The Emacs 28 compiler warns about dwarf-mode.el:
Warning (comp): dwarf-mode.el:180:32: Warning: Unused lexical argument `ignore'
This is easily fixed by prepending "_" to the parameter's name.
binutils/ChangeLog
2022-12-19 Tom Tromey <tromey@adacore.com>
* dwarf-mode.el (dwarf-do-refresh): Avoid compiler warning.
quick_symbol_functions::relocated is only needed for psymtabs, and
there it is only needed for Rust. However, because we've switched the
DWARF reader away from psymtabs, this means there's no longer a need
for this method at all.
MI version 1 is long since obsolete. Several years ago, I filed
PR mi/23170 for this. I think it's finally time to remove this.
Any users of MI 1 can and should upgrade to a newer version.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=23170
Commit b5661ff2 ("gdb: fix possible use-after-free when
executing commands") used lookup_cmd_exact () to lookup
command again after its execution to avoid possible
use-after-free error.
However this change broke test gdb.base/define.exp which
defines a post-hook for subcommand ("target testsuite").
In this case, lookup_cmd_exact () returned NULL because
there's no command 'testsuite' in top-level commands.
This commit fixes this case by looking up the command again
using the original command line via lookup_cmd ().
Approved-By: Simon Marchi <simon.marchi@efficios.com>
PR 29914
* dwarf.c (fetch_indexed_value): Fail if the section is not big
enough to contain a header size field.
(display_debug_addr): Fail if the computed address size is too big
or too small.
There's no point in even trying the Java test when gprofng was built
without Java support, and when the building of the constituents of the
testcase also would fail. On such systems this converts the respective
tests from "unresolved" to "unsupported", making the overall testsuite
run no longer report failure just because of this.
Casting pointers to unsigned int is generally problematic and hence
compilers tend to warn about such. While here they're used only in
fprintf(), it still seems better to omit such casts, even if only to
avoid setting bad precedents.
A backslash used to indicate line continuation (in a macro definition
here) is not supposed to be followed by blanks or other white space; the
end-of-line indicator is to follow immediately.
The signal handling tests spend most of their time in the signal
handlers, and hence for profile output to match anything in program
output, the respective name fields need to hold the handler function
names. This converts both respective tests from "unresolved" to actually
succeeding.
In order for so_syn.so and so_syx.so to be able to access the main
program's "testtime" variable, that variable needs exposing in the
dynamic symbol table. Since this is a test program only, do it the brute
force way and simply expose all global symbols.
This allows the bootstrap test to run if you have a symlink somewhere
in the build path directory. $ld depends on $base_dir which is set
via tcl [pwd], collapsing the symlink like /usr/bin/pwd, while $objdir
contains the symlink.
* testsuite/ld-bootstrap/bootstrap.exp: Normalize paths when
checking for ld build directory.
Now that the GDB 13 branch has been created,
this commit bumps the version number in gdb/version.in to
14.0.50.DATE-git
For the record, the GDB 13 branch was created
from commit 71c90666e601c511a5f495827ca9ba545e4cb463.
Also, as a result of the version bump, the following changes
have been made in gdb/testsuite:
* gdb.base/default.exp: Change $_gdb_major to 14.
This patch removes the bfd_malloc in default_indirect_link_order and
bfd_simple_get_relocated_section_contents, pushing the allocation down
to bfd_get_relocated_section_contents. The idea is to make use of the
allocation done with sanity checking in bfd_get_full_section_contents,
which is called by bfd_generic_get_relocated_section_contents.
Doing this exposed a bug in bfd_get_full_section_contents. With
relaxation it is possible that an input section rawsize is different
to the section size. In that case we want to use the larger of
rawsize (the on-disk size for input sections) and size.
* reloc.c (bfd_generic_get_relocated_section_contents),
* reloc16.c (bfd_coff_reloc16_get_relocated_section_contents),
* coff-alpha.c (alpha_ecoff_get_relocated_section_contents),
* coff-sh.c (sh_coff_get_relocated_section_contents),
* elf-m10200.c (mn10200_elf_get_relocated_section_contents),
* elf-m10300.c (mn10300_elf_get_relocated_section_contents),
* elf32-avr.c (elf32_avr_get_relocated_section_contents),
* elf32-cr16.c (elf32_cr16_get_relocated_section_contents),
* elf32-crx.c (elf32_crx_get_relocated_section_contents),
* elf32-h8300.c (elf32_h8_get_relocated_section_contents),
* elf32-nds32.c (nds32_elf_get_relocated_section_contents),
* elf32-sh.c (sh_elf_get_relocated_section_contents),
* elfxx-mips.c (_bfd_elf_mips_get_relocated_section_contents):
Handle NULL data buffer.
