| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
| |
These are useful functions for callers who want to supress all output
from p11-kit library.
|
|
|
|
|
|
|
|
| |
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.
|
|
|
|
|
|
| |
ie: when streror_s is missing in msvcrt.dll
https://bugs.freedesktop.org/show_bug.cgi?id=76594
|
|
|
|
|
|
|
|
| |
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()
|
|
|
|
| |
https://bugs.freedesktop.org/show_bug.cgi?id=76594
|
|
|
|
|
| |
We can't use automake 1.10 as serial-tests is not supported
there.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
when buliding for Windows (mingw).
This issue has been reported in bug #76594
a. Moved vasprintf before asprintf
b. Added prototypes for each of them
Thanks,
pa
Signed-off-by: Pavel A <pavel.aronsky@daynix.com>
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The proxy module would unload the PKCS#11 modules it was proxying
when C_Finalize() was called. However if a caller in another thread
was inside of a PKCS#11 function at the time, this would cause
a crash.
Change things around so that underlying modules are finalized during
the proxy C_Finalize() but not released/unloaded until the proxy
module itself is unloaded.
https://bugs.freedesktop.org/show_bug.cgi?id=74919
|
|
|
|
|
|
|
|
| |
We should return CKR_CRYPTOKI_NOT_INITIALIZED rather than
assert() when proxy PKCS#11 functions are called before the
module is initialized.
https://bugs.freedesktop.org/show_bug.cgi?id=74919
|
|
|
|
|
|
|
|
|
|
|
| |
Check the return value of mmap() correctly.
Empty files cannot be mmap'd so we implement some
work around code for that.
https://bugs.freedesktop.org/show_bug.cgi?id=74773
Signed-off-by: Stef Walter <stef@thewalter.net>
|
| |
|
|
|
|
|
|
| |
Because Windows is really properly screwed up.
https://bugs.freedesktop.org/show_bug.cgi?id=74149
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
This is an integration test that the extract and blacklist
functionality basics work.
More integration tests should follow, at which point we should
place the various generic testing bits into their own file.
|
|
|
|
| |
Naturally this doesn't apply to tarballs
|
| |
|
|
|
|
|
|
|
| |
This fixes an issue where a blacklist in one token wasn't properly
skipping anchors being extracted with extract-compat
https://bugs.freedesktop.org/show_bug.cgi?id=73558
|
|
|
|
|
|
| |
This allows simpler lookups.
https://bugs.freedesktop.org/show_bug.cgi?id=73558
|
|
|
|
|
|
| |
This gives a little broader testing of the enumerator
https://bugs.freedesktop.org/show_bug.cgi?id=73558
|
| |
|
|
|
|
| |
A simple wrapper for C_GetAttributeValue()
|
| |
|
| |
|
|
|
|
| |
Reported-by: Tijl Coosemans <tijl@FreeBSD.org>
|
|
|
|
|
|
|
|
|
| |
In recent versions of glibc this is true and prevents linking with
pthreads when it is not necessary.
Tweaked by Stef Walter
Signed-off-by: Stef Walter <stef@thewalter.net>
|
|
|
|
|
|
|
|
|
| |
It's not only unsed, but also causes build fail because CFLAGS
for tests does not contain LIBTASN1_CFLAGS.
Signed-off-by: Stef Walter <stef@thewalter.net>
https://bugs.freedesktop.org/show_bug.cgi?id=71379
|
|
|
|
|
|
| |
Related to the following bug:
https://bugs.freedesktop.org/show_bug.cgi?id=69314
|
|
|
|
|
| |
We use the FAKED_MODE environment variable as a way to detect
fakeroot.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
When the 'trust anchor' tool changes something, run
'trust extract-compat' after that point
|
|
|
|
|
| |
This will change once the spec has a specific attribute and code
to signify deletability.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
| |
The actual command is 'trust extract-compat'. Make installed placeholder
script reflect this. We still support the old placeholder script
if it is present.
|
| |
|
|
|
|
| |
Also prevent --store from storing an anchor multiple times
|
|
|
|
| |
Lists with PKCS#11 URI's and some basic fields.
|
| |
|
|
|
|
| |
Because we want to use this same logic for listing trust
|
|
|
|
|
| |
So that validation/storage logic doesn't kick in if a file was
removed outside of p11-kit trust module.
|
|
|
|
| |
This allows a token to remove the file if desired
|