Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: Nicolas Blanc <nicolas.blanc@intel.com>
To: gdb-patches@sourceware.org
Cc: lgustavo@codesourcery.com, nicolas.blanc@intel.com
Subject: [PATCH v14 0/5] remove-symbol-file & add-symbol-file
Date: Mon, 29 Jul 2013 08:28:00 -0000	[thread overview]
Message-ID: <1375086490-14352-1-git-send-email-nicolas.blanc@intel.com> (raw)

Adjust Patch 3/5 to reflect modification of struct target_section
from upstream. Thanks Luis for pointing this out.

The patch set was tested on x86_64 (Fedora 15) and x86 (Ubuntu 12.04).
No regression was detected.

Regards,

Nicolas

Changes between v13 and v3

* symfile.c (add_target_sections_of_objfile): Reflect modification
  of struct target_section from upstream.
  (symfile_free_objfile): Reflect modification of struct target_section
  from upstream.

Changes between v12 and v13

* Changed "is_elf_target" to "is_known_elf_target" (Patch 4/5)
  according to Doug's proposal
  (http://sourceware.org/ml/gdb-patches/2013-07/msg00426.html).

* Removed dependency to <elf.h> according to Mark's suggestion
  (http://sourceware.org/ml/gdb-patches/2013-07/msg00415.html).

Changes between v11 and v12

* symfile.c (add_target_sections_of_objfile): Reflect modification
  of struct target_section from upstream.

* gdb.base/sym-file-main.c: Move loader into a separate file as suggested
  and implemented the other recommendations from Doug.

* gdb.base/sym-file-loader.c: New file.

* gdb.base/sym-file-loader.h: New file.

* lib/gdb.exp (is_elf_target): New function.

Changes between v10 and v11:

* Changed documentation for add-symbol-file according to feedback (Patch 3):
  "keeps adding to" -> "is kept in addition to".

* Fixed indentation issues in test (sym-file-main.c from Patch 2).

Changes between v9 and v10:

* Dropped "USAGE" in "USAGE: remove-symbol-file FILENAME" according to suggestion.

* Fixed comment according to suggestion:
  "any reference to any objfile" -> "all references to any objfile"

* Added argument names to the prototype of is_addr_in_objfile according to suggestion.

* Fixed indentation of "macro" loops according to feedback.

* Fixed indentation issues.

Changes between v8 and v9:

* Patch 4: New patch (fix  add-symbol-file for 'info files').
* Patch 5: New patch (test add-symbol-file and 'info files');

* sym-file.exp: Fixed test name string.

Changes between v7 and v8:

* sym-file.exp: Call gdb_get_line_number with ${libfile}.c intead of $libsrc.

* Fixed typo in commit message for Patch 2/3.

Changes between v6 and v7:

* Changed disable_breakpoints_in_freed_objfile () from feedback:
  Use is_breakpoint () and is_tracepoint ().
  Add call to mark_breakpoint_location_modified ().

* Changed command description according to suggestion.

* Changed commit log message for new functions according to feedback.

* Fixed nits in gdb/solib.c and testsuite/gdb.base/sym-file.exp.

Changes between v5 and v6:

* Changed syntax of the command according to feedback:
    remove-symbol-file FILENAME
    remove-symbol-file -a ADDRESS

* Updated the documentation for the new syntax.

* Updated the test for the new syntax.

* Added check to only remove objfiles from the current program space.

Changes between v4 and v5:

* The address parameter can now be any address mapping into one of the sections
  of the file to remove. This is more flexible and more judicious because the symbol
  file may have no text section, as pointed out in some reviews.

* Move is_addr_in_objfile from breakpoint.c to objfiles.c. This new function
  is now used in both symfile.c and breakpoint.c.

* is_addr_in_objfile returns now 0 if OBJFILE is NULL.

* Updated the documentation to reflect the command change.

* Removed double blank lines.

* Reworded the log entries in the commit messages, as suggested.

* The test was simplified using prepare_for_testing, as suggested.

Changes between v3 and v4:

* gdb/symfile.c: add missing space character in output string.