* bfd.c (bfd_get_section_alloc_size): New function.
* bfd-in2.h: Regenerate.
* compress.c (bfd_get_full_section_contents): Correct section
malloc size.
* linker.c (default_indirect_link_order): Don't malloc memory
here before calling bfd_get_relocated_section_contents.
* simple.c (bfd_simple_get_relocated_section_contents): Likewise.
That's this line in elf_parse_notes:
while (p < buf + size)
* elf.c (_bfd_elf_make_section_from_shdr): Don't call
elf_parse_notes when sh_size is zero.
Commit 5aa0f10c424e added a per_xvec_warn array to provide support for
warnings from elf_object_p (and a later patch for warnings from
pe_bfd_object_p) to be cached and then only printed if the target
matches. It was quite limited in the style of message supported, only
one message could be printed, and didn't really meet the stated aim of
only warning when a target matches: There are many other errors and
warnings that can be emitted by functions called from elf_object_p.
So this patch extends the error handler functions to support printing
to a string buffer, extends per_xvec_warn to support multiple errors/
warnings, and hooks this all into bfd_check_format_matches. If
bfd_check_format_matches succeeds then any errors/warnings are printed
for the matching target. If bfd_check_format_matches fails either due
to no match or to multiple matches and only one target vector produced
errors, then those errors are printed.
* bfd.c (MAX_ARGS): Define, use throughout.
(print_func): New typedef.
(_bfd_doprnt): Add new print param. Replace calls to fprintf
with print.
(PRINT_TYPE): Similarly.
(error_handler_fprintf): Renamed from error_handler_internal.
Use _bfd_get_error_program_name. Add fprintf arg. Move code
setting up args..
(_bfd_doprnt_scan): ..to here. Add ap param.
(struct buf_stream): New.
(err_sprintf): New function.
(error_handler_bfd): New static variable.
(error_handler_sprintf): New function.
(_bfd_set_error_handler_caching): New function.
(_bfd_get_error_program_name): New function.
* elfcode.h (elf_swap_shdr_in): Use _bfd_error_handler in
warning messages.
(elf_object_p): Likewise.
* format.c (print_warnmsg): New function.
(clear_warnmsg): Rewrite.
(null_error_handler): New function.
(bfd_check_format_matches): Ignore warnings from recursive calls
checking first element of an archive. Use caching error handler
otherwise. Print warnings on successful match, or when only one
target has emitted warnings/errors.
* peicode.h (pe_bfd_object_p): Use _bfd_error_handler in
warning messages.
* targets.c (per_xvec_warn): Change type of array elements.
(struct per_xvec_message): New.
(_bfd_per_xvec_warn): Rewrite.
* Makefile.am (LIBBFD_H_FILES): Add bfd.c.
* Makefile.in: Regenerate.
* bfd-in2.h: Regenerate.
* libbfd.h: Regenerate.
Add a new test to check that .cfi_negate_ra_state on aarch64 is handled
well (a non-empty SFrame section with valid SFrame FREs is generated).
ChangeLog:
* testsuite/gas/cfi-sframe/cfi-sframe-aarch64-2.d: New test.
* testsuite/gas/cfi-sframe/cfi-sframe-aarch64-2.s: Likewise.
* testsuite/gas/cfi-sframe/cfi-sframe.exp: Adjust the list
accordingly.
In the textual dump of the SFrame section, when an SFrame FRE recovers a
mangled RA, use string "[s]" in the output to indicate that the return
address is a signed (mangled) one.
ChangeLog:
* libsframe/sframe-dump.c (dump_sframe_func_with_fres): Postfix
with "[s]" if RA is signed with authorization code.
DW_CFA_AARCH64_negate_ra_state in aarch64 is multiplexed with
DW_CFA_GNU_window_save in the DWARF format.
Remove the common-empty-4 testcase because the generated SFrame section
will not be be empty anymore. A relevant test will be added in a later
commit.
