summaryrefslogtreecommitdiff
path: root/p11-kit/modules.c
Commit message (Collapse)AuthorAgeFilesLines
* rpc: Add PKCS#11 module that connects to socketDaiki Ueno2017-02-171-0/+10
| | | | | | | | | | | | | | | This patch adds a PKCS#11 module that connects to the p11-kit server exposed on the filesystem. The filename of the socket is determined in the following order: - $P11_KIT_SERVER_ADDRESS, if the envvar is available - $XDG_RUNTIME_DIR/p11-kit/pkcs11, if the envvar is available - /run/$(id -u)/p11-kit/pkcs11, if /run/$(id -u) exists - /var/run/$(id -u)/p11-kit/pkcs11, if /var/run/$(id -u) exists - ~/.cache/p11-kit/pkcs11. Note that the program loading this module may have called setuid() and secure_getenv() which we use for fetching envvars could return NULL.
* build: Make libffi closure optionalDaiki Ueno2017-01-241-18/+8
| | | | | | | | libffi's closure support is not available on all platforms and may fail at run time if running under a stricter SELinux policy. Fallback to pre-compiled closures if it is not usable. https://bugs.freedesktop.org/show_bug.cgi?id=97611
* modules: Remove redundant NULL checkPankaj2016-12-131-1/+1
| | | | https://bugs.freedesktop.org/show_bug.cgi?id=93588
* modules: Reset the init count on fork()Daiki Ueno2016-10-311-0/+4
| | | | | | | | | | | | | | | Reset mod->init_count when forkid has changed. Otherwise C_Finalize does not get called. GCC's asan spotted this: Direct leak of 48 byte(s) in 1 object(s) allocated from: #0 0x7f89bc7bfe20 in malloc (/lib64/libasan.so.3+0xc6e20) #1 0x7f89bc47a1f1 in p11_dict_new ../common/dict.c:278 #2 0x7f89bc42143d in managed_C_Initialize ../p11-kit/modules.c:1477 #3 0x7f89bc464c72 in binding_C_Initialize ../p11-kit/virtual.c:121 #4 0x7f89bc1b0a51 in ffi_closure_unix64_inner (/lib64/libffi.so.6+0x5a51) #5 0x7f89bc1b0dbf in ffi_closure_unix64 (/lib64/libffi.so.6+0x5dbf) #6 0x7f89bc44f9e8 in rpc_C_Initialize ../p11-kit/rpc-server.c:691
* modules: Fix memleak when loading remote moduleDaiki Ueno2016-10-311-0/+2
| | | | | | | | | | | | Make sure to call p11_virtual_uninit() on managed module. Otherwise the associated lower_module will not be released. GCC's asan spotted this: Direct leak of 56 byte(s) in 1 object(s) allocated from: #0 0x7f6c5368dfe0 in calloc (/lib64/libasan.so.3+0xc6fe0) #1 0x4436ba in p11_rpc_client_init ../p11-kit/rpc-client.c:2082 #2 0x42c147 in p11_rpc_transport_new ../p11-kit/rpc-transport.c:850 #3 0x415d95 in setup_module_for_remote_inlock ../p11-kit/modules.c:411
* rpc: Send x-init-reserved to remote moduleDaiki Ueno2016-10-051-13/+17
| | | | | | | Signed-off-by: Stef Walter <stefw@redhat.com> * Fixed up indentation https://bugs.freedesktop.org/show_bug.cgi?id=80519
* Fix leak when C_Initialize() is called from childDaiki Ueno2016-08-221-0/+2
| | | | | | | The test case added for bug 90289 (commit c73edd00) revealed that some of the C_Initialize() implementations do not consider the case where it is called from the parent process and then from the child process, without calling C_Finalize() in between.
* Fix typos flagged by codespellDaiki Ueno2016-08-221-2/+2
|
* Doc: p11_kit_module_load accepts a filename arg.Andreas Metzler2016-02-241-1/+4
| | | | | | | | p11_kit_module_load() hands on the module_path argument to load_module_from_file_inlock() which accepts relative paths, prepending P11_MODULE_PATH. Update API documentation accordingly. https://lists.freedesktop.org/archives/p11-glue/2016-February/000587.html
* Avoiding redundant checkPankaj2016-01-051-3/+1
| | | | https://bugs.freedesktop.org/show_bug.cgi?id=93587
* p11-kit: Remove unused pointerPankaj2015-10-211-2/+0
| | | | https://bugs.freedesktop.org/show_bug.cgi?id=92532
* p11-kit: Fix warnings related to use dangling pointerPankaj2015-10-201-2/+6
| | | | https://bugs.freedesktop.org/show_bug.cgi?id=92551
* Added p11_kit_module_get_filename()Nikos Mavrogiannopoulos2015-07-141-0/+45
| | | | | | | | | | | That function allows to obtain the filename used by the PKCS #11 module. That is the filename used by dlopen(). Note that we don't provide p11_kit_module_for_filename() because it would have to deal with filename equivalences. Signed-off-by: Stef Walter <stefw@redhat.com> * Fixed up whitespace
* p11-kit: Use pthread_atfork() in a safe mannerStef Walter2014-10-031-43/+12
| | | | | | | | | | | | | Instead of trying to perform actions in pthread_atfork() which are not async-signal-safe, just increment a counter so we can later tell if the process has forked. Note this does not make it safe to mix threads and forking without immediately execing. This is a far broader problem that p11-kit, however we now do the right thing when fork+exec is used from a thread. https://bugs.freedesktop.org/show_bug.cgi?id=84567
* p11-kit: Remove the 'isolated' option for nowStef Walter2014-10-011-13/+0
| | | | | | This option was not completed in time, and as implemented suffers from limitations that the module is not really completely isolated as it still runs under the same user id as the calling process.
