* 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>
* agent/agent.h (opt): Add field allow_external_cache.
* agent/call-pinentry.c (start_pinentry): Act upon new var.
* agent/gpg-agent.c (oNoAllowExternalCache): New.
(opts): Add option --no-allow-external-cache.
(parse_rereadable_options): Set this option.
--
Pinentry 0.9.2 may be build with libsecret support and thus an extra
checkbox is displayed to allow the user to get passwords out of an
libsecret maintained cache. Security aware user may want to avoid
this feature and may do this at runtime by enabling this option.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/call-pinentry.c (start_pinentry): Add more strings.
--
We do this so that translations of these strings will be available at
the time a pinentry implements features which require these strings.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (oDebugPinentry): New.
(opts): Add --debug-pinentry.
(parse_rereadable_options): Set that option.
* agent/call-pinentry.c (start_pinentry): Pass option to
assuan_set_flag.
--
This option is quite useful to see the IPC between gpg-agent and
Pinentry. Note that "debug 1024" is also required.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/call-pinentry.c (PINENTRY_STATUS_PASSWORD_FROM_CACHE): New
constant.
(pinentry_status_cb): Add it to *FLAGS if PASSWORD_FROM_CACHE was
provided.
(agent_askpin): Pass "OPTION allow-external-password-cache" to the
pinentry. Always pass SETKEYINFO to the pinentry. If there is no
stable identifier, then use "--clear". If the password is incorrect
and PINENTRY_STATUS_PASSWORD_FROM_CACHE is set in *PINENTRY_STATUS,
then decrement PININFO->FAILED_TRIES.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* agent/call-pinentry.c (pinentry_status_cb): Or in
PINENTRY_STATUS_CLOSE_BUTTON; don't overwrite *FLAG.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* agent/call-pinentry.c (PINENTRY_STATUS_CLOSE_BUTTON): New constant.
(PINENTRY_STATUS_PIN_REPEATED): Likewise.
(close_button_status_cb): Rename from this...
(pinentry_status_cb): ... to this. Use the constants.
(agent_askpin): Rename local variable from close_button to
pinentry_status. Use symbolic constants rather than magic numbers.
--
Signed-off-by: Neal H. Walfield <neal@g10code.com>
* common/logging.h: Rename JNLIB_LOG_* to GPGRT_LOG_*.
* common/mischelp.h: Rename JNLIB_GCC_* to GPGRT_GCC_*.
--
JNLIB has no more meaning. Thus we switch to a GPGRT_ prefix in
anticipation that some code may eventually be moved to libgpg-error.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/call-pinentry.c (agent_askpin): Add args keyinfo and
cache_mode. Change all callers to pass (NULL,0) for them. Send
SETKEYINFO command.
* agent/findkey.c (unprotect): Pass the keygrip and the cache_mode for
the new args.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/cvt-openpgp.c (apply_protection): Strip leading zeroes from
opaque MPIs to comply with the OpenPGP spec.
--
This patch is the protected private key counterpart to commit ab17f7b.
Thanks to andy_s for describing the problem.
GnuPG-bug-id: 1853
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command-ssh.c (ssh_search_control_file): Check S before
upcasing it.
--
In contradiction to the comment we did not check the length of HEXGRIP
and thus the GPG_ERR_INV_LENGTH was never triggered.
Detected by Stack 0.3:
bug: anti-simplify
model: |
%cmp8 = icmp ne i32 %i.0, 40, !dbg !986
--> false
stack:
- /home/wk/s/gnupg/agent/command-ssh.c:1226:0
ncore: 2
core:
- /home/wk/s/gnupg/agent/command-ssh.c:1225:0
- buffer overflow
- /home/wk/s/gnupg/agent/command-ssh.c:1225:0
- buffer overflow
* agent/call-pinentry.c (unlock_pinentry): Add error logging. Map
error source of uncommon errors to Pinentry.
--
With this change it is possible to detect whether an error like
GPG_ERR_ASS_INV_RESPONSE has its origin in a call to Pinentry or comes
from another part of gpg-agent.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/pksign.c (agent_pksign_do): Use int.
* scd/app-openpgp.c (get_public_key): Likewise.
--
On 64-bit architecture, int and size_t might be different.
For the first argument for '%b', int is expected.
* common/host2net.h (buf16_to_ulong, buf16_to_uint): New.
(buf16_to_ushort, buf16_to_u16): New.
(buf32_to_size_t, buf32_to_ulong, buf32_to_uint, buf32_to_u32): New.
--
Commit 91b826a388 was not enough to
avoid all sign extension on shift problems. Hanno Böck found a case
with an invalid read due to this problem. To fix that once and for
all almost all uses of "<< 24" and "<< 8" are changed by this patch to
use an inline function from host2net.h.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/homedir.c (get_default_pinentry_name): New.
(gnupg_module_name): Use that for the default pinentry.
(gnupg_module_name_flush_some): New.
* agent/gpg-agent.c (agent_sighup_action): Flush some module names.
* agent/call-pinentry.c (start_pinentry): Do not modify
opt.pinentry_program.
--
The idea with this change is that under Windows we can install a
simple native Windows pinentry as "pinentry-basic" and a full GUI
version may then later install pinentry-gtk etc which would then
automatically be used.
Unfortunately installing another pinentry from a different package
would clobber the GnuPG core directory which is not nice. To fix that
we would need to agree on standard installation directories for GUIs
to also look there.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/cvt-openpgp.c (extract_private_key): Add arg "arraysize".
Make sure that R_FLAGS and R_CURVE are set to NULL.
--
Given that extract_private_key is not file local it is good to have some
extra asserts to protect against future wrong use.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/cvt-openpgp.c (extract_private_key): New.
(convert_to_openpgp): Use extract_private_key.
* agent/findkey.c (agent_public_key_from_file): Use
extract_private_key.
--
This patch add support of ECC key with a curve name and flags. Since
same functionality is also needed for convert_to_openpgp, it was
factored out into the extract_private_key function.