ChangeLog:
* gas/gen-sframe.c (sframe_v1_set_fre_info): Add new argument
for mangled_ra_p.
(sframe_set_fre_info): Likewise.
(output_sframe_row_entry): Handle mangled_ra_p.
(sframe_row_entry_new): Reset mangled_ra_p.
(sframe_row_entry_initialize): Initialize mangled_ra_p.
(sframe_xlate_do_gnu_window_save): New definition.
(sframe_do_cfi_insn): Handle DW_CFA_GNU_window_save.
* gas/gen-sframe.h (struct sframe_row_entry): New member.
(struct sframe_version_ops): Add a new argument for
mangled_ra_p.
* gas/testsuite/gas/cfi-sframe/cfi-sframe.exp: Remove test.
* gas/testsuite/gas/cfi-sframe/common-empty-4.d: Removed.
* gas/testsuite/gas/cfi-sframe/common-empty-4.s: Removed.
Use the last remaining bit in the 'SFrame FRE info' word to store whether
the RA is signed/unsigned with PAC authorization code: this bit is named
as the "mangled RA" bit. This bit is still unused for x86-64.
The behaviour of the mangled-RA info bit in SFrame format closely
follows the behaviour of DW_CFA_AARCH64_negate_ra_state in DWARF. During
unwinding, whenever an SFrame FRE with non-zero "mangled RA" bit is
encountered, it means the upper bits of the return address contain Pointer
Authentication code. The unwinder, hence, must use appropriate means to
restore LR correctly in such cases.
include/ChangeLog:
* sframe.h (SFRAME_V1_FRE_INFO_UPDATE_MANGLED_RA_P): New macro.
(SFRAME_V1_FRE_MANGLED_RA_P): Likewise.
As of 1bcb0708f229 ("gdb/linux-nat: Check whether /proc/pid/mem is
writable"), GDB checks if /proc/pid/mem is writable. This is done
early at GDB startup, in order to get a consistent warning, instead of
a warning that depends on whenever GDB writes to inferior memory.
PR gdb/29907 points out that some build systems (like QEMU's,
apparently) may call 'gdb --version' to check GDB's presence & its
version on the system, and that Gentoo's build process has sandboxing
which blocks the /proc/pid/mem access and thus GDB warns, which
results in build fails.
To help with that, this patch delays the /proc/pid/mem check until we
start or attach to an inferior. Ends up potentially emiting a warning
close where we already emit other ptrace- and /proc- related warnings,
which just Feels Right.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29907
Change-Id: I5537653ecfbbe76a04ab035e40e59d09b4980763
Once in a while I run into:
...
FAIL: gdb.threads/detach-step-over.exp: \
breakpoint-condition-evaluation=host: target-non-stop=off: non-stop=off: \
displaced=off: iter 1: all threads running
...
In can easily reproduce this by doing:
...
# Wait a bit, to give time for the threads to hit the
# breakpoint.
- sleep 1
return true
...
Fix this by counting the running threads in a loop, effectively allowing 10
seconds (instead of 1) for the threads to start running, but only sleeping if
needed.
Reduces total execution time from 1m27s to 56s.
Tested on x86_64-linux.
When the source program contains a goto label, it turns out it's
actually pretty hard for a user to find out more about that label.
For example:
(gdb) p some_label
No symbol "some_label" in current context.
(gdb) disassemble some_label
No symbol "some_label" in current context.
(gdb) x/10i some_label
No symbol "some_label" in current context.
(gdb) break some_label
Breakpoint 2 at 0x401135: file /tmp/py-label-symbol-value.c, line 35.
In all cases, some_label is a goto label within the current frame.
Only placing a breakpoint on the label worked.
This all seems a little strange to me, it feels like asking about a
goto label would not be an unreasonable thing for a user to do.
This commit doesn't fix any of the above issues, I mention them just
to provide a little context for why the following issue has probably
not been seen before.
It turns out there is one way a user can access the symbol for a goto
label, through the Python API:
python frame = gdb.selected_frame()
python frame_pc = frame.pc()
python block = gdb.current_progspace().block_for_pc(frame_pc)
python symbol,_ = gdb.lookup_symbol('some_label', block, gdb.SYMBOL_LABEL_DOMAIN)
python print(str(symbol.value()))
../../src/gdb/findvar.c:204: internal-error: store_typed_address: Assertion `type->is_pointer_or_reference ()' failed.