* p11-kit: Fix various noise/issues highlighted by clangStef Walter2014-08-111-1/+0
|
* p11-kit: Remove use after free in debug output code pathStef Walter2014-08-081-1/+1
|
* p11-kit: Tweak last commit, handle the not-forked caseStef Walter2014-08-071-1/+4
| | | | | When we hadn't forked, but were just not initialized, still return CKR_CRYPTOKI_NOT_INITIALIZED from managed modules.
* p11-kit: Don't complain about C_Finalize called in wrong processStef Walter2014-08-071-6/+13
| | | | | | When C_Finalize is called in the wrong process, it's often because of a caller unaware of forking. This is a painful area of PKCS#11, but at least for C_Finalize, lets not complain loudly about it.
* p11-kit: Add a new 'isolate' pkcs11 config optionStef Walter2014-07-081-12/+32
| | | | | | | This sets 'remote' appropriately to run the module in a separate process. https://bugs.freedesktop.org/show_bug.cgi?id=80472
* rpc: Implement execution of another tool to transport PKCS#11 RPCStef Walter2014-07-081-15/+55
|
* modules: Make config file and module configs overridable by testsStef Walter2014-07-081-4/+11
|
* p11-kit: Handle managed modules correctly when forkingStef Walter2014-07-041-5/+9
| | | | | | | | Correctly allow reinitialization when a process forks. We don't yet implement checks on all entry points of a managed module, but this allows callers to call C_Initialize again after forking, as outlined by the PKCS#11 v2 spec.
* p11-kit: Fix corrupted list when initialization of modules failStef Walter2014-06-251-0/+2
| | | | | | | | This fixes the function call p11_kit_module_initialize() to correctly rearrange the modules array when initializing a module fails. Also fixes p11_kit_modules_load_and_initialize()
* Fix various memory leaks exposed by 'make leakcheck'Stef Walter2013-07-231-0/+2
|
* p11-kit: Add P11_KIT_MODULE_TRUSTED flagStef Walter2013-07-041-17/+31
| | | | | A new flag to pass to p11_kit_modules_load() and related functions which limits loaded modules to ones with "trust-policy: yes".
* Bump the version for deprecated function documentationStef Walter2013-05-211-9/+9
|
* Further reorganization of the core module trackingStef Walter2013-05-211-188/+161
| | | | | | | | | * Keep the module ownership apart from the tracking of module function pointers, since these are only relevant for unmanaged modules. * Less assumptions that each module has a raw unmanaged module function pointer. * More clarity in the naming of dictionaries tracking the modules.
* Add the log-calls module config optionStef Walter2013-05-211-14/+30
| | | | | If 'log-calls = yes' is set then all the PKCS#11 modules are logged to stderr.
* Manage C_CloseAllSessions function for multiple callersStef Walter2013-05-211-4/+198
| | | | | | Make C_CloseAllSessions work for different callers. Track the sessions that each caller opens and close just those when C_CloseAllSessiosn is called.
* Update the proxy module to use managed PKCS#11 modulesStef Walter2013-05-211-2/+3
| | | | | | | | | | Each time C_GetFunctionList is called on the proxy module, a new managed PKCS#11 set of functions is returned. These are all cleaned up when the module is unloaded. We want the proxy module to continue to work even without the highly recommended libffi. For that reason we still keep the old behavior of sharing state in the proxy module.
* p11-kit: Managed PKCS#11 module loadingStef Walter2013-05-211-142/+1224
| | | | | | | | | Support a new managed style module loading for PKCS#11 modules. This allows us to better coordinate between multiple callers of the same PKCS#11 modules and provide hooks into their behavior. This meant redoing the public facing API. The old methods are now deprecated, marked and documented as such.
* Implement valgrind's hellgrind checks for threading problemsStef Walter2013-05-151-1/+1
| | | | | | | And cleanup our locks/locking model. There's no need to use recursive locks, especially since we can't use them on all platforms. In addition adjust taking of locks during initialization so that there's no chance of deadlocking here.
