Home - Waterfall Grid T-Grid Console Builders Recent Builds Buildslaves Changesources - JSON API - About

Console View

Tags: MAIL TEST aarch32 aarch64 aix cc-with-index centos7 debian fedora freebsd freebsd10 i686 index m32 m64 mingw32 native-extended-gdbserver native-extended-gdbserver-m32 native-extended-gdbserver-m64 native-gdbserver native-gdbserver-m32 native-gdbserver-m64 netbsd power8 ppc64be ppc64le rhel s390x solaris solaris11 sparcv9 ubuntu x86_64
Legend:   Passed Failed Warnings Failed Again Running Exception Offline No data

MAIL TEST aarch32 aarch64 aix cc-with-index centos7 debian fedora freebsd freebsd10 i686 index m32 m64 mingw32 native-extended-gdbserver native-extended-gdbserver-m32 native-extended-gdbserver-m64 native-gdbserver native-gdbserver-m32 native-gdbserver-m64 netbsd power8 ppc64be ppc64le rhel s390x solaris solaris11 sparcv9 ubuntu x86_64
John Baldwin
Add a new 'info proc files' subcommand of 'info proc'.
This command displays a list of open file descriptors.


* defs.h (enum info_proc_what) [IP_FILES]: New value.
* infcmd.c (info_proc_cmd_files): New function.
(_initialize_infcmd): Register 'info proc files' command.
GDB Administrator
Automatic date update in version.in
Tom Tromey
Check for gmp when checking for mpfr
There was a report on irc that the gdb check for mpfr failed when only
static libraries are available.  The issue is that mpfr depends on
gmp, but this is not handled explicitly by gdb.

Ideally upstream would switch to pkg-config.  Or even more ideally, we
would incorporate pkg-config into the compiler and not mess with any
of this.

Meanwhile, this changes gdb's configure to add gmp to the link line
when checking for mpfr.

2018-09-17  Tom Tromey  <tom@tromey.com>

* configure: Rebuild.
* configure.ac: Use gmp as a library dependency when checking for
Pedro Alves
[gdb/Python] Eliminate find_inferior_object
Commit 00431a78b28f ("Use thread_info and inferior pointers more
throughout") removed the declaration of find_inferior_object, but
missed removing the definition.

2018-09-17  Pedro Alves  <palves@redhat.com>

* python/py-inferior.c (find_inferior_object): Delete.
Tom Tromey
Remove munmap_listp_free_cleanup
This removes munmap_listp_free_cleanup, replacing it with a
std::unique_ptr at one spot and an explicit delete in another.  It
seemed simplest to completely change this data structure.

2018-09-18  Tom Tromey  <tom@tromey.com>

* compile/compile-object-run.c (do_module_cleanup): Use delete.
* compile/compile-object-load.c (struct munmap_list): Move to
header file.
(munmap_list::add): Rename from munmap_list_add; rewrite.
(munmap_list::~munmap_list): Rename from munmap_list_free.
(munmap_listp_free_cleanup): Remove.
(compile_object_load): Update.
* compile/compile-object-load.h (struct munmap_list): Move from
compile-object-load.c.  Rewrite.
GDB Administrator
Automatic date update in version.in
Simon Marchi
Carry the gnulib getcwd backport as a patch

  e2fc52e7457 ("Fix PR gdb/23558: Use system's 'getcwd' when cross-compiling GDB")

backported some changes from a future gnulib version to our import.
However, this means that every time someone wants to change our gnulib
import (e.g. add a module), they must make sure not to include that
backported change.  It also means that someone running the
update-gnulib.sh script without changes will get some diffs and wonder

Instead, I suggest we carry that backport as a patch applied by the
update-gnulib.sh script after running the import tool.  It will make it
clear what backport or local modification we have and should make
running update-gnulib.sh give a reproducible result.

There is a hunk in the configure file in this patch, this is because the
commit that backported the getcwd bits didn't include the re-generated

Note: you'll need this patch as well to get deterministic results:

  Generate aclocal-m4-deps.mk more deterministically and portably.


* patches/0001-Fix-PR-gdb-23558-Use-system-s-getcwd-when-cross-comp.patch:
New file.
* update-gnulib.sh: Apply patch.
* configure: Re-generate.
Sergio Durigan Junior
Expect optional "arch=" when executing "-stack-list-frames" on gdb.arch/amd64-invalid-stack-middle.exp
While regression-testing GDB on Fedora Rawhide, I saw the following
output when running gdb.arch/amd64-invalid-stack-middle.exp's
"-stack-list-frames" test:

  (gdb) interpreter-exec mi "-stack-list-frames"

This test is currently failing on Rawhide.  However, this output is
almost the same as I get on my local Fedora 28 machine (where the test
is passing):

  (gdb) interpreter-exec mi "-stack-list-frames"

With the exception that there's an "arch=" keyword on Fedora Rawhide's
version.  This is because, on Rawhide, I've compiled GDB with
"--enable-targets=xyz,kqp,etc.", while locally I haven't.