The problem is that label symbols are created using the
builtin_core_addr type, which is a pure integer type.
When GDB tries to fetch the value of a label symbol then we end up in
findvar.c, in the function language_defn::read_var_value, in the
LOC_LABEL case. From here store_typed_address is called to store the
address of the label into a value object with builtin_core_addr type.
The problem is that store_typed_address requires that the destination
type be a pointer or reference, which the builtin_core_addr type is
not.
Now it's not clear what type a goto label address should have, but
GCC has an extension that allows users to take the address of a goto
label (using &&), in that case the result is of type 'void *'.
I propose that when we convert the CORE_ADDR value to a GDB value
object, we use builtin_func_ptr type instead of builtin_core_addr,
this means the result will be of type 'void (*) ()'. The benefit of
this approach is that when gdbarch_address_to_pointer is called the
target type will be correctly identified as a pointer to code, which
should mean any architecture specific adjustments are done correctly.
We can then cast the new value to 'void *' type with a call to
value_cast_pointer, this should not change the values bit
representation, but will just update the type.
After this asking for the value of a label symbol works just fine:
(gdb) python print(str(symbol.value()))
0x401135 <main+35>
And the type is maybe what we'd expect:
(gdb) python print(str(symbol.value().type))
void *
Replace the use of struct buffer in linux-osdata.c with std::string.
There is no change in the logic, so there should be no user-visible
change.
Change-Id: I27f53165d401650bbd0bebe8ed88221e25545b3f
Approved-By: Pedro Alves <pedro@palves.net>
Add a version of buffer_xml_printf (defined in gdbsupport/buffer.{c,h})
that appends to an std::string, rather than a struct buffer. Call it
"string" rather than "buffer" since it operates on an std::string rather
than a buffer. And call it "appendf" rather than "printf", since it
appends to and does not replace the string's content. This mirrors
string_appendf.
Place the new version in gdbsupport/xml-utils.h.
The code is a direct copy of buffer_xml_printf. The old version is
going to disappear at some point, which is why I didn't do any effort to
share code.
Change-Id: I30e030627ab4970fd0b9eba3b7e8cec78fa561ba
Approved-By: Pedro Alves <pedro@palves.net>
This commit:
commit 53cf95c3389a3ecd97276d322e4a60fe3396a201
Date: Wed Dec 14 14:17:44 2022 +0000
gdb: make more use of make_target_connection_string
Introduced a couple of inefficient uses of std::string, both of which
are fixed in this commit.
There should be no user visible changes after this commit.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
This commit adds a test that creates a Python command that redefines
itself during its execution. This is to test use-after-free in
execute_command ().
This test needs run with ASan enabled in order to fail when it
should.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
PR gdb/28947
The address_significant gdbarch setting was introduced as a way to remove
non-address bits from pointers, and it is specified by a constant. This
constant represents the number of address bits in a pointer.
Right now AArch64 is the only architecture that uses it, and 56 was a
correct option so far.
But if we are using Pointer Authentication (PAuth), we might use up to 2 bytes
from the address space to store the required information. We could also have
cases where we're using both PAuth and MTE.
We could adjust the constant to 48 to cover those cases, but this doesn't
cover the case where GDB needs to sign-extend kernel addresses after removal
of the non-address bits.
This has worked so far because bit 55 is used to select between kernel-space
and user-space addresses. But trying to clear a range of bits crossing the
bit 55 boundary requires the hook to be smarter.
The following patch renames the gdbarch hook from significant_addr_bit to
remove_non_address_bits and passes a pointer as opposed to the number of
bits. The hook is now responsible for removing the required non-address bits
and sign-extending the address if needed.
While at it, make GDB and GDBServer share some more code for aarch64 and add a
new arch-specific testcase gdb.arch/aarch64-non-address-bits.exp.
Bug-url: https://sourceware.org/bugzilla/show_bug.cgi?id=28947
Approved-By: Simon Marchi <simon.marchi@efficios.com>
While 6fdb723799e2 ("gas: re-work line number tracking for macros and
their expansions") was meant to leave generated Dwarf as is, it really
didn't (and the testcase intended to catch that wasn't covering the case
which broke). Its adjustment to buffer_and_nest() didn't go far enough,
leading to the "linefile" directive inserted at the top to also be
processed later in the PR gas/16908 workaround (which clearly isn't
intended - it's being put there for processing during macro expansion
only). That unnoticed flaw in turn led me to work around it by a
(suspicious to me already at the time) conditional in as_where().
