* g10/tofu.c (initdb): Add column effective_policy to the bindings
table.
(record_binding): New parameters effective_policy and set_conflict.
Save the effective policy. If SET_CONFLICT is set, then set conflict
according to CONFLICT. Otherwise, preserve the current value of
conflict. Update callers.
(get_trust): Don't compute the effective policy here...
(get_policy): ... do it here, if it was not cached. Take new
parameters, PK, the public key, and NOW, the time that the operation
started. Update callers.
(show_statistics): New parameter PK. Pass it to get_policy. Update
callers.
(tofu_notice_key_changed): New function.
* g10/gpgv.c (tofu_notice_key_changed): New stub.
* g10/import.c (import_revoke_cert): Take additional argument CTRL.
Pass it to keydb_update_keyblock.
* g10/keydb.c (keydb_update_keyblock): Take additional argument CTRL.
Update callers.
[USE_TOFU]: Call tofu_notice_key_changed.
* g10/test-stubs.c (tofu_notice_key_changed): New stub.
* tests/openpgp/tofu.scm: Assume that manually setting a binding's
policy to auto does not cause the tofu engine to forget about any
conflict.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
We now store the computed policy in the tofu DB (in the
effective_policy column of the bindings table) to avoid computing it
every time, which is expensive. Further, policy is never overridden
in case of a conflict. Instead, we detect a conflict if CONFLICT is
not empty.
This change is backwards compatible to existing DBs. The only minor
incompatibility is that unresolved conflicts won't be automatically
resolved in case we import a direct signature, or cross signatures.
* g10/tofu.c (struct tofu_dbs_s): Remove unused field
record_binding_update2. Replace register_insert with
register_signature and register_encryption.
(tofu_register_signature): Don't use dbs->s.register_insert, but
dbs->s.register_signature.
(tofu_register_encryption): Don't use dbs->s.register_insert, but
dbs->s.register_encryption.
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* g10/tofu.c (busy_handler): Fix the flags and utime is not needed.
--
The argument flags must include one of O_RDONLY, O_WRONLY, or O_RDWR.
Adding O_TRUNC, the file is updated. So, utime is not needed.
Fixes-commit: b2e1b17efa952afcf7aeec8b15e9d0088dba587a
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/tofu.c (busy_handler): Add third argument which is mandatory for
O_CREATE flag.
--
Reported-by: Kristian Fiskerstrand
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/gpg.c (oOverrideSessionKeyFD): New.
(opts): Add option --override-session-key-fd.
(main): Handle that option.
(read_sessionkey_from_fd): New.
--
The override-session-key feature was designed to mitigate the effect
of the British RIP act by allowing to keep the private key private and
hand out only a session key. For that use case the leaking of the
session key would not be a problem. However there are other use
cases, for example fast re-decryption after an initial decryption,
which would benefit from concealing the session key from other users.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/gpgv.c (oEnableSpecialFilenames): New.
(opts): Add option --enable-special-filenames.
(main): Implement that option.
--
This is the same option we have in gpg. It allows to use commands
like
gpgv --enable-special-filenames -- '-&3' - <msg 3<msg.sig
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/options.h (CO_DE_VS): New.
(GNUPG): Also allow CO_DE_VS.
* g10/gpg.c (oDE_VS): New.
(parse_compliance_option): Add "de-vs".
(set_compliance_option): Set "de-vs".
* g10/misc.c (compliance_option_string): Return a description string.
(compliance_failure): Ditto.
* g10/keygen.c (ask_algo): Take care of CO_DE_VS.
(get_keysize_range): Ditto.
(ask_curve): Add new field to CURVES and trun flags into bit flags.
Allow only Brainpool curves in CO_DE_VS mode.
--
As of now this compliance mode only restricts the set of algorithms
and curves which can be created.
Signed-off-by: Werner Koch <wk@gnupg.org>
--
kbnode_t has replaced KBNODE for new code years ago, but that should
be documented. No bulk changes please to keep git blame easy to read.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/free-packet.c (free_attributes): Turn function into a nop for a
NULL arg.
(free_user_id): Ditto.
(free_compressed): Ditto.
(free_encrypted): Ditto.
(free_plaintext): Ditto.
(release_public_key_parts): Avoid extra check for NULL.
* g10/getkey.c (get_best_pubkey_byname): Ditto.
--
This change avoid surprises because it is common that function named
like free and taking a pointer also have similar semantics.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/getkey.c (get_best_pubkey_byname): Use the node returned by
'getkey_next' instead of doing another lookup.
Signed-off-by: Justus Winter <justus@g10code.com>
* g10/getkey.c (finish_lookup): Clarify that we do not return a
reference.
(lookup): Clarify the relation between RET_KEYBLOCK and RET_FOUND_KEY.
