AMD GPUs: Add inline assembly code for md5crypt/sha256crypt, PDF 1.7, 7-Zip, RAR3, Samsung Android and Windows Phone 8+
This commit is contained in:
@@ -18,6 +18,7 @@
|
||||
## Improvements
|
||||
##
|
||||
|
||||
- AMD GPUs: Add inline assembly code for md5crypt/sha256crypt, PDF 1.7, 7-Zip, RAR3, Samsung Android and Windows Phone 8+
|
||||
- Blake Kernels: Optimize BLAKE2B_ROUND() 64 bit rotates giving a 5% performance increase
|
||||
- Brain Session: Adds hashconfig specific opti_type and opts_type parameters to hashcat session computation to cover features like -O and -M
|
||||
- Kernel Threads: Use warp size / wavefront size query instead of hardcoded values as base for kernel threads
|
||||
@@ -28,6 +29,7 @@
|
||||
## Technical
|
||||
##
|
||||
|
||||
- ADL: Updated support for AMD Display Library to 14.0, updated datatypes and added support for OverDrive 7 and 8 based GPUs
|
||||
- Commandline: Throw an error if separator character given by the user with -p option is not exactly 1 byte
|
||||
- Kernel Cache: Add kernel threads into hash computation which is later used in the kernel cache filename
|
||||
- HIP Kernels: Got rid of hip/hip_runtime.h dependancy to enable more easy integration of the HIP backend on Windows
|
||||
|
||||
Reference in New Issue
Block a user