Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

gt; > cache has an Owned block, the requester always ends up in Modified, but I > > believe that's only the case when the requester set needsWritable(). I > > call these places out specifically below. > > Andreas Hansson wrote: > Thanks for the review. I was hoping more peo

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

failed store > > conditional) we mark it dirty anyway just to keep ourselves sane. > > Andreas Hansson wrote: > It is really conveying that we are pending a modified resp when we call > markInService. However, the MSHR pendingWritable flag is indeed only caring > about p

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

failed store > > conditional) we mark it dirty anyway just to keep ourselves sane. > > Andreas Hansson wrote: > It is really conveying that we are pending a modified resp when we call > markInService. However, the MSHR pendingWritable flag is indeed only caring > about p

Re: [gem5-dev] Review Request 3205: mem: support for gpu-style RMWs in ruby

> On Dec. 24, 2015, 9:18 a.m., Andreas Hansson wrote: > > This is a major change to the memory system, and I find the description > > rather misleading. It would be good to get some more examples, and make > > sure that this works as a general solution for "in-mem

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

ough. I'll change it to say that we respond without removing writable, as we rely on all other copies being invalidated. - Andreas --- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.or

Re: [gem5-dev] Review Request 3205: mem: support for gpu-style RMWs in ruby

em5.org/r/3205/#comment6726> const src/mem/request.hh (line 576) <http://reviews.gem5.org/r/3205/#comment6725> stick to nullptr src/mem/request.hh (line 580) <http://reviews.gem5.org/r/3205/#comment6727> const - Andreas Hanss

Re: [gem5-dev] Review Request 3251: mem: Do not use sender state to track forwarded snoops in cache

which cache actually attached the sender > state. Instead we use the same mechanism as the crossbar, and keep > track of the requests that have outstanding snoops. > > > Diffs > - > > src/mem/cache/cache.hh 8ed230a0b28c > src/mem/cache/cache.cc 8ed230a0b28c >

Re: [gem5-dev] Review Request 3251: mem: Do not use sender state to track forwarded snoops in cache

d9a0136ab8cc src/mem/cache/cache.cc d9a0136ab8cc Diff: http://reviews.gem5.org/r/3251/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3255: mem: Do not alter cache block state on uncacheable snoops

orrible in-place update). Patches are on the way. - Andreas --- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3255/#review7745 ----------- O

Re: [gem5-dev] Review Request 3185: mem: misc flags for AMD gpu model

end up with, given the limited use. Could we at least wrap all these in an HSA comment section so that it is clear that they are really only used by the APU? - Andreas Hansson On Dec. 23, 2015, 4:50 p.m., Tony Gutierrez wrote: > > --- &g

Re: [gem5-dev] Review Request 3205: mem: support for gpu-style RMWs in ruby

717) <http://reviews.gem5.org/r/3205/#comment6701> Why? That seems very odd - Andreas Hansson On Dec. 23, 2015, 6:34 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > ht

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

> > > as below, this only applies if the request had needsWritable() set, > > correct? which I think is not necessarily the case at this point... you are right, i have updated the comment to reflect that we set the flag in the check before if (respons

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

d9a0136ab8cc src/mem/simple_mem.cc d9a0136ab8cc src/mem/serial_link.cc d9a0136ab8cc src/mem/abstract_mem.cc d9a0136ab8cc src/cpu/o3/cpu.cc d9a0136ab8cc src/dev/dma_device.cc d9a0136ab8cc Diff: http://reviews.gem5.org/r/3254/diff/ Testing --- Thanks, Andreas Hansson

Re: [gem5-dev] Review Request 3256: mem: Only forward the non-writable flag if truly needed

, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3255: mem: Do not alter cache block state on uncacheable snoops

the line without modifying the state (even if responding). Diffs (updated) - src/mem/cache/cache.cc d9a0136ab8cc Diff: http://reviews.gem5.org/r/3255/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org

Re: [gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

d9a0136ab8cc src/mem/addr_mapper.cc d9a0136ab8cc src/mem/bridge.cc d9a0136ab8cc src/mem/cache/base.hh d9a0136ab8cc src/mem/cache/blk.hh d9a0136ab8cc src/mem/cache/cache.hh d9a0136ab8cc Diff: http://reviews.gem5.org/r/3254/diff/ Testing --- Thanks, Andreas Hansson

