aboutsummaryrefslogtreecommitdiffstats
path: root/agent (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* agent: Extend agent_card_getattr with KEYGRIP.NIIBE Yutaka2020-01-133-5/+11
| | | | | | | | | | * agent/agent.h (struct card_key_info_s): KEYGRIP null terminated. (agent_card_getattr): Add KEYGRIP argument. * agent/call-scd.c (agent_card_getattr): Handle KEYGRIP argument. (card_keyinfo_cb): Make KEYGRIP null terminated. * agent/command.c (cmd_readkey): Follow the change. Signed-off-by: NIIBE Yutaka <[email protected]>
* scd: Use a scdaemon internal key to protect the PIN cache IPC.Werner Koch2020-01-092-65/+49
| | | | | | | | | | | | | | | | | | | * agent/call-scd.c (handle_pincache_put): Do not decrypt. (handle_pincache_get): New. (inq_needpin): Call it. * scd/command.c (set_key_for_pincache): New. (pincache_put): Restructure and set key. (pincache_get): Ditto. -- This change does away with encrypting and decrypting the cached PIN in the agent. There is no need for this we can simply cache the cryptogram and let scdaemon decrypt after retrieving it from the agent. This way we do not need to share a key between gpg-agent and scdaemon and further the cached content is invalid after the scdaemon process has been replaced. Signed-off-by: Werner Koch <[email protected]>
* agent: SSH: SCD KEYINFO to list available keys.NIIBE Yutaka2020-01-095-116/+39
| | | | | | | | | | | | | | * agent/agent.h (agent_card_cardlist): Remove. (agent_card_keyinfo): Add CAP argument. * agent/call-scd.c (card_cardlist_cb): Remove. (agent_card_cardlist): Remove. (agent_card_keyinfo): Support CAP constraint. * agent/command-ssh.c (card_key_list): Remove. (ssh_handler_request_identities): Use SCD KEYINFO command. * agent/command.c (cmd_keyinfo): Follow the API change. * agent/divert-scd.c (ask_for_card): Likewise. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: First changes to support a PIN cache for scdaemon.Werner Koch2020-01-074-29/+217
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * agent/agent.h (CACHE_MODE_PIN): New. * agent/cache.c (housekeeping): Special handling of new new mode. (agent_flush_cache): Ditto. Add arg 'pincache_only' and change caller. (agent_put_cache): Support new mode. (agent_get_cache): Ditto. * agent/call-scd.c (wait_child_thread): Flush the entire PIN cache. (start_scd): Ditto. (agent_card_killscd): Ditto. (handle_pincache_put): New. Uses a dummy encryption key for now. (pincache_put_cb): New. (inq_needpin): Prepare for PINCACHE_GET inquiry. (learn_status_cb): Handle the PINENTRY_PUT status line. (get_serialno_cb): Ditto (agent_card_pksign): Ditto. (padding_info_cb): Ditto. (agent_card_readcert): Ditto. (agent_card_readkey): Ditto. (agent_card_writekey): Ditto. (card_getattr_cb): Ditto. (card_cardlist_cb): Ditto. (card_keyinfo_cb): Ditto. (pass_status_thru): Ditto. -- Take care: This is not finished. Signed-off-by: Werner Koch <[email protected]>
* agent: Improve --debug-pinentry diagnosticsWerner Koch2019-11-251-19/+35
| | | | | | | | | | | | | | * agent/call-pinentry.c (atfork_cb): Factor code out to ... (atfork_core): new. -- We convey certain envvars directly via the environment to Pinentry and thus they don't show up in the Assuan logging. Because we better don't call a logging function in an atfork handle, this patch splits the code up and uses the same code to display what was done in at fork after the connection has been established. Signed-off-by: Werner Koch <[email protected]>
* dirmngr: Fixed typo in recently added diagnostic.Werner Koch2019-11-181-1/+1
| | | | --
* agent: Fix ask_for_card to allow a key on multiple cards.NIIBE Yutaka2019-09-051-83/+36
| | | | | | | | * agent/divert-scd.c (ask_for_card): Don't use SERIALNO to select card, but use KEYGRIP. GnuPG-bug-id: 4695 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Fix an error path of agent_get_confirmation.NIIBE Yutaka2019-08-061-8/+9
