stepping.
When using an older GCC, the tests will fail to build and will be reported
as untested:
Jan Kratochvil [Thu, 12 Apr 2018 20:31:39 +0000 (22:31 +0200)]
Fix -D_GLIBCXX_DEBUG gdb-add-index regression
Fedora Rawhide started to use -D_GLIBCXX_DEBUG which made gdb-add-index
failing:
gdb: Out-of-bounds vector access while running gdb-add-index
https://bugzilla.redhat.com/show_bug.cgi?id=1540559
/usr/include/c++/7/debug/safe_iterator.h:270:
Error: attempt to dereference a past-the-end iterator.
Objects involved in the operation:
iterator "this" @ 0x0x7fffffffcb90 {
type = __gnu_debug::_Safe_iterator<__gnu_cxx::__normal_iterator<unsigned char*, std::__cxx1998::vector<unsigned char, gdb::default_init_allocator<unsigned char, std::allocator<unsigned char> > > >, std::__debug::vector<unsigned char, gdb::default_init_allocator<unsigned char, std::allocator<unsigned char> > > > (mutable iterator);
state = past-the-end;
references sequence with type 'std::__debug::vector<unsigned char, gdb::default_init_allocator<unsigned char, std::allocator<unsigned char> > >' @ 0x0x7fffffffcc50
}
/usr/include/c++/7/debug/vector:417:
Error: attempt to subscript container with out-of-bounds index 556, but
container only holds 556 elements.
Objects involved in the operation:
sequence "this" @ 0x0x2e87af8 {
type = std::__debug::vector<partial_symbol*, std::allocator<partial_symbol*> >;
}
The two -D_GLIBCXX_DEBUG regressions were made by:
commit bc8f2430e08cc2a520db49a42686e0529be4a3bc
Author: Jan Kratochvil <jan.kratochvil@redhat.com>
Date: Mon Jun 12 16:29:53 2017 +0100
Code cleanup: C++ify .gdb_index producer
commit af5bf4ada48ff65b6658be1fab8f9c8f8ab5f319
Author: Simon Marchi <simon.marchi@ericsson.com>
Date: Sat Oct 14 08:06:29 2017 -0400
Replace psymbol_allocation_list with std::vector
gdb/ChangeLog
2018-04-12 Jan Kratochvil <jan.kratochvil@redhat.com>
Tom Tromey [Thu, 29 Mar 2018 18:08:32 +0000 (12:08 -0600)]
Remove old univariant code from rust-lang.c
Since moving Rust enum handling into dwarf2read.c, some old code for
handling univariant enums in rust-lang.c has been obsolete. This
patch removes this code.
Tested on x86-64 Fedora 26, using rustc 1.23 (1.24 emits incorrect
DWARF for enums and so can't be used for this test).
Pedro Alves [Thu, 12 Apr 2018 16:47:59 +0000 (17:47 +0100)]
Fix Solaris build
This commit fixes a bit of rot in procfs.c caused by recent changes.
Specifically, the target_ops::to_detach change to pass down 'inferior
*' missed updating a forward declation, and the change to use
scoped_fd in more places missed removing one do_cleanups call.
src/gdb/procfs.c: In function ‘target_ops* procfs_target()’:
src/gdb/procfs.c:167:16: error: invalid conversion from ‘void (*)(target_ops*, const char*, int)’ to ‘void (*)(target_ops*, inferior*, int)’ [-fpermissive]
t->to_detach = procfs_detach;
^
src/gdb/procfs.c: In function ‘ssd* proc_get_LDT_entry(procinfo*, int)’:
src/gdb/procfs.c:1624:17: error: ‘old_chain’ was not declared in this scope
do_cleanups (old_chain);
^
src/gdb/procfs.c: At global scope:
src/gdb/procfs.c:90:13: error: ‘void procfs_detach(target_ops*, const char*, int)’ declared ‘static’ but never defined [-Werror=unused-function]
static void procfs_detach (struct target_ops *, const char *, int);
^
src/gdb/procfs.c:1923:1: error: ‘void procfs_detach(target_ops*, inferior*, int)’ defined but not used [-Werror=unused-function]
procfs_detach (struct target_ops *ops, inferior *inf, int from_tty)
^
gdb/ChangeLog:
2018-04-12 Pedro Alves <palves@redhat.com>
* procfs.c (procfs_detach): Make forward declaration's prototype
match definition's protototype.
(proc_get_LDT_entry): Remove stale do_cleanups call.
Igor Tsimbalist [Mon, 9 Apr 2018 10:58:50 +0000 (12:58 +0200)]
Enable Intel WAITPKG instructions.
Intel has disclosed a set of new instructions for Tremont processor.
The spec is
https://software.intel.com/en-us/intel-architecture-instruction-set-extensions-programming-reference
This patch enables Intel WAITPKG instructions.
Pedro Alves [Wed, 11 Apr 2018 10:35:58 +0000 (11:35 +0100)]
File I/O file handles after target closes
A future patch will propose making the remote target's target_ops be
heap-allocated (to make it possible to have multiple instances of
remote targets, for multiple simultaneous connections), and will
delete/destroy the remote target at target_close time.
That change trips on a latent problem, though. File I/O handles
remain open even after the target is gone, with a dangling pointer to
a target that no longer exists. This results in GDB crashing when it
calls the target_ops backend associated with the file handle:
#0 0x00007f79338570a0 in main_arena () at /lib64/libc.so.6
#1 0x0000000000858bfe in target_fileio_close(int, int*) (fd=1, target_errno=0x7ffe0499a4c8)
at src/gdb/target.c:2980
#2 0x00000000007088bd in gdb_bfd_iovec_fileio_close(bfd*, void*) (abfd=0x1a631b0, stream=0x223c9d0)
at src/gdb/gdb_bfd.c:353
#3 0x0000000000930906 in opncls_bclose (abfd=0x1a631b0) at src/bfd/opncls.c:528
#4 0x0000000000930cf9 in bfd_close_all_done (abfd=0x1a631b0) at src/bfd/opncls.c:768
#5 0x0000000000930cb3 in bfd_close (abfd=0x1a631b0) at src/bfd/opncls.c:735
#6 0x0000000000708dc5 in gdb_bfd_close_or_warn(bfd*) (abfd=0x1a631b0) at src/gdb/gdb_bfd.c:511
#7 0x00000000007091a2 in gdb_bfd_unref(bfd*) (abfd=0x1a631b0) at src/gdb/gdb_bfd.c:615
#8 0x000000000079ed8e in objfile::~objfile() (this=0x2154730, __in_chrg=<optimized out>)
at src/gdb/objfiles.c:682
#9 0x000000000079fd1a in objfile_purge_solibs() () at src/gdb/objfiles.c:1065
#10 0x00000000008162ca in no_shared_libraries(char const*, int) (ignored=0x0, from_tty=1)
at src/gdb/solib.c:1251
#11 0x000000000073b89b in disconnect_command(char const*, int) (args=0x0, from_tty=1)
at src/gdb/infcmd.c:3035
This goes unnoticed in current master, because the current remote
target's target_ops is never destroyed nowadays, so we end up calling:
remote_hostio_close -> remote_hostio_send_command
which gracefully fails with FILEIO_ENOSYS if remote_desc is NULL
(because the target is closed).