* gdb/doc/gdb.texinfo: add usage example.

Changes between v2 and v3:

* gdb/breakpoints.c (disable_breakpoints_in_freed_objfile):
  Changed name of disable_breakpoints_in_free_objfile to disable_breakpoints_in_freed_objfile.
  Call observer_breakpoint_modified at most one time per breakpoint.

* gdb/objfiles.h (struct objfile): Add comment for low_addr.

* gdb/solib.c (remove_user_added_objfile): Don't check OBJF_SHARED anymore to avoid leaving dangling pointers.

* gdb/symfile.c (add_symbol_file_command): Parse and evaluate the address parameter only once.

* gdb/symfile.c (remove_symbol_file_command): Don't use gdb_buildargv anymore. Ask the user for confirmation before removing a file.

* gdb/doc/gdb.texinfo: The parameter of the command is now referred to as "the address of the text section".

* gdb/NEWS: Move the entry for remove-symbol-file to "Changes since GDB 7.6". The address parameter is now referred to as "the beginning of the text section".

* Fixed letter case of variables in comments.

* Fixed indentation typos.


Nicolas Blanc (5):
  New remove-symbol-file command.
  Documentation for the remove-symbol-file command.
  'add-symbol-file' should update the current target sections.
  Function is_known_elf_target.
  Test adding and removing a symbol file at runtime.

 gdb/NEWS                                 |    6 +
 gdb/breakpoint.c                         |   67 ++++++-
 gdb/doc/gdb.texinfo                      |   26 ++-
 gdb/doc/observer.texi                    |    4 +
 gdb/objfiles.c                           |   23 ++
 gdb/objfiles.h                           |    2 +
 gdb/printcmd.c                           |   15 +-
 gdb/solib.c                              |   22 ++
 gdb/symfile.c                            |  150 +++++++++++++-
 gdb/testsuite/gdb.base/sym-file-lib.c    |   26 +++
 gdb/testsuite/gdb.base/sym-file-loader.c |  353 ++++++++++++++++++++++++++++++
 gdb/testsuite/gdb.base/sym-file-loader.h |   99 +++++++++
 gdb/testsuite/gdb.base/sym-file-main.c   |   79 +++++++
 gdb/testsuite/gdb.base/sym-file.exp      |  160 ++++++++++++++
 gdb/testsuite/lib/gdb.exp                |   15 ++
 15 files changed, 1034 insertions(+), 13 deletions(-)
 create mode 100644 gdb/testsuite/gdb.base/sym-file-lib.c
 create mode 100644 gdb/testsuite/gdb.base/sym-file-loader.c
 create mode 100644 gdb/testsuite/gdb.base/sym-file-loader.h
 create mode 100644 gdb/testsuite/gdb.base/sym-file-main.c
 create mode 100644 gdb/testsuite/gdb.base/sym-file.exp

-- 
1.7.6.5


             reply	other threads:[~2013-07-29  8:28 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-29  8:28 Nicolas Blanc [this message]
2013-07-29  8:28 ` [PATCH v14 5/5] Test adding and removing a symbol file at runtime Nicolas Blanc
2013-07-29  8:45   ` Blanc, Nicolas
2013-07-29  8:28 ` [PATCH v14 2/5] Documentation for the remove-symbol-file command Nicolas Blanc
2013-07-29  8:28 ` [PATCH v14 4/5] Function is_known_elf_target Nicolas Blanc
2013-07-29  8:28 ` [PATCH v14 1/5] New remove-symbol-file command Nicolas Blanc
2013-07-29  8:28 ` [PATCH v14 3/5] 'add-symbol-file' should update the current target sections Nicolas Blanc
2013-09-04 13:02 ` [PATCH v14 0/5] remove-symbol-file & add-symbol-file Blanc, Nicolas

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1375086490-14352-1-git-send-email-nicolas.blanc@intel.com \
    --to=nicolas.blanc@intel.com \
    --cc=gdb-patches@sourceware.org \
    --cc=lgustavo@codesourcery.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox