* sm/gpgsm.c (oKeyServer_deprecated): New.
(opts): Assign "ldapserver" to the new option and mark it as obsolete.
--
We want to use "ldapserver" in dirmngr but need to keep using
"keyserver" in gpgsm for existant versions of Kleopatra etc.
GnuPG-bug-id: 5801
* dirmngr/dirmngr.c: Reorder two option groups.
* tools/gpgconf-comp.c (known_options_gpgsm): Rename "keyserver" to
"ldapserver" and set level to invisible.
(known_options_dirmngr): Add "ldapserver" at the basic level.
* sm/gpgsm.c (opts): No more help text for "ldapserver".
--
GnuPG-bug-id: 5800
* sm/certdump.c (parse_dn_part): Translate OID to "Callsign"
* sm/keylist.c (oidtranstbl): Some more OIDs.
--
This is Ham thingy to make it easier to read LotW certificates.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (pwri_parse_pbkdf2): Use int for digest algo.
(pwri_decrypt): Use int for cipher algo and digest algo.
--
Fixes-commit: 02029f9eab
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* sm/gpgsm.c (opts): Add option --ldapserver and make --keyserver an
alias.
--
We should use "keyserver" for OpenPGP and thus it is better to allow
for "ldapserver" here - it is the same convention as now used in
dirmngr.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit d6df1bf849)
* common/keyserver.h: Remove.
* sm/gpgsm.h (struct keyserver_spec): Remove.
(opt): Change keyserver to a strlist_t.
* sm/gpgsm.c (keyserver_list_free): Remove.
(parse_keyserver_line): Remove.
(main): Store keyserver in an strlist.
* sm/call-dirmngr.c (prepare_dirmngr): Adjust for the strlist. Avoid
an ambiguity in dirmngr by adding a prefix if needed.
* g10/options.h (struct keyserver_spec): Move definition from
keyserver.h to here. Remove most fields.
* g10/keyserver.c (free_keyserver_spec): Adjust.
(cmp_keyserver_spec): Adjust.
(parse_keyserver_uri): Simplify.
(keyidlist): Remove fakev3 arg which does not make any sense because
we don't even support v3 keys.
--
We now rely on the dirmngr to parse the keyserver specs. Thus a bad
specification will not be caught immediately. However, even before
that dirmngr had stricter tests.
Signed-off-by: Werner Koch <wk@gnupg.org>
Ported-from: 9f586700ec
* sm/keydb.c (struct keydb_local_s): Add field saved_search_result.
(keydb_push_found_state): Implement for keyboxd.
(keydb_pop_found_state): Ditto.
(keydb_get_cert): Do not release the cert so that the function can be
used again to get the same cert. This is the same behaviour as in
pubring.kbx mode.
* sm/certchain.c, sm/import.c: Improve some error messages.
Signed-off-by: Werner Koch <wk@gnupg.org>
* tests/cms/samplemsgs/: Add sample messages.
* sm/gpgsm.c (main): Use gpgrt_fcancel on decryption error.
* sm/decrypt.c (decrypt_gcm_filter): New.
(gpgsm_decrypt): Use this filter if requested. Check authtag.
--
Note that the sample message pwri-sample.gcm.p7m is broken: The
authtag is duplicated to the authEncryptedContentInfo. I used a
temporary code during testing hack to that test message out.
* sm/keylist.c (list_cert_raw): Print the OpenPGP fpr.
--
This is useful for debugging for example if an OpenPGP key is used to
create an X.509 cert.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/keydb.h (GETPASSWORD_FLAG_SYMDECRYPT): New.
(passphrase_to_dek_ext): Remove this obsolete prototype.
* g10/passphrase.c (passphrase_get): Add arg flags. Use new flag
value.
(passphrase_to_dek): Add arg flags and pass it on.
* g10/mainproc.c (proc_symkey_enc): Use new flag.
* sm/decrypt.c (pwri_decrypt): Use "passphrase".
--
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 03f83bcda5)
Note that we keep on using the term "passphrase" although "password"
would be better. There are however so many occurance of this and
given it is a bike shedding topic we fix that in the PO files.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (pwri_decrypt): Add arg ctrl. Ask for passphrase.
* sm/export.c (export_p12): Mark string as translatable.
* sm/import.c (parse_p12): Ditto.
--
This is finishes the support for PWRI.
The N_() marks are added so that we don't rely of the side-effect of
having the same strings in protect-tool.c
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (pwri_parse_pbkdf2): New.
(pwri_decrypt): New.
(prepare_decryption): Support pwri.
(gpgsm_decrypt): Test for PWRI. Move IS_DE_VS flag to DFPARM.
--
Note that this is not finished because we need to implement a password
callback. For now "abc" is used as passwort.
Latest libksba is also required to return the required info.
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/kbxserver.c (cmd_search): Use the openpgp hack for calling
classify_user_id.
* kbx/backend-sqlite.c (run_select_statement): Remove angle brackets
in exact addrspec mode.
* g10/call-keyboxd.c (keydb_search): Do not duplicate the left angle
bracket.
* sm/keydb.c (keydb_search): Ditto.
--
Note that the openpgp hack flag of classify_user_id is actually a
misnomer because we actually hack a round a problem in gpgsm. And it
is only over there that we don't set it there. In keyboxd the flag
should be set. And we need to remove the angle brackets of course
because that is how we create the addrspec column values.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/gpg.c (main): Don't use the default log file from common.conf.
* sm/gpgsm.c (main): Ditto.
--
That was acutally not intended and contradicts the description in
doc/example/common.conf.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/keylist.c (do_show_certs): New.
(gpgsm_show_certs): New.
* sm/gpgsm.c (aShowCerts): New.
(opts): Add --show-certs.
(main): Call gpgsm_show_certs.