* agent/divert-scd.c (divert_pkdecrypt): Support ECDH.
* scd/app-openpgp.c (get_algo_byte, store_fpr): Support ECDH.
(send_key_attr): Support ECDH. Fix EdDSA algorithm value.
(retrieve_key_material): Initialize fields.
(get_public_key, ecc_writekey, do_writekey): Support ECDH.
(ecdh_writekey): Remove.
(do_decipher): Support ECDH.
(parse_algorithm_attribute): Support ECDH. Fix EdDSA.
--
Following the gpg-agent protocol, SCDaemon's counter part is now
implemented.
* agent/gpg-agent.c (finalize_rereadable_options): New.
(main, reread_configuration): Call it.
--
This change should help to avoid surprising behaviour.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command-ssh.c (setup_ssh_env): Move code to ...
* agent/gpg-agent.c (agent_copy_startup_env): .. new function. Change
calllers.
* agent/command.c (start_command_handler): Call that fucntion for
restricted connections.
--
A remote connection is and should not be able to setup the local
session environment. However, unless --keep-display is used we would
be left without an environment and thus pinentry can't be used. The
fix is the same as used for ssh-agent connection: We use the default
environment as used at the startup of the agent.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_learn): Add option --sendinfo.
* agent/learncard.c (agent_handle_learn): Add arg "send" andsend
certifciate only if that is set.
* g10/call-agent.c (agent_scd_learn): Use --sendinfo. Make INFO
optional.
(agent_learn): Remove.
* g10/keygen.c (gen_card_key): Replace agent_learn by agent_scd_learn.
--
The requirement of using --card-status on the first use of card on a
new box is a bit annoying but the alternative of always checking
whether a card is available before a decryption starts does not sound
promising either.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/scdaemon.c (ENAMETOOLONG): New.
(redir_socket_name): New.
(cleanup): Take care of a redirected socket.
(main): Pass redir_socket_name to create_server_socket.
(create_socket_name): Remove superfluous length check.
(create_server_socket): Add arg r_redir_name and implement
redirection. Replace assert for older Assuan by an error message.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (ENAMETOOLONG): New.
(redir_socket_name, redir_socket_name_extra)
(redir_socket_name_ssh): New.
(remove_socket): Take care of the redir names.
(main): Pass the redir names to create_server_socket.
(create_socket_name): Remove length check - that is anyway done later.
(create_server_socket): Add arg r_redir_name and implement redirection
if Libassuan is at least 2.14.
* agent/agent.h (opt): Add field extra_socket.
(server_control_s): Add field restricted.
* agent/command.c: Check restricted flag on many commands.
* agent/gpg-agent.c (oExtraSocket): New.
(opts): Add option --extra-socket.
(socket_name_extra): New.
(cleanup): Cleanup that socket name.
(main): Implement oExtraSocket.
(create_socket_name): Add arg homedir and change all callers.
(create_server_socket): Rename arg is_ssh to primary and change
callers.
(start_connection_thread): Take ctrl as arg.
(start_connection_thread_std): New.
(start_connection_thread_extra): New.
(handle_connections): Add arg listen_fd_extra and replace the
connection starting code by parameterized loop.
* common/asshelp.c (start_new_gpg_agent): Detect the use of the
restricted mode and don't fail on sending the pinentry environment.
* common/util.h (GPG_ERR_FORBIDDEN): New.
* gl/: Remove entire tree.
* configure.ac: Remove gnulib tests and the gl/ Makefile.
(setenv): Add to AC_CHECK_FUNCS.
* autogen.rc (extra_aclocal_flags): Set to empty.
* Makefile.am (ACLOCAL_AMFLAGS): Remove -I gl/m4
(SUBDIRS): Remove gl/.
* agent/Makefile.am (common_libs): Remove ../gl/gnulib.a
* common/Makefile.am (t_common_ldadd): Ditto.
* dirmngr/Makefile.am (dirmngr_LDADD): Ditto.
(dirmngr_ldap_LDADD, dirmngr_client_LDADD): Ditto.
* g10/Makefile.am (needed_libs): Ditto.
* g13/Makefile.am (g13_LDADD): Ditto.
* kbx/Makefile.am (kbxutil_LDADD): Ditto.
($(PROGRAMS)): Ditto.
* scd/Makefile.am (scdaemon_LDADD): Ditto.
* sm/Makefile.am (common_libs): Ditto.
* tools/Makefile.am (common_libs, commonpth_libs): Ditto.
* agent/gpg-agent.c: Remove "mkdtemp.h"
* g10/exec.c: Ditto.
* scd/scdaemon.c: Ditto.
* tools/symcryptrun.c: Ditto.
* common/sysutils.c: Remove "setenv.h"
* common/t-timestuff.c: Use putenv if setenv is not available.
--
gnulib has always been a cause of trouble in GnuPG because we used
only a very few functions and the complex include machinery of gnulib
is quite complex and the cause for many build problems for example on
OS X. This is not gnulib's fault but due to our limited use of gnulib
and that we only rarely update the gnulib code to avoid regressions.
In part two we will address the functions
mkdtemp
setenv
unsetenv
strpbrk
which may bot be implemented on all platforms. They are not required
on a libc based system.
Signed-off-by: Werner Koch <wk@gnupg.org>
--
Also fixed some of my own copyright notices due to the termination of
my assignment. The one displayed by --version is kept at FSF because
we had contributors in 2014 with FSF assignments and it gives the FSF
some visibility.
* agent/command.c (cmd_export_key): Actually implement the cache_nonce
feature.
* g10/export.c (do_export_stream): Make use of a cache_nonce.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (oDebugQuickRandom): New.
(opts): New option --debug-quick-random.
(main): Use new option.
* common/asshelp.c (start_new_gpg_agent): Add hack to pass an
additional argument for the agent name.
* tests/openpgp/defs.inc: Pass --debug-quick-random to the gpg-agent
starting parameters.
* tests/openpgp/version.test: Ditto.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/agent.h (struct pin_entry_info_s): Add fields repeat_okay and
with_repeat.
