Searched hist:50 (Results 301 - 325 of 671) sorted by relevance
/gem5/src/mem/ | ||
H A D | dramsim2.hh | 10066:06a33d872798 Tue Feb 18 05:50:00 EST 2014 Andreas Hansson <andreas.hansson@arm.com> mem: Add a wrapped DRAMSim2 memory controller This patch adds DRAMSim2 as a memory controller by wrapping the external library and creating a sublass of AbstractMemory that bridges between the semantics of gem5 and the DRAMSim2 interface. The DRAMSim2 wrapper extracts the clock period from the config file. There is no way of extracting this information from DRAMSim2 itself, so we simply read the same config file and get it from there. To properly model the response queue, the wrapper keeps track of how many transactions are in the actual controller, and how many are stacking up waiting to be sent back as responses (in the wrapper). The latter requires us to move away from the queued port and manage the packets ourselves. This is due to DRAMSim2 not having any flow control on the response path. DRAMSim2 assumes that the transactions it is given are matching the burst size of the choosen memory. The wrapper checks to ensure the cache line size of the system matches the burst size of DRAMSim2 as there are currently no provisions to split the system requests. In theory we could allow a cache line size smaller than the burst size, but that would lead to inefficient use of the DRAM, so for not we fatal also in this case. |
/gem5/tests/ | ||
H A D | tests.py | diff 11543:b5435e0310c7 Mon Jun 20 09:50:00 EDT 2016 Andreas Sandberg <andreas.sandberg@arm.com> tests: Add a test command to get test status as an exit code Add a "test" command to tests.py that queries a test pickle file and returns different exit codes depending on the outcome of the tests in the file. The following exit codes can currently be returned: * 0: All tests were successful or skipped. * 1: General fault in the script such as incorrect parameters or failing to parse a pickle file. * 2: At least one test failed to run. This is what the summary formatter usually shows as a 'FAILED'. * 3: All tests ran correctly, but at least one failed to verify its output. When displaying test output using the summary formatter, such a test would show up as 'CHANGED'. The command can be invoked like this: ./tests/tests.py test `find build/ARM/tests/opt/ -name status.pickle` Change-Id: I7e6bc661516f38ff08dfda7c4359a1e10bf97864 Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> |
/gem5/util/ | ||
H A D | decode_packet_trace.py | diff 10065:58bf21ca88de Tue Feb 18 05:50:00 EST 2014 Andreas Hansson <andreas.hansson@arm.com> util: Enhance the error messages for packet encode/decode This patch adds a more verbose error message when the Python protobuf module cannot be loaded. |
/gem5/util/stats/ | ||
H A D | profile.py | diff 2006:3ca085495c69 Tue Nov 22 21:50:00 EST 2005 Nathan Binkert <binkertn@umich.edu> Major improvements in the graph output code. Mostly adding more options, making existing options more visible and dealing with holes in data better. util/stats/barchart.py: - move the options for BarChart to a base class ChartOptions so they can be more easily set and copied. - add an option to set the chart size (so you can adjust the aspect ratio) - don't do the add_subplot thing, use add_axes directly so we can affect the size of the figure itself to make room for the legend - make the initial array bottom floating point so we don't lose precision - add an option to set the limits on the y axis - use a figure legend instead of an axes legend so we can put the legend outside of the actual chart. Also add an option to set the fontsize of the legend. - initial hack at outputting csv files util/stats/db.py: don't print out an error when the run is missing from the database just return None, the error will be print elsewhere. util/stats/output.py: - make StatOutput derive from ChartOptions so that it's easier to set default chart options. - make the various output functions (graph, display, etc.) take the name of the data as a parameter instead of making it a parameter to __init__. This allows me to create the StatOutput object with generic parameters while still being able to specialize the name after the fact - add support for graph_group and graph_bars