Commit e5707b545e broke signing using minidriver on Windows.
More specifically changing #define SC_ALGORITHM_RSA_PAD_NONE from 0x00000000 to 0x00000001 caused a call to sc_pkcs1_encode() to fail as the padding algorithm was not specified anywhere in the CardSignData() implementation. It kind of worked as long as SC_ALGORITHM_RSA_PAD_NONE was 0x00000000, but the above mentioned commit broke this.
Now padding algorithm has to be explicitly specified, otherwise a call to sc_pkcs1_encode() will fail.
- rename 'keytype' in some OpenPGP-specific types to 'key_id'
because they key ID was what the field was used for
- introduce field 'algorithm' in the structures above
to indicate the key's algorithm: RSA, ...
- define constant SC_OPENPGP_KEYALGO_RSA and use it
- rename constants SC_OPENPGP_KEYFORMAT_* to SC_OPENPGP_KEYFORMAT_RSA_*
because they are RSA specific
List additional algorithms & attributes as supported only when the card
supports changing the algorithms attributes DOs and exposes this by having
the EXT_CAP_ALG_ATTR_CHANGEABLE capability set.
Using different algorithms and attributes requires changing the algorithm
attributes DOs. If that is not supported - as indicated by a missing
EXT_CAP_ALG_ATTR_CHANGEABLE capability - then only those algorithms
described by the current algorithms attributes DOs' contents can be used.
In addition simplify setting the flags.
* use variables if they are already there
* be a bit more explicit in logging
* more consistent tag format: %04X
* cleanup flag setting for _sc_card_add_rsa_alg()
card-piv.c
make sure the string is null terminated before passing it
to hex_to_bin routine, which expects it
pkcs15-cac.c
free cn_name on failure
pkcs11-tool.c
make sure the string is null terminated before passing it to
parse_certificate(), which expects it
A card driver may declare support for computing the padding on the card,
or else the padding will be applied locally in padding.c. All five
PKCS11 PSS mechanisms are supported, for signature and verification.
There are a few limits on what we choose to support, in particular I
don't see a need for arbitrary combinations of MGF hash, data hash, and
salt length, so I've restricted it (for the user's benefit) to the only
cases that really matter, where salt_len = hash_len and the same hash is
used for the MGF and data hashing.
------------------------------------------------------------------------
Reworked and extended in 2018 by Jakub Jelen <jjelen@redhat.com> against
current OpenSC master, to actually work with existing PIV cards:
* extended of missing mechanisms (SHA224, possibility to select MGF1)
* compatibility with OpenSSL 1.1+
* Removed the ANSI padding
* Formatting cleanup, error checking
Based on the original work from
https://github.com/NWilson/OpenSC/commit/42f3199e66
Signed-off-by: Jakub Jelen <jjelen@redhat.com>
modified: src/libopensc/pkcs15-infocamere.c
modified: src/libopensc/pkcs15-starcert.c
modified: src/pkcs15init/pkcs15-lib.c
Changed isf_acl to also need SO PIN for CREATE.
modified: src/pkcs15init/starcos.profile
* localize variables
* print errors to stderr
* release allocated resources
* return error code on error - improve non-interactive use
* do not show help on unknown commands when used non-interactively
* make 'interactive' a global variable
* set it when opensc was called with the SCRIPT argument
* document the behaviour in the manual page
Make interactive a global variable and set it in main.
When arguments are given, compare them like ambguous_match() does,
and show the matching ones only.
Add documentation of the 'help' command to the manual page.
In main loop on multiple matches, show help on matching commands only.
* most importantly: immediately return success on exact match
- this allows one command to be a prefix of another one
- it fixes the long-standing breakage between 'find' and 'find_tags'
* fail on second prefix-only match instead of waiting until the end
* check all parameters
* add comments
* inform caller on whether the match was ambiguous or there was no match
* move printing error messages to processing loopt in main()
Accept a file name as a second argument to the 'random' command
to allow storing the generated random bytes to the file given.
Forbid writing binary data to stdout in interactive mode.
