Age | Commit message (Collapse) | Author | Files | Lines |
|
dialog (#516)
To get the size of each device / partition on the system, the method 'GetDeviceSize()' in 'src/Core/Unix/CoreUnix.cpp' first opens
the device / partition using 'open()' function to get a File Descriptor, then retrieves its size using this File Descriptor.
Starting OS X 10.11 ("El Capitan"), a feature called "System Integrity Protection (SIP)" or less formally, "rootless mode" has been added.
This feature blocks access to certain critical aspects of the OS and Hardware by 3rd-Party programs.
Specifically, low-level access to the system disks, devices and partitions is forbidden ; namely functions like 'open()' for instance fail
with the error code : "EPERM = Operation Not Permitted".
Therefore, for system devices / partitions, 'GetDeviceSize()' fails because of the failure of the 'open()' function, and throws an exception,
which is then caught inside the method 'GetHostDevices()' in '/src/Core/Unix/FreeBSD/CoreFreeBSD.cpp' : this leads to the size of the
device / partition being set to '0'.
Therefore, in the constructor of 'DeviceSelectionDialog' in 'src/Main/Forms/DeviceSelectionDialog.cpp', when the size of a device is '0',
the whole device is skipped, leading to all of its partitions not being treated or shown, even though some of these partitions may have a size which is != 0.
This commit fixes the issue by :
1 - First, checking whether the device size is '0'. If it is the case, the code loops through all the devices partitions : if there is at least one partition
with a size != 0, the device is not skipped. Otherwise, it is.
2 - Then, if the size of the device is '0', the size of the device is not shown to avoid confusing the user.
Also, since the device is not usable, the 'OK' button is not active when the device is selected.
3 - Finally, if a partition's size is '0', it is not shown since it is not usable : we cannot open it.
Signed-off-by: El Mostafa IDRASSI <el-mostafa.idrassi@prestalab.net>
|
|
active 'sudo' session instead of the use of a 'dummy' password. (#513)
Signed-off-by: El Mostafa IDRASSI <el-mostafa.idrassi@prestalab.net>
|
|
Now, under Debian 10+ and Ubuntu 18.04+, we link against the GTK-3 version of wxWidgets (libwxgtk3.0-gtk3-0v5).
Under Debian 9- and Ubuntu 16.04, we link against the GTK-2 version of wxWidgets (libwxgtk3.0-0v5) which is the
only one available.
Also, we now have 2 separate RPM scripts : 'build_cmake_rpm_gtk2.sh' which builds wxWidgets and links it against GTK-2,
then links VeraCrypt against 'gtk2' package (typically to be used under CentOS 6) and 'build_cmake_rpm_gtk3' which builds
wxWidgets and links it against GTK-3, then links VeraCrypt against 'gtk3' package (typically to be used under CentOS 7+).
|
|
|
|
/usr/share/veracrypt/doc to the standard /usr/share/doc/veracrypt
|
|
|
|
packages names
|
|
|
|
CPack, and shell scripts which build then package VeraCrypt under CentOS and Debian/Ubuntu. (#511)
The DEB script builds VeraCrypt and links it against wxWidgets that comes with the distribution.
The RPM script awaits for wxWidgets-3.0.4 source code which it builds then links VeraCrypt statically to it.
Both scripts create the corresponding package after the build.
|
|
exist (#510)
|
|
|
|
version < 4.9 without -mssse3 option (SSSE3=1 when using make). (#507)
Compiling with -mxxx defines the corresponding macro of the intrinsics.
For example, -mssse3 defines __SSSE3__ macro to 1.
In GCC versions < 4.9, it is not possible to use and call x86 intrinsics only at runtime without
compiling the entire file with the -mxxx option.
For example, if we want to call SSSE3 intrinsics without compiling with -mssse3, the macro __SSSE3__ is not defined.
Therefore, when including <tmmintrin.h>, this results in "error "SSSE3 instruction set not enabled"" because of :
#ifndef __SSSE3__
# error "SSSE3 instruction set not enabled"
Since GCC 4.9, this has been fixed and it is possible to call x86 intrinsics from select functions in a file
that are tagged with the corresponding target attribute without having to compile the entire file with the -mxxx option.
