diff options
| author | Tor Aamodt <[email protected]> | 2010-10-10 21:19:52 -0800 |
|---|---|---|
| committer | Tor Aamodt <[email protected]> | 2010-10-10 21:19:52 -0800 |
| commit | b3ce70a797756285ea9b15b3e5cf515d8b6a2b63 (patch) | |
| tree | 5b84d880e646f80611cc5b01931fdde622c5e06c /configs | |
| parent | 0151c5c566dcbcbb962acf35c86a8e0f4a4a6eb6 (diff) | |
1. create function unit classes for SP, SFU, LD/ST.
2. refactor memory stage into a ld/st function unit
3. refactor memory access generation (moved into warp_inst_t class)
the above should make supporting fermi uarch much easier
passing CUDA 3.1 regression
still need to...
(a) update scoreboard to keep count of outstanding memory requests
and use operand collector for writebacks into register file
(b) add back shared memory pipeline delay
(c) remove use of MSHR's for non-cached global/local accesses
(d) replace texture cache with a split tag/data array pipe
(e) re-implement memory_partition stuff so it makes more sense
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7844]
Diffstat (limited to 'configs')
| -rw-r--r-- | configs/QuadroFX5800/gpgpusim.config | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/configs/QuadroFX5800/gpgpusim.config b/configs/QuadroFX5800/gpgpusim.config index 2805674..e5e29bc 100644 --- a/configs/QuadroFX5800/gpgpusim.config +++ b/configs/QuadroFX5800/gpgpusim.config @@ -4,7 +4,6 @@ -gpgpu_ptx_force_max_capability 11 # high level architecture configuration --gpgpu_sm_uarch GT200 -gpgpu_n_shader 30 -gpgpu_n_mem 8 -gpgpu_clock_domains 325.0:650.0:650.0:800.0 @@ -56,8 +55,7 @@ -gpgpu_ptx_save_converted_ptxplus 1 # enable operand collector --gpgpu_operand_collector 1 --gpgpu_operand_collector_num_units 6 +-gpgpu_operand_collector_num_units_sp 6 -gpgpu_operand_collector_num_units_sfu 8 -visualizer_enabled 0 |