Fix this by invalidating a target's file I/O handles when the target
is closed.
With this change, remote_hostio_send_command no longer needs to handle the
case of being called with a closed remote target, originally added here:
<https://sourceware.org/ml/gdb-patches/2008-08/msg00359.html>.
gdb/ChangeLog:
2018-04-11 Pedro Alves <palves@redhat.com>
* target.c (fileio_fh_t::t): Add comment.
(target_fileio_pwrite, target_fileio_pread, target_fileio_fstat)
(target_fileio_close): Handle a NULL target.
(invalidate_fileio_fh): New.
(target_close): Call it.
* remote.c (remote_hostio_send_command): No longer check whether
remote_desc is open.
Pedro Alves [Wed, 11 Apr 2018 10:29:39 +0000 (11:29 +0100)]
C++ify fileio_fh_t, replace VEC with std::vector
Preparation for the next patch.
- Replace VEC with std::vector.
- Rewrite a couple macros as methods/functions.
- While at it, rename fileio_fh_t::fd as fileio_fh_t::target_fd to
avoid confusion between target and host file descriptors.
gdb/ChangeLog:
2018-04-11 Pedro Alves <palves@redhat.com>
* target.c (fileio_fh_t): Make it a named struct instead of a
typedef.
(fileio_fh_t::is_closed): New method.
(DEF_VEC_O (fileio_fh_t)): Remove.
(fileio_fhandles): Now a std::vector.
(is_closed_fileio_fh): Delete.
(acquire_fileio_fd): Adjust. Rename parameters.
(release_fileio_fd): Adjust.
(fileio_fd_to_fh): Reimplement as a function instead of a macro.
(target_fileio_pwrite, target_fileio_pread, target_fileio_fstat)
(target_fileio_close): Adjust.
Simon Marchi [Tue, 10 Apr 2018 20:50:59 +0000 (16:50 -0400)]
Iterate by index in auto_load_safe_path_vec_update
As reported by Jan, we get this error when building with -D_GLIBCXX_DEBUG:
/usr/include/c++/7/debug/safe_iterator.h:297:
Error: attempt to increment a singular iterator.
Objects involved in the operation:
iterator "this" @ 0x0x7fffffffd140 {
type = __gnu_debug::_Safe_iterator<__gnu_cxx::__normal_iterator<std::unique_ptr<char, gdb::xfree_deleter<char> >*, std::__cxx1998::vector<std::unique_ptr<char, gdb::xfree_deleter<char> >, std::allocator<std::unique_ptr<char, gdb::xfree_deleter<char> > > > >, std::__debug::vector<std::unique_ptr<char, gdb::xfree_deleter<char> >, std::allocator<std::unique_ptr<char, gdb::xfree_deleter<char> > > > > (mutable iterator);
state = singular;
references sequence with type 'std::__debug::vector<std::unique_ptr<char, gdb::xfree_deleter<char> >, std::allocator<std::unique_ptr<char, gdb::xfree_deleter<char> > > >' @ 0x0x265db40
}
The bug was introduced by commit
commit e80aaf6183c6692ecc167bf26cbdc53f8f1a55f0
Author: Simon Marchi <simon.marchi@polymtl.ca>
Date: Fri Mar 2 23:22:06 2018 -0500
Make delim_string_to_char_ptr_vec return an std::vector
The problem is that we iterate using a range-based for on a vector to
which we push in the loop. Pushing to the vector invalidates the
iterator used in the loop. Instead, change the code to iterate by index
as was done in the previous code.
gdb/ChangeLog:
* auto-load.c (auto_load_safe_path_vec_update): Iterate by
index.
The testcase has two processes, parent and child fork. The problem is
that the child exits itself if it notices the parent is gone, but the
testcase .exp does not expect that.
I first wrote a patch that handled the different combinations of
non-stop/detach-on-fork/follow-fork/schedule-multiple, making the .exp
file know when to expect the child to exit itself vs when to kill it
explicitly, but the result was that the code to kill the parent and
child was getting about as large as the test code that is the actual
point of the testcase, above the kills.
So I scratched that approach and came up with a simpler patch --
simply make the child not exit itself when the parent exits.
The .exp file is going to kill both parent and child explicitly, and,
main() already calls alarm() as a safeguard. I don't think we lose
anything.
gdb/testsuite/ChangeLog:
2018-04-10 Pedro Alves <palves@redhat.com>
* gdb.base/fork-running-state.c (fork_child): Don't exit if parent
exits. Instead loop running forever.
(fork_parent): Run forever too.
Simon Marchi [Mon, 9 Apr 2018 19:45:39 +0000 (15:45 -0400)]
Add selftests for range_contains and insert_into_bit_range_vector
Add some selftests for these two functions. To to make it easier to
compare sequences of ranges, add operator== and operator!= to compare
two gdb::array_view, and add operator== in struct range.
Simon Marchi [Mon, 9 Apr 2018 19:40:45 +0000 (15:40 -0400)]
Use an std::vector for inline_states
This patch replaces VEC(inline_state) with std::vector<inline_state> and
adjusts the code that uses it.
gdb/ChangeLog:
* common/gdb_vecs.h (unordered_remove): Add overload that takes
an iterator.
* inline-frame.c: Include <algorithm>.
(struct inline_state): Add constructor.
(inline_state_s): Remove.
(DEF_VEC_O(inline_state_s)): Remove.
(inline_states): Change type to std::vector.
(find_inline_frame_state): Adjust to std::vector.
(allocate_inline_frame_state): Remove.
(clear_inline_frame_state): Adjust to std::vector.
(skip_inline_frames): Adjust to std::vector.
Simon Marchi [Mon, 9 Apr 2018 19:16:19 +0000 (15:16 -0400)]
Remove VEC(tsv_s), use std::vector instead
This patch removes VEC(tsv_s), using an std::vector instead. I C++ified
trace_state_variable a bit in the process, using std::string for the
name. I also thought it would be nicer to pass a const reference to
target_download_trace_state_variable, since we know it will never be
NULL. This highlighted that the make-target-delegates script didn't
handle references well, so I adjusted this as well. It will surely be
useful in the future.
gdb/ChangeLog:
* tracepoint.h (struct trace_state_variable): Add constructor.
<name>: Change type to std::string.
* tracepoint.c (tsv_s): Remove.
(DEF_VEC_O(tsv_s)): Remove.
(tvariables): Change to std::vector.
(create_trace_state_variable): Adjust to std::vector.
(find_trace_state_variable): Likewise.
(find_trace_state_variable_by_number): Likewise.
(delete_trace_state_variable): Likewise.
(trace_variable_command): Adjust to std::string.
(delete_trace_variable_command): Likewise.
(tvariables_info_1): Adjust to std::vector.
(save_trace_state_variables): Likewise.
(start_tracing): Likewise.
(merge_uploaded_trace_state_variables): Adjust to std::vector
and std::string.
* target.h (struct target_ops)
<to_download_trace_state_variable>: Pass reference to
trace_state_variable.
