aboutsummaryrefslogtreecommitdiff
path: root/src/gpgpu-sim/shader.h
AgeCommit message (Collapse)Author
2014-08-14Merging Power model into FermiTayler Hetherington
//depot/gpgpu_sim_research/fermi_power/distribution/... to //depot/gpgpu_sim_research/fermi/distribution/... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 14723]
2014-08-14Revision #2 of modifying the cache hierarchy.Tayler Hetherington
Separated the L1 and L2 cache access() implementations. Removed PRIVATE/SHARED cache scope configurations. Added WRITE_EVICT cache write policy. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 14109]
2014-08-14Modified the cache hierarchy, reorganized code to eliminate code ↵Tayler Hetherington
replication, implemented write allocate / write back policies in L2 cache, added configurable parameters in gpgpusim.config ("W" = Write Allocate, "N" = No write allocate -> "P" = Private, "S" = shared), modified the cache configuration lines to always be separated by ":" instead of ":" and ",", and modified L1 and L2 data cache to be "Write Back" caches instead of "Read Only". Still need to implement Ahmed's sectored cache implementation. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 14081]
2014-08-14Fixing compile error on my machineAndrew M. B. Boktor
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13703]
2014-08-14Fixing an error reported by valgrind.Tim Rogers
Is really harmless since you can call delete on primatives allocated with new[] without consequence (all delete [] does is ensure that the destructors will be called) Still, it is always good practice to de-allocate anything allocated with new[] with delete[] - just in case the type allocated changes and the person that changes it forgets to update the de-allocation. Really, this little array should be on the stack... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13414]
2014-08-14Fix for bug 9: Now querying the state of the pdom stack in call_imp and ↵Ayub Gubran
callp_imp using a core_t function, thus moving the querying function into the abstract model of the core instead of shader_core_ctx which represents the performance mode. This code simplify the querying and also avoid the unnecessary calling hierarchy that was used. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13375]
2014-08-14Adding a two level scheduler as described in the ISCA 2012 tutorialAndrew M. B. Boktor
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 13261]
2014-08-14Removing some bottlenecks that limit that peak-IPCAndrew M. B. Boktor
- FUs depended on the result bus to know if they are going to be used on a certain cycle, this is not the case anymore, occupied bitvectors are added - A configurable number of result buses is added (the number of buses is equal to the EX_WB pipe width) - Modified the Fermi config file to add two ports to the operand collector IPC with a theoretical limit of number_of_SMs*64 is achievable using this configuration [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 12349]
2014-08-14Adding configurable instruction latencies and initiation intervalsAndrew M. B. Boktor
The observed latencies are reduced by 5 to account for other stages in the SM pipeline Eventually this should be calibrated against the microbenchmarks [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 12310]
2014-08-14This changelist adds the following:Andrew M. B. Boktor
1. A configurable number of functional units within each SM 2. A configurable pipeline widths (i.e. Issue width, writeback width ...). Merging //depot/gpgpu_sim_research/fermi_replay/distribution/src/... to //depot/gpgpu_sim_research/fermi/distribution/src/... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 12091]
2014-08-14-Bug 146 fix (Adding perfect memory interface)Ahmed El-Shafiey
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11877]
2014-08-14Changed arch_rech type to store 16 registers, 8 input and 8 output. 8 inputs ↵Inderpreet Singh
because one can have 4 input operands and 4 register operands in a surface store instruction. Fixed arch_regs for memory instructions being ignored in the pre-decode statge. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11576]
2014-08-14Adding a check for copmliance between the runtime simulation config and ↵Wilson Fung
MAX_THREAD_PER_SM. Also decreased LOCAL_MEM_SIZE_MAX to 8kB to make it fit within our allotted memory space (otherwise the simulator may mistaken global memory access as local memory accesses). [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11520]
2014-08-14Changes needed for the new fermi configs to work.Andrew M. B. Boktor
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11511]
2014-08-14Adding option 'gpgpu_simt_core_sim_order' which allow the user to specify ↵Wilson Fung
the order in which cores are simulator per cycle. Also adding support for calling function with empty parameter list. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11489]
2014-08-14Grouped all instruction counting code into a common member function in ↵Wilson Fung
shader_core_ctx. Now m_num_sim_insn counts scalar thread instructions. A new counter is added for warp instructions. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11472]
2014-08-14Fixed the stat collection for gpgpu_n_shmem_insn. See Bug 128 for more ↵Wilson Fung
detail. For verification, I added a directed test with a pre-calculated number of shared memory instructions. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11454]
2014-08-14Integrating the pure functional simulationAyub Gubran
