* scd/apdu.c (apdu_send_direct): Use lock_slot.
--
Cherry-pick the master commit of:
f808012ac2cf67ec563da178d963f300a7f2564d
With trylock_slot, it may return SW_HOST_BUSY. This may occur when
apdu_get_status is called by scd_update_reader_status_file.
Simply using lock_slot is much easier for user of apdu_send_direct.
GnuPG-bug-id: 5831
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (select_a_reader): Only SPRx32 is in the white list.
--
GnuPG-bug-id: 5644
Fixes-commit: 752422a792cecf459b37f517d634bcf272292b14
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (pcsc): Add flag context_valid, remove count.
(close_pcsc_reader): Use new flag instead of looking at magic context
value.
(pcsc_init): Set new flag.
(open_pcsc_reader): Use new flag.
(apdu_init): Clear new flag.
* scd/apdu.c: Remove assert.h. Replace all assert by log_assert.
--
The previous fix 192113552faa98f40cc91fe014ec55861474626c did not
help, thus the new hypothesis is that PC/SC might return a valid
context with the value -1. We now use a dedicated flag to track the
validity of the context.
The reference counting seems to be superfluous and is a relict due to
backporting from 2.3. Removed.
* scd/apdu.c: Include membuf.h.
(pcsc): Add reader_list field.
(open_pcsc_reader): Fill that field.
(apdu_get_reader_list): New.
* scd/command.c: Remove header ccid-driver.h.
(pretty_assuan_send_data): New.
(cmd_getinfo): Print all reader names.
--
Note that depending on the card backend (ccid or PC/SC) it might be
necessary to first send a reset followed by SERIALNO to get an updated
list of reader. Or well send KILLSCD.
The pretty printing of Assuan data lines does only work if you connect
direct to scdaemon because the wrapper in gpg-agent does not know
about this and combines the Assuan lines again.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/apdu.c (close_pcsc_reader): Don't ref-count if the context is
invalid.
(open_pcsc_reader): Compare the context against -1 which is our
indicator for an invalid context.
--
I got a crash report for Windows:
DBG: chan_0x000000a4 <- RESTART
DBG: chan_0x000000a4 -> OK
DBG: chan_0x000000a4 <- SERIALNO
DBG: open_pcsc_reader(portstr=(null))
reader slot 0: not connected
DBG: open_pcsc_reader => slot=0
DBG: enter: apdu_connect: slot=0
pcsc_connect failed: invalid PC/SC error code (0x6)
reader slot 0: not connected
DBG: leave: apdu_connect => sw=0x1000b
DBG: enter: apdu_close_reader: slot=0
DBG: enter: apdu_disconnect: slot=0
DBG: leave: apdu_disconnect => sw=0x0
Ohhhh jeeee: Assertion "pcsc.count > 0" in
close_pcsc_reader failed (...2.2.28/scd/apdu.c:817)
no smartcard reader was connected but the box might sport a virtual
reader. This patch should make it more robust.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/apdu.c (pcsc): New variable.
(struct reader_table_s): Remove pcsc.context from member.
(pcsc_get_status, connect_pcsc_card): Use pcsc.context.
(close_pcsc_reader): Release pcsc.context here with reference count.
(apdu_open_one_reader): Move API loading to ...
(pcsc_init): new.
(apdu_open_one_reader): Remove.
(apdu_open_reader): Call open_pcsc_reader instead of
apdu_open_one_reader.
(open_pcsc_reader): Call pcsc_init if needed. Call close_pcsc_reader
instead of pcsc_release_context. Make reader parsing more robust.
(apdu_init): Initialize pcsc.count and pcsc.context.
--
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
Backported-from-master: 1080e91efd60cb41c2d6dbafaee810e5967a3161)
The backport also adds some other chnages as described above.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/apdu.c (pcsc_cancel): New.
(pcsc_init): Load new function.
(connect_pcsc_card): Use it after a removed card error.
--
Backported-from-master: 8d81fd7c01e8dfacc719ff190f8e364014e32fdf
* scd/scdaemon.h (opt): Add field opcsc_shared.
* scd/scdaemon.c (opcscShared): New.
(opts): Add "--pcsc-shared".
(main): Set flag.
* scd/apdu.c (connect_pcsc_card): Use it.
(pcsc_get_status): Take flag in account.
* scd/app-openpgp.c (verify_chv2): Do not auto verify chv1 in shared
mode.
--
This option should in general not be used. The patch tries to limit
bad effects but using shared mode is somewhat dangerous depending on
the other PC/SC users.
(cherry picked from commit 5732e7a8e97cebf8e850c472e644e2a9b040836f)
* scd/ccid-driver.h (CCID_DRIVER_ERR_USB_*): New error codes.