* target-debug.h (target_debug_print_const_trace_state_variable_r): New.
* target-delegates.c: Re-generate.
* mi/mi-interp.c (mi_tsv_created): Adjust to std::string.
(mi_tsv_deleted): Likewise.
* mi/mi-main.c (mi_cmd_trace_frame_collected): Likewise.
* remote.c (remote_download_trace_state_variable): Change
pointer to reference and adjust.
* make-target-delegates (parse_argtypes): Handle references.
(write_function_header): Likewise.
(munge_type): Likewise.
Simon Marchi [Mon, 9 Apr 2018 17:31:06 +0000 (13:31 -0400)]
Adapt and integrate string_view tests
The previous patch copied the string_view tests from libstdc++. This
patch adjusts them in a similar way that the libstdc++ optional tests
are integrated in our unit test suite.
Not all tests are used, some of them require language features not
present in c++11. For example, we can't use a string_view constructor
where the length is not explicit in a constexpr, because
std::char_traits::length is not a constexpr itself (it is in c++17
though). Nevertheless, a good number of tests are integrated, which
covers pretty well the string_view features.
The local modifications are done in the following patch, so that it's
easier to review them.
gdb/ChangeLog:
* unittests/basic_string_view/capacity/1.cc: New file.
* unittests/basic_string_view/capacity/empty_neg.cc: New file.
* unittests/basic_string_view/cons/char/1.cc: New file.
* unittests/basic_string_view/cons/char/2.cc: New file.
* unittests/basic_string_view/cons/char/3.cc: New file.
* unittests/basic_string_view/cons/wchar_t/1.cc: New file.
* unittests/basic_string_view/cons/wchar_t/2.cc: New file.
* unittests/basic_string_view/cons/wchar_t/3.cc: New file.
* unittests/basic_string_view/element_access/char/1.cc: New file.
* unittests/basic_string_view/element_access/char/2.cc: New file.
* unittests/basic_string_view/element_access/char/empty.cc: New file.
* unittests/basic_string_view/element_access/char/front_back.cc: New file.
* unittests/basic_string_view/element_access/wchar_t/1.cc: New file.
* unittests/basic_string_view/element_access/wchar_t/2.cc: New file.
* unittests/basic_string_view/element_access/wchar_t/empty.cc: New file.
* unittests/basic_string_view/element_access/wchar_t/front_back.cc: New file.
* unittests/basic_string_view/include.cc: New file.
* unittests/basic_string_view/inserters/char/1.cc: New file.
* unittests/basic_string_view/inserters/char/2.cc: New file.
* unittests/basic_string_view/inserters/char/3.cc: New file.
* unittests/basic_string_view/inserters/pod/10081-out.cc: New file.
* unittests/basic_string_view/inserters/wchar_t/1.cc: New file.
* unittests/basic_string_view/inserters/wchar_t/2.cc: New file.
* unittests/basic_string_view/inserters/wchar_t/3.cc: New file.
* unittests/basic_string_view/literals/types.cc: New file.
* unittests/basic_string_view/literals/values.cc: New file.
* unittests/basic_string_view/modifiers/remove_prefix/char/1.cc: New file.
* unittests/basic_string_view/modifiers/remove_prefix/wchar_t/1.cc: New file.
* unittests/basic_string_view/modifiers/remove_suffix/char/1.cc: New file.
* unittests/basic_string_view/modifiers/remove_suffix/wchar_t/1.cc: New file.
* unittests/basic_string_view/modifiers/swap/char/1.cc: New file.
* unittests/basic_string_view/modifiers/swap/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/compare/char/1.cc: New file.
* unittests/basic_string_view/operations/compare/char/13650.cc: New file.
* unittests/basic_string_view/operations/compare/char/2.cc: New file.
* unittests/basic_string_view/operations/compare/char/70483.cc: New file.
* unittests/basic_string_view/operations/compare/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/compare/wchar_t/13650.cc: New file.
* unittests/basic_string_view/operations/compare/wchar_t/2.cc: New file.
* unittests/basic_string_view/operations/copy/char/1.cc: New file.
* unittests/basic_string_view/operations/copy/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/data/char/1.cc: New file.
* unittests/basic_string_view/operations/data/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/find/char/1.cc: New file.
* unittests/basic_string_view/operations/find/char/2.cc: New file.
* unittests/basic_string_view/operations/find/char/3.cc: New file.
* unittests/basic_string_view/operations/find/char/4.cc: New file.
* unittests/basic_string_view/operations/find/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/find/wchar_t/2.cc: New file.
* unittests/basic_string_view/operations/find/wchar_t/3.cc: New file.
* unittests/basic_string_view/operations/find/wchar_t/4.cc: New file.
* unittests/basic_string_view/operations/rfind/char/1.cc: New file.
* unittests/basic_string_view/operations/rfind/char/2.cc: New file.
* unittests/basic_string_view/operations/rfind/char/3.cc: New file.
* unittests/basic_string_view/operations/rfind/wchar_t/1.cc: New file.
* unittests/basic_string_view/operations/rfind/wchar_t/2.cc: New file.
* unittests/basic_string_view/operations/rfind/wchar_t/3.cc: New file.
* unittests/basic_string_view/operations/string_conversion/1.cc: New file.
* unittests/basic_string_view/operations/substr/char/1.cc: New file.
* unittests/basic_string_view/operations/substr/wchar_t/1.cc: New file.
* unittests/basic_string_view/operators/char/2.cc: New file.
* unittests/basic_string_view/operators/wchar_t/2.cc: New file.
* unittests/basic_string_view/range_access/char/1.cc: New file.
* unittests/basic_string_view/range_access/wchar_t/1.cc: New file.
* unittests/basic_string_view/requirements/explicit_instantiation/1.cc: New file.
* unittests/basic_string_view/requirements/explicit_instantiation/char/1.cc: New file.
* unittests/basic_string_view/requirements/explicit_instantiation/char16_t/1.cc: New file.
* unittests/basic_string_view/requirements/explicit_instantiation/char32_t/1.cc: New file.
* unittests/basic_string_view/requirements/explicit_instantiation/wchar_t/1.cc: New file.
* unittests/basic_string_view/requirements/typedefs.cc: New file.
* unittests/basic_string_view/typedefs.cc: New file.
* unittests/basic_string_view/types/1.cc: New file.
Simon Marchi [Mon, 9 Apr 2018 17:31:04 +0000 (13:31 -0400)]
Add gdb::string_view
We had a few times the need for a data structure that does essentially
what C++17's std::string_view does, which is to give an std::string-like
interface (only the read-only operations) to an arbitrary character
buffer.
This patch adapts the files copied from libstdc++ by the previous patch
to integrate them with GDB. Here's a summary of the changes:
* Remove things related to wstring_view, u16string_view and
u32string_view (I don't think we need them, but we can always add them
later).
* Remove usages of _GLIBCXX_BEGIN_NAMESPACE_VERSION and
_GLIBCXX_END_NAMESPACE_VERSION.
* Put the code in the gdb namespace. I had to add a few "std::" in
front of std type usages.