* agent/call-pinentry.c (close_button_status_cb): Rewrite and check
for PIN_REPEAT. Change users to check only the relevant bit.
(agent_askpin): Support repeat logic of new Pinentries.
* agent/command-ssh.c (ssh_identity_register): Use the new repeat
feature.
* agent/genkey.c (agent_ask_new_passphrase): Ditto.
--
If we need to confirm a passphrase entry (e.g. for new passphrase) we
set a flag into the pinentry info block. The we try to use the new
pinentry command SETREPEATERROR; if that fails, we continue as usual.
If that succeeds we ask the pinentry to show the repeat (confirmation)
prompt and on successful return we set another flag in the pinentry
info block so that the caller can skip its own confirmation check. A
new status line from the pinentry indicates that the feature is
actually supported (it may not be supported on certain systems for
example when using the ncurses backend).
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/agent.h (opt): Remove field use_standard_socket.
* agent/command.c (cmd_killagent): Always allow killing.
* agent/gpg-agent.c (main): Turn --{no,}use-standard-socket and
--write-env-file into dummy options. Always return true for
--use-standard-socket-p. Do not print the GPG_AGENT_INFO envvar
setting or set that envvar.
(create_socket_name): Simplify by removing non standard socket
support.
(check_for_running_agent): Ditto.
* common/asshelp.c (start_new_gpg_agent): Remove GPG_AGENT_INFO use.
* common/simple-pwquery.c (agent_open): Ditto.
* configure.ac (GPG_AGENT_INFO_NAME): Remove.
* g10/server.c (gpg_server): Do not print the AgentInfo comment.
* g13/server.c (g13_server): Ditto.
* sm/server.c (gpgsm_server): Ditto.
* tools/gpgconf.c (main): Simplify by removing non standard socket
support.
--
The indented fix to allow using a different socket than the one in the
gnupg home directory is to change Libassuan to check whether the
socket files exists as a regualr file with a special keyword to
redirect to another socket file name.
* agent/cache.c (last_stored_cache_key): New.
(agent_get_cache): Allow NULL for KEY.
(agent_store_cache_hit): New.
* agent/findkey.c (unprotect): Call new function and try to use the
last stored key.
* g10/revoke.c (create_revocation): Add arg CACHE_NONCE and pass to
make_keysig_packet.
(gen_standard_revoke): Add arg CACHE_NONCE and pass to
create_revocation.
* g10/keygen.c (do_generate_keypair): Call gen_standard_revoke with
cache nonce.
--
This patch adds two features:
1. The key for the last passphrase successfully used for unprotecting
a key is stored away. On a cache miss the stored away passphrase is
tried as well. This helps for the common GPG use case of having a
signing and encryption (sub)key with the same passphrase. See the
code for more comments.
2. The now auto-generated revocation certificate does not anymore
popup a passphrase prompt. Thus for standard key generation the
passphrase needs to be given only once (well, two with the
confirmation).
* agent/cvt-openpgp.c (get_keygrip): Special case EdDSA.
(convert_secret_key): Ditto.
(convert_transfer_key): Ditto.
(apply_protection): Handle opaque MPIs.
(do_unprotect): Check FLAG_OPAQUE instead of FLAG_USER1 before
unpacking an opaque mpi.
--
The key transfer protocol between gpg and gpg-agent uses gcrypt
algorithm numbers which merge all ECC algorithms into one. Thus it is
not possible to use the algorithm number to determine the EdDSA
algorithm. We need to known that because Libgcrypt requires the
"eddsa" flag with the curve "Ed25519" to actually use the Ed25519
signature specification.
The last fix is for correctness; the first case won't be used anyway.
* agent/protect.c (PROT_CIPHER): Rename to GCRY_CIPHER_AES128 for
clarity.
(do_decryption): Add args prot_cipher and prot_cipher_keylen. USe
them instead of the hardwired values.
(agent_unprotect): Change to use a table of protection algorithms.
Add AES-256 variant.
--
This patch will make a possible future key protection algorithm
changes smoother. AES-256 is also allowed although there is currently
no way to encrypt using it.
* agent/cvt-openpgp.c (convert_to_openpgp): Use the curve name instead
of the curve parameters.
* g10/export.c (canon_pubkey_algo): Rename to ...
(canon_pk_algo): this. Support ECC.
(transfer_format_to_openpgp): Expect curve name.
* agent/cvt-openpgp.c (do_unprotect): Return an s-exp also for
non-protected keys.
(convert_from_openpgp_main): Do not call agent_askpin for a
non-protected key.
* agent/cvt-openpgp.c (try_do_unprotect_arg_s): Add field "curve".
(get_keygrip): Add and use arg CURVE.
(convert_secret_key): Ditto.
(convert_transfer_key): Ditto.
(get_npkey_nskey): New.
(prepare_unprotect): Replace gcrypt functions by
get_npkey_nskey. Allow opaque MPIs.
(do_unprotect): Use CURVE instead of parameters.
(convert_from_openpgp_main): Ditto.
(convert_to_openpgp): Simplify.
* g10/import.c (one_mpi_from_pkey): Remove.
(transfer_secret_keys): Rewrite to use the curve instead of the
parameters.
* g10/parse-packet.c (parse_key): Mark protected MPIs with USER1 flag.
* common/openpgp-oid.c (openpgp_curve_to_oid): Allow the use of
"NIST P-256" et al.
* g10/keygen.c (ask_curve): Add arg ALGO.
(generate_keypair): Rewrite the ECC key logic.
* tests/openpgp/ecc.test: Provide the "ecc" passphrase.
* agent/cvt-openpgp.c (get_keygrip, convert_secret_key)
(convert_transfer_key): Follow newer (>= 1.6) libgcrypt API, which
does not distinguish the detail.
(do_unprotect, convert_from_openpgp_main): Don't call
map_pk_openpgp_to_gcry, as it's the value of libgcrypt API already and
not the value defined by OpenPGP.
(convert_to_openpgp): It's "ecc".
