10 Commits

Author	SHA1	Message	Date
Vikram S. Adve	2786ece7fb	1. Make table size prime -- improves hashing performance vastly. ... 2. Reduce #hashes from 2 to 1 for initial inserts. 3. Fix incorrect assertion from last checkin. llvm-svn: 7417	2003-07-30 12:49:25 +00:00
Vikram S. Adve	f8049f93c2	Bug fix: after reallocating the hash table, we have to re-insert each value ... instead of copying table entries! llvm-svn: 7396	2003-07-29 20:01:01 +00:00
Chris Lattner	37dc32a4b2	Remove unneccesary #ifdefs ... llvm-svn: 7214	2003-07-21 19:23:31 +00:00
Chris Lattner	4d52e214bb	Fix warnings ... llvm-svn: 7213	2003-07-21 19:20:44 +00:00
Vikram S. Adve	6fa4b79587	Use uint32_t for table index and size: table will never be > 4GB. ... Also, make Pointer type depend on architecture. llvm-svn: 7170	2003-07-11 22:02:28 +00:00
Vikram S. Adve	3b7c2a4db7	Pointer hash table reallocation code seems never to have been tested! ... Unfortunately, reallocation also means that the pointer numbering will change, so increase table size to try to avoid it. llvm-svn: 7130	2003-07-08 18:42:44 +00:00
Joel Stanley	d1e35d03b3	Prevented inclusion of (non-existent) stdint.h on Sparc. ... llvm-svn: 6876	2003-06-24 02:46:47 +00:00
Chris Lattner	d10f1b56c6	Update to match the reality that is now. ... llvm-svn: 6362	2003-05-27 21:43:14 +00:00
Chris Lattner	008db290b8	* If hashing a pointer, cast it to a 64 bit number so gcc doesn't warn ... * Don't put extra stuff after #endif llvm-svn: 2671	2002-05-20 21:15:30 +00:00
Vikram S. Adve	be31fcf252	Runtime routines to support tracing. ... Currently includes code to hash pointers to sequence numbers. The hash table should be separated out into a separate file since that is generic. The rest of this code is pretty small. llvm-svn: 2655	2002-05-19 15:49:58 +00:00