2011-01-27 |
stephey | Removed vestigial code left over from "waiting queues... |
tree | commitdiff |
2010-11-18 |
yeom | provides 'corepin' flag for mergesort |
tree | commitdiff |
2010-11-18 |
bdemsky | change |
tree | commitdiff |
2010-11-18 |
yeom | add flag "corepin" for core pinning |
tree | commitdiff |
2010-11-18 |
jjenista | force oid increment to be a prime number |
tree | commitdiff |
2010-11-17 |
bdemsky | add header files |
tree | commitdiff |
2010-11-17 |
bdemsky | affinity call |
tree | commitdiff |
2010-11-17 |
yeom | grabs a lock when it tries to reset |
tree | commitdiff |
2010-11-17 |
yeom | assigns workerID to the workerTR |
tree | commitdiff |
2010-11-13 |
yeom | removes unnecessary LOCKXCHG operation to optimize. |
tree | commitdiff |
2010-11-13 |
yeom | fixes on hashStructure : when parent bin is ready,... |
tree | commitdiff |
2010-11-12 |
yeom | fix: retireHashTable clears its own readbin out of... |
tree | commitdiff |
2010-11-11 |
yeom | index field of rentry is only for RCR, add guard condit... |
tree | commitdiff |
2010-11-11 |
jjenista | have to consider an empty, ready read bin in front... |
tree | commitdiff |
2010-11-10 |
jjenista | coreprof in RCR, just one event for now showing begin... |
tree | commitdiff |
2010-11-10 |
jjenista | poolalloc with no poolfree should be benign, but it... |
tree | commitdiff |
2010-11-10 |
jjenista | back out pool alloc for now, fix it right when its... |
tree | commitdiff |
2010-11-10 |
jjenista | pool allocate RCR obj-conflict hash table bin items |
tree | commitdiff |
2010-11-10 |
jjenista | coreprof events, RCR stuff will be online onyl after... |
tree | commitdiff |
2010-11-10 |
jjenista | alter RCR params, remove unused macro |
tree | commitdiff |
2010-11-09 |
jjenista | no need to add one worker thread, let end user decide |
tree | commitdiff |
2010-11-09 |
jjenista | bug fix, was releasing references a little too early |
tree | commitdiff |
2010-11-09 |
bdemsky | changes |
tree | commitdiff |
2010-11-09 |
bdemsky | nasty bug...have to make sure we don't free a task... |
tree | commitdiff |
2010-11-09 |
bdemsky | changes |
tree | commitdiff |
2010-11-09 |
bdemsky | (1) make type of memPool more precise... |
tree | commitdiff |
2010-11-09 |
jjenista | generated traversers already release |
tree | commitdiff |
2010-11-09 |
jjenista | pruning unneeded fields |
tree | commitdiff |
2010-11-09 |
jjenista | just an explanation of LOCKXCH |
tree | commitdiff |
2010-11-08 |
jjenista | capture a todo item, we don't expect its a problem... |
tree | commitdiff |
2010-11-08 |
bdemsky | fix bug |
tree | commitdiff |
2010-11-08 |
bdemsky | changes: |
tree | commitdiff |
2010-11-05 |
yeom | changes: don't need to have resolvePointer opr for... |
tree | commitdiff |
2010-11-05 |
bdemsky | doesn't make sense for us to need a barrier here... |
tree | commitdiff |
2010-11-05 |
jjenista | just add the type to make sure it is 3, which is what... |
tree | commitdiff |
2010-11-05 |
jjenista | without this barrier the compiler can optimize variable... |
tree | commitdiff |
2010-11-05 |
jjenista | head pointer might be NULL, as is checked a few lines... |
tree | commitdiff |
2010-11-05 |
bdemsky | change |
tree | commitdiff |
2010-11-05 |
bdemsky | bug fix |
tree | commitdiff |
2010-11-05 |
bdemsky | bug fixes |
tree | commitdiff |
2010-11-05 |
bdemsky | changes |
tree | commitdiff |
2010-11-03 |
bdemsky | make sure 1 is long long...other bug fix |
tree | commitdiff |
2010-11-02 |
jjenista | new coreprof event, move workschedule dowork label... |
tree | commitdiff |
2010-11-02 |
jjenista | I misunderstood the REntry pointer, its a pointer to... |
tree | commitdiff |
2010-11-02 |
bdemsky | fix so that profiling doesn't throw an error |
tree | commitdiff |
2010-11-01 |
jjenista | update poolcreate args for deque version as well |
tree | commitdiff |
2010-11-01 |
bdemsky | clean stuff up a little |
tree | commitdiff |
2010-11-01 |
bdemsky | convert CAS -> XCHG when possible....no retry necessary... |
tree | commitdiff |
2010-10-29 |
jjenista | commit some useful debug stuff all turned off until... |
tree | commitdiff |
2010-10-29 |