This can be seen in <tmmintrin.h> which in recent versions (>= 4.9) contains :
#ifndef __SSSE3__
#pragma GCC push_options
#pragma GCC target("ssse3")
#define __DISABLE_SSSE3__
Since SSSE3 is only used under Windows for ChaCha256, this can be fixed by preceding '#include <tmmintrin.h>' with
#if defined (_MSC_VER) && !defined (TC_WINDOWS_BOOT).
See https://gcc.gnu.org/gcc-4.9/changes.html
|
|
initialized before us, then stop random generator after we finish in order to avoid consuming CPU because of periodic fast poll thread. Next time a critical operation that requires RNG is performed, it will be initialized again. We do this because since the addition of secure desktop support, every time secure desktop is displayed, the RNG fast poll thread was started even if the user will never perform any critical operation that requires random bytes.
|
|
already initialized before our call or not
|
|
since the it consumes too much CPU and FastPoll requires fast and minimal entropy gathering
|
|
YYY' (#506)
in order for 'export TC_VERSION := $(shell grep VERSION_STRING ../Common/Tcdefs.h | head -n 1 | cut -d'"' -f 2)'
in 'src/Main/Main.make' to actually return the version rather than '-CustomEFI'.
|
|
available
disk free space. Add a CLI switch to disable this check.
|
|
(#472)
As $(PWD) is not always the expected value and can result in failing the target.
|
|
This can be useful when crossbuilding or building in a chroot where using
uname -m would cause the wrong compiler options.
|
|
in order to fix LTO linking.
After switching to LTO for openSUSE Tumbleweed, veracrypt build failed with:
[ 185s] ../Crypto/Whirlpool.c:105:45: error: 'Whirlpool_C' causes a section type conflict with 'SHA256_K'
[ 185s] 105 | CRYPTOPP_ALIGN_DATA(16) static const uint64 Whirlpool_C[8*256+R] CRYPTOPP_SECTION_ALIGN16 = {
[ 185s] | ^
[ 185s] ../Crypto/Sha2.c:321:34: note: 'SHA256_K' was declared here
[ 185s] 321 | CRYPTOPP_ALIGN_DATA(16) uint_32t SHA256_K[64] CRYPTOPP_SECTION_ALIGN16 = {
[ 185s] | ^
[ 185s] lto-wrapper: fatal error: g++ returned 1 exit status
Aligning section types of Whirlpool_C and SHA256_K fixes this.
|
|
container creation
|
|
|
|
can't resume from Hibernation without RAM encryption keys (a chicken and egg situation)
|
|
disks on demand instead of testing all disks to reduce CPU usage.
|
|
some text truncation issues with non-English languages.
|
|
verify the integrity of binaries.
|
|
certificate.
|
|
that come with the following modifications: - Fix issue that was preventing Streebog hash from being selected manually during Pre-Boot authentication - Ensure that the correct Windows bootloader is executed when the user press ESCAPE - make the rescue disk boot machine directly from disk if "VeraCrypt" folder is missing. This make it easy to create a bootable disk for VeraCrypt from the rescue disk by just removing or renaming its "VeraCrypt" folder.
|
|
Favorite that uses VolumeID. Add command option to disable the period update of devices.
|
|
|
|
|
|
by not using the trick to put it reliably in foreground. The trick is based on a emulation of ALT+TAB but sometimes ALT key would remain pressed in case of Secure Desktop making it impossible to type the password (a workaround was to press CTRL key which reset the state of ALT key)
|
|
can be localized correctly
|
|
don't support them yet.
|
|
Non-Interactive
mode by using keyfiles passed as arguments and check the Non-Interactive
mode switch.
|
|
decryption wizard/
|
|
ChaCha20 and T1HA algorithms and which are not present nor used in UEFI bootloader
|
|
intrin.h header
|
|
* Revert previous commit
* Fix "Invalid characters..." issue by not using "foreach" macro
The "foreach" macro creates a copy of the container.
This copy is destroyed immediately after the iteration is completed.
C-strings pointers passed to the local array were invalidated
with destroying of "std::string"s contained in the copy.
|
|
|
|
|
|
|
|
protection if RAM encryption is enabled
|
|
password length increase applies only to non-system volumes.
|
|
|
|
enabled
|
|
|
|
driver
|
|
|
|
|