* limit buffer to SC_MAX_EXT_APDU_BUFFER_SIZE
* fix buffer length calculation to correctly calculate the available space
* add length checks when parsing passed data into buffer
* consistently show ellipsis for repeatable arguments as 3 dots
* embrace alternative mandatory arguments with curly braces
* use hyphens instead of spaces within non-literal arguments
for improved alignment with the manual page
Sc_asn1_read_tag can return SC_ERROR_ASN1_END_OF_CONTENTS
which indicates the tag and length are OK, but any value
is not completely contained in the buffer supplied. card-piv.c
can use this when reading just the beginning of a object to
determine the size of a buffer needed to hold the object.
As most of the drivers do not modify these, we can mark them as const.
Two drivers that we cannot convert are dnie and masktech.
section size
.data 35232 -> 25472
.data.rel.ro 36928 -> 46688
Remove aid_file and aidfile variables in card-piv.c. These are not needed
as piv_select_aid parses the returned data from a SELECT AID command.
In response to e-mail from X41 group on 6/11/2018.
On branch x41-piv-2
Changes to be committed:
modified: card-piv.c
md_pinpad_dlg_allow_cancel now defines whether or not the user is asked
before verifying the PIN on the PIN pad. This can be denied without
interaction with the PIN pad. A checkbox in the dialog allows the user
to change this setting, which is saved in the registry by the path of
the process.
This change fixes the progress bar to match the actual configured
timout. The progressbar now fills instead of running empty, which seemed
less frightening for most users.
This change also fixes some copy/paste errors in the documentation of
opensc.conf(5).
In Minidriver, when the DLL is called in multiple threads, this can
lead to a deinitialization of OpenSSL's OIDs in one thread making them
unavailable from other threads of the same process. As result, CVCs
cannot be veriefied anymore during chip authentication.
This removes code related to the old CAC 1 specification, while
preserving the CAC 2 functionality including CAC Alt token detection
for the tokens without CCC or ACA.
The detection based on SELECT APPLET APDU is improved to require also
the READ BUFFER APDU working, which should fail on misbehaving Java cards.
macOS 10.13 ships with ccid driver 1.4.27 (fixed in 1.4.29) and this version identifies these readers wrongly as pinpad readers.
Signed-off-by: Raul Metsma <raul@metsma.ee>
... for "standard" OpenPGP cards.
This gives more detailed information to the user on the detailed specs
the card adheres to.
In addition it fixes a long-standing annoyance that every standard 2.x
card matching the v2.0 ATR was announced as CryptoStick 1.2.
This ATR is not only used in the CryptoStick 1.2, but also also in
ZeitControl cards as well as NitroKeys, ...
To help debugging,
- replace plain return's after LOG_FUNC_CALLED()
has been called with LOG_FUNC_RETURN()
- use LOG_FUNC_CALLED() & LOG_FUNC_RETURN() pairs more often
Structure `x509cert_info` fields `subject` and `issuer`
are doubled in size up to 512 bytes.
We have to use dynamic memory allocation
to completely overcome the issue.
Relates to OpenSC/OpenSC#1412.
Make sure to have an empty line between information printed for individual
objects, but not in short mode.
This makes output of -D and -C more consistent.
This fixes a build failure with optimized ppc64el and new gcc builds
card-piv.c: In function ‘piv_validate_general_authentication.isra.3’:
card-piv.c:2390:9: error: ‘rbuflen’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
body = sc_asn1_find_tag(card->ctx, rbuf, rbuflen, 0x7c, &bodylen);
~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
'pkcs15-tool --read-ssh-key' is now able to read NIST ECC keys from card.
Only 256, 384 and 521 field lengths are supported (same as allowed in
ssh-keygen -t ecdsa). Issue #803 is partialy fixed by this patch.
Openssh PKCS11 interface patches for ECC are now available, please check
https://bugzilla.mindrot.org/show_bug.cgi?id=2474
* Refactor cac_properties_t structure to make its creation more readable
* Avoid manual allocation in cac_get_acr() and clean up bogus pointers
* Avoid bogus comments
* Properly check lengths of retrieved values
Previously, the code handled all the data objects as SimpleTLV,
which caused invalid encoding when we tried to merge TL + V buffers
into single PKCS#15 buffers.