Merging //depot/gpgpu_sim_research/fermi_ayoub/distribution/src/gpgpu-sim/gpu-sim.cc //depot/gpgpu_sim_research/fermi_ayoub/distribution/src/gpgpu-sim/gpu-sim.h //depot/gpgpu_sim_research/fermi_ayoub/distribution/src/gpgpu-sim/shader.cc //depot/gpgpu_sim_research/fermi_ayoub/distribution/src/gpgpu-sim/shader.h to //depot/gpgpu_sim_research/fermi/distribution/src/gpgpu-sim/... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 11287]
2014-08-14Fixing the atomics I broke with the insn count fixTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10754]
2014-08-14Fixing the varying instruction count when the cache configuration changes.Tim Rogers
The problem was 2-fold: 1) If all the lanes hit, we were not incrementing the instruction count since the instruction did not progress to the WB stage, the scoreboard relasing is done in the ldst_unit::cycle function for cache hits. Added code there to increment the instrcution count. 2) Loads were some lanes hit in cache and some lanes do not were not being comepletely counted. Only the lanes sent off to the memory system were being counted because we were setting the warp's active lanes to the access's acrtive lanes.... I am not sure why this code was there... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10737]
2014-08-14Integration change. CL 8980 - l1 cache stat printTim Rogers
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10726]
2014-08-14Fix for Bug 110 - integrates in CL 10258 from tm-test branchInderpreet Singh
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 10299]
2014-08-14change copyright notice to include authorsTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9872]
2011-06-29changing copyright to BSDTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9687]
2011-05-25Fix bug #100: local memory address translation returns multiple addressesInderpreet Singh
Fix bug #101: Coalescing allows multiple accesses per thread for local memory access This will break atomics which assume at most one thread per mem_fetch. It did not break scoreboard as that logic tracks mem_fetches at warp level, not thread level. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 9303]
2011-03-03refactor pipeline stage namesTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8544]
2011-02-01Added configurable schedulers!aturner
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8472]
2011-01-24Adds highly configurable opperand collectoraturner
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8407]
2011-01-20Integration change. Bug fixes from AMD-CMU trace gen branch. Wilson Fung
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8389]
2010-12-28- parameter memory and active threads now part of kernel_info_t:Tor Aamodt
Parameters are finalized at kernel launch, which means the contents of parameter memory are initialized. Kernel arguement names have a fixed order, hence same address should be assigned on subsequent kernel launches of same kernel in other streams provided the data size param_t::size of arguments for each kernel launch is identical (an assertion has been added to check this is true). - passing regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8303]
2010-12-28- Checkpointing new support for concurrent kernel execution (CUDA only, not ↵Tor Aamodt
OpenCL) This changelist adds full support for streams supported by a new class, stream_manager and enables concurrent execution of kernels from different streams. - fast_regression.sh fails for simpleMultiCopy, simpleStreams (other tests passing) ** Known issues ** - Kernel parameter passing is not done correctly for concurrent kernel execution (somehow concurrentKernels is not affected by this): the parameters are stored inside function_info, which is shared among parallel kernel launches so that the values passed into the launch are likely to get overwritten if multiple grids are launched in parallel streams. - Statistics are printed out whenever the simulation thread runs out of cuda commands (doesn't make sense to print out when a kernel ends during concurrent kernel execution). This will probably require further tweaking so as to be more compatible with data collection scripts. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8302]
2010-11-30integrate changes (makes code more modular, i would argue)Tor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8179]
2010-11-29make an explicit read operands stageTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8174]
2010-11-29integrate mask changesTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8170]
2010-11-28enabling L2 data cache... it is write through, write evict like L1.Tor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8154]
2010-11-28adding 1st level data cacheTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 8153]
2010-10-240.9756 correlation. Set L1T line size to 128 bytes... problem wasTor Aamodt
stalling to send four requests per warp into L1T tag lookup. If L1T is really 32B blocks (as per Henry's paper), this suggests banking of L1T needs to be modeled. Other changes: 1. bug fix in memory access generation for texture/const cache access 2. adding back memory latency measurement for visualizer [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7913]
2010-10-241. updates to .gdbinit fileTor Aamodt
2. update texture to bypass ROP-delay queue... correlation now 0.9592 [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7912]
2010-10-24add back per shader icount tracking for visualizerTor Aamodt
[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7910]
2010-10-241. adding top level configuration class and making shader and memory ↵Tor Aamodt
configuration components of this class. 2. clock memory pipeline no. subwarp times for each shader clock and increase rob-size for texture cache (trying to improve correlation, currently at 0.9218) 3. start to modify shader stats to add back features for visualizer (warp divergence distribution kind of working again) passing cuda 3.1 regression and ptxplus correlation tests [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7909]