--
I have been using libksba test programs for countless times to look at
certificates and I always wanted to add such a feature to gpgsm. This
is simply much more convenient.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/comopt.c, common/comopt.h: New.
* common/Makefile.am: Add them.
* g10/gpg.c: Include comopt.h.
(main): Also parse common.conf.
* sm/gpgsm.c: Include comopt.h.
(main): Set a flag for the --no-logfile option. Parse common.conf.
* tools/gpgconf-comp.c (known_options_gpg): Remove "use-keyboxd", add
pseudo option "use_keyboxd".
(known_pseudo_options_gpg): Add pseudo option "use_keyboxd".
(known_options_gpgsm): Remove "use-keyboxd".
* tests/openpgp/defs.scm (create-gpghome): Create common.conf.
* doc/examples/common.conf: New.
--
Note that --use-keybox still works but prints a warning. We will
eventually remove this option becuase it was marked as an experimental
feature anyway.
It would be too confusing if gpg and gpgsm use different key storages.
Further, other components (e.g. dirmngr or gpg-wks-client) which call
gpg or gpgsm need to be aware that the keyboxd is used and pass that
option on the command line. Now that common.conf is always read (even
if --no-options is used) those tools will work instantly.
* sm/gpgsm.c: Include npth.h.
(main): Init nPth.
--
Actually we currently do not use extra threads but we need to link to
the npth version of the common functions because there is an option to
use threads. Now, without initialized npth, we could experience a
crash in gnupg_sleep, which uses npth_usleep in the npth based common
functions.
Fixes-commit: 046f419f80
* sm/certchain.c (find_up_dirmngr): Print one diagnostic only in
verbose mode. Do not print issuer not found diags in quiet mode.
* sm/minip12.c (parse_bag_data): Add missing verbose condition.
--
GnuPG-bug-id: 4757
* sm/certchain.c (find_up): Get rid of the legacy return code -1 and
chnage var name rc to err.
(gpgsm_walk_cert_chain): Change var name rc to err.
(do_validate_chain): Get rid of the legacy return code -1.
--
This was detected while fixing
GnuPG-bug-id: 4757
* sm/encrypt.c (gpgsm_encrypt): Take care of --quiet.
* sm/gpgsm.c: Include minip12.h.
(set_debug): Call p12_set_verbosity.
* sm/import.c (parse_p12): Dump keygrip only in debug mode.
* sm/minip12.c (opt_verbose, p12_set_verbosity): New.
(parse_bag_encrypted_data): Print info messages only in verbose mode.
--
GnuPG-bug-id: 4757
* configure.ac (NEED_GPG_ERROR_VERSION): Rename to NEED_GPGRT_VERSION
and set to 1.41.
* common/sysutils.c (gnupg_access): Remove code for older gpgrt
versions.
* kbx/backend-sqlite.c: Ditto.
* sm/gpgsm.c (main): Ditto.
--
We already have a requirement for a newer Libgcrypt and thus we can
also require a more recent libgpgrt (aka libgpg-error) which was
released before Libgcrypt.
* sm/keylist.c (OID_FLAG_KP): New.
(key_purpose_map): Merge into ...
(oidtranstbl): this.
(get_oid_desc): New arg 'matchflag'. Use function in place of direct
access to key_purpose_map.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/compliance.c (gnupg_gcrypt_is_compliant): New.
(gnupg_rng_is_compliant): Also check library version.
* g10/mainproc.c (proc_encrypted): Use new function.
(check_sig_and_print): Ditto.
* sm/decrypt.c (gpgsm_decrypt): Ditto.
* sm/encrypt.c (gpgsm_encrypt): Ditto.
* sm/verify.c (gpgsm_verify): Ditto
--
This will eventually allow us to declare Libgcrypt 1.9 to be de-vs
compliant. GnuPG can use this information then for its own checks.
As of now GnuPG tests the version of the used library but that is a
bit cumbersome to maintain.
Signed-off-by: Werner Koch <wk@gnupg.org>
* configure.ac: Require at least Libgcrypt 1.9.0. Remove all
GCRYPT_VERSION_NUMBER dependent code.
--
Only Libgcrypt 1.9 implements EAX which is a mandatory algorithm in
RFC4880bis.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/openfile.c (try_make_homedir): Move core of the code to ...
* common/homedir.c (gnupg_maybe_make_homedir): new.
* sm/keydb.c (try_make_homedir): Implement using new function.
* common/homedir.c: Include i18n.h.
* po/POTFILES.in: Add common/homedir.c.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/sysutils.c (any8bitchar) [W32]: New.
(gnupg_open): New. Replace most calls to open by this.
* common/iobuf.c (any8bitchar) [W32]: New.
(direct_open) [W32]: Use CreateFileW if needed.
--
This is yet another step for full Unicode support on Windows.
GnuPG-bug-id: 5098
--
We need to use es_fopen on Windows to cope with non-ascii file names.
This is quite a large but fortunately straightforward change. At a
very few places we keep using stdio (for example due to the use of
popen).
GnuPG-bug-id: 5098
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/sysutils.c (gnupg_access): New. Replace all calls to access
by this wrapper.
* common/homedir.c (w32_shgetfolderpath): Change to return UTF-8
directory name.
(standard_homedir): Adjust for change.
(w32_commondir, gnupg_cachedir): Ditto.
--
Also use SHGetFolderPathW instead of SHGetFolderPathA on Windows.
This is required to correctly handle non-ascii filenames on Windows.
GnuPG-bug-id: 5098
* kbx/kbx-client-util.c (kbx_client_data_new): Add arg 'dlines'.
* g10/call-keyboxd.c (open_context): Set DLINES to true.
* sm/keydb.c (open_context): Ditto.