This change is using GET PROPERTIES APDU after applet selection
to explore objects, figure out encoding and check the status of
PKI objects initialization to avoid reading them.
The previous solution was just guessing AIDs of the PKI objects
and trying if they answer.
This solution is inspecting card based on the Service Applet Table
(listing all the applets on the card) and using GET PROPERTIES APDU
listing all the available OIDs of the applet.
This was successfully tested with standard CAC card
(with different ACA AID) and uninitialized HID Alt tokens with empty
certificates slots.
If cardos cards are initialized by other software and there is a pinref
without the msb set, also the pin verify works without that bit set.
This patch changes pin initialisation so that the pin is created in mf
which has the effect that pin verify works without | 0x80 to the
pin ref.
Signed-off-by: Andreas Kemnade <andreas@kemnade.info>
- limit length of data to write even in raw mode to the real length
- cluster variuable definitions
- restrict scope of variables
- introduce a variable length to make the purpose more obious
- start preprocessor directives at column one
- add comments where needed
- harmonize coding style: space after "if" and casts
Put the arguments passed to option -d into an array instead of only
storing the latest value.
During output, iterate over the values passed in via the option.
* new function sc_compacttlv_find_tag()
Add function sc_compacttlv_find_tag() to search for a tag in a
compact-TLV structure.
* OpenPGP: use sc_compacttlv_find_tag()
While doing so, fix a typo affection OpenPGP v3.x cards
opensc-notify doesn't propose a GUI that can be displayed to the users,
so it doesn't make sense to display it in the application list/launcher
Fixes: #1379
Instead ogf blindly using "%lu", use "%"SC_FORMAT_LEN_SIZE_T"u"
to cope with the various implementations.
This fixes a bug introduced in commit 20b1d829
Instead of simply searching for a trigger byte with the risk of
getting garbage, correctly parse historical bytes from ATR as well
as the "historical bytes" DO as compact TLV structures.
In addition
- prepare for additional data used in OpenPGP cards 3.x
- ignore [per the spec] chaining capability fo OpenPGP cards 1.x
This is also done in the official AusweisApp2 and avoids confusion with
other tokens that also have an EF.CardAccess and are capable of
verifying the PIN via PACE.
Fixes https://github.com/OpenSC/OpenSC/issues/1360
Let sc_get_challenge() do sc_lock() and loop through the card driver's
get_challenge() until enough bytes were collected. The card driver's
get_challenge() now returns the number of bytes collected (less or equal
than requested) or an error code.
- Allow more code re-use.
- PIV driver now uses ASN.1 parser for reading the random bytes
sc_pkcs15_verify_pin say:
/* if pin cache is disabled, we can get here with no PIN data.
* in this case, to avoid error or unnecessary pin prompting on pinpad,
* check if the PIN has been already verified and the access condition
* is still open on card.
*/
It then call sc_pkcs15_get_pin_info
A context specific login is used in PKCS#11 to force the user
to enter the PIN again and a verify command be sent to the card.
(Actually it could be a different value for the PINi depending on the card)
sc_pkcs15_get_pin_info will then call the card driver, but does not
say why it is testing the login status.sc_pkcs15_get_pin_info may return
SC_PIN_STATE_LOGGED_IN=1 and sc_pkcs15_verify_pin will then skip sending
the actual verify command to the card via _sc_pkcs15_verify_pin
To avoid this, sc_pkcs15_get_pin_info will set data.pin_type = pin_info->auth_method;
In the case of a context specific login, this is SC_AC_CONTEXT_SPECIFIC
and the card driver can take action and can return SC_PIN_STATE_LOGGED_IN=0
so the verify will be done.
The PIV driver card-piv.c does this. Other drivers could do something similar.
Date: MOn May 21 20:40:00 2018 -0500
On branch History-fixes
Changes to be committed:
modified: card-piv.c
modified: pkcs15-pin.c
If a PIV card does not have or support a Discovery Object and
is known to lose the login state when the PIV AID is selected,
nothing was done in piv_card_reader_lock_obtained.
If was_reset > 0 select the PIV AID to at least get the
PIV AID selected.
For other cards either reading the Discovery a object and/or
selecting the PIV AID will make sure the PIV AID is selected.
