| In cmp_ratio(), stabilize the sorting order differently from the way we happened to be doing it before. This results in subtly different charset files being generated for the same input data, but arguably the new stabilization is more logical, and it helps reduce the number of recalculations and thus speeds up charset file generation (now 16.5 seconds in the test case mentioned in the previous commit). |
|
More...
|
almost 12 years ago
|
| Optimized the charset file generation algorithms and code ("--make-charset"). On a certain test case, this has improved the time to generate a charset file from 55.5 to 19.5 seconds while generating the exact same file. |
|
More...
|
almost 12 years ago
|
| Source code indentation: consistently put the blocks after "if" on a new line |
|
More...
|
almost 12 years ago
|
| Dropped the remnants of old .chr file compatibility code |
|
More...
|
almost 12 years ago
|
| Revised incremental mode such that the counts of character indices grow independently for each position. This change breaks compatibility with older .chr files. |
|
More...
|
almost 12 years ago
|
| Fixed OpenMP-specific issues with the previous change: make stderr available in the parallel region, wrap the warning printing in "#pragma omp critical". |
|
More...
|
almost 12 years ago
|
| 1.7.9.8: - The formats interface has been enhanced to better support GPU implementations (in jumbo), as well as fast hashes on multi-CPU systems (not yet made use of). - Handle possible NULL returns from crypt() and crypt_r(). |
|
More...
|
almost 12 years ago
|
| Handle possible NULL returns from crypt() and crypt_r() |
|
More...
|
almost 12 years ago
|
| Revised the crypt_all() format method prototype and semantics, as well as the semantics of get_key() and some other format methods called after crypt_all(), such that crypt_all() may alter and/or generate candidate passwords on its own, including beyond its originally supplied candidate password count. |
|
More...
|
almost 12 years ago
|
| Revised the crypt_all() format method prototype and semantics such that it may optionally perform preliminary checking for possible matches. |
|
More...
|
almost 12 years ago
|
| Added reset(), a format method called whenever the set of password hashes being cracked changes, such as after self-test, but before actual cracking starts. Normally, this is a no-op since a format implementation shouldn't mess with the database unnecessarily. However, when there is a good reason to do so this may e.g. transfer the salts and hashes onto a GPU card. |
|
More...
|
almost 12 years ago
|
| Added support for format destructors (the done() method), which is important e.g. when running --test across many GPU-using formats, not to run out of GPU memory after testing only a subset of the formats. |
|
More...
|
almost 12 years ago
|
| Self-test each format just once (rather than before the start of each cracking mode) unless we're debugging. |
|
More...
|
almost 12 years ago
|
| 1.7.9.7: Fixed a bug introduced in 1.7.9.5 where --show would omit the first hex digit of LM hashes. Thanks to magnum for reporting this and providing the patch. (The actual fix to loader.c was committed separately, on 2012/08/23.) |
|
More...
|
almost 12 years ago
|
| Fixed a bug introduced in 1.7.9.5 where --show would omit the first hex digit of LM hashes. Thanks to magnum for reporting this and providing the patch. |
|
More...
|
over 12 years ago
|
| 1.7.9.6: - Specify the alignment of binary ciphertexts and salts explicitly. - Corrected the loading of hashes on a line on their own, which was broken in 1.7.9.5. Thanks to JimF and magnum. |
|
More...
|
over 12 years ago
|
| When binary() or salt() returns a misaligned pointer, print a warning and continue - don't fail the test. This is not necessarily a fatal error, and if the program does not crash on misaligned access, then there's no reason to expect incorrect behavior. |
|
More...
|
over 12 years ago
|
| Added a comment on a performance optimization in ldr_split_line(), where it would parse some fields only when we know we won't need the rest. |
|
More...
|
over 12 years ago
|
| Corrected the loading of hashes on a line on their own, which was recently broken with the introduction of prepare(). Thanks to JimF and magnum. |
|
More...
|
over 12 years ago
|
| Specify binary and salt alignment explicitly instead of guessing it based on their sizes. |
|
More...
|
over 12 years ago
|
| mem_alloc_tiny(): no longer assume that a pointer fits in "unsigned long" |
|
More...
|
over 12 years ago
|
| Revised the order of parameters of mem_alloc_copy() |
|
More...
|
over 12 years ago
|
| 1.7.9.5: |
|
More...
|
over 12 years ago
|
| Implemented the source() method for LM hashes. |
|
More...
|
over 12 years ago
|
| Added source() format method, which reconstructs the ASCII ciphertext from its binary (saltless only at this time). |
|
More...
|
over 12 years ago
|
| Added prepare() format method. Pass a pointer to the format itself to init(), prepare(), valid(), split(). |
|
More...
|
over 12 years ago
|
| Initialize all external mode variables to 0 by default. |
|
More...
|
almost 13 years ago
|
| 1.7.9.4: Fixed a bug in the Keyboard external mode (uninitialized variables on "--restore" or when minlength is greater than 1). |
|
More...
|
almost 13 years ago
|
| Dropped #include <math.h>: it was previously added for pow(), but we've since replaced it with own powi(). |
|
More...
|
about 13 years ago
|
| Use the MAYBE_INLINE macro. |
|
More...
|
about 13 years ago
|