[gem5-dev] changeset in gem5: configs: Make the default memtest behaviour m...

changeset 744e6074f6ae in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=744e6074f6ae description: configs: Make the default memtest behaviour more complex Add functional and uncacheable accesses by default. diffstat: configs/example/memtest.py | 4 ++-- 1

[gem5-dev] changeset in gem5: mem: Fix memory allocation bug in deferred sn...

changeset f4ad5be63ba8 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=f4ad5be63ba8 description: mem: Fix memory allocation bug in deferred snoop handling This patch fixes a corner case in the deferred snoop handling, where requests ended up being

[gem5-dev] Review Request 3259: mem: Do not allocate space for packet data if not needed

a3b41de1c4f1 Diff: http://reviews.gem5.org/r/3259/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3257: mem: Prevent Cache wrongly deleting requests

> On Dec. 13, 2015, 11:28 a.m., Andreas Hansson wrote: > > src/mem/cache/mshr.cc, line 389 > > <http://reviews.gem5.org/r/3257/diff/1/?file=52243#file52243line389> > > > > How about just swapping these two around? I have a sneaking suspicion > > th

[gem5-dev] Review Request 3258: mem: Fix memory allocation bug in deferred snoop handling

in use. Diffs - src/mem/cache/cache.cc 33434d6cbd20 src/mem/cache/mshr.cc 33434d6cbd20 Diff: http://reviews.gem5.org/r/3258/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman

Re: [gem5-dev] Review Request 3257: mem: Prevent Cache wrongly deleting requests

switch these around, so that we check if (pkt->isInvalidate()) first, and then the else block would cover isPendingDirty? - Andreas Hansson On Dec. 10, 2015, 7:36 a.m., Chun-Chen Hsu wrote: > > --- > This is an automaticall

Re: [gem5-dev] Review Request 3257: mem: Prevent Cache wrongly deleting requests

> On Dec. 13, 2015, 11:28 a.m., Andreas Hansson wrote: > > src/mem/cache/mshr.cc, line 389 > > <http://reviews.gem5.org/r/3257/diff/1/?file=52243#file52243line389> > > > > How about just swapping these two around? I have a sneaking suspicion > > th

Re: [gem5-dev] Cron <m5test@zizzer> /z/m5/regression/do-regression quick

Tony, I suspect this is your RubyTester patch breaking things... Andreas On 11/12/2015, 09:17, "gem5-dev on behalf of Cron Daemon" wrote: >* >build/ALPHA/tests/opt/quick/se/60.rubytest/alpha/linux/rubytest-ruby >FAILED!

Re: [gem5-dev] Review Request 3234: mem: remove acq/rel cmds from packet and add mem fence req

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3234/#review7718 --- Ship it! Ship It! - Andreas Hansson On Dec. 7, 2015, 9:06 p.m., Tony

[gem5-dev] Review Request 3253: mem: Explicitly check MSHR snoops for cases not dealt with

transformed into a read ex, upgrade or invalidation). Diffs - src/mem/cache/mshr.cc 8ed230a0b28c Diff: http://reviews.gem5.org/r/3253/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org

[gem5-dev] Review Request 3252: mem: Remove unused cache squash functionality

/mshr.hh 8ed230a0b28c src/mem/cache/mshr.cc 8ed230a0b28c src/mem/cache/mshr_queue.hh 8ed230a0b28c src/mem/cache/mshr_queue.cc 8ed230a0b28c Diff: http://reviews.gem5.org/r/3252/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing

[gem5-dev] Review Request 3254: mem: Make cache terminology easier to understand

/serial_link.cc 8ed230a0b28c src/mem/simple_mem.cc 8ed230a0b28c src/mem/snoop_filter.cc 8ed230a0b28c src/mem/tport.cc 8ed230a0b28c Diff: http://reviews.gem5.org/r/3254/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5

[gem5-dev] Review Request 3256: mem: Only forward the non-writable flag if truly needed

know that the response is the only writable copy in the system (the request was sent out with NeedsWritable set). Diffs - src/mem/cache/cache.cc 8ed230a0b28c Diff: http://reviews.gem5.org/r/3256/diff/ Testing --- Thanks, Andreas Hansson

[gem5-dev] Review Request 3251: mem: Do not use sender state to track forwarded snoops in cache