This is easy to fix: we just have to extend the regexp and expect for
the optional "arch=" keyword there.  It's what this patch does.  With
it applied, the test now passes everywhere.


2018-09-18  Sergio Durigan Junior  <sergiodj@redhat.com>

* gdb.arch/amd64-invalid-stack-middle.exp: Expect optional
"arch=" keyword when executing "-stack-list-frames".
Simon Marchi
python: Make gdb.execute("show commands") work (PR 23669)
Since commit

  56bcdbea2bed ("Let gdb.execute handle multi-line commands")

trying to use a command like gdb.execute("show commands") in Python
fails.  GDB ends up trying to run the "commands" command.

The reason is that GDB gets confused with the special "commands"
command.  In process_next_line, the lookup_cmd_1 function returns the
cmd_list_element representing the "commands" sub-command of "show".
Lower, we check the cmd_list_element to see if it matches various
control commands by name, including the "commands" command.  This is
where we wrongfully conclude that the executed command must be
"commands", when in reality it was "show commands".

The fix proposed in this patch removes the comparisons by name, instead
comparing the cmd_list_element object by pointer with the objects
created at initialization time.

Tested on the buildbot, though on a single builder (Fedora-x86_64-m64).


PR python/23669
* breakpoint.c (commands_cmd_element): New.
(_initialize_breakpoint): Assign commands_cmd_element.
* breakpoint.h (commands_cmd_element): New.
* cli/cli-script.c (while_cmd_element, if_command,
define_cmd_element): New.
(command_name_equals): Remove.
(process_next_line): Compare commands by pointer, not by name.
(_initialize_cli_script): Assign the various cmd_list_element
* compile/compile.c (compile_cmd_element): New.
(_initialize_compile): Assign compile_cmd_element.
* compile/compile.h (compile_cmd_element): New.
* guile/guile.c (guile_cmd_element): New.
(install_gdb_commands): Assign guile_cmd_element.
* guile/guile.h (guile_cmd_element): New.
* python/python.c (python_cmd_element): New.
(_initialize_python): Assign python_cmd_element.
* python/python.h (python_cmd_element): New.
* tracepoint.c (while_stepping_cmd_element): New.
(_initialize_tracepoint): Assign while_stepping_cmd_element.
* tracepoint.h (while_stepping_cmd_element): New.


PR python/23669
* gdb.python/python.exp: Test gdb.execute("show commands").
GDB Administrator
Automatic date update in version.in
Tamar Christina
Fix Aarch64 bug in warning filtering.
This fixes a small bug with the warning filtering code, which when a line has
generated a warning and a template decode error (due to the way templates are
resolved) which would not have been emitted and warnings are being suppressed
with -W it would erroneously emit the error.

I have no testcase for this because the only places we generate warnings during
encoding/decoding now is using msr/mrs and system registers.  They don't have a
template that would trigger this.

However an upcoming patch series will have tests in it which would expose this bug.


* config/tc-aarch64.c (output_operand_error_report): Apply filtering to
current instead of head message.
Tom Tromey
Do not pass -DNDEBUG to Python compilations in development mode
The Python CFLAGS include -DNDEBUG.  This was apparently done
intentionally -- setting the flags is done manually because, according
to a comment, python-config passes too many things to the compiler
(which is true).

Per PR python/20445, this patch changes configure so that -DNDEBUG is
only used by release builds.  This probably doesn't have very much
effect in practice, but I did see that some Python headers use assert,
so perhaps it will give some safety.

Tested by rebuilding and re-running gdb.python/*.exp on x86-64 Fedora 28.

2018-09-17  Tom Tromey  <tom@tromey.com>

PR python/20445:
* configure: Rebuild.
* configure.ac: Conditionally use -DNDEBUG for Python.
Alan Modra
Tweak map file output for pei386_auto_import
* ldmain.c (add_archive_element): Handle auto-inport symbols
when printing map.
John Baldwin
Use KF_PATH to verify the size of a struct kinfo_file.
fbsd_core_vnode_path needs to use the offset of the kf_path member of
struct kinfo_file as the minimum size of a struct kinfo_file object.
However, it was using KVE_PATH instead due to a copy and paste bug.

While here, fix another copy and paste bug in the error message for a
truncated kinfo_file object.


* fbsd-tdep.c (fbsd_core_vnode_path): Use KF_PATH instead of
Nick Clifton
Add a warning to the bfd library for when it encounters an ELF file with an invalid section size.
PR 23657
* elfcode.h (elf_swap_shdr_in): Generate a warning message if an
ELF section has contents and size larger than the file size.
John Baldwin
Document the 'info proc files' command.

* NEWS: Mention 'info proc files' command.


* gdb.texinfo (Process Information): Document "info proc files"
GDB Administrator