to be applied to multiple configuration groups. This results in a cross product of the groups to be generated and used. - flush the html file output as we go so that we can load the file while graphs are still being generated. - make the proxy a parameter to the graph function so the proper system's data can be graphed - for any groups or bars that are completely missing, remove them from the graph. This way, if we decide not to do a set of runs, there won't be holes in the data. - output eps and ps by default in addition to the png. util/stats/profile.py: - clean up the data structures that are used to store the function profile information and try our best to avoid keeping extra data around that isn't used. - make get() return None if a job is missing so we know it was missing rather than the all zeroes thing. - make the function profile categorization stuff total up to 100% - Fixup the x-axis and y-axis labels. - fix the dot file output stuff. util/stats/stats.py: support the new options stuff for StatOutput |
/gem5/configs/example/ | ||
H A D | garnet_synth_traffic.py | diff 11688:725fef71f376 Wed Oct 26 14:50:00 EDT 2016 Andreas Hansson <andreas.hansson@arm.com> config: Break out base options for usage with NULL ISA This patch breaks out the most basic configuration options into a set of base options, to allow them to be used also by scripts that do not involve any ISA, and thus no actual CPUs or devices. The patch also fixes a few modules so that they can be imported in a NULL build, and avoid dragging in FSConfig every time Options is imported. |
/gem5/src/cpu/pred/ | ||
H A D | bi_mode.cc | 10244:d2deb51a4abf Mon Jun 30 13:50:00 EDT 2014 Anthony Gutierrez <atgutier@umich.edu> cpu: implement a bi-mode branch predictor |
H A D | 2bit_local.cc | 6226:f1076450ab2b Fri Jun 05 00:50:00 EDT 2009 Nathan Binkert <nate@binkert.org> move: put predictor includes and cc files into the same place |
H A D | 2bit_local.hh | 6226:f1076450ab2b Fri Jun 05 00:50:00 EDT 2009 Nathan Binkert <nate@binkert.org> move: put predictor includes and cc files into the same place |
H A D | bi_mode.hh | 10244:d2deb51a4abf Mon Jun 30 13:50:00 EDT 2014 Anthony Gutierrez <atgutier@umich.edu> cpu: implement a bi-mode branch predictor |
/gem5/src/cpu/minor/ | ||
H A D | decode.cc | 10259:ebb376f73dd2 Wed Jul 23 17:09:00 EDT 2014 Andrew Bardsley <Andrew.Bardsley@arm.com> cpu: `Minor' in-order CPU model This patch contains a new CPU model named `Minor'. Minor models a four stage in-order execution pipeline (fetch lines, decompose into macroops, decompose macroops into microops, execute). The model was developed to support the ARM ISA but should be fixable to support all the remaining gem5 ISAs. It currently also works for Alpha, and regressions are included for ARM and Alpha (including Linux boot). Documentation for the model can be found in src/doc/inside-minor.doxygen and its internal operations can be visualised using the Minorview tool utils/minorview.py. Minor was designed to be fairly simple and not to engage in a lot of instruction annotation. As such, it currently has very few gathered stats and may lack other gem5 features. Minor is faster than the o3 model. Sample results: Benchmark | Stat host_seconds (s) ---------------+--------v--------v-------- (on ARM, opt) | simple | o3 | minor | timing | timing | timing ---------------+--------+--------+-------- 10.linux-boot | 169 | 1883 | 1075 10.mcf | 117 | 967 | 491 20.parser | 668 | 6315 | 3146 30.eon | 542 | 3413 | 2414 40.perlbmk | 2339 | 20905 | 11532 50.vortex | 122 | 1094 | 588 60.bzip2 | 2045 | 18061 | 9662 70.twolf | 207 | 2736 | 1036 |