* More compatible path munging and handling codeStef Walter2013-04-031-35/+3
| | | | | | | | | | Centralize the path handling code, so we can remove unixy assumptions and have a chance of running on Windows. The current goal is to run all the tests on Windows. Includes some code from LRN <lrn1986@gmail.com> https://bugs.freedesktop.org/show_bug.cgi?id=63062
* Separate library init from message codeStef Walter2013-04-031-0/+1
| | | | | | | | | | | Put library init/uninit code its into their own statically linked library so that they don't get linked into the p11-kit executable. Refactor the message code so that the library initialization can plug in its per thread message buffer. https://bugs.freedesktop.org/show_bug.cgi?id=63046
* p11-kit: New priority option and change trust-policy optionStef Walter2013-03-151-0/+47
| | | | | | | | | | | | | | * Sort loaded modules appropriately using the 'priority' option. This allows us to have a predictable order for callers, when callers iterate through modules. * Modules default to having an 'priority' option of '0'. * If modules have the same order value, then sort by name. * The above assumes the role of ordering trust-policy sources. * Change the trust-policy option to a boolean * Some of this code will be rearranged when the managed branch is merged. https://bugs.freedesktop.org/show_bug.cgi?id=61978
* Rename p11_module_xxx() compat functions to p11_dl_xxx()Stef Walter2013-03-031-5/+10
| | | | | | | For clarity. In addition, make p11_dl_close() able to be used as a destroyer callback. Also make p11_dl_error() return an allocated string
* Add a /usr/share/p11-kit/modules directory for package module configsStef Walter2013-02-051-1/+3
| | | | | | * Try to make /etc/pkcs11/modules for administrator use * Override the old pkg-config variables to help packages start using the new location
* Use the stdbool.h C99 bool typeStef Walter2013-01-231-16/+16
| | | | | | It was getting really wild knowing whether a function returning an int would return -1 on failure or 0 or whether the int return value was actually a number etc..
* Move debug and library code into the common/ subdirectoryStef Walter2013-01-091-158/+160
| | | | | | | | | | | Start using p11_ as our internal prefix rather than _p11_. We explicitly export p11_kit_ so this is fine as far as visibility. Move the threading, mutex, and module compat, dict, and array code into the common directory too. Take this opportunity to clean up a bit of internal API as well, since so many lines are being touched internally.
* Guarantee that the key is freed when replacedStef Walter2013-01-071-1/+5
| | | | | | | | | * When setting a key in a map that already exists, then free the old key and replace with the new one. * Fix related bug where key was not properly allocated * Add tests for this https://bugs.freedesktop.org/show_bug.cgi?id=59087
* Refuse to load the p11-kit-proxy.so as a registered moduleStef Walter2012-09-181-4/+10
| | | | | | | | | | * Since p11-kit-proxy.so is a symlink to the libp11-kit.so library we check that we are not calling into our known CK_FUNCTION_LIST for the proxy code. * Although such a configuration is invalid, detecting this directly prevents strange initialization loop issues that are hard to debug. https://bugs.freedesktop.org/show_bug.cgi?id=55052
* Don't fail initialization if last initialized module failsStef Walter2012-09-181-4/+6
| | | | | | | | | | * We weren't resetting the result code after a failure, so even though failures for critical modules didn't interrupt the initialization loop, the result still leaked to callers. * Also print an error message clearly indicating that a module failed to initialize, regardless of whether critical or not. https://bugs.freedesktop.org/show_bug.cgi?id=55051
* Use Windows thread ids instead of handles for comparisonsStef Walter2012-07-261-3/+3
| | | | | | | | | | * It seems that the HANDLE's returned from GetCurrentThread are often equal for two threads. GetCurrentThreadID doesn't have this problem. * Separate our cross platform thread_t and thread_id_t types even though on unix they're the same thing. https://bugzilla.gnome.org/show_bug.cgi?id=44740
* Fix compiler warning about uninitialized variableStef Walter2012-07-241-1/+1
|
* If a module is not marked 'critical' then ignore failureStef Walter2012-06-271-3/+11
| | | | | * Ignore failure when initializing registered modules when 'critical' is not set on a module.
* Preconditions to check for input probs and out of memoryStef Walter2012-05-131-43/+42
| | | | | | * We don't try to guarantee completely robust and problem free behavior in cases where the caller or process isn't behaving. We consider these to be outside of our control.
* Use gcc extensions to check varargs during compileStef Walter2012-05-131-2/+5
| | | | | * Add macros GNUC_PRINTF and GNUC_NULL_TERMINATED to check correct printf and NULL terminated style varargs
* Add tests for enable-in and disable-inStef Walter2012-04-021-2/+2
|
* Fix crasher when a duplicate module is presentStef Walter2012-04-011-6/+7
|
generated by cgit v1.1 at 2024-10-17 08:23:04 +0000