| | | | | | | * agent/call-pinentry.c (agent_get_confirmation): Make sure unlock_pinentry is always called. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Relax the handling of pinentry error for keyboard grab.NIIBE Yutaka2019-07-111-1/+11
| | | | | | | | * agent/call-pinentry.c (start_pinentry): It's not fatal when pinentry doesn't support no-grab/grab option. GnuPG-bug-id: 4587 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Close a dialog cleanly when gpg/ssh is killed for CONFIRM.NIIBE Yutaka2019-06-281-22/+60
| | | | | | | | | | | | | | | | * agent/call-pinentry.c (watch_sock_start): Factor out from do_getpin. (watch_sock_end): Likewise. (do_getpin): Use those functions. (agent_get_confirmation): Likewise. (popup_message_thread): Likewise. -- Pinentry's dialog for confirmation should be also closed cleanly, as well as the dialog for pin-input. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Allow TERM="".NIIBE Yutaka2019-06-041-1/+1
| | | | | | | | | | * agent/call-pinentry.c (start_pinentry): When TERM is none, don't send OPTION ttytype to pinentry. -- GnuPG-bug-id: 4137 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Add pinentry_loopback_confirm declaration.NIIBE Yutaka2019-06-041-2/+5
| | | | | | * agent/agent.h (pinentry_loopback_confirm): New. Signed-off-by: NIIBE Yutaka <[email protected]>
* g10,agent: Support CONFIRM for --delete-key.NIIBE Yutaka2019-06-042-1/+38
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * agent/call-pinentry.c (agent_get_confirmation): Add call of pinentry_loopback_confirm. (agent_popup_message_start): Likewise. (agent_popup_message_stop): Return if it's loopback mode. * agent/command.c (pinentry_loopback_confirm): New. * g10/call-agent.c (default_inq_cb): Support "CONFIRM" inquery when PINENTRY_MODE_LOOPBACK mode. (confirm_status_cb): New. (agent_delete_key): Supply confirm_status_cb to set the description string for confirmation. -- In the Assuan communication, we introduce new interaction: [gpg] [gpg-agent] --- CMD: PKDECRYPT --> <-- STATUS: SETDESC "..." <-- STATUS: SETOK "..." <-- STATUS: SETNOTOK "..." <-- INQUERY: CONFIRM 0/1 (0 for display, 1 for user query) --- INQUERY-result: --> <-- RESULT: ... GnuPG-bug-id: 3465 Signed-off-by: NIIBE Yutaka <[email protected]>
* Return better error code for some getinfo IPC commands.Werner Koch2019-06-031-3/+3
| | | | | | | | | | | | | * 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 <[email protected]>
* agent: Add A-flag for KEYINFO output for card.NIIBE Yutaka2019-05-291-4/+29
| | | | | | | | | | | | | | | * agent/command.c (do_one_keyinfo): Add ON_CARD argument to put A-flag. (cmd_keyinfo): Call agent_card_keyinfo to offer additional information if it's on card. -- This is a modification in gpg-agent, intended for better enum_secret_keys in gpg frontend. GnuPG-bug-id: 4244 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Make an MD encoding function more robust.Werner Koch2019-05-281-7/+12
| | | | | | | | | | | * agent/pksign.c (do_encode_md): Use ascii_tolower and avoid uninitalized TMP in the error case. -- This is just in case libgcrypt ever returns an algorithm name longer than 15 bytes. Signed-off-by: Werner Koch <[email protected]>
* agent: Remove unused agent_show_message.NIIBE Yutaka2019-05-282-72/+6
| | | | | | | | | | | | * agent/call-pinentry.c (agent_show_message): Remove. * agent/genkey.c (take_this_one_anyway): Rename from take_this_one_anyway2. Remove a dead path calling agent_show_message. (check_passphrase_constraints): Use take_this_one_anyway. -- Fixes-commit: 2778c6f8f40d73272075ce04c07097f65c94054e Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Stop scdaemon after reload when disable_scdaemon.NIIBE Yutaka2019-05-233-0/+13