* Change __throw_out_of_range_fmt() for error().
* Make gdb::string_view an alias of std::string_view when building
with >= c++17.
* Remove a bunch of constexpr, because they are not valid in c++11
(e.g. they are not a single return line).
* Use std::common_type<_Tp>::type instead of std::common_type_t<_Tp>,
because c++11 doesn't have the later.
* Remove the #pragma GCC system_header, since that silences some
warnings that we might want to have if we're doing something not
correctly.
* Remove operator ""sv. It would need a lot of work to make all
supported compilers happy, and we can easily live without it.
* Remove operator<<. It is implemented using __ostream_insert (a
libstdc++ internal). Bringing it in might be possible, but I don't
think that would be worth the effort, since we don't really use
streams at the moment.
* Replace internal libstdc++ asserts ( __glibcxx_assert and
__glibcxx_requires_string_len) with gdb_assert.
* Remove hash helpers, because they use libstdc++ internal functions.
If we need them we always import them later.
The string_view class in cli/cli-script.c is removed and its usage
replaced with the new gdb::string_view.
gdb/ChangeLog:
* common/gdb_string_view.h: Remove libstdc++ implementation
details, adjust to gdb reality.
* common/gdb_string_view.tcc: Likewise.
* cli/cli-script.c (struct string_view): Remove.
(user_args) <m_args>: Change element type to gdb::string_view.
(user_args::insert_args): Adjust.
Simon Marchi [Mon, 9 Apr 2018 17:31:03 +0000 (13:31 -0400)]
Update ax_cv_cxx_compile_cxx.m4
This file provides the AX_CXX_COMPILE_STDCXX macro. In the context of
the following patch, I wanted to build and test GDB in c++17 mode. The
version of the macro we have in the repo does not support detecting
c++17 compilers, but the upstream version has been updated to do so.
Since we have local modifications to the file, I had to reconcile our
modifications and the updated upstream version (which was relatively
straightforward).
gdb/ChangeLog:
* ax_cxx_compile_stdcxx.m4: Sync with upstream.
* configure: Re-generate.
Clearly commit 76727919ceb5 ("Convert observers to C++") edited
gdbarch.c directly instead of gdbarch.sh. This fixes it.
gdb/ChangeLog:
2018-04-09 Pedro Alves <palves@redhat.com>
* gdbarch.sh: Include "observable.h" instead of "observer.h".
(set_target_gdbarch): Call
gdb::observers::architecture_changed.notify instead of
observer_notify_architecture_changed.
binutils/testsuite: Fix a crash with STN_UNDEF in relocation
Verify that `strip' completes successfully and a correct relocation
entry is copied for a relocation encountered with the STN_UNDEF symbol
index.
binutils/
* testsuite/binutils-all/strip-15.d: New test.
* testsuite/binutils-all/strip-15rel.s: New test source.
* testsuite/binutils-all/strip-15rela.s: New test source.
* testsuite/binutils-all/strip-15mips64.s: New test source.
* testsuite/binutils-all/objcopy.exp: Run the new test.
MIPS64/BFD: Fix a crash with STN_UNDEF in relocation
Prevent a null BFD pointer dereference and a resulting segmentation
fault in `mips_elf64_write_rel' or `mips_elf64_write_rela':
Program received signal SIGSEGV, Segmentation fault.
0x0000000000437690 in mips_elf64_write_rela (abfd=0x71e130, sec=0x720700,
rela_hdr=0x721ff8, count=0x7fffffffb82c, data=0x7fffffffb88c)
at .../bfd/elf64-mips.c:4123
4123 if ((*ptr->sym_ptr_ptr)->the_bfd->xvec != abfd->xvec
4124 && ! _bfd_elf_validate_reloc (abfd, ptr))
in the MIPS64 (n64 MIPS) ELF backend whenever the STN_UNDEF symbol index
is retrieved from the `r_sym' field of a relocation seen in input while
running `objcopy' or `strip'. The reason for the null BFD pointer is
that internally in BFD an STN_UNDEF symbol reference resolves to an
absolute zero symbol that does not have a BFD associated. Check the
pointer then before using it, like the generic ELF backend does in
`elf_write_relocs'.
This complements the same change made for generic ELF bundled with:
("Add SH ELF support."), which also updated RELA only and not REL (which
has been since fixed with: commit 947216bf8f34 ("ELF reloc code tidy"),
<https://sourceware.org/ml/binutils/2002-11/msg00727.html>).
bfd/
* elf64-mips.c (mips_elf64_write_rel): Handle a NULL BFD pointer
in the BFD symbol referred by the relocation.
(mips_elf64_write_rela): Likewise.
binutils/testsuite: Verify the handling of invalid `r_sym' in relocation
Verify that `strip' terminates gracefully and a correct error message is
produced for a relocation encountered with an invalid symbol index. No
single relocation number is valid across all targets we support, so pick
a few numbers to choose from depending on the target.
binutils/
* testsuite/binutils-all/strip-14.d: New test.
* testsuite/binutils-all/strip-14rel.s: New test source.
* testsuite/binutils-all/strip-14rela.s: New test source.
* testsuite/binutils-all/strip-14mips64.s: New test source.
* testsuite/binutils-all/objcopy.exp: Run the new test.
MIPS64/BFD: Fix a crash with invalid `r_sym' in relocation
Prevent an out-of-range access and a possible segmentation fault in
`mips_elf64_slurp_one_reloc_table':
Program received signal SIGSEGV, Segmentation fault.
mips_elf64_slurp_one_reloc_table (abfd=0x71bd90, asect=0x71cf70,
rel_hdr=<value optimized out>, reloc_count=1,
relents=<value optimized out>, symbols=0x7218c0, dynamic=0)
at .../bfd/elf64-mips.c:3758
3757 ps = symbols + rela.r_sym - 1;
3758 s = *ps;
in the MIPS64 (n64 MIPS) ELF backend whenever an invalid symbol index is
retrieved from the `r_sym' field of a relocation seen in input while
running `objcopy' or `strip'. Issue an error instead, like the generic
ELF backend does, taking code from `elf_slurp_reloc_table_from_section',
except for relocation types that do not refer to a symbol.
This complements commit 1f70368c21a8 ("Stop objdump crash on corrupt
reloc table"), <https://sourceware.org/ml/binutils/2002-09/msg00332.html>,
and commit 05a487dc8c39 ("make check fails on i686-linux-gnu"),
<https://sourceware.org/ml/binutils/2002-09/msg00340.html>, where the
generic ELF backend code comes from.
bfd/
* elf64-mips.c (mips_elf64_slurp_one_reloc_table): Issue an
error for out-of-range `r_sym' values.
Alan Modra [Sun, 8 Apr 2018 23:55:10 +0000 (09:25 +0930)]
PowerPC inline PLT call support
In addition to the existing relocs we need two more to mark all
instructions in the call sequence, PLTCALL on the call itself (plus
the toc restore insn for ppc64), and PLTSEQ on others. All
relocations in a particular sequence have the same symbol.
Marking sequences like this allows the linker to convert them to nops
and a direct call if the target symbol turns out to be local.
