cpython

Commit Graph

Author	SHA1	Message	Date
Eric V. Smith	6ba5665fc7	Fix typo in comment.	12 years ago
Raymond Hettinger	74fc8c47f6	Add comments to frozenset_hash(). Also, provide a minor hint to the compiler on how to group the xors.	12 years ago
Raymond Hettinger	e259f13874	Minor code clean-up. Keep the C-API all in one section.	12 years ago
Raymond Hettinger	710a67edfc	Note that LINEAR_PROBES can be set to zero.	13 years ago
Raymond Hettinger	4ef0528b97	Minor beautification. Put updates and declarations in a more logical order.	13 years ago
Raymond Hettinger	0ce1953bf7	When LINEAR_PROBES=0, let the compiler remove the dead code on its own.	13 years ago
Raymond Hettinger	c70a2b7bb9	Make the linear probe sequence clearer.	13 years ago
Raymond Hettinger	8408dc581e	Issue 18771: Make it possible to set the number linear probes at compile-time.	13 years ago
Raymond Hettinger	742d8716ff	Put the defines in the logical section and fix indentation.	13 years ago
Raymond Hettinger	583cd03fd1	Minor code beautification.	13 years ago
Raymond Hettinger	4ea9080da9	Improve code clarity by removing two unattractive macros.	13 years ago
Raymond Hettinger	8f8839e10a	Remove the freelist scheme for setobjects. The setobject freelist was consuming memory but not providing much value. Even when a freelisted setobject was available, most of the setobject fields still needed to be initialized and the small table still required a memset(). This meant that the custom freelisting scheme for sets was providing almost no incremental benefit over the default Python freelist scheme used by _PyObject_Malloc() in Objects/obmalloc.c.	13 years ago
Raymond Hettinger	04fd9dd52b	Small rearrangement to bring together the three functions for probing the hash table.	13 years ago
Raymond Hettinger	ae7b00e2d3	Move the overview comment to the top of the file.	13 years ago
Raymond Hettinger	c56e0e3980	Minor touchups.	13 years ago
Raymond Hettinger	69492dab07	Factor-out the common code for setting a KeyError.	13 years ago
Raymond Hettinger	a35adf5b09	Instead of XORed indicies, switch to a hybrid of linear probing and open addressing. Modern processors tend to make consecutive memory accesses cheaper than random probes into memory. Small sets can fit into L1 cache, so they get less benefit. But they do come out ahead because the consecutive probes don't probe the same key more than once and because the randomization step occurs less frequently (or not at all). For the open addressing step, putting the perturb shift before the index calculation gets the upper bits into play sooner.	13 years ago
Raymond Hettinger	6c3c1ccd1b	Update copyright.	13 years ago
Raymond Hettinger	95c0d67581	Further reduce the cost of hash collisions by inspecting an additional nearby entry.	13 years ago
Raymond Hettinger	afe890923f	Tighten-up the lookkey() logic and beautify the code a bit. Use less code by moving many of the steps from the initial lookup into the main search loop. Beautify the code but keep the overall logic unchanged.	13 years ago
Antoine Pitrou	9d95254bb7	Issue #18772 : fix the gdb plugin after the set implementation changes	13 years ago
Raymond Hettinger	bfc1e1a9cd	Add the same dummy type that is used in dictionaries.	13 years ago
Raymond Hettinger	fcf3b500ba	Issue 18797: Remove unneeded refcount adjustments for dummy objects. It suffices to keep just one reference when the object is created.	13 years ago
Raymond Hettinger	5bb1b1dd6f	Hoist the global dummy lookup out of the inner loop for set_merge().	13 years ago
Raymond Hettinger	929cbac307	Remove a redundant hash table probe (this was artifact from an earlier draft of the patch).	13 years ago