| | | | | | | | | | * agent/call-scd.c (agent_card_killscd): New. * agent/gpg-agent.c (agent_sighup_action): Call agent_card_killscd. -- GnuPG-bug-id: 4326 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: For SSH key, don't put NUL-byte at the end.NIIBE Yutaka2019-05-211-2/+2
| | | | | | | | | | * agent/command-ssh.c (ssh_key_to_protected_buffer): Update the length by the second call of gcry_sexp_sprint. -- GnuPG-bug-id: 4502 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent,scd: Scan and load all public keys for availability.NIIBE Yutaka2019-05-161-7/+14
| | | | | | | * agent/divert-scd.c (ask_for_card): Scan by SERIALNO command. * scd/app-openpgp.c (do_with_keygrip): Make sure to load pubkey. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Support scdaemon operation using KEYGRIP.NIIBE Yutaka2019-05-155-34/+215
| | | | | | | | | | | | | | | | | | | | | * agent/agent.h (struct card_key_info_s): New. (divert_pksign, divert_pkdecrypt): New API. * agent/call-scd.c (card_keyinfo_cb): New. (agent_card_free_keyinfo, agent_card_keyinfo): New. * agent/divert-scd.c (ask_for_card): Having GRIP argument, ask scdaemon with agent_card_keyinfo. (divert_pksign, divert_pkdecrypt): Ditto. * agent/pkdecrypt.c (agent_pkdecrypt): Supply GRIP. * agent/pksign.c (agent_pksign_do): Ditto. -- We are going to relax the requirment for SERIALNO of card. It's OK, when a card doesn't have recorded SERIALNO. If a card has a key with GRIP, it can be used. GnuPG-bug-id: 2291, 4301 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Replace most assert by log_assert.Werner Koch2019-05-1417-60/+42
| | | | --
* agent: correct length for uri and comment on 64-bit big-endian platformsDaniel Kahn Gillmor2019-05-141-2/+5
| | | | | | | | | | | | | | | | | | | | | | | | | * agent/findkey.c (agent_public_key_from_file): pass size_t as int to gcry_sexp_build_array's %b. -- This is only a problem on big-endian systems where size_t is not the same size as an int. It was causing failures on debian's s390x, powerpc64, and sparc64 platforms. There may well be other failures with %b on those platforms in the codebase, and it probably needs an audit. Once you have a key in private-keys-v1.d/$KEYGRIP.key with a comment or a uri of reasonable length associated with it, this fix can be tested with: gpg-agent --server <<<"READKEY $KEYGRIP" On the failing platforms, the printed comment will be of length 0. Gnupg-bug-id: 4501 Signed-off-by: Daniel Kahn Gillmor <[email protected]>
* dirmngr: Add a CSRF expection for pm.meWerner Koch2019-05-091-2/+2
| | | | | | -- Also comment typo fix.
* agent: If a Label is make sure that label is part of the prompt.Werner Koch2019-05-071-0/+40
| | | | | | | | | | | | | | * agent/findkey.c (has_comment_expando): New. (agent_key_from_file): Modify DESC_TEXT. -- A Label entry in the keyfile is always set manually and thus we can assume that the user wants to have this label in the prompt. In case the prompt template does not demand a comment this patch appends a comment to thhe template. This is a common case for on-disk keys used by gpg. Signed-off-by: Werner Koch <[email protected]>
* agent: Allow the use of "Label:" in a key file.Werner Koch2019-05-072-25/+82
| | | | | | | | | | | | | | | | | | | | | | | | | | | | * agent/findkey.c (linefeed_to_percent0A): New. (read_key_file): Add optional arg 'keymeta' and change all callers. (agent_key_from_file): Prefer "Label:" over the comment for protected keys. -- If in the extended key format an item Label: This is my key is found, "This is my key" will be displayed instead of the comment intially recorded in the s-expression. This is pretty useful for the ssh keys because often there is only the original file name recorded in the comment. If no Label is found or it is empty the S-expression comment is used. To show more than one line, the standard name-value syntax can be used, for example: Label: The Ssh key <blank line> <space>I registered on fencepost. Signed-off-by: Werner Koch <[email protected]>