H A D | fetch2.hh | 10259:ebb376f73dd2 Wed Jul 23 17:09:00 EDT 2014 Andrew Bardsley <Andrew.Bardsley@arm.com> cpu: `Minor' in-order CPU model This patch contains a new CPU model named `Minor'. Minor models a four stage in-order execution pipeline (fetch lines, decompose into macroops, decompose macroops into microops, execute). The model was developed to support the ARM ISA but should be fixable to support all the remaining gem5 ISAs. It currently also works for Alpha, and regressions are included for ARM and Alpha (including Linux boot). Documentation for the model can be found in src/doc/inside-minor.doxygen and its internal operations can be visualised using the Minorview tool utils/minorview.py. Minor was designed to be fairly simple and not to engage in a lot of instruction annotation. As such, it currently has very few gathered stats and may lack other gem5 features. Minor is faster than the o3 model. Sample results: Benchmark | Stat host_seconds (s) ---------------+--------v--------v-------- (on ARM, opt) | simple | o3 | minor | timing | timing | timing ---------------+--------+--------+-------- 10.linux-boot | 169 | 1883 | 1075 10.mcf | 117 | 967 | 491 20.parser | 668 | 6315 | 3146 30.eon | 542 | 3413 | 2414 40.perlbmk | 2339 | 20905 | 11532 50.vortex | 122 | 1094 | 588 60.bzip2 | 2045 | 18061 | 9662 70.twolf | 207 | 2736 | 1036 |
H A D | pipeline.cc | 10259:ebb376f73dd2 Wed Jul 23 17:09:00 EDT 2014 Andrew Bardsley <Andrew.Bardsley@arm.com> cpu: `Minor' in-order CPU model This patch contains a new CPU model named `Minor'. Minor models a four stage in-order execution pipeline (fetch lines, decompose into macroops, decompose macroops into microops, execute). The model was developed to support the ARM ISA but should be fixable to support all the remaining gem5 ISAs. It currently also works for Alpha, and regressions are included for ARM and Alpha (including Linux boot). Documentation for the model can be found in src/doc/inside-minor.doxygen and its internal operations can be visualised using the Minorview tool utils/minorview.py. Minor was designed to be fairly simple and not to engage in a lot of instruction annotation. As such, it currently has very few gathered stats and may lack other gem5 features. Minor is faster than the o3 model. Sample results: Benchmark | Stat host_seconds (s) ---------------+--------v--------v-------- (on ARM, opt) | simple | o3 | minor | timing | timing | timing ---------------+--------+--------+-------- 10.linux-boot | 169 | 1883 | 1075 10.mcf | 117 | 967 | 491 20.parser | 668 | 6315 | 3146 30.eon | 542 | 3413 | 2414 40.perlbmk | 2339 | 20905 | 11532 50.vortex | 122 | 1094 | 588 60.bzip2 | 2045 | 18061 | 9662 70.twolf | 207 | 2736 | 1036 |
H A D | dyn_inst.hh | 10259:ebb376f73dd2 Wed Jul 23 17:09:00 EDT 2014 Andrew Bardsley <Andrew.Bardsley@arm.com> cpu: `Minor' in-order CPU model This patch contains a new CPU model named `Minor'. Minor models a four stage in-order execution pipeline (fetch lines, decompose into macroops, decompose macroops into microops, execute). The model was developed to support the ARM ISA but should be fixable to support all the remaining gem5 ISAs. It currently also works for Alpha, and regressions are included for ARM and Alpha (including Linux boot). Documentation for the model can be found in src/doc/inside-minor.doxygen and its internal operations can be visualised using the Minorview tool utils/minorview.py. Minor was designed to be fairly simple and not to engage in a lot of instruction annotation. As such, it currently has very few gathered stats and may lack other gem5 features. Minor is faster than the o3 model. Sample results: Benchmark | Stat host_seconds (s) ---------------+--------v--------v-------- (on ARM, opt) | simple | o3 | minor | timing | timing | timing ---------------+--------+--------+-------- 10.linux-boot | 169 | 1883 | 1075 10.mcf | 117 | 967 | 491 20.parser | 668 | 6315 | 3146 30.eon | 542 | 3413 | 2414 40.perlbmk | 2339 | 20905 | 11532 50.vortex | 122 | 1094 | 588 60.bzip2 | 2045 | 18061 | 9662 70.twolf | 207 | 2736 | 1036 |
H A D | dyn_inst.cc | 10259:ebb376f73dd2 Wed Jul 23 17:09:00 EDT 2014 Andrew Bardsley <Andrew.Bardsley@arm.com> cpu: `Minor' in-order CPU model This patch contains a new CPU model named `Minor'. Minor models a four stage in-order execution pipeline (fetch lines, decompose into macroops, decompose macroops into microops, execute). The model was developed to support the ARM ISA but should be fixable to support all the remaining gem5 ISAs. It currently also works for Alpha, and regressions are included for ARM and Alpha (including Linux boot). Documentation for the model can be found in src/doc/inside-minor.doxygen and its internal operations can be visualised using the Minorview tool utils/minorview.py. Minor was designed to be fairly simple and not to engage in a lot of instruction annotation. As such, it currently has very few gathered stats and may lack other gem5 features. Minor is faster than the o3 model. Sample results: Benchmark | Stat host_seconds (s) ---------------+--------v--------v-------- (on ARM, opt) | simple | o3 | minor | timing | timing | timing ---------------+--------+--------+-------- 10.linux-boot | 169 | 1883 | 1075 10.mcf | 117 | 967 | 491 20.parser | 668 | 6315 | 3146 30.eon | 542 | 3413 | 2414 40.perlbmk | 2339 | 20905 | 11532 50.vortex | 122 | 1094 | 588 60.bzip2 | 2045 | 18061 | 9662 70.twolf | 207 | 2736 | 1036 |