* agent/gpg-agent.c (map_pk_openpgp_to_gcry): Remove.
* g10/call-agent.c (agent_pkdecrypt): Fix off-by-one error.
* g10/pubkey-enc.c (get_it): Fix swapping the fields error.
* agent/gpg-agent.c (main): Make sure homedir is absolute.
* common/asshelp.c (lock_spawning): Create lock file with an absolute
name.
(start_new_gpg_agent): Use an absolute name for the socket and pass
option --homedir to the agent.
(start_new_dirmngr): Use an absolute name for the --homedir.
--
This patch makes gpg's --homedir option behave again like in older
versions. This is done by starting a new agent for each different
home directory. Note that this assumes --use-standard-socket is used
which is the default for 2.1.
* agent/cvt-openpgp.c (convert_to_openpgp): Fix use
gcry_sexp_extract_param.
* g10/export.c (do_export_stream): Provide a proper prompt to the
agent.
--
NB: The export needs more work, in particular the ECC algorithms.
* agent/command-ssh.c (ssh_read_key_public_from_blob): Use
es_fopenmem.
(ssh_handler_request_identities): Ditto.
(ssh_request_process): Ditto.
--
es_fopenmem is easier to understand than the more general function
es_mopen. Thus we better use the former for clarity.
* agent/command-ssh.c (ssh_key_type_spec): Add field name.
(ssh_key_types): Add human readable names.
(add_control_entry): Add arg SPEC and print key type as comment.
(ssh_identity_register): Add arg SPEC.
(ssh_handler_add_identity): Add var SPEC and pass ssh_receive_key.
* agent/command-ssh.c (SPEC_FLAG_IS_EdDSA): New.
(ssh_key_types): Add entry for ssh-ed25519.
(ssh_identifier_from_curve_name): Move to the top.
(stream_read_skip): New.
(stream_read_blob): New.
(ssh_signature_encoder_rsa): Replace MPIS array by an s-exp and move
the s-exp parsing to here.
(ssh_signature_encoder_dsa): Ditto.
(ssh_signature_encoder_ecdsa): Ditto.
(ssh_signature_encoder_eddsa): New.
(sexp_key_construct): Rewrite.
(ssh_key_extract): Rename to ...
(ssh_key_to_blob): .. this and rewrite most of it.
(ssh_receive_key): Add case for EdDSA.
(ssh_convert_key_to_blob, key_secret_to_public): Remove.
(ssh_send_key_public): Rewrite.
(ssh_handler_request_identities): Simplify.
(data_sign): Add rename args. Add new args HASH and HASHLEN. Make
use of es_fopenmen and es_fclose_snatch. Remove parsing into MPIs
which is now doe in the sgnature encoder functions.
(ssh_handler_sign_request): Take care of Ed25519.
(ssh_key_extract_comment): Rewrite using gcry_sexp_nth_string.
--
To make the code easier readable most of the Ed25591 work has been
done using a new explicit code path. Warning: Libgcrypt 1.6.1 uses a
non optimized implementation for Ed25519 and timing attacks might be
possible.
While working on the code I realized that it could need more rework;
it is at some places quite baroque and more complicated than needed.
Given that we require Libgcrypt 1.6 anyway, we should make more use of
modern Libgcrypt functions.
* agent/call-scd.c (start_scd): Replace int by assuan_fd_t.
(start_pinentry): Ditto.
* common/asshelp.c (start_new_gpg_agent): Replace int by assuan_fd_t.
* common/dotlock.c (GNUPG_MAJOR_VERSION): Include stringhelp.h for
prototypes on Windows and some other platforms.
* common/logging.c (fun_writer): Declare addrbuf only if needed.
* g10/decrypt.c (decrypt_message_fd) [W32]: Return not_implemented.
* g10/encrypt.c (encrypt_crypt) [W32]: Return error if used in server
mode.
* g10/dearmor.c (dearmor_file, enarmor_file): Replace GNUPG_INVALID_FD
by -1 as temporary hack for Windows.
* g10/export.c (do_export): Ditto.
* g10/revoke.c (gen_desig_revoke, gen_revoke): Ditto.
* g10/sign.c (sign_file, clearsign_file, sign_symencrypt_file): Ditto.
* g10/server.c (cmd_verify, gpg_server) [W32]: Return an error.
--
The gpg server mode is not actual working and thus we can avoid the
warnings by explicitly disabling the mode. We keep it working under
Unix, though.
* agent/command-ssh.c (setup_ssh_env): Fix env setting.
--
gniibe reported this to gnupg-devel on 2012-07-04:
[...]
(2) UPDATESTARTUPTTY doesn't work to switch TTY for pinentry for
SSH.
[...]
Current implementation:
In the function start_command_handler_ssh, the logic puts
priority on ctrl->session_env which is initialized by
agent_init_default_ctrl. There are always GPG_TTY and TERM
defined, because lines around 968 in gpg-agent.c, it says:
/* Make sure that we have a default ttyname. */
While UPDATESTARTUPTTY updates opt.startup_env, it doesn't
affect at all.
Here is a patch to point the issue. Tested and works for me.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 9f5578c29a)
* agent/gpg-agent.c (opts, main): Add option --no-allow-mark-trusted.
Put this option into the gpgconf-list.
(main): Enable opt.allow_mark_trusted by default.
* tools/gpgconf-comp.c (gc_options_gpg_agent): Replace
allow-mark-trusted by no-allow-mark-trusted.
* agent/trustlist.c (agent_marktrusted): Always set the "relax" flag.
--
These changes have been in effect for the Gpg4win Windows version
since 2011-01-24 and thus first released with Gpg4win 2.1.0. Given
the current state of PKIX it does not make any sense to lure the Unix
user into false security by making it harder to trust self-signed or
CAcert certificates.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 90b419f3e9)
Resolved conflicts:
NEWS
agent/gpg-agent.c
* agent/gpg-agent.c [W32]: Include Several Windows header.
(opts): Change help text for enable-ssh-support.