* agent: Put Token lines into the key files.Werner Koch2019-05-039-32/+121
| | | | | | | | | | | | | | | * agent/findkey.c (write_extended_private_key): Add args serialno and keyref. Write a Token line if that does not yet exist. (agent_write_private_key): Add args serialno and keyref and change all callers. (agent_write_shadow_key): Skip leading spaces. * agent/keyformat.txt: Improve extended key format docs. -- Noet that the extended key forma is the defaqult in 2.3. This patch is a first step to better handle tokens which carray the same key. Signed-off-by: Werner Koch <[email protected]>
* agent: Allow other ssh fingerprint algos in KEYINFO.Werner Koch2019-03-261-4/+20
| | | | | | | * agent/command.c (cmd_keyinfo): Allow for --ssh-fpr=ALGO. Default to the standard algo. Signed-off-by: Werner Koch <[email protected]>
* agent: Re-introduce --enable-extended-key-format.Werner Koch2019-03-063-4/+15
| | | | | | | | | | | | | * agent/gpg-agent.c (oEnableExtendedKeyFormat): Re-introduce. (parse_rereadable_options): Handle it in a special way. * agent/protect.c (agent_protect): Be safe and set use_ocb only to 1 or 0. * tools/gpgconf-comp.c: Add --enable-extended-key-format again. -- This is required for backward compatible with profiles. Signed-off-by: Werner Koch <[email protected]>
* agent: Default to extended key format.Werner Koch2019-03-061-9/+13
| | | | | | | | | | | | | | | * agent/gpg-agent.c (oDisableExtendedKeyFormat, oNoop): New. (oEnableExtendedKeyFormat): Remove. (opts): Make --enable-extended-key-format a dummy option. Add disable-extended-key-format. (parse_rereadable_options): Implement oDisableExtendedKeyFormat. -- Extended key format is supported since vesion 2.1.12 which should have long been replaced by a newer version inh all installations. Thus for 2.3 we will make use of the extended-key-format by default. Signed-off-by: Werner Koch <[email protected]>
* agent: Fix detection of exit of scdaemon.NIIBE Yutaka2019-03-061-62/+80
| | | | | | | | | | | | | | * agent/call-scd.c (start_scd): Acquire START_SCD_LOCK for SCD_LOCAL_LIST. Move common case code to fast path. Release START_SCD_LOCK before calling unlock_scd. When new CTX is allocated, clear INVALID flag. (agent_reset_scd): Serialize the access to SCD_LOCAL_LIST by START_SCD_LOCK. -- GnuPG-bug-id: 4377 Signed-off-by: NIIBE Yutaka <[email protected]>
* scd:piv: Implement import of private keys for Yubikeys.Werner Koch2019-03-052-3/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | * scd/app-piv.c (concat_tlv_list): Add arg 'secure' and adjust callers. (writekey_rsa, writekey_ecc): New. (do_writekey): New. (do_writecert): Provide a better error message for an empty cert. (app_select_piv): Register do_writekey. * scd/iso7816.c (iso7816_send_apdu): New. * scd/app-common.h (APP_WRITEKEY_FLAG_FORCE): New. * agent/command.c (cmd_keytocard): Make the timestamp optional. * tools/card-call-scd.c (inq_writekey_parms): Remove. (scd_writekey): Rewrite. * tools/gpg-card.c (cmd_writekey): New. (enum cmdids): Add cmdWRITEKEY. (dispatch_command, interactive_loop): Call cmd_writekey. -- This has been tested with gpgsm and RSA keys. For ECC keys only partly tested using the sample OpenPGP nistp256 and nistp384 keys because gpgsm does not yet support ECC certificates and thus we can't write the certificates to the cert object after a writekey. Note that they nevertheless show up in "gpgcard list" because gpg-card searches for them in gpg and gpgsm. However, this does not work completely. Signed-off-by: Werner Koch <[email protected]>
* agent: Minor change to the KEYTOCARD command.Werner Koch2019-03-054-71/+56
| | | | | | | | | | * agent/command.c (cmd_keytocard): Make timestamp optional. Use modern parser function. * agent/call-scd.c (agent_card_writekey): Rename an arg and for clarity return gpg_error_t instead of int. * agent/divert-scd.c (divert_writekey): Ditto. Signed-off-by: Werner Koch <[email protected]>