--
This allows to compile time switch between the D-line and the
fd-passing data communication between gpg/gpgsm and keyboxd. A quick
test with about 3000 OpenPGP keys showed that D-lines are only 10%
slower than the fd-passing based implementation. Given that the
thread adds extra complexity we go for now with the D-line approach.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certdump.c (parse_dn_part): Fix parser flaw.
--
This could in theory result in reading bytes after a after Nul in a
string and thus possible segv on unallocated memory or reading other
parts of the memory. However, it is harmless because the rfc2253
strings have been constructed by libksba.
GnuPG-bug-id: 5037
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/call-agent.c (gpgsm_agent_genkey): Pass --timestamp option.
(gpgsm_agent_import_key): Ditto.
* g10/call-agent.c (agent_genkey): Add arg timestamp and pass it on.
(agent_import_key): Ditto.
* g10/import.c (transfer_secret_keys): Pass the creation date to the
agent.
* g10/keygen.c (common_gen): Ditto.
--
Having the creation time in the private key file makes it a lot easier
to re-create an OpenPGP public keyblock in case it was accidentally
lost.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/openpgp-oid.c (map_openpgp_pk_to_gcry): Use cast for enum
conversion.
* dirmngr/dns-stuff.c (get_dns_srv): Use explicit conversion from
int to float.
* sm/gpgsm.c (parse_keyserver_line): Initialize ERR.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* g10/gpg.c (oChUid): New.
(opts): Add --chuid.
(main): Implement --chuid. Delay setting of homedir until the new
chuid is done.
* sm/gpgsm.c (main): Delay setting of homedir until the new chuid is
done.
* tools/gpg-card.c (oChUid): New.
(opts): Add --chuid.
(changeuser): New helper var.
(main): Implement --chuid.
* tools/gpg-connect-agent.c (oChUid): New.
(opts): Add --chuid.
(main): Implement --chuid.
--
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.c (oChUid, opts): New option --chuid.
(main): Implement option.
--
This option will at least be useful for Scute.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/keylist.c (list_cert_colon): Emit a new "fp2" record.
(list_cert_raw): Print the SHA2 fingerprint.
(list_cert_std): Ditto.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/genkey.c (agent_ask_new_passphrase): No qualitybar.
* g10/call-agent.c (agent_get_passphrase): Ditto.
* sm/call-agent.c (gpgsm_agent_ask_passphrase): Ditto.
--
The concept of a passphrase quality indicator is anyway questionable
because user are smart enough to trick them out and they also tend to
limit the actually used entropy.
Except for the red/green switching (to show whether constraints are
fulfilled) our qualitybar is pretty bad and thus worse than none.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/compliance.h (PK_ALGO_FLAG_RSAPSS): New.
* common/compliance.c (gnupg_pk_is_compliant): Add arg alog_flags and
test rsaPSS. Adjust all callers.
(gnupg_pk_is_allowed): Ditto.
* sm/misc.c (gpgsm_ksba_cms_get_sig_val): New wrapper function.
(gpgsm_get_hash_algo_from_sigval): New.
* sm/certcheck.c (gpgsm_check_cms_signature): Change type of sigval
arg. Add arg pkalgoflags. Use the PK_ALGO_FLAG_RSAPSS.
* sm/verify.c (gpgsm_verify): Use the new wrapper and new fucntion to
also get the algo flags. Pass algo flags along.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.c (main) <aSign,aEncr>: Uses gpgrt_fcancel on error.
--
When creating a signature or encrypting and the respective key is not
available or the user canceled the PIN entry gpgsm prints the initial
part of the message due to internal buffering in gpgrt. By using
gpgrt_fcancel we can avoid this at least as long as the data is less
than the standard buffer size (which is currently 8k). If is not a
complete solution but the best we can do easily. Outputting to the
tty is anyway more of a testing aid than for real use.
This makes use of the new gpgrt_fcancel API.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certdump.c: Include membuf.h.
(gpgsm_print_serial_decimal): New.
* sm/keylist.c (list_cert_raw): Print s/n also in decimal
(list_cert_std): Ditto.
--
Many CA's print the serial number in decimal on their cards.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certcheck.c (do_encode_md): Take care of nistp521.
--
That curve is a bit odd in that it does not match a common hash digest
length. We fix that here for just this case instead of writing more
general code to support all allowed cases (i.e. hash shorter than Q).
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/misc.c (transform_sigval): Support EdDSA.
* sm/certreqgen.c (create_request): Support EdDSA cert creation.
* sm/certcheck.c (gpgsm_check_cert_sig): Map some ECC algo OIDs to
hash algos.
* sm/call-agent.c (struct sethash_inq_parm_s): New.
(sethash_inq_cb): New.
(gpgsm_agent_pksign): Add mode to pass plain data for EdDSA.
--
Tested using a parameter file
Key-Type: EdDSA
Key-Length: 1024
Key-Grip: 09D9AE3D494F7888C93BE5106AD8A734A87617F0
Key-Usage: sign
Serial: random
Name-DN: CN=dummy test ed25519
where the keygrip is from a gpg generated Ed25519 key. ECDSA was
tested using
Key-Type: ECDSA
Key-Length: 1024
Key-Grip: 8E06A180EFFE4C65B812150CAF19BF30C0689A4C
Key-Usage: sign
Serial: random
Name-DN: CN=dummy test nistp256
and RSA using
Key-Type: RSA
Key-Length: 2048
Key-Grip: C6A6390E9388CDBAD71EAEA698233FE5E04F001E
Key-Usage: sign
Serial: random
Name-DN: CN=dummy test rsa
The command used in all cases is
gpgsm -v --gen-key --batch a.parm >a.crt
gpgsm -v --import <a.crt
More support, in particular in the user interface, is required and
will follow soon.