://reviews.gem5.org/r/3251/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

[gem5-dev] Review Request 3249: configs: Make the default memtest behaviour more complex

11257:55f345f251a2 --- configs: Make the default memtest behaviour more complex Add functional and uncacheable accesses by default. Diffs - configs/example/memtest.py 8ed230a0b28c Diff: http://reviews.gem5.org/r/3249/diff/ Testing --- Thanks, Andreas Hansson

[gem5-dev] Review Request 3250: mem: Fix cache sender state handling and add clarification

clarified. Diffs - src/mem/cache/cache.cc 8ed230a0b28c Diff: http://reviews.gem5.org/r/3250/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3234: mem: remove acq/rel cmds from packet and add mem fence req

> On Dec. 7, 2015, 10:04 p.m., Andreas Hansson wrote: > > No objections...but could you give a hint as to how they will be used? > > Tony Gutierrez wrote: > These are currently used by packets that encapsulate acq/rel requests. > Instead of using specific acq/req cmds

Re: [gem5-dev] Review Request 3234: mem: remove acq/rel cmds from packet and add mem fence req

? - Andreas Hansson On Dec. 7, 2015, 9:06 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

Re: [gem5-dev] Review Request 3241: cpu: fix unitialized variable which may cause assertion failure

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3241/#review7688 --- Ship it! Ship It! - Andreas Hansson On Dec. 4, 2015, 11:11 a.m., Pau

Re: [gem5-dev] Review Request 3031: config: Enable elastic trace capture and replay in se/fs

> On Nov. 6, 2015, 7:50 a.m., Nilay Vaish wrote: > > configs/example/etrace_replay.py, lines 107-109 > > > > > > I think these are the only lines that extra > > over what se.py has. Why not just move these to se.py? >

Re: [gem5-dev] Review Request 3031: config: Enable elastic trace capture and replay in se/fs

1, which I think is a good example > script. > > > Since the etrace_replay script calls Simulation.run(), I hardly think > that whatever you wrote > above is right. Other than se.py and fs.py, we do not maintain the > scripts in configs/example that well.

[gem5-dev] changeset in gem5: stats: Bump stats to match current behaviour

changeset 135c16fa409d in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=135c16fa409d description: stats: Bump stats to match current behaviour diffstat: tests/long/fs/10.linux-boot/ref/x86/linux/pc-o3-timing/stats.txt |14 +-

[gem5-dev] changeset in gem5: mem: Fix search-replace issues in DRAMPower w...

changeset 1b9331fd8966 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=1b9331fd8966 description: mem: Fix search-replace issues in DRAMPower wrapper license Fix a number of unintentional insertions of 'const'. diffstat: src/mem/drampower.cc | 8

Re: [gem5-dev] Review Request 3235: stats: make sim_insts and sim_ops respect stats reset

executed. Perhaps hostSeconds shouldn't be reset as well and then the calculation would be correct? sim_insts isn't supposed to be used for anything other than simulator level measurement. If you want instructions executed on a cpu you should look at cpu.numInsts. - Andreas Hansson On Nov. 24

[gem5-dev] changeset in gem5: config: Minor fixes to the DRAM utilisation s...

at 0 diff -r c6461e8dfc0a -r 2981e399c816 util/dram_sweep_plot.py --- a/util/dram_sweep_plot.py Sun Nov 22 05:10:16 2015 -0500 +++ b/util/dram_sweep_plot.py Sun Nov 22 05:10:17 2015 -0500 @@ -38,10 +38,8 @@ # Authors: Andreas Hansson try: - from mpl_toolkits.mplot3d import Axes3D

[gem5-dev] changeset in gem5: cpu: Fix memory leak in traffic generator

changeset c6461e8dfc0a in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=c6461e8dfc0a description: cpu: Fix memory leak in traffic generator In cases where we discard the packet, make sure to also delete it and the associated request. diffstat:

Re: [gem5-dev] Review Request 3180: mem: add request types for acquire and release

. > Honestly I think the original author of this patch may have only added the > Command for ACQ/REL because they needed something to pass to the Packet() > ctor and no other Commands fit, at least that is my best guess. > > Andreas Hansson wrote: > What is the big pictur

Re: [gem5-dev] Review Request 3180: mem: add request types for acquire and release