* agent: PKSIGN should return signature in same format for card.NIIBE Yutaka2019-02-271-46/+36
| | | | | | | | | | | | | | * agent/pksign.c (agent_pksign_do): -- It's best to keep same data format by libgcrypt. For card (due to historical reasons), gpg-agent or scdaemon used to prefix 0x00 when it starts 0x80, so that it can be parsed signed MPI as well as unsigned MPI. It used to do nothing for preceding zeros. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Fix for suggested Libgcrypt use.Werner Koch2019-02-251-1/+16
| | | | | | | | | | | | * agent/divert-scd.c (divert_pkdecrypt): Skip a flags parameter. -- The libgcrypt docs say that a "flags" parameter should always be used in the input of pkdecrypt. Thus we should allow that parameter also when parsing an s-expression to figure out the algorithm for use with scdaemon. Signed-off-by: Werner Koch <[email protected]>
* agent: Terminate pinentry process gracefully, by watching socket.NIIBE Yutaka2019-02-191-4/+81
| | | | | | | | | | | | | | * agent/call-pinentry.c (watch_sock): New. (do_getpin): Spawn the watching thread. -- While we don't have npth_cancel (and it's difficult to implement it correctly), this is a kind of best compromise allowing a thread's polling when pinentry is active. GnuPG-bug-id: 2011 Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Minor change for pinentry status handling.NIIBE Yutaka2019-02-191-11/+7
| | | | | | | | * agent/call-pinentry.c (struct entry_parm_s): Add status. (do_getpin): Use param->status. (agent_askpin): Copy param->status. to pininfo. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Factor out the getpin interaction.NIIBE Yutaka2019-02-191-42/+37
| | | | | | | * agent/call-pinentry.c (do_getpin): New. (agent_askpin, agent_get_passphrase): Use do_getpin. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Clear bogus pinentry cache, when it causes an error.NIIBE Yutaka2019-01-283-21/+39
| | | | | | | | | | | | | | | | * agent/agent.h (PINENTRY_STATUS_*): Expose to public. (struct pin_entry_info_s): Add status. * agent/call-pinentry.c (agent_askpin): Clearing the ->status before the loop, let the assuan_transact set ->status. When failure with PINENTRY_STATUS_PASSWORD_FROM_CACHE, it returns soon. * agent/findkey.c (unprotect): Clear the pinentry cache, when it causes an error. -- GnuPG-bug-id: 4348 Signed-off-by: NIIBE Yutaka <[email protected]>
* gpg: Move S2K encoding function to a shared file.Werner Koch2019-01-261-3/+1
| | | | | | | | | | | | | | | | * g10/passphrase.c (encode_s2k_iterations): Move function to ... * common/openpgp-s2k.c: new file. Remove default intialization code. * common/openpgpdefs.h (S2K_DECODE_COUNT): New to keep only one copy. * g10/call-agent.c (agent_get_s2k_count): Change to return the count and print an error. * agent/protect.c: Include openpgpdefs.h * g10/card-util.c (gen_kdf_data): Adjust for changes * g10/gpgcompose.c: Include call-agent.h. (sk_esk): Adjust for changes. * g10/passphrase (passphrase_to_dek): Adjust for changes. * g10/main.h (S2K_DECODE_COUNT): Remove macro. Signed-off-by: Werner Koch <[email protected]>
* agent: Support --mode=ssh option for CLEAR_PASSPHRASE.NIIBE Yutaka2019-01-251-8/+10
| | | | | | | | | * agent/command.c (cmd_clear_passphrase): Add support for SSH. -- GnuPG-bug-id: 4340 Signed-off-by: NIIBE Yutaka <[email protected]>
* common: Extend function percent_data_escape.Werner Koch2019-01-241-1/+1
| | | | | | | | | | | | * common/percent.c (percent_data_escape): Add new args prefix and plus_escape. * agent/command.c (cmd_put_secret): Adjust for changed function * common/t-percent.c (test_percent_data_escape): Extend test for the prefix. (test_percent_data_escape_plus): new test for the plus escaping. Signed-off-by: Werner Koch <[email protected]>
* ssh: Simplify the curve name lookup.Werner Koch2019-01-171-84/+54