GnuPG-bug-id: 4888
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certdump.c (gpgsm_get_serial): New.
* sm/certcheck.c (gpgsm_check_cert_sig): Support EdDSA signatures.
--
Note that this does not work with the self-signed RFC-8410 sample
certificate; see the code for comments.
The Ed488 case has not been tested due to a lack of support in
Libgcrypt.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.h (struct certlist_s): Add helper field pk_algo.
* sm/sign.c (gpgsm_sign): Store the public key algo. Take the hash
algo from the curve. Improve diagnostic output in verbose mode.
--
GnuPG-bug-id: 4098
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (ecdh_decrypt): Support
dhSinglePass-stdDH-sha1kdf-scheme. Factor key derive code out to ...
(ecdh_derive_kek): new global function. Allow for hashs shorter than
the key.
(hash_ecc_cms_shared_info): Make file-only.
* sm/encrypt.c (ecdh_encrypt): Replace derive code by a call to the
new ecdh_derive_kek. Add test code to create data using
dhSinglePass-stdDH-sha1kdf-scheme.
* sm/gpgsm.h (opt): Add member force_ecdh_sha1kdf.
* sm/gpgsm.c: Add option --debug-force-ecdh-sha1kdf.
--
I did some test against
Governikus_Signer: 2.9.2.0
MCard security Provider: 2.4.0
Algorithm catalog from: 05.12.2019
using a Signature Card v2.0 and a nistp256 certificate. Encrypting
with Governikus used the stdDH-sha1kdf scheme which we can now
decrypt. Encrypting with GPGSM uses for that curve the recommended
scheme sha256kdf but Governikus was not able to decrypt this (no
usable error message). Encrypting using stdDH-sha1kdf with GPGSM by
using the new --debug-force-ecdh-sha1kdf option showed that
Governikus was able to decrypt this.
FWIW: RFC5753 (Use of Elliptic Curve Cryptography (ECC) Algorithms)
has this requirement:
Implementations that support EnvelopedData with the
ephemeral-static ECDH standard primitive:
- MUST support the dhSinglePass-stdDH-sha256kdf-scheme key
agreement algorithm, the id-aes128-wrap key wrap algorithm, and
the id-aes128-cbc content encryption algorithm; and
which Governikus seems not to fulfill.
GnuPG-bug-id: 4098
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/sexputil.c (cipher_mode_to_string): New.
* sm/decrypt.c (prepare_decryption): Show cipher algo and mode.
(gpgsm_decrypt): Show key algo and fingerprint
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/sexputil.c (pubkey_algo_to_string): New.
* sm/certcheck.c (do_encode_md): Replace GCRY_PK_ECDSA by GCRY_PK_ECC.
* sm/certreqgen-ui.c (check_keygrip): Add all ECC algorithms.
* sm/gpgsm.c (our_pk_test_algo): Also allow EdDSA.
* sm/verify.c (gpgsm_verify): Map ECC algo to ECDSA. Use new pubkey
algo name function
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/verify.c (gpgsm_verify): Print the used algorithms.
--
Note that we now use the full fingerprint instead of the certificate
id. This better aligns with what we do in gpg.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/fingerprint.c (gpgsm_get_key_algo_info): Factor code out to ...
(gpgsm_get_key_algo_info2): new.
(gpgsm_pubkey_algo_string): New.
* sm/keylist.c (list_cert_colon): Put curve into field 17
(list_cert_raw): Print the unified key algotithm string instead of the
algo and size.
(list_cert_std): Ditto.
--
It is important to known whether a 256 bit ECC uses a NIST or a
Brainpool curve.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (ecdh_decrypt): Add arg nbits and detect bare secret.
(prepare_decryption): Add arg nbits and pass on.
(gpgsm_decrypt): Pass size of curve to prepare_decryption.
--
GnuPG-bug-id: 4098
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/decrypt.c (hash_ecc_cms_shared_info): Make global.
* sm/encrypt.c (ecdh_encrypt): New.
(encrypt_dek): Add arg PK_ALGO and support ECDH.
(gpgsm_encrypt): Pass PK_ALGO.
--
Note: This has only been tested with a messages created and decrypted
by GnuPG.
GnuPG-bug-id: 4098
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/minip12.c [TEST]: Remove test code. Include util.h, tlv.h. and
openpgpdefs.h. Remove the class and tag constants and replace them by
those from tlv.h.
(builder_add_oid, builder_add_mpi): New.
(build_key_sequence): Rename to ...
(build_rsa_key_sequence): this.
(build_ecc_key_sequence): New.
(p12_build): Call RSA or ECC builder.
(p12_raw_build): Ditto.
* sm/export.c (gpgsm_p12_export): Use correct armor header for ECC.
(sexp_to_kparms): Support ECC.
* sm/t-minip12.c: New to replace the former TEST code in minip12.h.
--
GnuPG-bug-id: 4921
* sm/decrypt.c: Include tlv.h.
(string_from_gcry_buffer): New.
(hash_ecc_cms_shared_info): New.
(ecdh_decrypt): New.
(prepare_decryption): Support ECDH. Add arg pk_algo.
(gpgsm_decrypt): Lift some variables from an inner code block.
--
Note: This has only been tested with a single messages created by
OpenSSL and taken from the Mozilla bug tracker. In particular the
code to included UserKeyingMaterial (ukm) has not been tested.
GnuPG-bug-id: 4098
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certchain.c (find_up): Disable external lookups in offline mode.
Always allow AKI lookup if CRLs are also enabled.
--
GnuPG-bug-id: 4898
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/call-dirmngr.c (gpgsm_dirmngr_lookup): Add optional arg URL and
adjust all callers.
* sm/certchain.c (oidstr_caIssuers): New.
(struct find_up_store_certs_s): Add additional fields.