/gem5/src/mem/cache/prefetch/ | ||
H A D | queued.hh | diff 13551:f352df8e2863 Sat Nov 17 18:50:00 EST 2018 Javier Bueno <javier.bueno@metempsy.com> mem-cache: virtual address support for prefetchers Prefetchers can be configured to operate with virtual or physical addreses. The option can be configured through the "use_virtual_addresses" parameter of the Prefetcher object. Change-Id: I4f8c3687988afecc8a91c3c5b2d44cc0580f72aa Reviewed-on: https://gem5-review.googlesource.com/c/14416 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> |
/gem5/src/mem/ruby/common/ | ||
H A D | SConscript | diff 8492:1ad244a20877 Mon Aug 08 11:50:00 EDT 2011 Nilay Vaish<nilay@cs.wisc.edu> BuildEnv: Eliminate RUBY as build environment variable This patch replaces RUBY with PROTOCOL in all the SConscript files as the environment variable that decides whether or not certain components of the simulator are compiled. |
/gem5/src/arch/arm/isa/formats/ | ||
H A D | sve_2nd_level.isa | diff 14029:744989da399f Fri Feb 23 08:50:00 EST 2018 Giacomo Gabrielli <giacomo.gabrielli@arm.com> arch-arm: Treat SVE prefetch instructions as no-ops Change-Id: Ife0424e274dd65d6dc4f6e5cc5e37d17b03be0d8 Signed-off-by: Giacomo Gabrielli <giacomo.gabrielli@arm.com> Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/13522 Tested-by: kokoro <noreply+kokoro@google.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> |
/gem5/src/arch/arm/ | ||
H A D | miscregs_types.hh | diff 14094:921238acf537 Thu Jul 18 05:50:00 EDT 2019 Giacomo Travaglini <giacomo.travaglini@arm.com> arch-arm: Add HPD bit for TCR_EL2/EL3 This is controlling Hierarchical Permissions for EL3 and for EL2 when HCR_EL2.E2H = 0. Change-Id: I9d6615ff4980cc56a28dc32088cf1524155f0fa4 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Ciro Santilli <ciro.santilli@arm.com> Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19609 Tested-by: kokoro <noreply+kokoro@google.com> |
/gem5/src/mem/protocol/ | ||
H A D | SConscript | diff 8492:1ad244a20877 Mon Aug 08 11:50:00 EDT 2011 Nilay Vaish<nilay@cs.wisc.edu> BuildEnv: Eliminate RUBY as build environment variable This patch replaces RUBY with PROTOCOL in all the SConscript files as the environment variable that decides whether or not certain components of the simulator are compiled. |
/gem5/src/dev/arm/ | ||
H A D | gic_v3_its.hh | diff 14168:2a96e30b9400 Wed Aug 14 12:50:00 EDT 2019 Giacomo Travaglini <giacomo.travaglini@arm.com> dev-arm: Add GITS_PIDR2 register to the ITS memory map The GITS Peripheral Identification Register #2 bits assignments are the same as those for GICD_PIDR2. Change-Id: I235008a383e08dd557d899cb3aa18202ef943f8b Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/20254 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Tested-by: kokoro <noreply+kokoro@google.com> |
H A D | gic_v3_distributor.hh | diff 14168:2a96e30b9400 Wed Aug 14 12:50:00 EDT 2019 Giacomo Travaglini <giacomo.travaglini@arm.com> dev-arm: Add GITS_PIDR2 register to the ITS memory map The GITS Peripheral Identification Register #2 bits assignments are the same as those for GICD_PIDR2. Change-Id: I235008a383e08dd557d899cb3aa18202ef943f8b Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/20254 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Tested-by: kokoro <noreply+kokoro@google.com> |
/gem5/src/arch/x86/ | ||
H A D | X86LocalApic.py | diff 9162:019047ead23b Tue Aug 21 05:50:00 EDT 2012 Andreas Hansson <andreas.hansson@arm.com> Device: Remove overloaded pio_latency parameter This patch removes the overloading of the parameter, which seems both redundant, and possibly incorrect. The PciConfigAll now also uses a Param.Latency rather than a Param.Tick. For backwards compatibility it still sets the pio_latency to 1 tick. All the comments have also been updated to not state that it is in simticks when it is not necessarily the case. |