If multiple applications are using the card, this will allow
the first one to select the AID, and any others that handle
a reset will not cause interference wit the first.
On branch History-fixes
Changes to be committed:
modified: card-piv.c
&& is replaced by || in the test of valid key references
for retired keys found in the Historic object.
For retired keys, the user_consent flag was being set by default.
Thus a C_Login(CKU_CONTEXT_SPECIFIC) would be required.
NIST 800-73 only requires PIN_Always on the Sign Key.
To extend the usefullnes of "retired keys" on non government
issued PIV-like cards, code had already been added
to use the certificate keyUsage flags to override the NIST
defined key usage flags. The NONREPUDATION flag is now used
to set the user_consent flag.
So rather then always requiring C_Login(CKU_CONTEXT_SPECIFIC)
for any retured key, the code only requires it for non government
cards where teh certificate has NONREPUDATION.
Changes to be committed:
modified: card-piv.c
modified: pkcs15-piv.c
* Previously, it was dependent on ATR blocks, but it did
not allow enrolling various types of cards without knowning
their ATR in advance.
* Improved documnetation for this option in configuration files
Resolves: #1265
Using the forced-driver prevents parsing of additional constructions
in configuration files (for example flags based on ATRs). This
implementation replaces transparently the existing list defined in
card_drivers.
Resolves: #1266
The following errors occured during a compilation using gcc 8:
In function »gids_create_file.constprop«,
inserted by »gids_save_certificate.isra.8« beicard-gids.c:1548:7:
card-gids.c:465:2: Error: »strncpy« output may be truncated copying 8 bytes from a string of length 8 [-Werror=stringop-truncation]
strncpy(record->filename, filename, 8);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
pkcs15-oberthur.c: In function »sc_pkcs15emu_oberthur_add_prvkey«:
pkcs15-oberthur.c:741:5: Error: »strncpy« output may be truncated copying 254 bytes from a string of length 254 [-Werror=stringop-truncation]
strncpy(kobj.label, objs[ii]->label, sizeof(kobj.label) - 1);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
* reader-pcsc: Do not temporarily set SC_READER_REMOVED on all readers
Fixes#1324.
* reader-cryptotokenkit: Do not temporarily set SC_READER_REMOVED on all readers
See #1324.
It was used to make pkcs11-tool work with vendor defined PKCS#11
modules. If this behavior is still desired, pass the define
ZERO_CKAID_FOR_CA_CERTS during the build
Don't pretend that we're capable of performing memory locking. The
implementation of that, `sc_mem_alloc_secure()` (also removed), was
almost unused anyway.
Workaround to not detect the MUSCLE applet as whatever other card driver
selects an AID first. MUSCLE applet will always return 9000, which will
confuse other card drivers. Since this bug is not going to go away any
time soon, we simply detect the MUSCLE applet first and hope that other
applets don't have a similar bug.
* Initial version of pkcs11 testsuite
* Refactor test cases to several files, clean up awful and unused stuff
* Static mechanism list based on the actual token offer
* Get rid of magic numbers
* Documentation
* License update based on the original project
* Verbose readme
* Cleanup unused code, long lines and method order
* Typo; More verbose errors
* Use fallback mechanisms
* Refactor object allocation and certificate search
* PKCS11SPY mentioned, more TODO
* add SHA mechanisms
* Do not try to Finalize already finalized cryptoki
* Add more flags and mechanisms
* Do not list table for no results
* Logical order of the tests (regression last)
* read ALWAYS_AUTHENTICATE from correct place
* ALWAYS_AUTHENTICATE for decryption
* Test EC key length signature based on the actual key length
* Shorten CKM_ list output, add keygen types detection
* Skip decrypting on non-supported mechanisms
* Fail hard if the C_Login fails
* Reorganize local FLAGS_ constants
* Test RSA Digest mechanisms
* Correct mechanisms naming, typos
* Do not attempt to do signature using empty keys
* CKM_ECDSA_SHA1 support
* Correct type cast when getting attributes
* Report failures from all mechanisms
* Standardize return values, eliminate complete fails, documentation interface
* Wait for slot event test
* Add switch to allow interaction with a card (WaitForSlotEvent)
* At least try to verify using C_Verify, if it fails, fall back to openssl