Raymond Hettinger	ae9e616a00	Issue 18772: Restore set dummy object back to unicode and restore the identity checks in lookkey(). The Gdb prettyprint plugin depended on the dummy object being displayable. Other solutions besides a unicode object are possible. For now, get it back up and running. The identity checks in lookkey() need to be there to prevent the dummy object from leaking through Py_RichCompareBool() into user code in the rare circumstance where the dummy's hash value exactly matches the hash value of the actual key being looked up.	13 years ago
Raymond Hettinger	3c0a4f5def	Issue18771: Reduce the cost of hash collisions for set objects.	13 years ago
Raymond Hettinger	07351a0449	Remove the else-clause because the conditions are no longer mutually exclusive.	13 years ago
Raymond Hettinger	237b34b074	Use a known unique object for the dummy entry. This lets us run PyObject_RichCompareBool() without first needing to check whether the entry is a dummy.	13 years ago
Raymond Hettinger	8ad3919577	Hoist the global "dummy" lookup outside of the reinsertion loop.	13 years ago
Antoine Pitrou	9ed5f27266	Issue #18722 : Remove uses of the "register" keyword in C code.	13 years ago
Raymond Hettinger	c629d4c9a2	Replace outdated optimization with clearer code that compiles better. Letting the compiler decide how to optimize the multiply by five gives it the freedom to make better choices for the best technique for a given target machine. For example, GCC on x86_64 produces a little bit better code: Old-way (3 steps with a data dependency between each step): shrq $5, %r13 leaq 1(%rbx,%r13), %rax leaq (%rax,%rbx,4), %rbx New-way (3 steps with no dependency between the first two steps which can be run in parallel): leaq (%rbx,%rbx,4), %rax # i5 shrq $5, %r13 # perturb >>= PERTURB_SHIFT leaq 1(%r13,%rax), %rbx # 1 + perturb + i5	13 years ago
Raymond Hettinger	c86d7e989c	Silence compiler warning for an unused declaration	13 years ago
Antoine Pitrou	5e946bacef	Fix compilation warning with gcc 4.8 (unused typedef)	13 years ago
Gregory P. Smith	27cbcd6241	Fix the internals of our hash functions to used unsigned values during hash computation as the overflow behavior of signed integers is undefined. In practice we require compiling everything with -fwrapv which forces overflow to be defined as twos compliment but this keeps the code cleaner for checkers or in the case where someone has compiled it without -fwrapv or their compiler's equivalent. Found by Clang trunk's Undefined Behavior Sanitizer (UBSan). Cleanup only - no functionality or hash values change.	13 years ago
Ezio Melotti	0e1af282b8	Fix typo.	14 years ago
David Malcolm	49526f48fc	Issue #14785 : Add sys._debugmallocstats() to help debug low-level memory allocation issues	14 years ago
Antoine Pitrou	a701388de1	Rename _PyIter_GetBuiltin to _PyObject_GetBuiltin, and do not include it in the stable ABI.	14 years ago
Kristján Valur Jónsson	31668b8f7a	Issue #14288 : Serialization support for builtin iterators.	14 years ago
Antoine Pitrou	093ce9cd8c	Issue #6695 : Full garbage collection runs now clear the freelist of set objects. Initial patch by Matthias Troffaes.	14 years ago
Benjamin Peterson	2b50a01d11	remove unused variable	14 years ago
Benjamin Peterson	e3b5eda8be	remove unused variable	14 years ago
Petri Lehtinen	e0aa803714	Fix the return value of set_discard (issue #10519 )	14 years ago
Petri Lehtinen	a39de111a5	Fix the return value of set_discard (issue #10519 )	14 years ago
Petri Lehtinen	5acc27ebe4	Avoid unnecessary recursive function calls (closes #10519 )	14 years ago
Petri Lehtinen	5f4d870668	Avoid unnecessary recursive function calls (closes #10519 )	14 years ago
Martin v. Löwis	bd928fef42	Rename _Py_identifier to _Py_IDENTIFIER.	14 years ago
Martin v. Löwis	1ee1b6fe0d	Use identifier API for PyObject_GetAttrString.	15 years ago
Martin v. Löwis	d63a3b8beb	Implement PEP 393.	15 years ago
Mark Dickinson	57e683e53e	Issue #1621 : Fix undefined behaviour in bytes.__hash__, str.__hash__, tuple.__hash__, frozenset.__hash__ and set indexing operations.	15 years ago

1 2 3 4

194 Commits (abdc16ee1bb9819ecc6500a8f1d8720fc2072643)