(find_up_store_certs_cb): Store the fingerprint.
(find_up_via_auth_info_access): New.
(find_up): Try the AIA URI first.
--
Note that --auto-issuer-key-retrieve is required to use that.
GnuPG-bug-id: 4898
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/validate.c (hash_algo_from_buffer): New.
(uint_from_buffer): New.
(check_cert_sig): Support rsaPSS.
* sm/certcheck.c (gpgsm_check_cert_sig): Fix small memory leak on
error.
--
Yes, I know that there is a lot of code duplication. In fact some of
the code is ugly and it would be better if we enhance Libgcrypt to
guarantee that returned memory buffers via gcry_sexp_extract_param are
allways Nul terminated and we should also enhance that function to
directly extract into an unsigned int or char *.
GnuPG-bug-id: 4538
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certdump.c (format_name_writer): Take care of a flush request.
--
We won't see a flush here so that fix does not solve a real bug but we
want to be correct. Note that this function seems to predate the
es_fopenmem function and thus in master (2.3) we should replace it
entirely.
GnuPG-bug-id: 4495
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from 2.2 commit 8d0d61aca3)
* sm/certchain.c (is_cert_still_valid): Shortcut if tehre is no DP.
* common/audit.c (proc_type_verify): Print "n/a" if a cert has no
distribution point.
* sm/gpgsm.h (opt): Add field enable_issuer_based_crl_check.
* sm/gpgsm.c (oEnableIssuerBasedCRLCheck): New.
(opts): Add option --enable-issuer-based-crl-check.
(main): Set option.
--
If the issuer does not provide a DP and the user wants such an issuer,
we expect that a certificate does not need revocation checks. The new
option --enable-issuer-based-crl-check can be used to revert to the
old behaviour which requires that a suitable LDAP server has been
configured to lookup a CRL by issuer.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certreqgen.c (proc_parameters): Remove checking GCRY_PK_RSA.
--
This is an initial change to support ECC key generation.
GnuPG-bug-id: 4888
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* common/gc-opt-flags.h (GC_OPT_FLAG_RUNTIME): Move to ...
* tools/gpgconf-comp.c: here.
(known_options_scdaemon): Remove "options".
(known_options_dirmngr): Remove "options".
(known_options_gpgsm): Remove "options".
(known_options_gpg): Remove "options" and "keyserver".
(struct gc_option_s): Rename active t gpgconf_list.
(gc_component_list_options): Do not act upon active.
(option_check_validity): Ditto.
(is_known_option): Make it work correctly for unknown options.
(retrieve_options_from_program): Use renamed flag gpgconf_list only to
detect duplicated items from --gpgconf-list. Do not set runtime.
Only e set the options if set by --gpgconf-list; never clear them.
* agent/gpg-agent.c: Simplify the --gpgconf-list output.
* dirmngr/dirmngr.c: Ditto.
* g10/gpg.c: Ditto.
* kbx/keyboxd.c: Ditto.
* scd/scdaemon.c: Ditto.
* sm/gpgsm.c: Ditto.
* tests/openpgp/gpgconf.scm: Use "compliance" instead of "keyserver"
for the string arg test.
--
There is no need to read the list of options from the components
unless they convey a default value. It is better to consult only the
list we have in gpgconf-comp.c to decide on whether an option should
be displayed. Right, this might mess up thing if a newer gpgconf
version is used with an older component, but we already print warnings
in this case and in general we do not want to support this anymore -
the times of gpg 1.4. and 2.0 are long over now.
GnuPG-bug-id: 4788
Signed-off-by: Werner Koch <wk@gnupg.org>
* tools/gpgconf-comp.c: Remove all regular option descriptions. They
are now read in from the component. Also remove a few meanwhile
obsolete options.
* agent/gpg-agent.c: Add option description which were only set in
gpgconf-comp.c.
* dirmngr/dirmngr.c: Ditto.
* scd/scdaemon.c: Ditto.
* sm/gpgsm.c: Ditto.
* g10/gpg.c: Ditto.
--
This second part removes all regular option descriptions because they
can be read from the components. A few were missing in the components
and thus moved to there.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/gpgv.c: Use new option parser.
* sm/gpgsm.c: Ditto.
* agent/gpg-agent.c: Ditto.
(opts): Add option --no-options.
--
This is the next part of changes. The latest libgpg-error is required
so that that re-reading options (SIGHUP) works.
GnuPG-bug-id: 4788
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/util.h: Remove argparse.h.
* common/argparse.c: Undef GPGRT_ENABLE_ARGPARSE_MACROS.
* configure.ac (GPGRT_ENABLE_ARGPARSE_MACROS): Define.
* agent/gpg-agent.c: Undef GPGRT_ENABLE_ARGPARSE_MACROS and include
argparse.h. Do this also for all main modules which use our option
parser except for gpg. Replace calls to strusage by calls to
gpgrt_strusage everywhere.
* g10/gpg.c (opts): Change type to gpgrt_opt_t. Flag oOptions and
oNoOptions with ARGPARSE_conffile and ARGPARSE_no_conffile.
(main): Change type of pargs to gpgrt_argparse_t. Rework the option
parser to make use of the new gpgrt_argparser.
--
This is not yet finished but a make check works. gpg has the most
complex and oldest option handling and thus this is the first
migration target. SE-Linux checks and version-ed config files are
missing and will be added later.
GnuPG-bug-id: 4788
Signed-off-by: Werner Koch <wk@gnupg.org>
No functional changes, just fixing minor spelling issues.