* Get rid of function_pointers
* Get rid of additional newline
* Share always_authenticate() function between the test cases
* Refactor Encrypt&decrypt test to functions
* Do not overwrite bits if they are not provided by CKA, indentation
* Cleanup and Break to more functions Sign&Verify test
* CKM_RSA_X_509 sign and verify with openssl padding
* More TODO's
* Proper abstracted padding with RSA_X_509 mechanism
* Add ongoing tasks from different TODO list
* Update instructions. Another todo
* Variables naming
* Increase mechanism list size, use different static buffers for flags and mechanism names
* nonstandard mechanism CKM_SHA224_RSA_PKCS supported by some softotkens
* Get rid of loop initial declarations
* Loop initial declaration, typos, strict warnings
* Move the p11test to the new folder to avoid problems with dynamically linked opensc.so
* Update path in README
* Possibility to validate the testsuite agains software tokens
* Add possibility to select slot ID on command-line (when there are more cards present)
* Clean up readme to reflect current options and TODOs
* Do not attempt to use keys without advertised sign&verify bits to avoid false positives
* Get and present more object attributes in readonly test; refactor table
* New test checking if the set of attributes (usage flags) is reasonable
* Test multipart signatures. There is not reasonable mechanism supporting multipart encryption
* Use PKCS#11 encryption if possible (with openssl fallback)
* Identify few more mechanisms (PSS) in the lest
* Resize table to fit new mechanisms
* Remove initial loop declaration from multipart test
* Use pkcs11-tool instead of p11tool form most of the operations (master have most of the features)
* Preparation for machine readable results
* Refactor log variables out of the main context, try to export generic data
* Do not write to non-existing FD if not logging
* Export missing data into the log file in JSON
* Store database in json
* Sanity check
* Avoid uninitialized structure fields using in state structure
* Dump always_authenticate attribute too
* Manual selection of slots with possibility to use slots without tokens
* Do not free before finalizing
* Proper cleanup of message in all cases
* Proper allocation and deallocation of messages
* Sanitize missing cases (memory leaks)
* Suppressions for testing under valgrind
* Better handling message_lengt during sign&verify (avoid invalid access)
* Suppress another PCSC error
* Do not use default PIN. Fail if none specified
* Sanitize initialization. Skip incomplete key pairs
* Add missing newline in errors
* Fix condition for certificate search
* Avoid several calls for attributes of zero length
* Handle if the private key is not present on the card
* Improve memory handling, silent GCC warning of 'unused' variable
* Fail early with missing private key, cleanup the messages
* Use correct padding for encryption
* Cache if the card supports Verify/Encrypt and avoid trying over and over again
* Loosen the condition for the Usage flags
* OpenSSL 1.1.0 compatibility
* Add missing mechanisms
* Do not require certificates on the card and pass valid data for RSA_PKCS mechanisms
* Add missing PIN argument in runtest.sh
* Add OpenSSL < 1.1 comatible bits
* Add SHA2 ECDSA mechanisms handling
* Use public key from PKCS#11 if the certificate is missing (or compare it with certificate)
* Avoid long definitions in OpenSSL compat layer
* In older OpenSSL, the header file is ecdsa.h
* Add missing config.h to apply compat OpenSSL layer
* ASN1_STRING_get0_data() is also new in 1.1.0
* Return back RSA_X_509 mechanism
* Drop bogus CKM_* in the definitions
* Drop CKM_SHA224_RSA_PKCS as it is already in pkcs11.h
* Update documentation
* Use NDEBUG as intended
* typos, cleanup
* Typos, cleanup, update copyright
* Additional check for OpenCryptoki, generate more key types on soft tokens
* Prepare for RSA-PSS and RSA-OAEP
* Use usage&result flags for the tests, gracefully ignore PSS&OAEP
* pkcs11.h: Add missing definitions for PSS
* PSS and OAEP tests
readonly: Typos, reformat
* Working version, memory leak
* Tweak message lengths for OAEP and PSS
* Skip tests that are not aplicable for tokens
* configure.ac: New switch --enable-tests
Do not attempt to build tests if cmocka is not available or
--enable-tests is provided. It makes also more lightweight release
builds out of the box (or with --disable-tests).