When the call is __tls_get_addr, each relocation shown above is paired
with an R_PPC*_TLSLD or R_PPC*_TLSGD reloc to additionally mark the
sequence for possible TLS optimization. The TLSLD or TLSGD relocs are
emitted first.
include/
* elf/ppc.h (R_PPC_PLTSEQ, R_PPC_PLTCALL): Define.
* elf/ppc64.h (R_PPC64_PLTSEQ, R_PPC64_PLTCALL): Define.
bfd/
* elf32-ppc.c (ppc_elf_howto_raw): Add PLTSEQ and PLTCALL howtos.
(is_plt_seq_reloc): New function.
(ppc_elf_check_relocs): Handle PLTSEQ and PLTCALL relocs.
(ppc_elf_tls_optimize): Handle inline plt call sequence.
(ppc_elf_relax_section): Handle PLTCALL reloc.
(ppc_elf_relocate_section): Nop out inline plt call sequence when
resolving locally.
* elf64-ppc.c (ppc64_elf_howto_raw): Add R_PPC64_PLTSEQ and
R_PPC64_PLTCALL entries. Comment R_PPC64_TOCSAVE.
(has_tls_get_addr_call): Correct comment.
(is_branch_reloc): Add PLTCALL.
(is_plt_seq_reloc): New function.
(ppc64_elf_check_relocs): Handle PLT16_LO_DS reloc. Set
has_tls_reloc for R_PPC64_TLSGD and R_PPC64_TLSLD. Create plt
entry for R_PPC64_PLTCALL.
(ppc64_elf_tls_optimize): Handle inline plt call sequence.
(ppc_type_of_stub): Handle PLTCALL reloc.
(toc_adjusting_stub_needed): Likewise.
(ppc64_elf_relocate_section): Set "can_plt_call" for PLTCALL
reloc insn. Nop out inline plt call sequence when resolving
locally. Handle __tls_get_addr inline plt call optimization.
elfcpp/
* powerpc.h (R_POWERPC_PLTSEQ, R_POWERPC_PLTCALL): Define.
gold/
* powerpc.cc (Target_powerpc::Track_tls::maybe_skip_tls_get_addr_call):
Handle inline plt sequence relocs.
(Stub_table::Plt_stub_key::Plt_stub_key): Likewise.
(Target_powerpc::Scan::reloc_needs_plt_for_ifunc): Likewise.
(Target_powerpc::Relocate::relocate): Likewise.
Alan Modra [Sun, 8 Apr 2018 23:52:53 +0000 (09:22 +0930)]
Support PLT16 relocs against local symbols
Necessary if gcc is to use PLT16 relocs to implement -mlongcall, and
there isn't a good technical reason why local symbols should be
excluded from PLT16 support. Non-ifunc local symbol PLT entries go in
a separate section to other PLT entries. In a fixed position
executable they won't need to be relocated, and in a PIE or shared
library I chose to not implement lazy relocation.
bfd/
* elf64-ppc.c (LOCAL_PLT_ENTRY_SIZE): Define.
(struct ppc_stub_hash_entry): Add symtype field.
(PLT_KEEP): Define.
(struct ppc_link_hash_table): Add pltlocal and relpltlocal.
(create_linkage_sections): Create pltlocal and relpltlocal.
(ppc64_elf_check_relocs): Allow PLT relocs on local symbols.
Set PLT_KEEP.
(ppc64_elf_adjust_dynamic_symbol): Keep PLT entries for inline calls.
(allocate_dynrelocs): Allocate pltlocal and relpltlocal.
(ppc64_elf_size_dynamic_sections): Size pltlocal and relpltlocal.
Keep PLT entries for inline calls against locals.
(ppc_build_one_stub): Use pltlocal as appropriate.
(ppc_size_one_stub): Likewise.
(ppc64_elf_size_stubs): Set symtype.
(build_global_entry_stubs_and_plt): Init pltlocal and write
relpltlocal for globals.
(write_plt_relocs_for_local_syms): Likewise for local syms.
(ppc64_elf_relocate_section): Support PLT for local syms.
* elf32-ppc.c (PLT_KEEP): Define.
(struct ppc_elf_link_hash_table): Add pltlocal and relpltlocal.
(ppc_elf_create_glink): Create pltlocal and relpltlocal.
(ppc_elf_check_relocs): Allow PLT relocs on local symbols.
Set PLT_KEEP. Adjust update_local_sym_info call.
(ppc_elf_adjust_dynamic_symbol): Keep PLT entries for inline calls.
(allocate_dynrelocs): Allocate pltlocal and relpltlocal.
(ppc_elf_size_dynamic_sections): Size pltlocal and relpltlocal.
(ppc_elf_relocate_section): Support PLT16 relocs for local syms.
(write_global_sym_plt): Init pltlocal and write relpltlocal.
(ppc_finish_symbols): Likewise for locals.
ld/
* emulparams/elf32ppc.sh (OTHER_RELRO_SECTIONS_2): Add .branch_lt.
(OTHER_GOT_RELOC_SECTIONS): Add .rela.branch_lt.
* testsuite/ld-powerpc/elfv2so.d: Update for symbol/stub reordering.
* testsuite/ld-powerpc/relbrlt.d: Likewise.
* testsuite/ld-powerpc/relbrlt.s: Likewise.
* testsuite/ld-powerpc/tlsso.r: Likewise.
* testsuite/ld-powerpc/tlstocso.r: Likewise.
gold/
* powerpc.cc (Target_powerpc::lplt_): New variable.
(Target_powerpc::lplt_section): Associated accessor.
(Target_powerpc::plt_off): Handle local non-ifunc symbols.
(Target_powerpc::make_lplt_section): New function.
(Target_powerpc::make_local_plt_entry): New function.
(Powerpc_relobj::do_relocate_sections): Write out lplt.
(Output_data_plt_powerpc::first_plt_entry_offset): Zero for lplt.
(Output_data_plt_powerpc::add_local_entry): New function.
(Output_data_plt_powerpc::do_write): Ignore lplt.
(Target_powerpc::make_iplt_section): Make lplt first.
(Target_powerpc::make_brlt_section): Make .branch_lt relro.
(Target_powerpc::Scan::local): Handle PLT16 relocs.
Alan Modra [Sun, 8 Apr 2018 23:51:59 +0000 (09:21 +0930)]
Rearrange PLT reloc output on powerpc
The current scheme where we output PLT relocs for global symbols in
finish_dynamic_symbol, and PLT relocs for local symbols when
outputting stubs does not work if PLT entries are to be used for
inline PLT sequences against non-dynamic globals or local symbols.
bfd/
* elf64-ppc.c (ppc_build_one_stub): Move output of PLT relocs
for local symbols to..
(write_plt_relocs_for_local_syms): ..here. New function.
(ppc64_elf_finish_dynamic_symbol): Move output of PLT relocs for
global symbols to..
(build_global_entry_stubs_and_plt): ..here. Rename from
build_global_entry_stubs.
(ppc64_elf_build_stubs): Always call build_global_entry_stubs_and_plt.
Call write_plt_relocs_for_local_syms.
* elf32-ppc.c (get_sym_h): New function.