---
Most of these were identified from the command line by running:
codespell \
--ignore-words-list fpr,stati,keyserver,keyservers,asign,cas,iff,ifset \
--skip '*.po,ChangeLog*,help.*.txt,*.jpg,*.eps,*.pdf,*.png,*.gpg,*.asc' \
doc g13 g10 kbx agent artwork scd tests tools am common dirmngr sm \
NEWS README README.maint TODO
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
* common/util.h (EXTERN_UNLESS_MAIN_MODULE): Add the definion only
here but now without the Norcroft-C. Change all other places where it
gets defined.
* common/iobuf.h (iobuf_debug_mode): Declare unconditionally as
extern.
* common/iobuf.c (iobuf_debug_mode): Define it here.
* agent/gpg-agent.c (INCLUDED_BY_MAIN_MODULE): Define here and also in
all main modules of all other programs.
* g10/main.h: Put util.h before the local header files.
--
This change is required for use with gcc/ld's LTO feature which does
not allow common blocks. Further gcc 10 will make -fno-common the
default and thus this chnage is always needed. What a pitty.
Co-authored-by: Tomáš Mráz
GnuPG-bug-id: 4831
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 21d9bd8b87)
- Applied respective chnages also to gpg-card and keyboxd.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/server.c (do_listkeys): Implement new option.
--
This option can be used by clients who can only provide a DER encoded
form of the issuer. For example in PKCS#11 providers.
Testing:
Put the DER encoded issuer DN into a file, say issuer.der.
The run
gpg-connect-agent -E -- gpgsm --server
> /definqfile ISSUER_DER issuer.der
> list-keys --issuer-der 01020304
and if the local keyring has a certifictate with that issuer and a s/n
of 0x01020304 that certificate will be listed.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/sysutils.c (inhibit_set_foregound_window): New var.
(gnupg_inhibit_set_foregound_window): New func.
(gnupg_allow_set_foregound_window): Use var.
* g10/gpg.c (main): Inhibit in loopback mode.
* sm/gpgsm.c (main): Ditto.
--
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.h (struct server_control_s): Add field 'current_time'.
* sm/certchain.c (find_up_search_by_keyid): Detect a corner case.
Also simplify by using ref-ed cert objects in place of an anyfound
var.
--
See the code for a description of the problem. Tested using the certs
from the bug report and various command lines
gpgsm --faked-system-time=XXXX --disable-crl-checks \
-ea -v --debug x509 -r 0x95599828
with XXXX being 20190230T000000 -> target cert too young
with XXXX being 20190330T000000 -> okay
with XXXX being 20190830T000000 -> okay, using the long term cert
with XXXX being 20220330T000000 -> target cert expired
The --disabled-crl-checks option is required because in our a simple
test setting dirmngr does not know about the faked time.
GnuPG-bug-id: 4696
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.h (struct keyserver_spec): Add field use_ldaps.
* sm/gpgsm.c (parse_keyserver_line): Parse flags.
* sm/call-dirmngr.c (prepare_dirmngr): Send ldaps flag to the dirmngr.
* dirmngr/dirmngr.h (struct ldap_server_s): Add field use_ldaps.
* dirmngr/ldapserver.c (ldapserver_parse_one): Parse flags.
* dirmngr/ldap.c (start_cert_fetch_ldap): Call wrapper with --tls.
* dirmngr/dirmngr_ldap.c: New option --tls.
(fetch_ldap): Make use of that option.
--
There was no way to specify an LDAPS server in
dirmngr_ldapserver.socnf or with gpgsm's --keyserver option. This
patch fixes this. Eventually we should allow to replace host and port
by a partial URI in the same way ldap_initialize does it. For backward
compatibility we do not yet do that.
Although the dirmngr code accepts an URL (eg. taken from a
certificate), I can't see how the scheme was ever used. Thus the
patch also detects an ldaps scheme and uses this. That part has not
been tested, though.
Signed-off-by: Werner Koch <wk@gnupg.org>
--
Resolved Conflicts:
* common/asshelp.c: Keep the new code in master for spawing under
Windows.
* g10/Makefile.am: Keep all new file.
* g10/photoid.c: Pass CTRL to pct_expando.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/keydb.c (keydb_add_resource): Call keybox_compress.
* sm/keydb.c (keydb_add_resource): Release the lock after a compress.
--
Note that in gpgsm we already did the compress run but we didn't
released the lock on the file. This might have been a reason for some
strange hangs.
GnuPG-bug-id: 4644
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/call-agent.c (agent_pkdecrypt): accept but do not require
NUL-terminated data from the agent.
* sm/call-agent.c (gpgsm_agent_pkdecrypt): accept but do not require
NUL-terminated data from the agent.
GnuPG-bug-id: 4652
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
* sm/call-agent.c (gpgsm_scd_pksign): Cast to integer for %b.
--
This fix is needed on big endian machine where size_t is bigger
than integer.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* sm/decrypt.c: Use TMP_RC for ksba_cms_get_issuer_serial,
and return the last error when no key is available.
Fix the error report with TMP_RC for second call of
ksba_cms_get_issuer_serial.
GnuPG-bug-id: 4561
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* sm/certlist.c (gpgsm_add_to_certlist): Add diagnostic and fold two
similar branches.
--
Without this patch gpgsm printed:
gpgsm[23045]: DBG: chan_6 <- RECIPIENT edward.tester@demo.gnupg.com
gpgsm[23045]: certificate is not usable for encryption
gpgsm[23045]: certificate is good
with this patch a
gpgsm[23045]: looking for another certificate
is inserted into the log.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_getinfo): Return GPG_ERR_FALSE as boolean False.
* g13/server.c (cmd_getinfo): Ditto.
* sm/server.c (cmd_getinfo): Ditto.
--
GPG_ERR_FALSE was introduced with libgpg-error 1.21 and we now require
a later version for gnupg 2. Thus we can switch to this more
descriptive code.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certlist.c (cert_usage_p): Add arg 'silent' and change all
callers.