Check arguments. Actually release the node if it is not returned.
Signed-off-by: Justus Winter <justus@g10code.com>
* g10/getkey.c (getkey_next): Fix invocation of 'lookup'. If we want
to use RET_FOUND_KEY, RET_KEYBLOCK must be valid.
Fixes-commit: 8ea72a776a88f3c851e812d258355be80caa1bc1
Signed-off-by: Justus Winter <justus@g10code.com>
* g10/getkey.c (pubkey_cmp): Make a copy of the user id.
(get_best_pubkey_byname): Free the user ids.
Signed-off-by: Justus Winter <justus@g10code.com>
* g10/gpg.c (oCompliance): New.
(opts): Add "--compliance".
(parse_tofu_policy): Use a generic description string for "help".
(parse_compliance_option): New.
(main): Add option oCompliance. Factor out code for compliance
setting to ...
(set_compliance_option): new.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/keydb.c (keydb_add_resource): Make ANY_REGISTERED
file-global. Write a STATUS_ERROR.
(maybe_create_keyring_or_box): Check for non-accessible but existant
file.
(keydb_search): Write a STATUS_ERROR if no keyring has been registered
but continue to return NOT_FOUND.
* sm/keydb.c (keydb_add_resource): Rename ANY_PUBLIC to ANY_REGISTERED
and make file-global. Write a STATUS_ERROR.
(keydb_search): Write a STATUS_ERROR if no keyring has been registered
but continue to return NOT_FOUND. Also add new arg CTRL and change
all callers to pass it down.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/keygen.c (parse_algo_usage_expire): Use a different error
message for an unknown algorithm name.
--
GnuPG-bug-id: 2832
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/getkey.c (get_best_pubkey_byname): If 'get_pubkey_byname' does
not return a getkey context, then it can return at most one key,
therefore there is nothing to rank. Also, always initialize '*retctx'
to be on the safe side.
GnuPG-bug-id: 2828
Fixes: ab89164be02012f1bf159c971853b8610e966301
Signed-off-by: Justus Winter <justus@g10code.com>
--
This fixes a few
if (foo)
/* A comment
with several
lines. */
{
}
Which has the problem that the block is visually not related to the
"if" and might thus falsely be considered a standalone block.
Also adds a asterisk on the left side of longer comments.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/call-pinentry.c (start_pinentry): Get flavor and version and
pass it to agent_inq_pinentry_launched.
* agent/command.c (agent_inq_pinentry_launched): Add arg EXTRA.
* g10/server.c (gpg_proxy_pinentry_notify): Print a new diagnostic.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/getkey.c (struct pubkey_cmp_cookie): New type.
(key_is_ok, uid_is_ok, subkey_is_ok): New functions.
(pubkey_cmp): Likewise.
(get_best_pubkey_byname): Likewise.
* g10/keydb.h (get_best_pubkey_byname): New prototype.
* g10/keylist.c (locate_one): Use the new function.
* g10/pkclist.c (find_and_check_key): Likewise.
* tests/openpgp/Makefile.am (XTESTS): Add new test.
(TEST_FILES): Add new files.
* tests/openpgp/key-selection.scm: New file.
* tests/openpgp/key-selection/0.asc: Likewise.
* tests/openpgp/key-selection/1.asc: Likewise.
* tests/openpgp/key-selection/2.asc: Likewise.
* tests/openpgp/key-selection/3.asc: Likewise.
* tests/openpgp/key-selection/4.asc: Likewise.
--
When a name resembling a mail address is given to either --locate-keys
or --recipient, rank the search results and use only the most relevant
key.
This also lets us query which key will be used for encryption using
--locate-keys. However, note that --locate-keys may also return keys
incapable of encryption, though it will prefer keys that have an
encryption subkey.
GnuPG-bug-id: 2359
Signed-off-by: Justus Winter <justus@g10code.com>
* common/stringhelp.c: Include limits.h.
(compare_version_strings): Change semantics to behave like strcmp.
Include the patch lebel in the comparison. Allow checking a single
version string.
* common/t-stringhelp.c (test_compare_version_strings): Adjust test
vectors and a few new vectors.
* g10/call-agent.c (warn_version_mismatch): Adjust to new sematics.
* g10/call-dirmngr.c (warn_version_mismatch): Ditto.
* sm/call-agent.c (warn_version_mismatch): Ditto.
* sm/call-dirmngr.c (warn_version_mismatch): Ditto.
Signed-off-by: Werner Koch <wk@gnupg.org>
There were several different variant spellings of "passphrase". This
should fix them all for all English text.
I did notice that po/it.po contains multiple instances of
"passhprase", which also looks suspect to me, but i do not know
Italian, so i did not try to correct it.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
* g10/tofu.c: Include <time.h>, <utime.h>, <fcntl.h> and <unistd.h>.
