DeforaNetworks/hashcat
3
0
Fork 0
Code Releases Activity

OpenCL Kernels: Refactored kernel thread management from native to maximum per kernel

Browse Source
This commit is contained in:
jsteube
2018-02-11 21:29:57 +01:00
parent d38608b9bc
commit c4f30220a0
8 changed files with 403 additions and 467 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
docs/changes.txt
View File

@@ -63,8 +63,7 @@
- OpenCL Kernels: Add general function declaration keyword (inline) and some OpenCL runtime specific exceptions for NV and CPU devices
- OpenCL Kernels: Replace variables from uXX to uXXa if used in __constant space
- OpenCL Kernels: Use a special kernel to initialize the password buffer used during autotune measurements, to reduce startup time
- OpenCL Kernels: Use the kernel local buffer size as additional reference in order to limit the thread-count
- OpenCL Kernels: Thread-count is switched from native to maximum - as a consequence we assume host memory pool of 2GB per GPU
- OpenCL Kernels: Refactored kernel thread management from native to maximum per kernel
- OpenCL Runtime: Add current timestamp to OpenCL kernel source in order to force OpenCL JiT compiler to recompile and not use the cache
- OpenCL Runtime: Enforce to use OpenCL version 1.2 to restrain OpenCL runtimes to make use of the __generic address space qualifier
- OpenCL Runtime: Updated rocm detection