| | | | | | | | | | | | | | | | | | | | | * agent/command-ssh.c (struct ssh_key_type_spec): Add field alt_curve_name. (ssh_key_types): Add some alternate curve names. (ssh_identifier_from_curve_name): Lookup also bey alternative names and return the canonical name. (ssh_key_to_blob): Simplify the ECDSA case by using gcry_pk_get_curve instead of the explicit mapping. (ssh_receive_key): Likewise. Use ssh_identifier_from_curve_name to validate the curve name. Remove the reverse mapping because since GnuPG-2.2 Libgcrypt 1.7 is required. (ssh_handler_request_identities): Log an error message. -- This change will make it easier to support other curves, in particular those from tokens. Libgcrypt has a large list of alias names which we now use to to make the mapping more flexible. Signed-off-by: Werner Koch <[email protected]>
* agent: Make the S2K calibration time runtime configurabe.Werner Koch2018-12-113-7/+34
| | | | | | | | | | | | | | | | | | * agent/protect.c (s2k_calibration_time): New file global var. (calibrate_s2k_count): Use it here. (get_calibrated_s2k_count): Replace function static var by ... (s2k_calibrated_count): new file global var. (set_s2k_calibration_time): New function. * agent/gpg-agent.c (oS2KCalibration): New const. (opts): New option --s2k-calibration. (parse_rereadable_options): Parse that option. -- Note that using an unrelistic high value (like 60000) takes quite some time for calibration. GnuPG-bug-id: 3399 Signed-off-by: Werner Koch <[email protected]>
* scd: Add strerror to new error message.Werner Koch2018-11-301-1/+1
| | | | | | * agent/call-scd.c (wait_child_thread): Add %s. Signed-off-by: Werner Koch <[email protected]>
* agent: Better serialization for scdaemon access.NIIBE Yutaka2018-11-271-19/+16
| | | | | | | | | | | | | * agent/call-scd.c (unlock_scd): Move lock before accessing IN_USE. (wait_child_thread): Add log_info for Windows, and fixed log_error message. -- The old code is still valid with cooperate threads, but this is better. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Have a thread to wait for the child process of scdaemon.NIIBE Yutaka2018-11-263-89/+97
| | | | | | | | * agent/call-scd.c (wait_child_thread): New. (start_scd): Create a thread for wait_child_thread. (agent_scd_check_aliveness): Remove. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Defer calling assuan_release when it's still in use.NIIBE Yutaka2018-11-261-22/+40
| | | | | | | | | | * agent/call-scd.c (struct scd_local_s): Remove LOCK, introduce IN_USE and INVALID flags. (unlock_scd): Call assuan_release when CTX is invalid. (start_scd): Set IN_USE. (agent_scd_check_aliveness): Don't call assuan_release when it's in use. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Clean up SCDaemon management.NIIBE Yutaka2018-11-261-8/+1
| | | | | | | | * agent/call-scd.c (struct scd_local_s): Remove ctrl_backlink. (start_scd): Don't assign to the field. (agent_scd_check_aliveness): Fix typo in comment. Signed-off-by: NIIBE Yutaka <[email protected]>
* agent: Simplify agent_popup_message_stop.NIIBE Yutaka2018-11-141-8/+0
| | | | | | | | | | | | | | | | | | | | | | * agent/call-pinentry.c (agent_popup_message_stop): Just kill it. -- By checking if it's alive or not, we can lower a risk of sending SIGINT to a wrong process on unusual condition when PID is re-used to a different process. That's true, however, since it's alive usually, simply sending SIGINT is enough here. Note that here is a race condition for detecting if process is active or not; A process can die just after being detected alive. Moreover, when the process of pinentry accidentally died already, it should have caused return of assuan_transact and the thread of popup_message_thread likely already set popup_finished=1. Signed-off-by: NIIBE Yutaka <[email protected]>
generated by cgit v1.2.3 (git 2.25.1) at 2025-06-14 19:27:32 +0000