/gem5/src/arch/alpha/ | ||
H A D | interrupts.hh | diff 7720:65d338a8dba4 Sun Oct 31 03:07:00 EDT 2010 Gabe Black <gblack@eecs.umich.edu> ISA,CPU,etc: Create an ISA defined PC type that abstracts out ISA behaviors. This change is a low level and pervasive reorganization of how PCs are managed in M5. Back when Alpha was the only ISA, there were only 2 PCs to worry about, the PC and the NPC, and the lsb of the PC signaled whether or not you were in PAL mode. As other ISAs were added, we had to add an NNPC, micro PC and next micropc, x86 and ARM introduced variable length instruction sets, and ARM started to keep track of mode bits in the PC. Each CPU model handled PCs in its own custom way that needed to be updated individually to handle the new dimensions of variability, or, in the case of ARMs mode-bit-in-the-pc hack, the complexity could be hidden in the ISA at the ISA implementation's expense. Areas like the branch predictor hadn't been updated to handle branch delay slots or micropcs, and it turns out that had introduced a significant (10s of percent) performance bug in SPARC and to a lesser extend MIPS. Rather than perpetuate the problem by reworking O3 again to handle the PC features needed by x86, this change was introduced to rework PC handling in a more modular, transparent, and hopefully efficient way. PC type: Rather than having the superset of all possible elements of PC state declared in each of the CPU models, each ISA defines its own PCState type which has exactly the elements it needs. A cross product of canned PCState classes are defined in the new "generic" ISA directory for ISAs with/without delay slots and microcode. These are either typedef-ed or subclassed by each ISA. To read or write this structure through a *Context, you use the new pcState() accessor which reads or writes depending on whether it has an argument. If you just want the address of the current or next instruction or the current micro PC, you can get those through read-only accessors on either the PCState type or the *Contexts. These are instAddr(), nextInstAddr(), and microPC(). Note the move away from readPC. That name is ambiguous since it's not clear whether or not it should be the actual address to fetch from, or if it should have extra bits in it like the PAL mode bit. Each class is free to define its own functions to get at whatever values it needs however it needs to to be used in ISA specific code. Eventually Alpha's PAL mode bit could be moved out of the PC and into a separate field like ARM. These types can be reset to a particular pc (where npc = pc + sizeof(MachInst), nnpc = npc + sizeof(MachInst), upc = 0, nupc = 1 as appropriate), printed, serialized, and compared. There is a branching() function which encapsulates code in the CPU models that checked if an instruction branched or not. Exactly what that means in the context of branch delay slots which can skip an instruction when not taken is ambiguous, and ideally this function and its uses can be eliminated. PCStates also generally know how to advance themselves in various ways depending on if they point at an instruction, a microop, or the last microop of a macroop. More on that later. Ideally, accessing all the PCs at once when setting them will improve performance of M5 even though more data needs to be moved around. This is because often all the PCs need to be manipulated together, and by getting them all at once you avoid multiple function calls. Also, the PCs of a particular thread will have spatial locality in the cache. Previously they were grouped by element in arrays which spread out accesses. Advancing the PC: The PCs were previously managed entirely by the CPU which had to know about PC semantics, try to figure out which dimension to increment the PC in, what to set NPC/NNPC, etc. These decisions are best left to the ISA in conjunction with the PC type itself. Because most of the information about how to increment the PC (mainly what type of instruction it refers to) is contained in the instruction object, a new advancePC virtual function was added to the StaticInst class. Subclasses