2010-10-211. rewriting memory access generation code (from scratch), why not...Tor Aamodt
passing CUDA 3.1 and ptxplus correlation, but correlation still bad (0.62)... after debugging 1 to get it working with ptxplus, problem is very clear: shared and constant cache accesses not occuring for operations that combine these with ALU operations. TODO: have a "read-operands" stage, which somehow combines operand collector register reading with shared and const memory accesses... [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7895]
2010-10-19adding texture cache model with fragment fifo for latency hidingTor Aamodt
passing CUDA 3.1 regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7886]
2010-10-18Re-designed cache model:Tor Aamodt
- read only cache model with integrated mshrs (no L1D, yet); new cache interface should be easily extendable to support texture cache with latency fifo and separate tag/data arrays, though this is not yet added (currently tags and data arrays are not decoupled for texture) - new partition model using the above removes all old MSHRs, L1D etc... passing CUDA 3.1 regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7875]
2010-10-161. creating cache_config object to encapsulate cache configuration informationTor Aamodt
(and parse it before creating the simulator objects). 2. creating core_config to hold only features of a shader_core that are high level enough either (a) the functional simulator needs to know about them, or (b) they affect memory *access* generation. 3. in config files only (so far) separate out notion of write-{through,back}, from notion of when a line is allocated... will use this to distinguish different types of caches. passing CUDA 3.1 regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7870]
2010-10-161. moving address decoding into a class (and out of cache entirely)Tor Aamodt
2. moving DRAM timing parameters into memory_config class 3. removing some likely useless statistics 4. other cleaning up of code passing CUDA 3.1 regression [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7868]
2010-10-121. adding simt_core_cluster, which models a TPC or (for fermi) GPC...Tor Aamodt
this gives us a place to stick caches shared among shader cores but on the shader side of the interconnect... maybe move the clock boundary code here? after integrating booksim 2 code? 2. added a pending write table to ldst_unit rather than scoreboard ... rationale is that ld/st unit needs to process register writes once it is done it can notify scoreboard once. 3. re-enabled shared memory delay (use pipeline within ldst_unit) 4. re-enabling operand collector writeback for all instruction types 5. disable MSHRs in this change list passing CUDA 3.1 regression next? texture cache, then redo mshrs? [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7845]
2010-10-101. create function unit classes for SP, SFU, LD/ST.Tor Aamodt
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]
2010-10-09Refactoring:Tor Aamodt
1. Moving mem_access_t to abstract_hardware_model and making set (queue) of accesses part of warp_inst_t. I.e., treat set of accesses as an ISA concept rather than a hardware organization concept. This is only partly "done"... logic for computing accesses is still part of shader_core_ctx in this CL. Given number of warp_inst_t accessors for accessq, now seems like we might even want to move some memory stage code into warp_inst_t class. How those accesses make it to memory system is the hardware concept. 2. Making warp_inst_t an explicit arguement of subroutines used in memory stage... The eventual goal here is (likely) to refactor memory into a hardware block... i.e., have function units be a class that contains some set of pipeline stages internally and some set of input/output "ports". 3. Moving accessor functions is_load, is_store; is_const, is_local into class declaration (where they belong). 4. Removing code for selecting pipeline uarch (might add it back later, but first want a clean GT200 organization). In particular, removing option to have an operand collector -- now you MUST have the operand collector. 5. Removing more deadcode from prior changes (fixed delay queue related) Scripts/configs: 6. Correlation script not printing out exit condition when hardware launch fails 7. Update config files to have proper compute model selected [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7834]
2010-10-081. modify shader_core_ctx::execute_pipe() to model instruction throughput ↵Tor Aamodt
correctly (off-by-one error) 2. adding code to dump_pipeline to display reason for memory stage stalling 3. removing dead code resulting from prior changes correlation vs. GT200 is 0.95 (need to add back modeling of memory writebacks, shared memory latency) [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7831]
2010-10-081. refactoring cuda api code for loading PTX (removing external PTX loading ↵Tor Aamodt
entirely) 2. some bug fixes for warp_inst_t 3. creating a new class, gpgpu_t, which contains the functional "memory" state visible to all threads running on a GPU (doing this as part of my continuing effort to hunt down and eradicate every global variable that is not the top level "the gpu") 4. other misc. changes Almost passing CUDA 3.1 regression? oclHistogram keeps failing under torque, but does not fail when run on the command line from the same directory. [git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7827]
generated by cgit v1.3 (git 2.53.0) at 2026-07-16 18:55:52 +0000