(ppc_elf_relax_section): Use get_sym_h.
(ppc_elf_relocate_section): Move output of PLT relocs and glink
stubs for local symbols to..
(ppc_finish_symbols): ..here. New function.
(ppc_elf_finish_dynamic_symbol): Move output of PLT relocs for
global syms to..
(write_global_sym_plt): ..here. New function.
* elf32-ppc.h (ppc_elf_modify_segment_map): Delete attribute.
(ppc_finish_symbols): Declare.
ld/
* ppc32elf.em (ppc_finish): Call ppc_finish_symbols.
Alan Modra [Sun, 8 Apr 2018 23:51:10 +0000 (09:21 +0930)]
PowerPC PLT16 relocations
The PowerPC64 ELFv2 ABI and the PowerPC SysV ABI support a number of
relocations that can be used to create and access a PLT entry.
However, the relocs are not well defined. The PLT16 family of relocs
talk about "the section offset or address of the procedure linkage
table entry". It's plain that we do need a relative address when PIC
as otherwise we'd have dynamic text relocations, but "section offset"
doesn't specify which section. The most obvious one, ".plt", isn't
that useful because there is no readily available way of addressing
the start of the ".plt" section. Much more useful would be "the
GOT/TOC-pointer relative offset of the procedure linkage table entry",
and I suppose you could argue that is a "section offset" of sorts.
For PowerPC64 it is better to use the same TOC-pointer relative
addressing even when non-PIC, since ".plt" may be located outside the
range of a 32-bit address. However, for ppc32 we do want an absolute
address when non-PIC as a GOT pointer may not be set up. Also, for
ppc32 PIC we have a similar situation to R_PPC_PLTREL24 in that the
GOT pointer is set to a location in the .got2 section and we need to
specify the .got2 offset in the PLT16 reloc addend.
This patch supports PLT16 relocations using these semantics. This is
not an ABI change for ppc32 since the relocations were not previously
supported by GNU ld, but is for ppc64 where some of the PLT16 relocs
were supported. I'm not particularly concerned since the old ppc64
PLT16 reloc semantics made them almost completely useless.
bfd/
* elf32-ppc.c (ppc_elf_check_relocs): Handle PLT16 relocs.
(ppc_elf_relocate_section): Likewise.
* elf64-ppc.c (ppc64_elf_check_relocs): Handle PLT16_LO_DS.
(ppc64_elf_relocate_section): Likewise. Correct PLT16
resolution to plt entry relative to toc pointer.
gold/
* powerpc.cc (Target_powerpc::plt_off): New functions.
(is_plt16_reloc): New function.
(Stub_table::plt_off): Use Target_powerpc::plt_off.
(Stub_table::plt_call_size): Use plt_off.
(Stub_table::do_write): Likewise.
(Target_powerpc::Scan::get_reference_flags): Return RELATIVE_REF
for PLT16 relocations.
(Target_powerpc::Scan::reloc_needs_plt_for_ifunc): Return true
for PLT16 relocations.
(Target_powerpc::Scan::global): Make a PLT entry for PLT16 relocations.
(Target_powerpc::Relocate::relocate): Support PLT16 relocations.
(Powerpc_scan_relocatable_reloc::global_strategy): Return RELOC_SPECIAL
for ppc32 plt16 relocs.
Alan Modra [Sun, 8 Apr 2018 23:41:44 +0000 (09:11 +0930)]
PowerPC indirect calls to __tls_get_addr
It is possible to construct indirect calls to __tls_get_addr in
assembly that confuse TLS optimization. (PowerPC gcc doesn't support
such calls, ignoring -mlongcall for __tls_get_addr.) This patch fixes
the problem by requiring a TLSLD or TLSGD marker reloc before any insn
in an indirect call to __tls_get_addr will be optimized. They also
need additional marker relocs defined in a later patch, so don't
expect the optimization to work just yet. The point here is to
prevent mis-optimization of indirect calls without any marker relocs.
The presense of a marker reloc is tracked by a new bit in the tls_mask
field of ppc_link_hash_entry and the corresponding lgot_masks unsigned
char array for local symbols. Since the field is only 8 bits, we've
run out of space. However, tracking TLS use for variables, and
tracking IFUNC for functions are independent, and bits can be reused.
TLS_TLS is always set for TLS usage, so can be used to select the
meaning of the other bits. This patch does that even for elf32-ppc.c
which hasn't yet run out of space in the field.
* elf64-ppc.c (TLS_TLS, TLS_GD, TLS_LD, TLS_TPREL, TLS_DTPREL,
TLS_TPRELGD, TLS_EXPLICIT): Renumber. Test TLS_TLS throughout
file when other TLS flags are tested in a mask.
(TLS_MARK, NON_GOT): Define.
(PLT_IFUNC): Redefine, and test TLS_TLS throughout file as well.
(update_local_sym_info): Don't create got entry when NON_GOT.
(ppc64_elf_check_relocs): Pass NON_GOT with PLT_IFUNC.
Set TLS_MARK.
(get_tls_mask): Do toc lookup if tls_mask is just TLS_MARK.
(ppc64_elf_relocate_section): Likewise.
(ppc64_elf_tls_optimize): Don't attempt to optimize indirect
__tls_get_addr calls lacking a marker reloc.
* elf32-ppc.c (TLS_TLS, TLS_GD, TLS_LD, TLS_TPREL, TLS_DTPREL,
TLS_TPRELGD): Renumber. Update comment.
(TLS_MARK, NON_GOT): Define.
(PLT_IFUNC): Redefine, and test TLS_TLS throughout file as well.
(update_local_sym_info): Don't create got entry when NON_GOT.
(ppc_elf_check_relocs): Pass NON_GOT with PLT_IFUNC.
Set TLS_MARK.
(ppc_elf_tls_optimize): Don't attempt to optimize indirect
__tls_get_addr calls lacking a marker reloc.
Simon Marchi [Sat, 7 Apr 2018 18:08:56 +0000 (14:08 -0400)]
Fix gdb.mi/mi-stack.exp when gcc generates a stack protector
I see some failures in the gdb.mi/mi-stack.exp test. The test runs to
the callee4 function:
int callee4 (void)
{
int A=1;
int B=2;
int C;
int D[3] = {0, 1, 2};
C = A + B;
return 0;
}
and expects to be stopped at the A=1 line. However, when gcc generates
some stack protection code, it will stop at the { instead, as shown by
this disassembly (after I did "break callee4" and "run"):
(gdb) disassemble /s
Dump of assembler code for function callee4:
/home/simark/src/binutils-gdb/gdb/testsuite/gdb.mi/mi-stack.c:
26 {
0x00005555555546ca <+0>: push %rbp
0x00005555555546cb <+1>: mov %rsp,%rbp
0x00005555555546ce <+4>: sub $0x20,%rsp
=> 0x00005555555546d2 <+8>: mov %fs:0x28,%rax
0x00005555555546db <+17>: mov %rax,-0x8(%rbp)
0x00005555555546df <+21>: xor %eax,%eax
27 int A=1; /* callee4 begin */
0x00005555555546e1 <+23>: movl $0x1,-0x20(%rbp)
28 int B=2;
0x00005555555546e8 <+30>: movl $0x2,-0x1c(%rbp)
The rest of the test relies on execution stopping on the A=1, so many things
fail after that. This patch uses mi_continue_to_line instead, to stop at the
A=1 line precisely.
gdb/testsuite/ChangeLog:
* gdb.mi/mi-stack.exp (test_stack_frame_listing): Use
mi_continue_to_line.
