Commit Graph

349 Commits (e7704497025335336c62379b53940878fd75add2)

Author SHA1 Message Date
Jens Steube e770449702 OpenCL Runtime: Use our own yielding technique for synchronizing rather than vendor specific
6 years ago
Jens Steube ae577410d0 OpenCL Device: Do a real query on OpenCL local memory type instead of just assuming it
6 years ago
Jens Steube 06b39dc3d5 Startup: Show OpenCL runtime initialization message (per device)
6 years ago
Jens Steube 566b1c67b5 Small improvement in --speed-only and --progress-only accuracy
6 years ago
Jens Steube bb758d66e8 Revert "Simplify some code"
6 years ago
Jens Steube 2d15fa07e7 Simplify some code
6 years ago
Jens Steube a402822646 Fixed output of --speed-only and --progress-only in case a fast hash is used in combination with --slow-candidates
6 years ago
Jens Steube 83632a83e4 Memory: Limit maximum host memory to allocate depending on bitness
6 years ago
Jens Steube d6fe2c7cd4 The hashcat brain
6 years ago
Jens Steube 24ce7bacae Updated Intel OpenCL runtime version check
6 years ago
Jens Steube 5ee3e502f8 Fixed missing wordlist encoding in combinator mode
6 years ago
R. Yushaev 5c87720acc Add SHA3 and Keccak
6 years ago
jsteube 52c52836e3 Add new option --slow-candidates which allows hashcat to generate passwords on-host
6 years ago
Jens Steube 466ea8eaba Fixed detection of unique ESSID in WPA-PMKID-* parser
6 years ago
Jens Steube a134ea52b3 Fixed a race condition in status view leading to out-of-bound reads
6 years ago
jsteube a5746548e8 Allow use of hash-mode 7900, 10700 and 13731 on AMD devices after workaround
6 years ago
jsteube dad05d9f69 Testing: Workaround some AMD OpenCL runtime segmentation faults
6 years ago
jsteube 9f54c3dd14 Transfer only a few byte should be enough to force the runtime to actually allocate the memory
6 years ago
jsteube 242d991def Fix memleak in memory evaluation
6 years ago
jsteube f9b13035f2 Limit evaluate available memory check to nvidia runtime
6 years ago
jsteube 86fc587182 Be a bit more sensible while try to evaluate available memory
6 years ago
jsteube e67195aa11 Try to evaluate available OpenCL device memory and use this information instead of total available OpenCL device memory for autotune
6 years ago
jsteube c1edc5d562 Increase max PWS space to 1GB to allow higher -n values for GPU with lots of compute units
6 years ago
jsteube c1ebfe486f Refactor --progress-only and --speed-only calculation
6 years ago
jsteube 88ebca40b8 Added hash-mode 16800 = WPA-PMKID-PBKDF2
6 years ago
jsteube 32d6b3e10e OpenCL kernels: Add '-pure' prefix to kernel filenames to avoid problems caused by reusing existing hashcat installation folder
6 years ago
jsteube a8060f4946 Rewritten handling of OpenCL JiT build options to fix warnings with GCC 8
6 years ago
jsteube ca1115a1ee No longer need to use 32 threads on second dimension for bitsliced algorithms
6 years ago
Jens Steube 872f82ab8d Get rid of thread limit for bitsliced kernels
6 years ago
Jens Steube aa82d8d34d Re-enable CPU optimizations and some CPU case in thread management
6 years ago
Jens Steube fe321105fe No need to copy empty compression buffers
6 years ago
Jens Steube c7ae5fb7f0 Do not run more run a kernel with more threads than workitems
6 years ago
Jens Steube 07f119c28b Fix -a 7 of fast hashes in optimized mode
6 years ago
Jens Steube d965475b28 Replace a missing while() with round_up_multiple_64()
6 years ago
Jens Steube 1af8e29a4b Add round_up_multiple_32() and round_up_multiple_64()
6 years ago
jsteube 98b20526f4 Fix missing cleanup for new AUX kernels
6 years ago
jsteube 8053a05594 Use kernel specific max thread counts for helper kernel, not device specific
6 years ago
jsteube fe4413797e OpenCL Kernels: Use three separate comparison kernels (depending on keyver) for WPA instead of one
6 years ago
Jens Steube 31da893a40 Enforce 2GB host memory per GPU limit
6 years ago
jsteube 91e522d094 Add special marker for speed only mode
6 years ago
jsteube 240652a0ff Move speed timer initialization from run_cracker() to run_copy()
6 years ago
jsteube c4f30220a0 OpenCL Kernels: Refactored kernel thread management from native to maximum per kernel
6 years ago
Jens Steube d38608b9bc Add get_kernel_preferred_wgs_multiple() for future use
6 years ago
Jens Steube 1f442b5b0c Add get_kernel_local_mem_size() for future use
6 years ago
jsteube ce8f445ebe Fix a few warnings and divide by zero (speed_pos)
6 years ago
Jens Steube d656e9c3a4 OpenCL Kernels: Use the kernel local buffer size as additional reference in order to limit the thread-count
6 years ago
jsteube a7a15df911 OpenCL Kernels: Improve performance preview accuracy in --benchmark, --speed-only and --progress-only mode
6 years ago
Jens Steube fd89cac052 No longer need to warm-up OpenCL device for benchmark with new thread settings
6 years ago
Jens Steube fa379074b6 Fixed a logic error in storing temporary progress for slow hashes,leading to invalid speeds in status view
6 years ago
Jens Steube b0ff625844 OpenCL Kernels: Thread-count is switched from native to maximum - as a consequence we assume host memory pool of 2GB per GPU
6 years ago