(tofu_dbs_s): Add fields want_lock_file and want_lock_file_ctime.
(begin_transaction): Only yield if DBS->WANT_LOCK_FILE_CTIME has
changed since we took the lock. Don't use gpgrt_yield to yield, but
sleep for 100ms. After taking the batch lock, update
DBS->WANT_LOCK_FILE_CTIME. Also take the batch lock the first time we
take the real lock. When taking the real lock, use immediate not
deferred mode to avoid gratuitious aborts.
(end_transaction): When dropping the outermost real lock, drop the
batch lock.
(busy_handler): New function.
(opendbs): Set the busy handler to it when opening the DB. Initialize
CTRL->TOFU.DBS->WANT_LOCK_FILE.
(tofu_closedbs): Free DBS->WANT_LOCK_FILE.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
By default, SQLite defers transactions until they are actually needed.
A consequence of this is that if we have two readers and both decide
to do a write, then one has to abort. To avoid this problem, we can
make the outermost transaction an immediate transaction. This has the
disadvantage that we only allow a single reader at a time, but at
least we don't have gratuitous aborts anymore.
A second problem is that SQLite apparently doesn't actually create a
queue of waiters. The result is that doing a sched_yield between
dropping and retaking the batch transaction is not enough to allow the
other process to make progress. Instead, we need to wait a
while (emperically: 100ms seems reasonable). To avoid waiting when
there is no contention, we use a new file's timestamp to signal that
there is a waiter.
* g10/tofu.c (build_conflict_set): If CONFLICT_SET contains a single
element, don't bother to check for cross sigs. Add parameter PK.
Update callers.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* g10/build-packet.c (build_sig_subpkt_from_sig): Always write the new
Issuer Fingerprint sub-packet.
* g10/mainproc.c (check_sig_and_print): Always consider that
sub-packet.
--
The specs for this sub-packet have been pushed to the OpenPGP WG's
repo today.
See-also: https://mailarchive.ietf.org/arch/msg/\
openpgp/GvPo2eSL9GW9WcGhOocY7KBa9FY
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/call-dirmngr.c (gpg_dirmngr_ks_get): Add arg 'quick'.
(gpg_dirmngr_wkd_get): Ditto.
* g10/keyserver.c (keyserver_get): Add arg 'quick'.
(keyserver_get_chunk): Add arg 'quick'.
(keyserver_import_fprint): Ditto. Change callers to pass 0 for it.
(keyserver_import_keyid): Ditto.
(keyserver_import_wkd): Ditto.
* g10/mainproc.c (check_sig_and_print): Call the 3 fucntions with
QUICK set.
--
Note that this option has not yet been implemented by dirmngr.
Dirmngr will simply ignore it for now.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/ecdh.c (pk_ecdh_encrypt_with_shared_point): Returns error when
it's short. Clarify the format. Handle other prefixes correctly.
--
With the scdaemon's change, there is no case NBYTES < SECRET_X_SIZE.
This fixes the break of ECDH with X25519.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/ecdh.c (pk_ecdh_encrypt_with_shared_point): Improve handling of
ECDH shared point format.
--
This handles the case where the result comes from scdaemon.
Signed-off-by: Arnaud Fontaine <arnaud.fontaine at ssi.gouv.fr>
* scd/app-openpgp.c (send_key_attr): Use curve instead of OID.
(ecdh_params): New.
(ecc_read_pubkey): Use ecdh_params. Use curve name.
(ecc_writekey): Likewise.
(ecc_curve): Rename from ecc_oid.
(parse_algorithm_attribute): Use ecc_curve.
* g10/call-agent.c (learn_status_cb): Use openpgp_is_curve_supported to
intern the curve name string.
* g10/card-util.c (card_status): Conver curve name to alias for print.
--
Now, sdcaemon answer for KEY-ATTR is in the canonical curve name
instead of the alias. Since it is used of key generation for
card encryption key with backup, it should be canonical name.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/keygen.c (gen_card_key): Add back ALGO as the second argument.
Don't get ALGO by KEY-ATTR by this function. It's caller to provide
ALGO. Don't do that by both of caller and callee.
(generate_keypair): Only put paramerters needed. Use parameters
for ALGO to call gen_card_key.
(generate_card_subkeypair): Get ALGO and call gen_card_key with it.
--
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/keydb.c (maybe_create_keyring_or_box): Open in binary mode.
--
This fixes keybox corruption on windows.
Signed-off-by: Andre Heinecke <aheinecke@intevation.de>
* g10/keygen.c (generate_keypair): For card key generation, fill
parameters by KEY-ATTR.
* scd/app-openpgp.c (ecc_read_pubkey): OID should be freed at last,
after its reference by OIDBUF is finished.
(ecc_writekey): Likewise.
--
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