(opts, main): Add option --enable-putty-support
(putty_support, PUTTY_IPC_MAGIC, PUTTY_IPC_MAXLEN): New for W32.
(agent_init_default_ctrl): Add and asssert call.
(putty_message_proc, putty_message_thread): New.
(handle_connections) [W32]: Start putty message thread.
* common/sysutils.c (w32_get_user_sid): New for W32 only
* tools/gpgconf-comp.c (gc_options_gpg_agent): Add
--enable-ssh-support and --enable-putty-support. Make the
configuration group visible at basic level.
* agent/command-ssh.c (serve_mmapped_ssh_request): New for W32 only.
--
This patch enables support for Putty. It has been tested with Putty
0.62 using an Unix created ssh key copied to the private-keys-v1.d
directory on Windows and with a manually crafted sshcontrol file. It
also works with a smartcard key.
May thanks to gniibe who implemented a proxy in Python to test the
putty/gpg-agent communication.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 9f32499f99)
Resolved conflicts:
NEWS
agent/agent.h
agent/gpg-agent.c: Convert from pth to npth.
common/sysutils.c
common/sysutils.h
* agent/command-ssh.c (file_to_buffer)
(ssh_handler_request_identities): Open streams in binary mode.
(start_command_handler_ssh): Factor some code out to ..
(setup_ssh_env): new function.
--
This is for now a theoretical fix because there is no ssh client yet
which uses the GnuPG style IPC. OpenSSL for Cygwin uses only a quite
similar one. gniibe suggested to implement that IPC style in
Libassuan so that a Cygwin version of OpenSSL may be used with GnuPG.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit ed056d67c7)
Also fixed one typo.
* agent/gpg-agent.c (main): s/ttyname/gnupg_ttyname/.
--
This was not triggered by the latest mingw runtime but that has other
problems and thus I reverted to the same we used for gpg4win 2.1 which
is Debian Wheezy.
* common/sexputil.c (get_pk_algo_from_canon_sexp): Change to return a
string.
* g10/keygen.c (check_keygrip): Adjust for change.
* sm/certreqgen-ui.c (check_keygrip): Likewise.
* agent/pksign.c (do_encode_dsa): Remove bogus map_pk_openpgp_to_gcry.
* g10/misc.c (map_pk_openpgp_to_gcry): Remove.
(openpgp_pk_test_algo): Change to a wrapper for openpgp_pk_test_algo2.
(openpgp_pk_test_algo2): Rewrite.
(openpgp_pk_algo_usage, pubkey_nbits): Add support for EdDSA.
(openpgp_pk_algo_name): Rewrite to remove need for gcry calls.
(pubkey_get_npkey, pubkey_get_nskey): Ditto.
(pubkey_get_nsig, pubkey_get_nenc): Ditto.
* g10/keygen.c(do_create_from_keygrip): Support EdDSA.
(common_gen, gen_ecc, ask_keysize, generate_keypair): Ditto.
* g10/build-packet.c (do_key): Ditto.
* g10/export.c (transfer_format_to_openpgp): Ditto.
* g10/getkey.c (cache_public_key): Ditto.
* g10/import.c (transfer_secret_keys): Ditto.
* g10/keylist.c (list_keyblock_print, list_keyblock_colon): Ditto.
* g10/mainproc.c (proc_pubkey_enc): Ditto.
* g10/parse-packet.c (parse_key): Ditto,
* g10/sign.c (hash_for, sign_file, make_keysig_packet): Ditto.
* g10/keyserver.c (print_keyrec): Use openpgp_pk_algo_name.
* g10/pkglue.c (pk_verify, pk_encrypt, pk_check_secret_key): Use only
OpenPGP algo ids and support EdDSA.
* g10/pubkey-enc.c (get_it): Use only OpenPGP algo ids.
* g10/seskey.c (encode_md_value): Ditto.
--
This patch separates Libgcrypt and OpenPGP public key algorithms ids
and in most cases completely removes the Libgcrypt ones. This is
useful because for Libgcrypt we specify the algorithm in the
S-expressions and the public key ids are not anymore needed.
This patch also adds some support for PUBKEY_ALGO_EDDSA which will
eventually be used instead of merging EdDSA with ECDSA. As of now an
experimental algorithm id is used but the plan is to write an I-D so
that we can get a new id from the IETF. Note that EdDSA (Ed25519)
does not yet work and that more changes are required.
The ECC support is still broken right now. Needs to be fixed.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (main): Defer setting of socket_name_ssh to avoid
removal of the socket when it will die in create_server_socket for
socket_name.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
Replace hardwired strings at many places with new macros from config.h
and use the new strusage macro replacement feature.
* common/asshelp.c (lock_spawning) [W32]: Change the names of the spawn
sentinels.
* agent/command.c (cmd_import_key): Use asprintf to create the prompt.
* agent/findkey.c (key_parms_from_sexp): Add algo name "ecc".
(agent_is_dsa_key): Ditto.
(agent_is_eddsa_key): New. Not finished, though.
* agent/pksign.c (do_encode_eddsa): New.
(agent_pksign_do): Use gcry_log_debug functions.
* agent/protect.c (agent_protect): Parse a flags parameter.
* g10/keygen.c (gpg_curve_to_oid): Move to ...
* common/openpgp-oid.c (openpgp_curve_to_oid): here and rename.
(oid_ed25519): New.
(openpgp_oid_is_ed25519): New.
(openpgp_oid_to_curve): New.
* common/t-openpgp-oid.c (test_openpgp_oid_is_ed25519): New.
* g10/build-packet.c (gpg_mpi_write): Write the length header also for
opaque MPIs.
(gpg_mpi_write_nohdr): New.
(do_key): Use gpg_mpi_write_nohdr depending on algorithm.
(do_pubkey_enc): Ditto.
* g10/ecdh.c (pk_ecdh_encrypt_with_shared_point): Use
gpg_mpi_write_nohdr.
* g10/export.c (transfer_format_to_openpgp):
* g10/keygen.c (ecckey_from_sexp): Return the error.