On Nov. 20, 2015, 1:33 p.m., Tony Gutierrez wrote: > > I am still not really understanding the request and the packet part. How > > are the two related (are they)? > > Tony Gutierrez wrote: > I have a patch to undo the packet ACQ/REL changes, and will post it soon. > Honestly I think the

Re: [gem5-dev] Style Checker: Includes Order Warning

;>I'm running 11220:c0ea80fed78f, which should be the tip. It seems to >>happen after a qpush, which is a commit. >> >>Joe >> >> >>From: gem5-dev <gem5-dev-boun...@gem5.org> on behalf of Andreas Hansson >><andreas.hans...@arm.c

Re: [gem5-dev] Style Checker: Includes Order Warning

Hi Joe, This should be fixed (by Andreas Sandberg). Are you using an up-to-date repo? Andreas On 19/11/2015, 06:48, "gem5-dev on behalf of Gross, Joe" wrote: >Hello, > >When working with mq patches in the gem5 repo, I get a lot of

Re: [gem5-dev] Review Request 3132: misc: Separate stats file is generated for SystemC-gem5 co-simulation

uot; stats dump. It seems like an unfortunate amount of code duplication. - Andreas Hansson On Oct. 16, 2015, 1:46 p.m., Abdul Mutaal Ahmad wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://r

Re: [gem5-dev] Review Request 3101: DrainManager/Checkpoint changes in SystemC coupling

On Sept. 10, 2015, 4:20 p.m., Abdul Mutaal Ahmad wrote: > > As you already mentioned, restoring from a checkpoint probably doesn't > > work. Looks good other than that. I'm Ok with this change if you add a > > warning when restoring from checkpoints, but please fix checkpointing in a > >

Re: [gem5-dev] Review Request 3183: mem: add boolean to disable PacketQueue's size sanity check

explaining why we are adding this? - Andreas Hansson On Nov. 12, 2015, 11:57 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

Re: [gem5-dev] Review Request 3185: mem: misc flags for AMD gpu model

> On Nov. 3, 2015, 10:51 a.m., Andreas Hansson wrote: > > In my view this patch really needs to be split: > > > > 1. The adding of write masks is an example of functionality that definitely > > warrants its own patch, and I would argue the write mask should go

Re: [gem5-dev] Review Request 3183: mem: add boolean to disable PacketQueue's size sanity check

> On Nov. 12, 2015, 11:03 p.m., Andreas Hansson wrote: > > src/mem/packet_queue.hh, line 135 > > <http://reviews.gem5.org/r/3183/diff/3/?file=51628#file51628line135> > > > > bool disable_sanity_check to comply with the style guide. please also > > a

Re: [gem5-dev] Review Request 3185: mem: misc flags for AMD gpu model

> On Nov. 12, 2015, 10:19 p.m., Tony Gutierrez wrote: > > Can we sort out http://reviews.gem5.org/r/3180/ before we get cracking with this one? - Andreas --- This is an automatically generated e-mail. To reply, visit:

Re: [gem5-dev] Review Request 3183: mem: add boolean to disable PacketQueue's size sanity check

gem5.org/r/3183/#comment6506> see comment from header on naming src/mem/packet_queue.cc (line 118) <http://reviews.gem5.org/r/3183/#comment6507> could swap these around, up to you - Andreas Hansson On Nov. 12, 2015, 11:01 p.

Re: [gem5-dev] Review Request 3190: ruby: fixed latency memory controller

> On Oct. 30, 2015, 10:14 p.m., Andreas Hansson wrote: > > To me this seems redundant. We already have SimpleMemory. Why not use that? Ping - Andreas --- This is an automatically generated e-mail. To reply, visit: http://rev

Re: [gem5-dev] Review Request 3183: mem: increase arbitrary limit on the packet queue

> On Oct. 30, 2015, 10:17 p.m., Andreas Hansson wrote: > > src/mem/packet_queue.cc, line 117 > > <http://reviews.gem5.org/r/3183/diff/1/?file=50969#file50969line117> > > > > That is 64 kB of implicit storage! As I've already mentioned many > >

Re: [gem5-dev] Review Request 3190: ruby: fixed latency memory controller

> On Oct. 30, 2015, 10:14 p.m., Andreas Hansson wrote: > > To me this seems redundant. We already have SimpleMemory. Why not use that? > > Andreas Hansson wrote: > Ping > > Matthew Poremba wrote: > The intention of this patch is to handle a large number of