* gdb.mi/mi-stack.c (callee4): Add comment.
Simon Marchi [Sat, 7 Apr 2018 17:53:42 +0000 (13:53 -0400)]
Remove some usages of get_dwarf2_per_objfile
This patch removes some usages of get_dwarf2_per_objfile, where we can
get hold of the dwarf2_per_objfile object in a simpler way. For
example, it's simpler (and slightly less work) to pass
dwarf2_per_objfile and get the objfile from it than to pass the objfile
and call get_dwarf2_per_objfile.
Ideally, get_dwarf2_per_objfile should only be used in the entry points
of the dwarf2 code, where we receive an objfile.
gdb/ChangeLog:
* dwarf2read.c (create_cus_from_index_list): Replace objfile arg
with dwarf2_per_objfile.
(create_cus_from_index): Likewise.
(create_signatured_type_table_from_index): Likewise.
(dwarf2_read_index): Likewise.
(dwarf2_initialize_objfile): Likewise.
(dwarf2_fetch_die_loc_sect_off): Get dwarf2_per_objfile from
per_cu rather than get_dwarf2_per_objfile.
Simon Marchi [Sat, 7 Apr 2018 17:53:40 +0000 (13:53 -0400)]
Replace dw2_get_cu/dw2_get_cutu with methods of dwarf2_per_objfile
Those two functions look like good candidates to become methods of
dwarf2_per_objfile. I did that, and added get_tu as well. When
replacing usages of dw2_get_cutu, I changed some instances to get_cutu
and others to get_cu, when appropriate (when we know we want a CU and
not a TU).
Simon Marchi [Sat, 7 Apr 2018 17:53:39 +0000 (13:53 -0400)]
Remove some unused variables in dwarf2read.c
Most of them are obvious. The ones in dwarf2_record_block_ranges are
less obvious, because it is a bit suspicious to have that many
variables unused. But after inspection, it seems like it dates from
commit 5f46c5a54825 ("Code cleanup: Split dwarf2_ranges_read to a
callback"), where dwarf2_record_block_ranges was made to use
dwarf2_ranges_process, which contains the same functionality.
This issue happens only with position-independent executables. Adding
the main objfile for the new inferior (the fork child) causes GDB to try
to reset the breakpoints. However, that new objfile has not been
relocated yet. So the breakpoint on "break_here" resolves to an
unrelocated address, from which we are trying to read/write to set a
breakpoint. Passing SYMFILE_DEFER_BP_RESET avoids that problem. The
executable is relocated just after, in the follow_fork_inferior
function.
The buildbot seems happy with this patch. I don't think it's necessary
to add a new test. Just changing this made many tests go from FAIL to
PASS on my machine, where gcc produces PIE executables by default. If
anything, I think we would need to add a board file that produces
position-independent executables, so that we can run all the tests with
PIE, even on machines where that is not the default.
gdb/ChangeLog:
* progspace.c (clone_program_space): Pass SYMFILE_DEFER_BP_RESET
to symbol_file_add_main.
Simon Marchi [Sat, 7 Apr 2018 17:48:05 +0000 (13:48 -0400)]
Implement write_async_safe for mi_console_file (PR 22299)
Enabling "set debug lin-lwp 1" with the MI interpreter doesn't work.
When the sigchld_handler function wants to print a debug output
("sigchld\n"), it uses ui_file_write_async_safe. This ends up in the
default implementation of ui_file::write_async_safe, which aborts GDB.
This patch implements the write_async_safe method for mi_console_file.
The "normal" MI output is line buffered, which means the output
accumulates in m_buffer until a \n is written, at which point it's
flushed in m_raw. The implementation of write_async_safe provided by
this patch bypasses this buffer and writes directly to m_raw. There are
two reasons for this:
(1) Appending to m_buffer (therefore to an std::string) is probably not
async-safe, as it may allocate memory.
(2) We may have a partial output already in m_buffer, so that would lead
to some nested MI output, not so great.
There is probably still a chance to have bad MI output, if
sigchld_handler is invoked in the middle of mi_console_file's flush, and
the line being flushed is only partially sent to m_raw. The solution
would probably be to block signals during flushing. Since this is only
used for debug output, I don't know if it's worth the effort to do that.
To implement write_async_safe, I needed to use the fputstrn_unfiltered,
which does the necessary escaping (e.g. replace \n with \\n). I started
by adding printchar's callback parameters to fputstrn_unfiltered, to be
able to pass async-safe versions of them. It's not easy to provide an
async-safe version of do_fprintf, but it turns out that we can easily
replace printchar's callbacks with a single do_fputc quite easily. The
async-safe version of do_fputc simply calls the underlying ui_file's
write_async_safe method.
gdb/ChangeLog:
PR mi/22299
* mi/mi-console.c (do_fputc_async_safe): New.
(mi_console_file::write_async_safe): New.
(mi_console_file::flush): Adjust calls to fputstrn_unfiltered.
* mi/mi-console.h (class mi_console_file) <write_async_safe>:
New.
* ui-file.c (ui_file::putstrn): Adjust call to
fputstrn_unfiltered.
* utils.c (printchar): Replace do_fputs and do_fprintf
parameters by do_fputc.
(fputstr_filtered): Adjust call to printchar.
(fputstr_unfiltered): Likewise.
(fputstrn_filtered): Likewise.
(fputstrn_unfiltered): Add do_fputc parameter, pass to
printchar.
* utils.h (do_fputc_ftype): New typedef.
(fputstrn_unfiltered): Add do_fputc parameter.
Simon Marchi [Sat, 7 Apr 2018 17:24:58 +0000 (13:24 -0400)]
Fix generation of x86-64 gdbarch with osabi none (PR 22979)
When a 64-bits (x86-64) gdbarch is created, it is first born as a
32-bits gdbarch in i386_gdbarch_init. The call gdbarch_init_osabi will
call the handler register for the selected (arch, osabi) pair, such as
amd64_linux_init_abi. The various amd64 handlers call amd64_init_abi,
which turns the gdbarch into a 64-bits one.
When selecting the i386:x86-64 architecture with no osabi, no such
handler is ever called, so the gdbarch stays (wrongfully) a 32-bits one.
My first idea was to manually call amd64_init_abi & al in
i386_gdbarch_init when the osabi is GDB_OSABI_NONE. However, this
doesn't work in a build of GDB where i386 is included as a target but
not amd64. My next option (implemented in this patch), is to allow
registering handlers for GDB_OSABI_NONE. I added two such handlers in
amd64-tdep.c, so now it works the same as for the "normal" osabis. It
required re-ordering things in gdbarch_init_osabi to allow running
handlers for GDB_OSABI_NONE.