jjenista | for original thread to become worker zero |
tree | commitdiff |
2010-10-29 |
bdemsky | bug fix....have to make sure we don't insert task into... |
tree | commitdiff |
2010-10-28 |
jjenista | even though man page says otherwise, addr and len to... |
tree | commitdiff |
2010-10-27 |
jjenista | add feature to memPool for detecting misuse of records... |
tree | commitdiff |
2010-10-27 |
jjenista | make sure mistakes are obvious with identifiable values |
tree | commitdiff |
2010-10-27 |
yeom | changes: to prevent unnecessary LOCKXCHG when the unres... |
tree | commitdiff |
2010-10-26 |
bdemsky | really bad bug |
tree | commitdiff |
2010-10-26 |
bdemsky | single ended queue work stealing scheduler |
tree | commitdiff |
2010-10-26 |
jjenista | couple of minor bug fixes |
tree | commitdiff |
2010-10-26 |
jjenista | change calloc to RUNMALLOC, added some optional debug... |
tree | commitdiff |
2010-10-26 |
bdemsky | optimize dispatch a little more |
tree | commitdiff |
2010-10-26 |
bdemsky | Changes... small fix to jim's bugfix...touching our... |
tree | commitdiff |
2010-10-25 |
bdemsky | fix bug w/ recycling + RCR |
tree | commitdiff |
2010-10-25 |
jjenista | this is a BUG, don't dereference the object pointer... |
tree | commitdiff |
2010-10-25 |
stephey | Added necessary variable initializations |
tree | commitdiff |
2010-10-25 |
bdemsky | make the traversers do something |
tree | commitdiff |
2010-10-25 |
jjenista | Changing OoOJava work scheduler to true work stealing... |
tree | commitdiff |
2010-10-25 |
bdemsky | changes |
tree | commitdiff |
2010-10-24 |
bdemsky | bug fixes |
tree | commitdiff |
2010-10-24 |
bdemsky | changes...handle races between coarse and fine grained... |
tree | commitdiff |
2010-10-23 |
bdemsky | changes |
tree | commitdiff |
2010-10-22 |
bdemsky | bug |
tree | commitdiff |
2010-10-22 |
bdemsky | changes |
tree | commitdiff |
2010-10-22 |
bdemsky | more core...fix traversers... |
tree | commitdiff |
2010-10-22 |
bdemsky | still some bugs...but stall site code generation is in |
tree | commitdiff |
2010-10-22 |
bdemsky | changes towards rcr working... |
tree | commitdiff |
2010-10-21 |
jjenista | implemented deque for work stealing |
tree | commitdiff |
2010-10-21 |
bdemsky | more changes....much is still missing |
tree | commitdiff |
2010-10-21 |
bdemsky | changes |
tree | commitdiff |
2010-10-21 |
bdemsky | changes |
tree | commitdiff |
2010-10-20 |
bdemsky | changes |
tree | commitdiff |
2010-10-20 |
bdemsky | changes |
tree | commitdiff |
2010-10-20 |
bdemsky | more work towards working version |
tree | commitdiff |
2010-10-19 |
bdemsky | changes |
tree | commitdiff |
2010-10-19 |
bdemsky | let the prefetch go through a register instead... |
tree | commitdiff |
2010-10-19 |
bdemsky | changes |
tree | commitdiff |
2010-10-19 |
jjenista | switch to a thread local stall sem for parent to give... |
tree | commitdiff |
2010-10-19 |
jzhou | Add support for 'static' fields in multicore gc version... |
tree | commitdiff |
2010-10-18 |
bdemsky | more changes towards oooJava version 2 |
tree | commitdiff |
2010-10-18 |
bdemsky | more changes towards new version of oooJava |
tree | commitdiff |
2010-10-14 |
jzhou | Add process of global thread queue in gc, not tested yet |
tree | commitdiff |
2010-10-14 |
jzhou | Synchronized block is now working. A simple test is... |
tree | commitdiff |
2010-10-14 |
jjenista | coreprof updates, nothing huge |
tree | commitdiff |
2010-10-14 |
jjenista | hash of OIDs should not bit shift |
tree | commitdiff |
2010-10-14 |
jzhou | Fix bug in termination of the multicore gc version... |
tree | commitdiff |
2010-10-14 |
jzhou | Add the runtime for multicore gc version w/o tasks... |
tree | commitdiff |
2010-10-13 |
jjenista | bug fixes for oooJava, some debug events for coreprof... |
tree | commitdiff |
2010-10-13 |
jjenista | added prefetch when grabbing a record for next one... |
tree | commitdiff |
2010-10-13 |
bdemsky | changes...plus a little optimization |
tree | commitdiff |
2010-10-13 |
bdemsky | changes |
tree | commitdiff |
2010-10-13 |
jzhou | Changes to accommodate the runtime for multicore gc... |
tree | commitdiff |
next |