(gpgsm_cert_use_sign_p): Add arg 'silent' and pass to cert_usage_p.
Change all callers.
* sm/sign.c (gpgsm_get_default_cert): Set SILENT when calling
gpgsm_cert_use_sign_p
--
GnuPG-bug-id: 4535
Signed-off-by: Werner Koch <wk@gnupg.org>
* kbx/keybox-init.c (keybox_lock): New arg TIMEOUT. Change all
callers to pass -1 when locking.
* sm/keydb.c (struct resource_item): Remove LOCKANDLE.
(struct keydb_handle): Add KEEP_LOCK.
(keydb_add_resource): Use keybox locking instead of a separate dotlock
for testing whether we can run a compress.
(keydb_release): Reset KEEP_LOCK.
(keydb_lock): Set KEEP_LOCK.
(unlock_all): Take care of KEEP_LOCK.
(lock_all): Use keybox_lock instead of dotlock fucntions.
(keydb_delete): Remove arg UNLOCK.
* sm/delete.c (delete_one): Adjust keydb_delete. Due to the KEEP_LOCK
the keydb_release takes care of unlocking.
--
This aligns the code more with g10/keydb.c and avoids the separate
calls to dotlock_take.
GnuPG-bug-id: 4505
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.c (opts): New options --authenticode and --attribute.
* sm/gpgsm.h (opt): Add vars authenticode and attribute_list.
* sm/sign.c (add_signed_attribute): New but inactive.
(gpgsm_sign): Use new options.
--
Because libksba 1.4 is not yet ready the new code is not yet active.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/call-agent.c (agent_scd_keypairinfo): Use --keypairinfo.
* sm/call-agent.c (gpgsm_agent_scd_keypairinfo): Ditto.
* scd/app-openpgp.c (do_getattr): Add attributes "$ENCRKEYID" and
"$SIGNKEYID".
* scd/app-piv.c (do_getattr): Ditto.
--
We already have $AUTHKEYID to locate the keyref of the key to be used
with ssh. It will also be useful to have default keyref for
encryption and signing. For example, this will allow us to repalce
the use of "OPENPGP.2" by a app type specific keyref.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/call-agent.c (scd_keypairinfo_status_cb): Also store the usage
flags.
* sm/call-agent.c (scd_keypairinfo_status_cb): Ditto.
* sm/certreqgen-ui.c (gpgsm_gencertreq_tty): Print the usage flags.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.c (main): Add special handling for bad keys in decrypt
mode.
--
The problem can easily be tested by adding --encrypt-to EXPIRED_KEY to
a decryption command. With that patch the errors are printed but
decryption continues and the process returns success unless other
errors occur.
GnuPG-bug-id: 4431
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/gpgsm.c (main): Call translate_sys2libc_fd_int to
convert the FDs.
--
This is required to actually pass gpgsm an fd on windows
and not a windows handle.
For the passphrase-fd this was already done.
* kbx/keybox-search-desc.h (KEYDB_SEARCH_MODE_FPR16)
(KEYDB_SEARCH_MODE_FPR20, KEYDB_SEARCH_MODE_FPR32): Remove. Switch
all users to KEYDB_SEARCH_MODE_FPR along with the fprlen value.
--
These search modes were added over time and there has until recently
be no incentive to remove the cruft. With the change for v5 keys I
finally went over all places and allowed the generic fingerprint mode
along with a given length of the fingerprint at all places.
Consequently the other modes can now be removed.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/userids.c (classify_user_id): Support 32 byte fingerprints.
* kbx/keybox-search-desc.h (KEYDB_SEARCH_MODE_FPR32): New.
(struct keydb_search_desc): Add field fprlen.
* kbx/keybox-defs.h (struct _keybox_openpgp_key_info): Add field
version and increase size of fpr to 32.
* kbx/keybox-blob.c: Define new version 2 for PGP and X509 blobs.
(struct keyboxblob_key): Add field fprlen and increase size of fpr.
(pgp_create_key_part_single): Allow larger fingerprints.
(create_blob_header): Implement blob version 2 and add arg want_fpr32.
(_keybox_create_openpgp_blob): Detect the need for blob version 2.
* kbx/keybox-search.c (blob_get_first_keyid): Support 32 byte
fingerprints.
(blob_cmp_fpr): Ditto.
(blob_cmp_fpr_part): Ditto.
(has_fingerprint): Add arg fprlen and pass on.
(keybox_search): Support KEYDB_SEARCH_MODE_FPR32 and adjust for
changed has_fingerprint.
* kbx/keybox-openpgp.c (parse_key): Support version 5 keys.
* kbx/keybox-dump.c (_keybox_dump_blob): Support blob version 2.
* g10/delkey.c (do_delete_key): Support KEYDB_SEARCH_MODE_FPR32.
* g10/export.c (exact_subkey_match_p): Ditto.
* g10/gpg.c (main): Ditto.
* g10/getkey.c (get_pubkey_byfprint): Adjust for changed
KEYDB_SEARCH_MODE_FPR.
* g10/keydb.c (keydb_search_desc_dump): Support
KEYDB_SEARCH_MODE_FPR32 and adjust for changed KEYDB_SEARCH_MODE_FPR.
(keydb_search): Add new arg fprlen and change all callers.
* g10/keyedit.c (find_by_primary_fpr): Ditto.
* g10/keyid.c (keystr_from_desc): Ditto.
* g10/keyring.c (keyring_search): Ditto.
* g10/keyserver.c (print_keyrec): Ditto.
(parse_keyrec): Ditto.
(keyserver_export): Ditto.
(keyserver_retrieval_screener): Ditto.
(keyserver_import): Ditto.
(keyserver_import_fprint): Ditto.
(keyidlist): Ditto.
(keyserver_get_chunk): Ditto.