* scd/apdu.h: New SW_HOST error codes.
* scd/apdu.c (host_sw_string): Print them
* scd/ccid-driver.c (map_libusb_error): New.
(ccid_open_usb_reader, bulk_in, abort_cmd): Map libusb error codes.
* scd/iso7816.c (map_sw): Map new codes to gpg-error.
--
Backport from master commit:
9a8d7e41bba1926158a21ebdda542241493ef983
This change will help to get low level error conditions from hipher
application code.
Signed-off-by: Werner Koch <wk@gnupg.org>
* common/logging.c (log_printhex): Chnage order of args. Make it
printf alike. Change all callers.
* configure.ac: Add -Wno-format-zero-length
--
This makes it consistent with modern libgpgrt logging and thus eases
back porting from newer GnuPG versions which use libgpgrt logging.
Signed-off-by: Werner Koch <wk@gnupg.org>
* scd/apdu.c (apdu_send_direct): New arg R_SW.
* scd/command.c (cmd_apdu): Ditto.
* scd/iso7816.c (iso7816_apdu_direct): New arg R_SW.
(iso7816_general_authenticate): New.
* scd/app-nks.c (get_chv_status, get_nks_version): Pass NULL for new
arg.
--
iso7816_general_authenticate will be used for the PIV card support.
The new arg to iso7816_apdu_direct and apdu_send_direct allows to get
the raw status word back without the need to handle an output buffer.
Signed-off-by: Werner Koch <wk@gnupg.org>
(cherry picked from commit 70bb5c7931598590b1acfae90bf4657f5911d2d3)
* scd/apdu.c (struct reader_table_s): Add field of current_state.
(new_reader_slot): Initialize current_state.
(pcsc_get_status): Keep the status in READER_TABLE array.
Return SW_HOST_NO_READER when PCSC_STATE_CHANGED.
* scd/scdaemon.c (handle_connections): Silence a warning.
--
To detect some change of card status, including suspend/resume
possibly, SCardGetStatusChange should be used keeping the
dwCurrentState field.
This change could improve situation for suspend/resume with Yubikey on
Windows. Even not, this is doing the Right Thing.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (open_ccid_reader): Fix error handling of ccid_get_atr.
* scd/app.c (select_application): Always kick the loop if new APP.
* scd/ccid-driver.c (ccid_open_usb_reader): Don't setup at open.
(ccid_slot_status): Setup interrupt transfer when !ON_WIRE.
--
We can use the interrupt transfer to be notified about card status
change. In this case, we don't need to issue PC_to_RDR_GetSlotStatus
command. This change improve the setup the notification; it should be
done after registration of APP.
When the setup is done just after opening the USB connection (before
issuing PC_to_RDR_IccPowerOn), a reader might notifies about no card
availability (because of not yet powered on), even though the card is
ready to be powered on.
GnuPG-bug-id: 3508
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (apdu_open_reader): Add an argument APP_EMPTY.
When CCID driver fails to open, try PC/SC if APP is nothing.
* scd/app.c (select_application): Supply arg if APP is nothing.
--
After scanning available card readers by CCID driver, scdaemon should
try PC/SC service if no APP is registered yet. Also, when the slot
is allocated for PC/SC (ccid.handle==NULL), it should not call
ccid_compare_BAI, otherwise scdaemon crashes.
Debian-bug-id: 852702, 854005, 854595, 854616
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (get_status_reader): Add ON_WIRE arg, here.
(ct_get_status, pcsc_get_status_direct, pcsc_get_status_wrapped)
(pcsc_get_status, get_status_ccid, my_rapdu_get_status): Likewise.
(reset_pcsc_reader_wrapped, open_pcsc_reader_wrapped): Follow the
change.
(apdu_get_status_internal): It's lower-level driver which judge
it's not needed. Otherwise, it can't detect the removal.
* scd/ccid-driver.c (ccid_slot_status): After the POWERED_OFF check,
we can skip sending GetSlotStatus packet on wire, when no need.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
Fixes-commit: c8eee4d396a751d41fd1ee1e1b87b851fca172a
* scd/apdu.c(struct reader_table_s): Add require_get_status.
(apdu_connect): Change return value meaning. Call apdu_reset here.
* scd/app.c (app_new_register): Add require_get_status.
(select_application): Use the return value of apdu_connect.
(scd_update_reader_status_file): Call update_fdset_for_usb with
checking all_have_intr_endp.
(app_list_start, app_list_finish): Remove.
* scd/ccid-driver.c (struct ccid_driver_s): Add transfer.
(intr_cb): Don't call libusb_transfer in this callback.
(ccid_require_get_status): New.
(do_close_reader): Call libusb_transfer here.
* scd/scdaemon.c (update_fdset_for_usb): Remove the first argument.