(gen_ecc): Repalce arg NBITS by CURVE.
(read_parameter_file): Add keywords "Key-Curve" and "Subkey-Curve".
(ask_curve): New.
(generate_keypair, generate_subkeypair): Use ask_curve.
(do_generate_keypair): Also pass curve name.
* g10/keylist.c (list_keyblock_print, list_keyblock_colon): Print
curve name.
* g10/parse-packet.c (mpi_read): Remove workaround for
Libcgrypt < 1.5.
(parse_key): Fix ECC case. Print the curve name.
* g10/pkglue.c (mpi_from_sexp): Rename to get_mpi_from_sexp.
(pk_verify, pk_check_secret_key): Add special case for Ed25519.
* g10/seskey.c (encode_md_value): Ditto.
* g10/sign.c (do_sign, hash_for, sign_file): Ditto.
--
Be warned that this code is subject to further changes and that the
format will very likely change before a release. There are also known
bugs and missing code.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/pksign.c (do_encode_dsa): Remove Libgcrypt version check
--
Now that we have decided on a release plan for Libgcrypt 1.6 and given
all the improvements it makes more sense to make use of these
improvements than to clutter the GnuPG code with workarounds for older
Libgcrypt versions.
* agent/pksign.c (rfc6979_hash_algo_string): New.
(do_encode_dsa) [Libgcrypt >= 1.6]: Make use of RFC-6979.
--
Now that we have a good (and not NSA/NIST demanded ;-) specification
on how to use DSA without a random nonce, we take advantage of it and
thus avoid pitfalls related to a misbehaving RNG during signature
creation.
Note that OpenPGP has the option of using a longer hash algorithm but
truncated to what is suitable for the used DSA key size. The hash
used as input to RFC-6979 will also be one with an appropriate digest
length but not a truncated one. This is allowed by RFC-6979.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/app-common.h (APP_DECIPHER_INFO_NOPAD): New.
* scd/app-openpgp.c (do_decipher): Add arg R_INFO.
* scd/app-nks.c (do_decipher): Add arg R_INFO as a dummy.
* scd/app.c (app_decipher): Add arg R_INFO.
* scd/command.c (cmd_pkdecrypt): Print status line "PADDING".
* agent/call-scd.c (padding_info_cb): New.
(agent_card_pkdecrypt): Add arg R_PADDING.
* agent/divert-scd.c (divert_pkdecrypt): Ditto.
* agent/pkdecrypt.c (agent_pkdecrypt): Ditto.
* agent/command.c (cmd_pkdecrypt): Print status line "PADDING".
* g10/call-agent.c (padding_info_cb): New.
(agent_pkdecrypt): Add arg R_PADDING.
* g10/pubkey-enc.c (get_it): Use padding info.
--
Decryption using a card never worked in gpg 2.1 because the
information whether the pkcs#1 padding needs to be removed was not
available. Gpg < 2.1 too this info from the secret sub key but that
has gone in 2.1.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_preset_passphrase, pinentry_loopback): Use %zu
in format string.
* scd/ccid-driver.c (ccid_get_atr): Ditto.
* agent/command-ssh.c (stream_read_string): Init arg STRING_SIZE to
avoid maybe_unitialized warning.
--
Actually the first one might have been a problem on big endian
machines.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command-ssh.c (struct control_file_s): Rename to
ssh_control_file_s.
(ssh_open_control_file, ssh_close_control_file)
(ssh_read_control_file, ssh_search_control_file): New.
(control_file_t): Rename and move to ...
* agent/agent.h (ssh_control_file_t): here.
* agent/command.c (do_one_keyinfo): Add args is_ssh, ttl, disabled,
and confirm. Rename unknown keytype indicator from '-' to 'X'. Extend
output.
(cmd_keyinfo): Add options --ssh-list and --with-ssh.
--
This extension allows the development of frontends to manage the
sshcontrol file.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/Makefile.am (gpg_agent_DEPENDENCIES): Remove cruft from wrong
resolve conflict 2013-04-25.
(gpg_agent_DEPENDENCIES): Remove obsolete gpg_agent_res_deps
(gpg_agent_LDFLAGS): Remove obsolete gpg_agent_res_ldflags.
--
This fixes 88e24341e5. Thanks to
Christian Aistleitner for pinpointing this bug. Am sorry for accusing
automake to have introduced this bug; I should have washed my own eyes.
GnuPG-bug-id: 1511
* agent/command.c (cmd_import_key): Add option --unattended.
* agent/cvt-openpgp.c (convert_transfer_key): New.
(do_unprotect): Factor some code out to ...
(prepare_unprotect): new function.
(convert_from_openpgp): Factor all code out to ...
(convert_from_openpgp_main): this. Add arg 'passphrase'. Implement
openpgp-native protection modes.
(convert_from_openpgp_native): New.
* agent/t-protect.c (convert_from_openpgp_native): New dummy fucntion
* agent/protect-tool.c (convert_from_openpgp_native): Ditto.
* agent/protect.c (agent_unprotect): Add arg CTRL. Adjust all
callers. Support openpgp-native protection.
* g10/call-agent.c (agent_import_key): Add arg 'unattended'.
* g10/import.c (transfer_secret_keys): Use unattended in batch mode.
--
With the gpg-agent taking care of the secret keys, the user needs to
migrate existing keys from secring.gpg to the agent. This and also
the standard import of secret keys required the user to unprotect the
secret keys first, so that gpg-agent was able to re-protected them
using its own scheme. With many secret keys this is quite some
usability hurdle. In particular if a passphrase is not instantly
available.
To make this migration smoother, this patch implements an unattended
key import/migration which delays the conversion to the gpg-agent
format until the key is actually used. For example:
gpg2 --batch --import mysecretkey.gpg
works without any user interaction due to the use of --batch. Now if
a key is used (e.g. "gpg2 -su USERID_FROM_MYSECRETKEY foo"), gpg-agent
has to ask for the passphrase anyway, converts the key from the
openpgp format to the internal format, signs, re-encrypts the key and
tries to store it in the gpg-agent format to the disk. The next time,
the internal format of the key is used.