* travis: Install cmocka if not available
* Do not build tests on Windows and make dist pass
* Try to install cmocka from apt and from brew
* Do not require sudo (cmocka from apt and brew works)
When loaded as PKCS#11 module, OpenSC will reuse the application
identifier for each instance. We don't want to put any burdon on the
primary GApplication, so we use a non unique ID.
fixes https://github.com/OpenSC/OpenSC/issues/1332
The PKCS#15 emulation layer for the CAC uses a single PIN. Set its
label to "PIN" (rather than the card type "CAC I" or "CAC II"), so
that the PIN label will be omitted from the token label, providing
more space for the card holder name instead.
This is intended to match the behavior used for PIV cards, which
was changed with commit 56c8f59b25.
* Handle errors as intended in sc_pkcs15emu_openpgp_add_data()
If a data object can be read, but it cannot be added to the PKCS#15
framework, return from this function with an error; do not continue
reading other data objects. Otherwise, do not return an error from
this function when a data object cannot be read or is empty.
Improve existing comments for clarity.
* Address other compiler warnings when using --disable-optimization
Certain variables that are not initialized when they are declared
prevent the build from completing, when --disable-optimization is
passed to ./configure.
1. In epass2003_set_security_env, remove unused code, add condition
check of ec and rsa
2. Line 1709 - add return check of hash_data
3. In epass2003_decipher API, the old sign using apdu.le = 256, now add
condition check of rsa
4. Line 2731-2734 - After login successful, need get session status,
adjust code, improve condition check of data->cmd.
Remove all the code related to the old GUI and PIN PAD.
This code was initially developed by Zetes and had the ability to
display a dialog to request the PIN to the user. It was also able to
manage some specific proprietary pin pads.
As the Belgian government/fedict has now its own implementation, all
these old crufts can be removed.
https://github.com/OpenSC/OpenSC/issues/1296
Some ActivIdentity CAC/PIV cards lose the login state when selecting
the PIV AID SC_CARD_TYPE_PIV_II_CAC and CI_PIV_AID_LOSE_STATE were added
so piv_card_reader_lock_obtained will try and do a SELECT PIV AID.
card->type is reset to its original value if piv_match_card_continued
fails to match a card as PIV.
pkcs15-piv.c now uses sc_card_ctl which checks card->ops->card_ctl for NULL.
closes https://github.com/OpenSC/OpenSC/pull/1307
fixes https://github.com/OpenSC/OpenSC/issues/1297
We can't check for `tag == SC_ASN1_TAG_EOC` directly, because this
would also be true for a tag of 0x80 (with `class ==
SC_ASN1_CLASS_CONSTRUCTED`). So what we do is we check for the output
buffer to be NULL!
fixes https://github.com/OpenSC/OpenSC/issues/1273
1. Buffer underrun in epass2003_decipher().
2. The parameter `data' in update_secret_key() must be constant.
(Discovered by Clang 4.0.0 on OpenBSD 6.2.)
Fixes#1286. The behaviour of pkcs11-tool will follow the standard -
send DER. If EC_POINT_NO_ASN1_OCTET_STRING is defined then it will
write plain bytes.
ATR and differences between 3.4/3.5 are based on
https://github.com/CardContact/sc-hsm-embedded
Actually, 3.5 is untested, but 3.4 is almost identical, so we hope that
for 3.5 everything is fine.
Luka Logar
Peter Marschall
Frank Morgner
gabrielmuller
Jakub Jelen
Raul Metsma
Nicholas Wilson
konstantinpersidskiy
Priit Laes
Alon Bar-Lev
alex-nitrokey
alex-nitrokey
asc
Doug Engert
Jakub Jelen
Eugene Bright
Leif Erik Wagner
Gianfranco Costamagna
Stanislav Brabec
Peter Popovec
ytoku
Andreas Kemnade
Laurent Bigonville
Peter Marschall
rmartinc
Florian Bezdeka
David Ward
Feitian Technologies
Mouse
sergioag
AnthonyA
Vadim Penzin
Igor Castañeda Ferreira