* g10/keydb.c (kid_list_t): New.
(kid_not_found_table, n_kid_not_found_table): New.
(kid_not_found_p, kid_not_found_insert, kid_not_found_flush): New.
(keydb_insert_keyblock): Flush the new cache.
(keydb_delete_keyblock): Ditto.
(keydb_update_keyblock): Ditto.
(keydb_search): Use the new cache.
(keydb_dump_stats): New.
* g10/gpg.c (g10_exit): Dump keydb stats.
--
What we do here is to keep track of key searches by long keyids (as
stored in all signatures) so that we do not need to scan the keybox
again after we already found that this keyid will result in
not-found. As soon as we change gpg to run as a co-process we should
store this table per session because other instances of gpg may have
updated the keybox without us knowing.
On a test ring with
gpg: 94721 good signatures
gpg: 6831 bad signatures
gpg: 150703 signatures not checked due to missing keys
gpg: 5 signatures not checked due to errors
gpg: keydb: kid_not_found_table: total: 14132
this new cache speeds a --check-sigs listing up from 28 minutes to
less than 3 minutes.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ldap-wrapper.c (ldap_wrapper_thread): Do not close the
stdout reader after EOF from read_log_data.
* dirmngr/crlcache.c (crl_cache_reload_crl): Close the reader before
the next iteration.
--
I assumed that the log_fd also has a reader object but that reader
object is used for stdout and needs to be closed by the consumer.
The real bug with the non-released ldap_wrapper control objects was
that when looping over distribution points we did not closed the used
reader object before the next iteration. Now, the test case had more
than one DP and thus we lost one reader object.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/ldap-wrapper.c (ldap_wrapper_thread): Move nfds computation
into the loop. Check the queue also on timeout. Close log_fd and
reader context on EOF or error.
--
The major bug here was that on an EOF of the log fd the log fd was not
closed and thus the final queue item removal could not work. Checking
the queue on a timeout is not really necessary but it help in case
there is a race condition lingering.
GnuPG-bug-id: 1838, 1978
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/dirmngr.c (pid_suffix_callback): Clear int_and_ptr_u before
use.
(start_connection_thread): Ditto.
(handle_connections): Ditto.
--
Example valgrind output:
==2921== Conditional jump or move depends on uninitialised value(s)
==2921== at 0x5BBDEF4: pthread_getspecific (pthread_getspecific.c:57)
==2921== by 0x40AAEE: pid_suffix_callback (dirmngr.c:614)
==2921== by 0x433F5A: do_logv (logging.c:684)
This is because on 64 bit systems "sizeof aptr > sizeof aint" and thus
Valgrind complains about this. It is no a real problem because we
don't use the unitialized bits.
Signed-off-by: Werner Koch <wk@gnupg.org>
--
Under Windows the file names are determined at runtime. To have
somewhat useful names in the manuals, we provide replacements using
the strings "INSTDIR" and "APPDATA" for the installation directory and
the user specific application data.
* agent/gpg-agent.c (start_connection_thread): Rename to ...
(do_start_connection_thread): this. Factor nonce checking out to ...
(start_connection_thread_std): this,
(start_connection_thread_extra): this,
(start_connection_thread_browser): and this.
--
Although not tested, the code did not worked on Windows becuase we
were checning the wrong nonce.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/agent.h (opt): Add field "browser_socket".
* agent/command.c (cmd_setkeydesc): Use a different message for
restricted==2.
* agent/gpg-agent.c (oBrowserSocket): New.
(opts): Add --browser-socket.
(socket_name_browser, redir_socket_name_browser): New.
(socket_nonce_browser): New.
(cleanup): Cleanup browser socket.
(main): Implement option.
(start_connection_thread_browser): New.
(handle_connections): Add arg listen_fd_browser and use it.
--
This is very similar to --extra-socket but intended to be used by a web
browser session. AS of now it only displays a different "Note: in
the Pinentry than --extra-socket but it may eventually be tweaked for
the use by browser extensions making use of gpg-agent.
It is marked experimental and and thus may be removed in later
versions.
To better support the different "client classes", it would be useful
to add corresponsing cache classes so that each class has its own
cache.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/agent.h (opt): Add field allow_emacs_pinentry.
* agent/call-pinentry.c (start_pinentry): Act upon new var.
* agent/gpg-agent.c (oAllowEmacsPinentry): New.
(opts): Add option --allow-emacs-pinentry.
(parse_rereadable_options): Set this option.
* tools/gpgconf-comp.c (gc_options_gpg_agent): Add new option.
--
gpgconf-comp and manual entry added by wk.
Signed-off-by: Werner Koch <wk@gnupg.org>
* doc/mkdefsinc.c: New.
* doc/Makefile.am: Include cmacros.am.
(EXTRA_DIST): Add mkdefsinc.c defsincdate.
(BUILT_SOURCES): Add defsincdate
(CLEANFILES): Add mkdefsinc and defs.inc.
(mkdefsinc): New rule.
(yat2m-stamp): Depend on defs.inc.
($(myman_pages) gnupg.7): Ditto.
(gnupg.texi): Remove rule to touch itself.
(dist-hook): New.
(defsincdate): New.
(defs.inc): New.
* doc/gnupg.texi: Remove inclusion of version.texi. Include defs.inc.
Also include defs.inc in all files used to build man files. Change
fixed directory names to those from defs.inc.
--
GnuPG-bug-id: 1661
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_getinfo): Add new sub-command.
--
The current output is:
> getinfo std_env_names
D GPG_TTY
D TERM
D DISPLAY
D XAUTHORITY
D XMODIFIERS
D GTK_IM_MODULE
D QT_IM_MODULE
D PINENTRY_USER_DATA
OK
Note that there is an invisible \x00 at the end of each line.
* g10/tdbio.c (tdbio_read_record): Fix returning of the error.
--
Actually the returned error will anyway be GPG_ERR_TRUSTDB but the old
code was not correct.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/tdbdump.c (list_trustdb): Add arg FP and change callers to pass
es_stdout.
* g10/tdbio.c (upd_hashtable): On a corrupted trustdb call
list_trustdb only in verbose > 1 mode and let it dump to stderr.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/signal.c: Remove.
* g10/main.h: Remove old function API.
* g10/tdbio.c: Use new API, even in the dead code.
--
We use common/signal.c now. The file g10/signal.c has been useless
since 2003-06-27. Now, the removal.
* agent/command.c (cmd_get_passphrase): Read from the user cache.
--
We used to read the passphrase with mode CACHE_MODE_NORMAL but we put
it into the cache with CACHE_MODE_USER. However, agent_get_cache does
not yet distinguish between them and thus this does not change
anything.
* agent/agent.h (agent_get_passphrase): Add arguments keyinfo and
cache_mode. Update callers.
* agent/call-pinentry.c (agent_get_passphrase): Add arguments keyinfo
and cache_mode. If KEYINFO and CACHE_MODE describe a cachable key,
then send SETKEYINFO to the pinentry.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* Makefile.am: Always build kbx/
* g10/Makefile.am (AM_CFLAGS): Include KSBA_CFLAGS.
--
Note that "make check" still prints a warning.
Signed-off-by: Werner Koch <wk@gnupg.org>