provide an implementation that moves around the right element of the PC with a minimal amount of decision making. In ISAs like Alpha, the instructions always simply assign NPC to PC without having to worry about micropcs, nnpcs, etc. The added cost of a virtual function call should be outweighed by not having to figure out as much about what to do with the PCs and mucking around with the extra elements. One drawback of making the StaticInsts advance the PC is that you have to actually have one to advance the PC. This would, superficially, seem to require decoding an instruction before fetch could advance. This is, as far as I can tell, realistic. fetch would advance through memory addresses, not PCs, perhaps predicting new memory addresses using existing ones. More sophisticated decisions about control flow would be made later on, after the instruction was decoded, and handed back to fetch. If branching needs to happen, some amount of decoding needs to happen to see that it's a branch, what the target is, etc. This could get a little more complicated if that gets done by the predecoder, but I'm choosing to ignore that for now. Variable length instructions: To handle variable length instructions in x86 and ARM, the predecoder now takes in the current PC by reference to the getExtMachInst function. It can modify the PC however it needs to (by setting NPC to be the PC + instruction length, for instance). This could be improved since the CPU doesn't know if the PC was modified and always has to write it back. ISA parser: To support the new API, all PC related operand types were removed from the parser and replaced with a PCState type. There are two warts on this implementation. First, as with all the other operand types, the PCState still has to have a valid operand type even though it doesn't use it. Second, using syntax like PCS.npc(target) doesn't work for two reasons, this looks like the syntax for operand type overriding, and the parser can't figure out if you're reading or writing. Instructions that use the PCS operand (which I've consistently called it) need to first read it into a local variable, manipulate it, and then write it back out. Return address stack: The return address stack needed a little extra help because, in the presence of branch delay slots, it has to merge together elements of the return PC and the call PC. To handle that, a buildRetPC utility function was added. There are basically only two versions in all the ISAs, but it didn't seem short enough to put into the generic ISA directory. Also, the branch predictor code in O3 and InOrder were adjusted so that they always store the PC of the actual call instruction in the RAS, not the next PC. If the call instruction is a microop, the next PC refers to the next microop in the same macroop which is probably not desirable. The buildRetPC function advances the PC intelligently to the next macroop (in an ISA specific way) so that that case works. Change in stats: There were no change in stats except in MIPS and SPARC in the O3 model. MIPS runs in about 9% fewer ticks. SPARC runs with 30%-50% fewer ticks, which could likely be improved further by setting call/return instruction flags and taking advantage of the RAS. TODO: Add != operators to the PCState classes, defined trivially to be !(a==b). Smooth out places where PCs are split apart, passed around, and put back together later. I think this might happen in SPARC's fault code. Add ISA specific constructors that allow setting PC elements without calling a bunch of accessors. Try to eliminate the need for the branching() function. Factor out Alpha's PAL mode pc bit into a separate flag field, and eliminate places where it's blindly masked out or tested in the PC. diff 3895:5e8f0e3aeca2 Mon Jan 08 20:50:00 EST 2007 Lisa Hsu <hsul@eecs.umich.edu> pagetable.hh: small fix so ALPHA_FS will build on macs interrupts.hh: small fix for alpha compile src/arch/alpha/interrupts.hh: small fix for alpha compile src/arch/alpha/pagetable.hh: small fix so ALPHA_FS will build on macs |
/gem5/src/cpu/ | ||