Having a "None" field in the vast majority of entries is needlessly
cluttering the overall table. Instead of this being a separate field,
use a representation matching that of Intel SDM and AMD PM for the main
use of the field: Append the value after a / as the separator.
[I sent this earlier today, but I don't see it in the archives.
Resending it through a different computer / SMTP.]
The use of the static buffer in command_line_input is becoming
problematic, as explained here [1]. In short, with this patch [2] that
attempt to fix a post-hook bug, when running gdb.base/commands.exp, we
hit a case where we read a "define" command line from a script file
using command_command_line_input. The command line is stored in
command_line_input's static buffer. Inside the define command's
execution, we read the lines inside the define using command_line_input,
which overwrites the define command, in command_line_input's static
buffer. After the execution of the define command, execute_command does
a command look up to see if a post-hook is registered. For that, it
uses a now stale pointer that used to point to the define command, in
the static buffer, causing a use-after-free. Note that the pointer in
execute_command points to the dynamically-allocated buffer help by the
static buffer in command_line_input, not to the static object itself,
hence why we see a use-after-free.
Fix that by removing the static buffer. I initially changed
command_line_input and other related functions to return an std::string,
which is the obvious but naive solution. The thing is that some callees
don't need to return an allocated string, so this this an unnecessary
pessimization. I changed it to passing in a reference to an std::string
buffer, which the callee can use if it needs to return
dynamically-allocated content. It fills the buffer and returns a
pointers to the C string inside. The callees that don't need to return
dynamically-allocated content simply don't use it.
So, it started with modifying command_line_input as described above, all
the other changes derive directly from that.
One slightly shady thing is in handle_line_of_input, where we now pass a
pointer to an std::string's internal buffer to readline's history_value
function, which takes a `char *`. I'm pretty sure that this function
does not modify the input string, because I was able to change it (with
enough massaging) to take a `const char *`.
A subtle change is that we now clear a UI's line buffer using a
SCOPE_EXIT in command_line_handler, after executing the command.
This was previously done by this line in handle_line_of_input:
/* We have a complete command line now. Prepare for the next
command, but leave ownership of memory to the buffer . */
cmd_line_buffer->used_size = 0;
I think the new way is clearer.
[1] https://inbox.sourceware.org/gdb-patches/becb8438-81ef-8ad8-cc42-fcbfaea8cddd@simark.ca/
[2] https://inbox.sourceware.org/gdb-patches/20221213112241.621889-1-jan.vrany@labware.com/
Change-Id: I8fc89b1c69870c7fc7ad9c1705724bd493596300
Reviewed-By: Tom Tromey <tom@tromey.com>
There are two places where unaligned loads were seen on aarch64:
- #1. access to the SFrame FRE stack offsets in the in-memory
representation/abstraction provided by libsframe.
- #2. access to the SFrame FRE start address in the on-disk representation
of the frame row entry.
For #1, we can fix this by reordering the struct members of
sframe_frame_row_entry in libsframe/sframe-api.h.
For #2, we need to default to using memcpy instead, and copy out the bytes
to a location for output.
SFrame format is an unaligned on-disk format. As such, there are other blobs
of memory in the on-disk SFrame FRE that are on not on their natural
boundaries. But that does not pose further problems yet, because the users
are provided access to the on-disk SFrame FRE data via libsframe's
sframe_frame_row_entry, the latter has its' struct members aligned on their
respective natural boundaries (and initialized using memcpy).
PR 29856 libsframe asan: load misaligned at sframe.c:516
ChangeLog:
PR libsframe/29856
* bfd/elf64-x86-64.c: Adjust as the struct members have been
reordered.
* libsframe/sframe.c (sframe_decode_fre_start_address): Use
memcpy to perform 16-bit/32-bit reads.
* libsframe/testsuite/libsframe.encode/encode-1.c: Adjust as the
struct members have been reordered.
include/ChangeLog:
PR libsframe/29856
* sframe-api.h: Reorder fre_offsets for natural alignment.
Don't delete the runtime-generated command files. This makes it easier
to reproduce tests by hand.
Change-Id: I4e53484eea216512f1c5d7dfcb5c464b36950946
Approved-By: Tom Tromey <tom@tromey.com>