* g10/keydb.c (keydb_search): Add new arg fprlen and change all
callers.
* sm/keydb.c (keydb_search_fpr): Adjust for changed
KEYDB_SEARCH_MODE_FPR.
--
This prepares the support for OpenPGP v5 keys. The new version 2 blob
format is needed for the longer fingerprints and we also use this
opportunity to prepare for storing the keygrip in the blob for faster
lookup by keygrip. Right now this is not yet functional.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/keylist.c (oidtranstbl): Add Yubikey OIDs.
(OID_FLAG_HEX): New.
(print_hex_extn): New.
(list_cert_raw): Make use of that flag.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/keylist.c (print_compliance_flags): Also check the diges_also.
--
A certificate with algorithm sha1WithRSAEncryption can be de-vs
compliant (e.g. if the next in the chain used sha256WithRSAEncryption
to sign it and RSA is long enough) but flagging it as such is useless
because that certificate can't be used because it will create
signatures using the non-compliant SHA-1 algorithm.
Well, it could be used for encryption. But also evaluating the
key-usage flags here would make it harder for the user to understand
why certain certificates are listed as de-vs compliant and others are
not.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/certreqgen.c (create_request): Fix for certmode.
--
When using an existing key from a card for certificate signing (in
contrast to the default of generating a CSR), the code tried to use
the same key for signing instead of the Signing-Key parameter. It is
perfectly okay to use the regular signing path via gpg-agent for
certificate creation - only self-signed certificates with a key on the
card require the direct use of the card key (via "SCD PKSIGN").
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/misc.c (transform_sigval): Allow for larger values of MDALGO and
PKALGO.
--
Libgcrypt already defines larger values for them, so we should be
prepared in case we use them in the future.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/call-agent.c (gpgsm_scd_pksign): Allow SHA512. Create proper
S-expression for EdDSA signature.
* sm/certreqgen.c (create_request): Force use of SHA512 when
using a ed25519 key.
* sm/misc.c (transform_sigval): Insert OID for ed25519.
--
GnuPG-bug-id: 4013
Signed-off-by: Damien Goutte-Gattat <dgouttegattat@incenp.org>
* sm/call-agent.c (gpgsm_scd_pksign): Identify type of signing key
and format resulting S-expression accordingly.
* sm/misc.c (transform_sigval): Support ECDSA signatures.
--
Current GpgSM implementation assumes card-based keys are RSA keys.
This patch introduces support for ECDSA keys.
By itself this patch is not sufficient, we also need support
from libksba.
GnuPG-bug-id: 4092
Signed-off-by: Damien Goutte-Gattat <dgouttegattat@incenp.org>
* sm/certreqgen-ui.c (gpgsm_gencertreq_tty): Get and show algo.
--
This extends the prompt to show something like
Serial number of the card: FF020001008A77F6
Available keys:
(1) 4130F84FA3704F4645924AEC3FFA48AD26D33656 PIV.9A nistp384
(2) AB2988FB8C227BCD5175BF92F66AA3A95AE83214 PIV.9E rsa2048
(3) DB7DDAEAA88534BA45CCD7A9B761425103EA2090 PIV.9C rsa2048
(4) BABB48C3D80ACCF9839F101DF2910966C8B988DF PIV.9D nistp256
Your selection? 1
Having the algorithm here is helpful in particular because right now
we support only RSA with X.509. Take care: PIV card based certificate
creation does not yet work.
Signed-off-by: Werner Koch <wk@gnupg.org>
* dirmngr/server.c (cmd_isvalid): Use option second arg to trigger
OCSP checkibng. Fallback to CRL if no default OCSP responder has been
configured.
* sm/call-dirmngr.c (gpgsm_dirmngr_isvalid): Adjust accordingly.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/cpr.c (write_status_failure): Make it print only once.
* g10/gpg.c (wrong_args): Bump error counter.
(g10_exit): Print a FAILURE status if we ever did a log_error etc.
(main): Use log_error instead of log_fatal at one place. Print a
FAILURE status for a bad option. Ditto for certain exit points so
that we can see different error locations.
--
This makes it easier to detect errors by tools which have no way to
get the exit code (e.g. due to double forking).
GnuPG-bug-id: 3872
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/server.c: Include shareddefs.h.
(option_handler): Add option.
--
This is required when running gpgsm in server mode as done by GPGME.
Noet that a command line option takes precedence.
Signed-off-by: Werner Koch <wk@gnupg.org>
* g10/gpg.c (oRequestOrigin): New const.
(opts): New option --request-origin.
(main): Parse that option.
* g10/options.h (struct opt): Add field request_origin.
* g10/call-agent.c (start_agent): Send option to the agent.
* sm/gpgsm.c (oRequestOrigin): New const.
(opts): New option --request-origin.
(main): Parse that option.
* sm/gpgsm.h (struct opt): Add field request_origin.
* sm/call-agent.c (start_agent): Send option to the agent.
Signed-off-by: Werner Koch <wk@gnupg.org>
* sm/export.c (sexp_to_kparms): Fix the computation of array[6],
which must be 'd mod (q-1)' but was 'p mod (q-1)'.
--
This bug is not serious but makes some consistency checks fail.
For example, 'openssl rsa -check' reports the following error:
$ gpgsm --out my.key --export-secret-key-raw 0xXXXXXXXX
$ openssl rsa -check -noout -inform DER -in my.key
RSA key error: dmq1 not congruent to d
--
Let me(wk) add this:
This bug was introduced with
Fixes-commit: 91056b1976
right at the start of GnuPG 2.1 in July 2010. Before that (in 2.0) we
used gpg-protect-tool which got it right. We probably never noticed
this because gpgsm, and maybe other tools too, fix things up during
import.
Signed-off-by: Werner Koch <wk@gnupg.org>
Werner Koch