Re: [gem5-dev] New System configuration

Hi Rohith, In these cases, the graphical visualisation of the system is your friend. Make sure you have py-dot installed, and check config.dot.pdf and config.dot.svg in the m5out directory. Andreas On 11/11/2015, 06:13, "gem5-dev on behalf of rohith mathew"

Re: [gem5-dev] Review Request 3181: cpu: Add store-access operations

unecessary. It would be great if we can get all these issue resolved. - Andreas Hansson On Oct. 30, 2015, 9:50 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://review

Re: [gem5-dev] Simulating a new system configuration

t to use coherency protocols, is classic memory system enough >or >do we need to go for ruby? >2)If I want to occupy some database operations in main memory for my HTM >implementation, which all files I should work with? > >Thanks in advance. > >regards, >Rohith > >

Re: [gem5-dev] Cron <m5test@zizzer> /z/m5/regression/do-regression quick

Could someone fry the build directory (or add the config scripts to the dependency list so that they are re-run)? Andreas On 06/11/2015, 08:08, "gem5-dev on behalf of Cron Daemon" wrote: >*

[gem5-dev] changeset in gem5: stats: Update stats to match cache changes

changeset b1bd4afb6b16 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=b1bd4afb6b16 description: stats: Update stats to match cache changes diffstat: tests/long/fs/10.linux-boot/ref/alpha/linux/tsunami-minor/stats.txt | 1647 +-

[gem5-dev] changeset in gem5: config: Update memtest to stress test cache c...

changeset 8149b36b8803 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=8149b36b8803 description: config: Update memtest to stress test cache clusivity This patch adds an new twist to the memtest cache hierarchy, in that it switches from mostly

[gem5-dev] changeset in gem5: mem: Add cache clusivity

changeset f8fdd931e674 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=f8fdd931e674 description: mem: Add cache clusivity This patch adds a parameter to control the cache clusivity, that is if the cache is mostly inclusive or exclusive. At the moment

[gem5-dev] changeset in gem5: mem: Do not treat CleanEvict as a write opera...

changeset 9eabb2bf349b in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=9eabb2bf349b description: mem: Do not treat CleanEvict as a write operation This patch changes the CleanEvict command type to not be considered a write. Initially it was made a

[gem5-dev] changeset in gem5: mem: Unify delayed packet deletion

changeset 0964165d1857 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=0964165d1857 description: mem: Unify delayed packet deletion This patch unifies how we deal with delayed packet deletion, where the receiving slave is responsible for deleting the

[gem5-dev] changeset in gem5: misc: Appease clang static analyzer

changeset 4237221d3e31 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=4237221d3e31 description: misc: Appease clang static analyzer A few minor fixes to issues identified by the clang static analyzer. diffstat: src/base/loader/dtb_object.cc

[gem5-dev] changeset in gem5: mem: Order packet queue only on matching addr...

changeset 6f8b2a005abb in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=6f8b2a005abb description: mem: Order packet queue only on matching addresses Instead of conservatively enforcing order for all packets, which may negatively impact the

[gem5-dev] changeset in gem5: mem: Add an option to perform clean writeback...

changeset 929fd978ab4e in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=929fd978ab4e description: mem: Add an option to perform clean writebacks from caches This patch adds the necessary commands and cache functionality to allow clean writebacks. This

[gem5-dev] changeset in gem5: config: Update memtest to stress test clean w...

changeset 6ef7d715d583 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=6ef7d715d583 description: config: Update memtest to stress test clean writebacks This patch adds yet another twist to the memtest cache hierarchy, in that the writeback_clean

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

(assuming we get rid of NeedsExclusive on > > WritebackClean), and use that to replace all the "== CleanEvict || == > > WritebackClean" expressions > > Andreas Hansson wrote: > I would prefer the addition of "IsEviction" to be in a separate follow-on

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

lly generated e-mail. To reply, visit: http://reviews.gem5.org/r/3156/#review7483 ----------- On Nov. 4, 2015, 12:52 p.m., Andreas Hansson wrote: > > --- > Th

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

fferently from all the other writebacks. > > Andreas Hansson wrote: > I am not sure I understand your question. Is it "why don't we just stick > it on the list of writebacks"? > > Steve Reinhardt wrote: > yes, exactly > > Andreas Hansson wrote: > It is al

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

> On Nov. 4, 2015, 6:41 a.m., Steve Reinhardt wrote: > > src/mem/cache/cache.hh, line 294 > > <http://reviews.gem5.org/r/3156/diff/2/?file=51181#file51181line294> > > > > why not MemCmd::Writeback? > > Andreas Hansson wrote: > We never fill wri

Re: [gem5-dev] Review Request 3195: gcc is not cc

195/#comment6366> Surely we should just rely on the environment variable here, and simply remove this line? - Andreas Hansson On Nov. 4, 2015, 2:36 p.m., Bjoern A. Zeeb wrote: > > --- > This is an automatically generated e-mail.

