index
:
weft
main
CPEN 511 Project - Weft: Improving SIMD Utilization through MIMD-like Co-issue and Thread Compaction
Davit Grigoryan
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
src
Age
Commit message (
Expand
)
Author
2014-08-14
Added read to precharge constraint - negligible effect to DRAM efficiency.
Inderpreet Singh
2014-08-14
release!?!
Tor Aamodt
2014-08-14
Fixed the DRAM timing model to add the write-read turn and write-precharge de...
Wilson Fung
2014-08-14
ready to release?
Tor Aamodt
2014-08-14
Fixes for atomic callbacks
arun
2014-08-14
Integration change from CL8943 to fix barrier behaviour.
Wilson Fung
2014-08-14
fixup some dangling references
Tor Aamodt
2014-08-14
change copyright notice to include authors
Tor Aamodt
2014-08-14
Fixing comment clobber from yesterday
Tim Rogers
2011-07-07
Integration change. Bringing in some changes from mem_divergence that allow f...
Tim Rogers
2011-06-30
version info
Tor Aamodt
2011-06-29
changing copyright to BSD
Tor Aamodt
2011-06-29
Adding back stat for memory divergence
Tim Rogers
2011-05-28
fix for bug 103
Tor Aamodt
2011-05-26
Another local memory address translation bug fix - it now adds an offset to p...
Inderpreet Singh
2011-05-25
Bug fix for local memory address translation that was made in tm-test branch ...
Inderpreet Singh
2011-05-25
Fix bug #100: local memory address translation returns multiple addresses
Inderpreet Singh
2011-03-03
refactor pipeline stage names
Tor Aamodt
2011-02-01
Added configurable schedulers!
aturner
2011-01-24
Adds highly configurable opperand collector
aturner
2011-01-20
Integration change. Bug fixes from AMD-CMU trace gen branch.
Wilson Fung
2011-01-02
integrate
Tor Aamodt
2011-01-02
integrate bug fix (passes fast regression)
Tor Aamodt
2010-12-28
the assertion
Tor Aamodt
2010-12-28
- parameter memory and active threads now part of kernel_info_t:
Tor Aamodt
2010-12-28
- Checkpointing new support for concurrent kernel execution (CUDA only, not O...
Tor Aamodt
2010-12-21
added support for negative .f64 operands in decuda_to_ptxplus
Jimmy Kwa
2010-12-20
fix from CL 8285
Tor Aamodt
2010-12-15
Added next instruction type to ptxplus, ".ff64". It's the same as ".f64" exce...
Jimmy Kwa
2010-11-30
integrate changes (makes code more modular, i would argue)
Tor Aamodt
2010-11-29
make an explicit read operands stage
Tor Aamodt
2010-11-29
integrate mask changes
Tor Aamodt
2010-11-28
bug fix for ptxplus w/ data cache disabled
Tor Aamodt
2010-11-28
enabling L2 data cache... it is write through, write evict like L1.
Tor Aamodt
2010-11-28
adding 1st level data cache
Tor Aamodt
2010-11-15
Cleaned up decuda_to_ptxplus code to remove compiler warnings
Jimmy Kwa
2010-10-24
0.9756 correlation. Set L1T line size to 128 bytes... problem was
Tor Aamodt
2010-10-24
1. updates to .gdbinit file
Tor Aamodt
2010-10-24
1. fix load imbalance issue (CTA's were filling up first core in a cluster be...
Tor Aamodt
2010-10-24
add back per shader icount tracking for visualizer
Tor Aamodt
2010-10-24
1. adding top level configuration class and making shader and memory configur...
Tor Aamodt
2010-10-22
enables global loads/stores for ptxplus
Tor Aamodt
2010-10-21
1. rewriting memory access generation code (from scratch), why not...
Tor Aamodt
2010-10-19
adding texture cache model with fragment fifo for latency hiding
Tor Aamodt
2010-10-18
update lru state on hit
Tor Aamodt
2010-10-18
Re-designed cache model:
Tor Aamodt
2010-10-16
1. creating cache_config object to encapsulate cache configuration information
Tor Aamodt
2010-10-16
1. refactoring histogram/logger so that classes are in header files
Tor Aamodt
2010-10-16
1. moving address decoding into a class (and out of cache entirely)
Tor Aamodt
2010-10-12
1. adding simt_core_cluster, which models a TPC or (for fermi) GPC...
Tor Aamodt
[next]