H A D | pc_event.cc | diff 7823:dac01f14f20f Sat Jan 08 00:50:00 EST 2011 Steve Reinhardt <steve.reinhardt@amd.com> Replace curTick global variable with accessor functions. This step makes it easy to replace the accessor functions (which still access a global variable) with ones that access per-thread curTick values. diff 7720:65d338a8dba4 Sun Oct 31 03:07:00 EDT 2010 Gabe Black <gblack@eecs.umich.edu> ISA,CPU,etc: Create an ISA defined PC type that abstracts out ISA behaviors. This change is a low level and pervasive reorganization of how PCs are managed in M5. Back when Alpha was the only ISA, there were only 2 PCs to worry about, the PC and the NPC, and the lsb of the PC signaled whether or not you were in PAL mode. As other ISAs were added, we had to add an NNPC, micro PC and next micropc, x86 and ARM introduced variable length instruction sets, and ARM started to keep track of mode bits in the PC. Each CPU model handled PCs in its own custom way that needed to be updated individually to handle the new dimensions of variability, or, in the case of ARMs mode-bit-in-the-pc hack, the complexity could be hidden in the ISA at the ISA implementation's expense. Areas like the branch predictor hadn't been updated to handle branch delay slots or micropcs, and it turns out that had introduced a significant (10s of percent) performance bug in SPARC and to a lesser extend MIPS. Rather than perpetuate the problem by reworking O3 again to handle the PC features needed by x86, this change was introduced to rework PC handling in a more modular, transparent, and hopefully efficient way. PC type: Rather than having the superset of all possible elements of PC state declared in each of the CPU models, each ISA defines its own PCState type which has exactly the elements it needs. A cross product of canned PCState classes are defined in the new "generic" ISA directory for ISAs with/without delay slots and microcode. These are either typedef-ed or subclassed by each ISA. To read or write this structure through a *Context, you use the new pcState() accessor which reads or writes depending on whether it has an argument. If you just want the address of the current or next instruction or the current micro PC, you can get those through read-only accessors on either the PCState type or the *Contexts. These are instAddr(), nextInstAddr(), and microPC(). Note the move away from readPC. That name is ambiguous since it's not clear whether or not it should be the actual address to fetch from, or if it should have extra bits in it like the PAL mode bit. Each class is free to define its own functions to get at whatever values it needs however it needs to to be used in ISA specific code. Eventually Alpha's PAL mode bit could be moved out of the PC and into a separate field like ARM. These types can be reset to a particular pc (where npc = pc + sizeof(MachInst), nnpc = npc + sizeof(MachInst), upc = 0, nupc = 1 as appropriate), printed, serialized, and compared. There is a branching() function which encapsulates code in the CPU models that checked if an instruction branched or not. Exactly what that means in the context of branch delay slots which can skip an instruction when not taken is ambiguous, and ideally this function and its uses can be eliminated. PCStates also generally know how to advance themselves in various ways depending on if they point at an instruction, a microop, or the last microop of a macroop. More on that later. Ideally, accessing all the PCs at once when setting them will improve performance of M5 even though more data needs to be moved around. This is because often all the PCs need to be manipulated together, and by getting them all at once you avoid multiple function calls. Also, the PCs of a particular thread will have spatial locality in the cache. Previously they were grouped by element in arrays which spread out accesses. Advancing the PC: The PCs were previously managed entirely by the CPU which had to know about PC semantics, try to figure out which dimension to increment the PC in, what to set NPC/NNPC, etc. These decisions are best left to the ISA in conjunction with the PC type itself. Because most of the information about how to increment the PC (mainly what type of instruction it refers to) is contained in the instruction object, a new advancePC virtual function was added to the StaticInst class. Subclasses provide an implementation that moves around the right