~bzr-pqm/bzr/bzr.dev : changes

~bzr-pqm/bzr/bzr.dev » Changes from revision 2474.1.30

Rev	Summary	Authors	Date
2474.1.30	Start working towards a parser which uses a Reader (producer... Start working towards a parser which uses a Reader (producer) rather than working on a list of fields. Currently slower than text.split('\0'), but should be possible to avoid the intermediate list entirely.	John Arbash Meinel	17 years ago
2474.1.29	Refactor, so that the inner _fields_to_entries function is t... Refactor, so that the inner _fields_to_entries function is the doing the path comparison, and it will re-use the dirname object, rather than copying a new string each time. This should have equivalent performance, but have a rather large memory savings, because we don't maintain N copies of the dirname for N files in that directory. It (theoretically) will speed up some comparisons, too, because the string hash, etc, will be properly cached.	John Arbash Meinel	17 years ago
2474.1.28	Ask the field converter to determine the current directory Ask the field converter to determine the current directory rather than parsing it out of the returned entry.	John Arbash Meinel	17 years ago
2474.1.27	Switching to direct access of members of the list drops us d... Switching to direct access of members of the list drops us down to 305ms	John Arbash Meinel	17 years ago
2474.1.26	Switch to using an offset rather than doing a list splice Switch to using an offset rather than doing a list splice	John Arbash Meinel	17 years ago
2474.1.25	Refactor into a helper function to make implementation clear... Refactor into a helper function to make implementation clearer This also improves performance to 319ms	John Arbash Meinel	17 years ago
2474.1.24	Unrolling into a direct loop drops us to 326ms Unrolling into a direct loop drops us to 326ms	John Arbash Meinel	17 years ago
2474.1.23	A C implementation of _fields_to_entry_0_parents drops the t... A C implementation of _fields_to_entry_0_parents drops the time from 400ms to 330ms for a 21k-entry tree	John Arbash Meinel	17 years ago
2474.1.22	Do the same renaming => py_ and c_ for _read_dirblocks Do the same renaming => py_ and c_ for _read_dirblocks	John Arbash Meinel	17 years ago
2474.1.21	Cleanup the multiple testing. Cleanup the multiple testing. Change the function names from both being 'bisect_dirblocks' to being py_bisect_dirblocks and c_bisect_dirblocks. And enable using the compiled form when it is available.	John Arbash Meinel	17 years ago
2474.1.20	Apply all of the tests for DirState.bisect_dirblock to the c... Apply all of the tests for DirState.bisect_dirblock to the compiled function.	John Arbash Meinel	17 years ago
2474.1.19	Clean up _cmp_dirblock_strings_alt to make it the default. Clean up _cmp_dirblock_strings_alt to make it the default. This improves bisect_dirblock_compiled by another 2x. So far the improvement is now 800ms => 100ms => 50ms with the current function.	John Arbash Meinel	17 years ago
2474.1.18	Add an integer-size comparison loop at the begining, and Add an integer-size comparison loop at the begining, and update the test suite to make sure we are properly exercising it.	John Arbash Meinel	17 years ago
2474.1.17	Using a custom loop seems to be the same speed, but is proba... Using a custom loop seems to be the same speed, but is probably easier to understand.	John Arbash Meinel	17 years ago
2474.1.16	Shave off maybe 10% by using the PyString_* macros instead o... Shave off maybe 10% by using the PyString_* macros instead of functions.	John Arbash Meinel	17 years ago
2474.1.15	No need to benchmark bisect_dirblock_compiled_cached No need to benchmark bisect_dirblock_compiled_cached The cache isn't used in the compiled form.	John Arbash Meinel	17 years ago
2474.1.14	Switching bisect_dirblocks remove the extra .split('/') Switching bisect_dirblocks remove the extra .split('/') This is a massive improvement (approx 8x). Since we avoid all the temporary lists, dictionary lookups etc. Now we just have a custom string comparison, which is quite fast.	John Arbash Meinel	17 years ago
2474.1.13	Now that we have bisect_dirblock working again, bring back Now that we have bisect_dirblock working again, bring back cmp_dirblock_strings.	John Arbash Meinel	17 years ago
2474.1.12	Clean up bisect_dirstate to not use temporary variables. Clean up bisect_dirstate to not use temporary variables.	John Arbash Meinel	17 years ago
2474.1.11	Avoid a Py_INCREF by using a void * Avoid a Py_INCREF by using a void *	John Arbash Meinel	17 years ago

Older »