--
With Gnuk Token, it works fine as expected. With Gemalto reader,
intr_cb is not called when card is removed. So, the macro
LIBUSB_WORKS_EXPECTED_FOR_INTERRUPT_ENDP is not defined yet.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (apdu_dev_list_finish): Don't call ccid_dev_scan_finish
with no table.
(apdu_open_reader): Only increment when it's zero.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (new_reader_slot): Lock is now in apdu_dev_list_start.
(close_pcsc_reader_direct, close_ccid_reader): RDRNAME is handled...
(apdu_close_reader): ... by this function now.
(apdu_prepare_exit): Likewise.
(open_ccid_reader): Open with dev_list.
(apdu_dev_list_start, apdu_dev_list_finish): New.
(apdu_open_one_reader): New.
(apdu_open_reader): Support multiple readers.
* scd/app.c (select_application): With SCAN, opening all readers
available, and register as new APP.
(app_write_learn_status): app->ref_count == 0 is valid for APP which is
not yet used.
(app_list_start, app_list_finish): New.
* scd/ccid-driver.c (struct ccid_driver_s): Remove RID and BCD_DEVICE.
Add BAI.
(parse_ccid_descriptor): BCD_DEVICE is now on the arguments.
(ccid_dev_scan, ccid_dev_scan_finish): New.
(ccid_get_BAI, ccid_compare_BAI, ccid_open_usb_reader): New.
(ccid_open_reader): Support multiple readers.
(ccid_set_progress_cb, ccid_close_reader): No RID any more.
--
With this change, multiple readers/tokens are supported by the internal
CCID driver of GnuPG. Until the changes of upper layers (scdaemon,
gpg-agent, and gpg front end), only a single reader is used, though.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (reader_table_lock, apdu_init): New.
(new_reader_slot): Serialize by reader_table_lock.
* scd/app.c (lock_app, unlock_app, app_new_register): Fix error code
usage.
(initialize_module_command): Call apdu_init.
* scd/scdaemon.c (main): Handle error for initialize_module_command.
--
This is a long standing bug. There are two different things; The
serialization of allocating a new SLOT, and the serialization of using
the SLOT. The latter was implemented in new_reader_slot by lock_slot.
However, the former was not done. Thus, there was a possible race where
a same SLOT is allocated to multiple threads.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (struct reader_table_s): Remove any_status, last_status,
status, and change_counter field.
(new_reader_slot, dump_reader_status, ct_activate_card, open_ct_reader)
(connect_pcsc_card, open_pcsc_reader_direct, open_pcsc_reader_wrapped)
(open_ccid_reader, apdu_reset): Follow the change.
(ct_dump_reader_status): Remove.
(apdu_get_status_internal, apdu_get_status): Remove CHANGED arg.
(apdu_connect): Follow the change.
* scd/command.c (struct vreader_s): Remove reset_failed, any, and
changed field.
(cmd_getinfo, update_reader_status_file): Follow the change.
--
In the past, scdaemon monitors card insertion (as well as removal), so
the code has been complicated, and there has been duplication in two
layers. Now, it only monitors card removal, it's now simplified.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (send_le): Fix the size, adding two for status
bytes to Le.
--
This is long standing bug. So far, Le was not exact value.
Since forthcoming change will introduce exact value of expected length
of response data, this change is needed.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (pcsc_vendor_specific_init): Only check vender ID.
--
Some other products by Cherry works with pinpad, although it only works
for smaller keys (RSA 1024). TPDU support is good for larger keys.
Signed-off-by: NIIBE Yutaka <gniibe@fsij.org>
* scd/apdu.c (pcsc_error_to_sw): map PCSC_E_NO_SERVICE and
PCSC_E_SERVICE_STOPPED to the internal SW_HOST_NO_READER error code.
--
Signed-off-by: Daniel Hoffend <dh@dotlan.net>
GnuPG-bug-id: 2167
In Windows 8 (and later), PC/SC service only runs when reader/token is
plugged in. After its removal, it returns PCSC_E_NO_SERVICE error.
This error should be handled as no reader. This comment is by gniibe.
* scd/apdu.c (apdu_close_reader): Call CLOSE_READER method even if we
got an error from apdu_disconnect.
* scd/app-common.h (no_reuse): Remove.
* scd/app.c (application_notify_card_reset): Deallocate APP here.
(select_application, release_application): Don't use NO_REUSE.
--
Reproducible scenario: Invoke gpg --card-edit session from a terminal.
Invoke another gpg --card-edit session from another. Remove a token.
Insert a token again. Type RET on both terminals. One of terminal
answers "Conflicting usage".
Perhaps, having NO_REUSE field was to avoid race conditions. Now,
APP can be safely deallocated by application_notify_card_reset.
Thanks to the2nd.