This patch has only been tested with the old demo keys, more tests
with other protection formats and no protection are needed.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_import_key): Pass 0 instead of KEYLEN to
gcry_sexp_canon_len.
--
We used to pass KEYLEN to the gcry_sexp_canon_len for no good reason:
convert_from_openpgp is guaranteed to return a valid canonical
S-expression and KEYLEN would thus act only as an upper limit. This
is not a problem because usually the original input key is longer than
the returned unprotected key. A future patch may change this
assertion and thus we better fix this bug now.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/gpg-agent.c (oDisableCheckOwnSocket): New.
(disable_check_own_socket): New.
(parse_rereadable_options): Set new option.
(check_own_socket): Implement new option.
Signed-off-by: Werner Koch <wk@gnupg.org>
* agent/command.c (cmd_keytocard): Decrement KEYDATALEN.
* agent/findkey.c (agent_public_key_from_file): Increment for ELEMS.
--
For ECDSA and ECDH, there are 6 elements.
* common/agent-opt.c: New.
* common/shareddefs.h: New.
* common/Makefile.am: Add new files.
* agent/agent.h: Include shareddefs.h.
(pinentry_mode_t): Factor out to shareddefs.h.
* agent/command.c (option_handler): Use parse_pinentry_mode.
* agent/command-ssh.c (SPEC_FLAG_IS_ECDSA): New.
(struct ssh_key_type_spec): Add fields CURVE_NAME and HASH_ALGO.
(ssh_key_types): Add types ecdsa-sha2-nistp{256,384,521}.
(ssh_signature_encoder_t): Add arg spec and adjust all callers.
(ssh_signature_encoder_ecdsa): New.
(sexp_key_construct, sexp_key_extract, ssh_receive_key)
(ssh_convert_key_to_blob): Support ecdsa.
(ssh_identifier_from_curve_name): New.
(ssh_send_key_public): Retrieve and pass the curve_name.
(key_secret_to_public): Ditto.
(data_sign): Add arg SPEC and change callers to pass it.
(ssh_handler_sign_request): Get the hash algo from SPEC.
* common/ssh-utils.c (get_fingerprint): Support ecdsa.
* agent/protect.c (protect_info): Add flag ECC_HACK.
(agent_protect): Allow the use of the "curve" parameter.
* agent/t-protect.c (test_agent_protect): Add a test case for ecdsa.
* agent/command-ssh.c (ssh_key_grip): Print a better error code.
--
The 3 standard curves are now supported in gpg-agent's ssh-agent
protocol implementation. I tested this with all 3 curves and keys
generated by OpenSSH 5.9p1.
Using existing non-ssh generated keys will likely fail for now. To fix
this, the code should first undergo some more cleanup; then the fixes
are pretty straightforward. And yes, the data structures are way too
complicated.
* agent/command-ssh.c: Remove dirent.h.
(control_file_s): Add struct item.
(rewind_control_file): New.
(search_control_file): Factor code out to ...
(read_control_file_item): New.
(ssh_handler_request_identities): Change to iterate over entries in
sshcontrol.
--
Formerly we scanned the private key directory for matches of entries
in sshcontrol. This patch changes it to scan the sshcontrol file and
thus considers only keys configured there. The rationale for this is
that it is common to have only a few ssh keys but many private keys.
Even if that assumption does not hold true, the scanning of the
sshcontrol file is faster than reading the directory and only then
scanning the ssh control for each directory entry.
* agent/command-ssh.c (SSH_CONTROL_FILE_NAME): New macro to replace
the direct use of the string.
(struct control_file_s, control_file_t): New.
(open_control_file, close_control_file): New. Use them instead of
using fopen/fclose directly.
* agent/gpg-agent.c (main): Pass new envar gnupg_SSH_AUTH_SOCK_by to
an invoked process.
--
This environment variable is useful for debugging if
--use-standard-socket is used (which is the default since 2.1).
Commonly you should have this in your init script (e.g. ~/.bashrc):
unset GPG_AGENT_INFO
unset SSH_AGENT_PID
SSH_AUTH_SOCK="${HOME}/.gnupg/S.gpg-agent.ssh"
export SSH_AUTH_SOCK
The problem is that gpg-agent won't be able to override the
SSH_AUTH_SOCK envvar if gpg-agent has been invoked as
gpg-agent --enable-ssh-support --daemon /bin/bash
To fix this you should instead use this code in the init script:
unset GPG_AGENT_INFO
unset SSH_AGENT_PID
if [ ${gnupg_SSH_AUTH_SOCK_by:-0} -ne $$ ]; then
export SSH_AUTH_SOCK="${HOME}/.gnupg/S.gpg-agent.ssh"
fi
This will work in all cases and thus allows to start gpg-agent for
testing purposes with a different homedir and use this gpg-agent as an
ssh-agent. Example:
GNUPGHOME=$(pwd) gpg-agent --enable-ssh-support --daemon /bin/bash
gnupg_SSH_AUTH_SOCK_by is set to the PID of the exec-ed process and
thus will work safely if called recursively.
* configure.ac (HAVE_BROKEN_TTYNAME): New ac_define set for Android
systems.
* common/util.h (gnupg_ttyname): New macro. Change all callers of
ttyname to use this macro instead.
(ttyname) [W32]: Rename to _gnupg_ttyname and use also if
HAVE_BROKEN_TTYNAME is defined.
* common/simple-pwquery.c (agent_send_all_options): Keep on using
ttyname unless HAVE_BROKEN_TTYNAME is set. This is because this file
may be used standalone.
* scd/command.c (MAXLEN_SETDATA): New.
(cmd_setdata): Add option --append.
* agent/call-scd.c (agent_card_pkdecrypt): Use new option for long
data.
* scd/app-openpgp.c (struct app_local_s): Add field manufacturer.
(app_select_openpgp): Store manufacturer.
(do_decipher): Print a note for broken cards.