Without this patch applied (but with the previous one*) :
(gdb) set osabi none
(gdb) set architecture i386:x86-64
The target architecture is assumed to be i386:x86-64
(gdb) p sizeof(void*)
$1 = 4
and now:
(gdb) set osabi none
(gdb) set architecture i386:x86-64
The target architecture is assumed to be i386:x86-64
(gdb) p sizeof(void*)
$1 = 8
* Before the previous patch, which fixed "set osabi none", this bug was
hidden because we didn't actually try to generate a gdbarch for no
osabi, it would always fall back on Linux. Generating the gdbarch for
amd64/linux did work.
gdb/ChangeLog:
PR gdb/22979
* amd64-tdep.c (amd64_none_init_abi): New function.
(amd64_x32_none_init_abi): New function.
(_initialize_amd64_tdep): Register handlers for x86-64 and
x64_32 with GDB_OSABI_NONE.
* osabi.c (gdbarch_init_osabi): Allow running handlers for the
GDB_OSABI_NONE osabi.
gdb/testsuite/ChangeLog:
PR gdb/22979
* gdb.arch/amd64-osabi.exp: New file.
This is because the value GDB_OSABI_UNKNOWN has an unclear role,
sometimes meaning "no osabi" and sometimes "please selected
automatically". Doing "set osabi none" sets the requested osabi to
GDB_OSABI_UNKNOWN, in which case gdbarch_info_fill overrides it with a
value from the target description, or the built-in default osabi. This
means that it's impossible to force GDB not to use an osabi with "set
osabi". Since my GDB's built-in default osabi is Linux, it always falls
back to GDB_OSABI_LINUX.
To fix it, I introduced GDB_OSABI_NONE, which really means "I don't want
any osabi". GDB_OSABI_UNKNOWN can then be used only for "not set yet,
please auto-detect". GDB_OSABI_UNINITIALIZED now seems unnecessary
since it overlaps with GDB_OSABI_UNKNOWN, so I think it can be removed
and gdbarch_info::osabi can be initialized to GDB_OSABI_UNKNOWN.
Simon Marchi [Sat, 7 Apr 2018 17:19:12 +0000 (13:19 -0400)]
Make target_read_alloc & al return vectors
This patch started by changing target_read_alloc_1 to return a
byte_vector, to avoid manual memory management (in target_read_alloc_1
and in the callers). To communicate failures to the callers, it
actually returns a gdb::optional<gdb::byte_vector>.
Adjusting target_read_stralloc was a bit more tricky, since it wants to
return a buffer of char, and not gdb_byte. Since you can't just cast a
gdb::byte_vector into a gdb::def_vector<char>, I made
target_read_alloc_1 templated, so both versions (that return vectors of
gdb_byte and char) are generated. Since target_read_stralloc now
returns a gdb::char_vector instead of a gdb::unique_xmalloc_ptr<char>, a
few callers need to be adjusted.
Further improve warnings for relocations referring to discarded sections.
Relocations referring to discarded sections are now treated as errors
instead of warnings.
Also with this patch, we will now print the section group signature and the
object file with the prevailing definition of that group along with the
name of the symbol that the relocation is referring to. This additional
information should be much more useful to anyone trying to track down
the source of such errors.
To do so, we now map each discarded section to the Kept_section info in
the Layout class, and defer the logic that maps a discarded section to
its counterpart in the kept group. This gives us the information we need
to identify the signature symbol given the discarded section, and the
name of the object file that provided the prevailing (i.e., first)
definition of that group.
gold/
* object.cc (Sized_relobj_file::include_section_group): Store
reference to Kept_section info for discarded comdat sections
regardless of size. Move size checking to map_to_kept_section.
(Sized_relobj_file::include_linkonce_section): Likewise.
(Sized_relobj_file::map_to_kept_section): Add section name parameter.
Insert size checking logic from above functions.
(Sized_relobj_file::find_kept_section_object): New method.
(Sized_relobj_file::get_symbol_name): New method.
* object.h (Sized_relobj_file::map_to_kept_section): Add section_name
parameter. Adjust all callers.
(Sized_relobj_file::find_kept_section_object): New method.
(Sized_relobj_file::get_symbol_name): New method.
(Sized_relobj_file::Kept_comdat_section): Replace object and shndx
fields with sh_size, kept_section, symndx, and is_comdat fields.
(Sized_relobj_file::set_kept_comdat_section): Replace kept_object
and kept_shndx parameters with is_comdat, symndx, sh_size, and
kept_section.
(Sized_relobj_file::get_kept_comdat_section): Likewise.
* target-reloc.h (enum Comdat_behavior): Change CB_WARNING to CB_ERROR.
Adjust all references.
(issue_undefined_symbol_error): New function template.
(relocate_section): Pass section name to map_to_kept_section.
Move discarded section code to new function above.
* aarch64.cc (Target_aarch64::scan_reloc_section_for_stubs): Move
declaration for gsym out one level. Call issue_discarded_error.
* arm.cc (Target_arm::scan_reloc_section_for_stubs): Likewise.
* powerpc.cc (Relocate_comdat_behavior): Change CB_WARNING to CB_ERROR.
Tom Tromey [Wed, 4 Apr 2018 03:52:31 +0000 (21:52 -0600)]
Remove value::next and value::released
This patch converts all_values to simply hold a list of references to
values. Now, there's no need to have a value record whether or not it
is released -- there is only a single reference-counting mechanism for
values. So, this also removes value::next, value::released, and
value_next.
gdb/ChangeLog
2018-04-06 Tom Tromey <tom@tromey.com>
* value.c (struct value) <released, next>: Remove.
(all_values): Now a std::vector.
(allocate_value_lazy): Update.
(value_next): Remove.
(value_mark, value_free_to_mark, release_value)
(value_release_to_mark): Update.
Tom Tromey [Wed, 4 Apr 2018 02:20:01 +0000 (20:20 -0600)]
Remove free_value_chain
This patch changes value_release_to_mark and fetch_subexp_value to
return a std::vector of value references, rather than relying on the
"next" field that is contained in a struct value. This makes it
simpler to reason about the returned values, and also allows for the
removal of free_value_chain.
gdb/ChangeLog
2018-04-06 Tom Tromey <tom@tromey.com>
* value.h (fetch_subexp_value, value_release_to_mark): Update.
(free_value_chain): Remove.
* value.c (free_value_chain): Remove.
(value_release_to_mark): Return a std::vector.
* ppc-linux-nat.c (num_memory_accesses): Change "chain" to a
std::vector.
(check_condition): Update.
* eval.c (fetch_subexp_value): Change "val_chain" to a
std::vector.
* breakpoint.c (update_watchpoint): Update.
(can_use_hardware_watchpoint): Change "vals" to a std::vector.
Tom Tromey [Wed, 4 Apr 2018 00:23:30 +0000 (18:23 -0600)]
Change value history to use value_ref_ptr
This simplifies the value history implementation by replacing the
current data structure with a std::vector, and by making the value
history simply hold a reference to each value.
gdb/ChangeLog
2018-04-06 Tom Tromey <tom@tromey.com>
projects / thirdparty / binutils-gdb.git / log
