| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
| |
* Do conditional compiling of Cryptography_setup_ssl_threads
* Check Cryptography_HAS_LOCKING_CALLBACKS before initing static locks
Check if compiling and initing locking callbacks is necessary
PEP8 fix
* Make test_crypto_lock_init more complete
|
| |
|
| |
And not expose an unprefixed name to anyone who dlopens us.
|
| |
|
| |
So here we need to make sure we don't simply include windows but only the parts that we want
|
| |
|
|
|
|
|
|
| |
* No more FUNCS/MACROS distinction
* change the docs to not talk about MACROS since they're gone
* remove out of date comment
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* pypy3 fix on macos using work from the pypy project
https://bitbucket.org/pypy/pypy/commits/198dc138680f96c391802fa1e77b8b6d2e0134e6?at=py3.5
* change abort error msg and fix wrong type
* oh windows
* remove an unused variable
* rename mutex1_t, use calloc, small style fixes
* calloc correctly
* (call)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* switch the PEM password callback to a C implementation
Calling from C to Python is fraught with edge cases, especially in
subinterpreter land. This commit moves the PEM password callback logic
into a small C function and then removes all the infrastructure for the
cffi callbacks (as we no longer have any)
* review feedback and update tests
* rename the struct
* aaand one more fix
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Use static callbacks with Python 3.x again
Static callbacks were disabled for Python 3.5+ to work around an issue
with subinterpreters, locking callbacks and osrandom engine. Locking
callback and osrandom engine were replaced with a C implementations in
version 1.6 and 1.7.
https://github.com/pyca/cryptography/issues/2970
Closes: #3348
Signed-off-by: Christian Heimes <christian@python.org>
* remove unused import
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Remove Python OpenSSL locking callback and replace it with one in C
The Python OpenSSL locking callback is unsafe; if GC is triggered
during the callback's invocation, it can result in the callback being
invoked reentrantly, which can lead to deadlocks. This patch replaces
it with one in C that gets built at compile time via cffi along with
the rest of the OpenSSL binding.
* fixes for some issues
* unused
* revert these changes
* these two for good measure
* missing param
* sigh, syntax
* delete tests that assumed an ability to mess with locks
* style fixes
* licensing stuff
* utf8
* Unicode. Huh. What it isn't good for, absolutely nothing.
|
| | |
|
| | |
|
|
|
cffi 1.4.0 will introduce a new API to create static callbacks. Contrary
to the old callback API, static callbacks no longer depend on libffi's
dynamic code generation for closures. Static code has some benefits over
dynamic generation. For example the code is faster. Also it doesn't need
writeable and executable memory mappings, which makes it compatible with
SELinux's deny execmem policy.
The branch depends on PR #2488.
https://bitbucket.org/cffi/cffi/issues/232/static-callbacks
Closes: #2477
Signed-off-by: Christian Heimes <cheimes@redhat.com>
|