--
Please note that I was not able to run a full test because I only have
broken cards (S/N < 346) available.
* findkey.c (agent_public_key_from_file): Fix use of
gcry_sexp_build_array.
--
A test case leading to a segv in Libgcrypt is
gpg-connect-agent \
"READKEY 9277C5875C8AFFCB727661C18BE4E0A0DEED9260" /bye
The keygrip was created by "monkeysphere s", which has a comment.
gcry_sexp_build_array expects pointers to the arguments which is quite
surprising. Probably ARG_NEXT was accidentally implemented wrongly.
Anyway, we can't do anything about it and thus need to fix the check
the users of this function.
Some-comments-by: Werner Koch <wk@gnupg.org>
The asymmetric quotes used by GNU in the past (`...') don't render
nicely on modern systems. We now use two \x27 characters ('...').
The proper solution would be to use the correct Unicode symmetric
quotes here. However this has the disadvantage that the system
requires Unicode support. We don't want that today. If Unicode is
available a generated po file can be used to output proper quotes. A
simple sed script like the one used for en@quote is sufficient to
change them.
The changes have been done by applying
sed -i "s/\`\([^'\`]*\)'/'\1'/g"
to most files and fixing obvious problems by hand. The msgid strings in
the po files were fixed with a similar command.
* agent/trustlist.c (clear_trusttable): New.
(agent_reload_trustlist): Use new function.
(read_trustfiles): Require to be called with lock held.
(agent_istrusted): Factor all code out to ...
(istrusted_internal): new. Add ALREADY_LOCKED arg. Make sure the
table islocked. Do not print TRUSTLISTFLAG stati if called internally.
(agent_marktrusted): Replace calls to agent_reload_trustlist by
explicit code.
--
In contrast to pth, npth does not use recursive mutexes by default.
However, the code in trustlist.c assumed recursive locks and thus we
had to rework it.
GnuPG requires that options are given before other arguments. This
can sometimes be confusing. We now print a warning if we found an
argument looking alike a long option without being preceded by the
stop option. This is bug#1343.
* common/argparse.h (ARGPARSE_FLAG_STOP_SEEN): New.
* common/argparse.c (arg_parse): Set new flag.
* g10/gpg.c (main): Print the warning.
* agent/gpg-agent.c (main): Ditto.
* dirmngr/dirmngr.c (main): Ditto.
* g13/g13.c (main): Ditto.
* scd/scdaemon.c (main): Ditto.
* sm/gpgsm.c (main): Ditto.
* tools/gpg-connect-agent.c (main): Ditto.
* tools/gpgconf.c (main): Ditto.
This is not yet fully implemented. It will eventually allow to
support pinpad equipped readers which do not support variable length
pin lengths.
* agent/protect.c (parse_shadow_info): Add optional arg R_PINLEN and
parse pinlen info. Change all callers to pass NULL for it.
* agent/command.c: Include asshelp.h.
(cmd_pkdecrypt, cmd_genkey, cmd_preset_passphrase)
(pinentry_loopback): Use print_assuan_status for INQUIRE_MAXLEN.
For the shared code parts it is cumbersome to pass an error sourse
variable to each function. Its value is always a constant for a given
binary and thus a global variable makes things a lot easier than the
former macro stuff.
* common/init.c (default_errsource): New global var.
(init_common_subsystems): Rename to _init_common_subsystems. Set
DEFAULT_ERRSOURCE.
* common/init.h: Assert value of GPG_ERR_SOURCE_DEFAULT.
(init_common_subsystems): New macro.
* common/util.h (default_errsource): Add declaration.
* kbx/keybox-defs.h: Add some GPG_ERR_SOURCE_DEFAULT trickery.
Since there isn't a way to prompt the user to insert the smartcard when
pinentry-mode=loopback, return GPG_ERR_CARD_NOT_PRESENT instead of
GPG_ERR_NO_PIN_ENTRY.
* agent/divert-scd.c (ask_for_card): Return GPG_ERR_CARD_NOT_PRESENT
when pinentry-mode=loopback.
Fix pinentry-mode=loopback when cancelling an inquire from scdaemon.
This is similar to commit 4f21f8d but for both protocol command
cancellation and pinentry cancellation.
* agent/call-scd.c (agent_card_pkdecrypt): Check for
GPG_ERR_ASS_CANCELED.
(agent_card_pksign): Ditto.
* agent/gpg-agent.c (get_agent_ssh_socket_name): Use
INVALID_HANDLE_VALUE instead of 0.
(handle_signal) [!HAVE_W32_SYSTEM]: Don't define.
(handle_connections): Port Windows code to NPTH.
* dirmngr/dirmngr.c (handle_connections): Port Windows code to NPTH.
* g13/g13.c (handle_connections): Port Windows code to NPTH.
* scd/scdaemon.c (handle_connections): Port Windows code to NPTH.
This status message is used to inform the client of the maximum length
of an inquired passphrase and is used in pinentry-mode=loopback.
* agent/command.c (pinentry_loopback): Send the INQUIRE_MAXLEN status
message before doing the inquire.
* agent/gpg-agent.c (main): Supply omitted "break" statement for
lc-messages option. Otherwise, control would fall through to the
following oXauthority case and use the same value there.
Similar to commit 29af488 but also fixes PKDECRYPT and PKSIGN.
* agent/call-scd.c (agent_card_pkdecrypt): Check for GPG_ERR_CANCELED
when returning from the PKDECRYPT operation of scdaemon and cancel the
inquire.
(agent_card_pksign): Ditto.
(cancel_inquire): New.
* agent/gpg-agent.c, dirmngr/dirmngr.c, g13/g13.c, scd/scdaemon.c
(USE_GCRY_THREAD_CBS): New macro, defined if
GCRY_THREAD_OPTION_VERSION is 0.
(fixed_gcry_pth_init) [!USE_GCRY_THREAD_CBS]: Don't define.
(main) [!USE_GCRY_THREAD_CBS]: Do not install thread callbacks.
Neal H. Walfield