element of the PC with a minimal amount of decision making. In ISAs like Alpha, the instructions always simply assign NPC to PC without having to worry about micropcs, nnpcs, etc. The added cost of a virtual function call should be outweighed by not having to figure out as much about what to do with the PCs and mucking around with the extra elements. One drawback of making the StaticInsts advance the PC is that you have to actually have one to advance the PC. This would, superficially, seem to require decoding an instruction before fetch could advance. This is, as far as I can tell, realistic. fetch would advance through memory addresses, not PCs, perhaps predicting new memory addresses using existing ones. More sophisticated decisions about control flow would be made later on, after the instruction was decoded, and handed back to fetch. If branching needs to happen, some amount of decoding needs to happen to see that it's a branch, what the target is, etc. This could get a little more complicated if that gets done by the predecoder, but I'm choosing to ignore that for now. Variable length instructions: To handle variable length instructions in x86 and ARM, the predecoder now takes in the current PC by reference to the getExtMachInst function. It can modify the PC however it needs to (by setting NPC to be the PC + instruction length, for instance). This could be improved since the CPU doesn't know if the PC was modified and always has to write it back. ISA parser: To support the new API, all PC related operand types were removed from the parser and replaced with a PCState type. There are two warts on this implementation. First, as with all the other operand types, the PCState still has to have a valid operand type even though it doesn't use it. Second, using syntax like PCS.npc(target) doesn't work for two reasons, this looks like the syntax for operand type overriding, and the parser can't figure out if you're reading or writing. Instructions that use the PCS operand (which I've consistently called it) need to first read it into a local variable, manipulate it, and then write it back out. Return address stack: The return address stack needed a little extra help because, in the presence of branch delay slots, it has to merge together elements of the return PC and the call PC. To handle that, a buildRetPC utility function was added. There are basically only two versions in all the ISAs, but it didn't seem short enough to put into the generic ISA directory. Also, the branch predictor code in O3 and InOrder were adjusted so that they always store the PC of the actual call instruction in the RAS, not the next PC. If the call instruction is a microop, the next PC refers to the next microop in the same macroop which is probably not desirable. The buildRetPC function advances the PC intelligently to the next macroop (in an ISA specific way) so that that case works. Change in stats: There were no change in stats except in MIPS and SPARC in the O3 model. MIPS runs in about 9% fewer ticks. SPARC runs with 30%-50% fewer ticks, which could likely be improved further by setting call/return instruction flags and taking advantage of the RAS. TODO: Add != operators to the PCState classes, defined trivially to be !(a==b). Smooth out places where PCs are split apart, passed around, and put back together later. I think this might happen in SPARC's fault code. Add ISA specific constructors that allow setting PC elements without calling a bunch of accessors. Try to eliminate the need for the branching() function. Factor out Alpha's PAL mode pc bit into a separate flag field, and eliminate places where it's blindly masked out or tested in the PC. |
/gem5/tests/long/fs/10.linux-boot/ref/alpha/linux/tsunami-o3-dual/ | ||
H A D | stats.txt | diff 10242:cb4e86c17767 Sun Jun 22 17:33:00 EDT 2014 Steve Reinhardt <steve.reinhardt@amd.com> stats: update for O3 changes Mostly small differences in total ticks, but O3 stall causes shifted significantly. 30.eon does speed up by ~6% on Alpha and ARM, and 50.vortex by 4.5% on ARM. At the other extreme, X86 70.twolf is 0.8% slower. diff 9348:44d31345e360 Fri Nov 02 12:50:00 EDT 2012 Ali Saidi <Ali.Saidi@ARM.com> update stats for preceeding changes |
Completed in 121 milliseconds