Re: [gem5-dev] Simulating a new system configuration

Hi Rohith, I am afraid I cannot help you with the HTM bit (there was some previous work posted on the review board years ago...), but when it comes to the system topology, the gem5 classic memory system is very flexible. I’d suggest to write your own config script, possibly based on

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

fferently from all the other writebacks. > > Andreas Hansson wrote: > I am not sure I understand your question. Is it "why don't we just stick > it on the list of writebacks"? > > Steve Reinhardt wrote: > yes, exactly It is all rather unfortunate, and largely d

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

src/mem/cache/mshr.hh 854e61d5390e src/mem/cache/mshr.cc 854e61d5390e src/mem/cache/mshr_queue.hh 854e61d5390e src/mem/cache/mshr_queue.cc 854e61d5390e Diff: http://reviews.gem5.org/r/3156/diff/ Testing --- Thanks, Andreas Hansson

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

(assuming we get rid of NeedsExclusive on > > WritebackClean), and use that to replace all the "== CleanEvict || == > > WritebackClean" expressions > > Andreas Hansson wrote: > I would prefer the addition of "IsEviction" to be in a separate follow-on

Re: [gem5-dev] Cron <m5test@zizzer> /z/m5/regression/do-regression quick

Hi Nilay, It was due to the typo in HMC.py. Should be resolved with your latest push. Andreas On 05/11/2015, 11:37, "gem5-dev on behalf of Nilay Vaish" wrote: >On Thu, 5 Nov 2015, Cron Daemon wrote: > >> *

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

(assuming we get rid of NeedsExclusive on > > WritebackClean), and use that to replace all the "== CleanEvict || == > > WritebackClean" expressions > > Andreas Hansson wrote: > I would prefer the addition of "IsEviction" to be in a separate follow-on

Re: [gem5-dev] Review Request 3196: tcp_iface.cc does not compile on FreeBSD

functionality (soon on the reviewboard). Thanks for pointing it out. - Andreas Hansson On Nov. 4, 2015, 2:39 p.m., Bjoern A. Zeeb wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

2d1d51615e0e src/mem/cache/mshr.hh 2d1d51615e0e src/mem/cache/mshr.cc 2d1d51615e0e src/mem/cache/mshr_queue.hh 2d1d51615e0e src/mem/cache/mshr_queue.cc 2d1d51615e0e Diff: http://reviews.gem5.org/r/3156/diff/ Testing --- Thanks, Andreas Hansson

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

> On Nov. 4, 2015, 7:11 a.m., Steve Reinhardt wrote: > > src/mem/packet.cc, line 93 > > <http://reviews.gem5.org/r/3158/diff/2/?file=51194#file51194line93> > > > > do we really want NeedsExclusive set here? > > Andreas Hansson wrote: > I

Re: [gem5-dev] Review Request 3172: mem: Use the packet delays and do not just zero them out

2d1d51615e0e src/mem/simple_mem.cc 2d1d51615e0e Diff: http://reviews.gem5.org/r/3172/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

- Andreas ------- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3158/#review7474 --- On Nov. 2, 2015, 10:44 a.m., Andreas Hansson wrote: >

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

hether to allocate or > > not. Couldn't we look at the commands as we add the targets to the MSHR? Is there such a thing? I will look into the possibility of moving it. I'm not terribly happy with how it is done currently. - Andreas --- This

Re: [gem5-dev] Review Request 3189: gpu: AMD's baseline GPU model

> On Oct. 30, 2015, 10:27 p.m., Andreas Hansson wrote: > > Impressive! > > > > Should it not be src/dev/gpu rather than src/gpu? > > > > Also, could you please make it a subdirectory in the gpu dir (as from > > above). There are more GPUs i

Re: [gem5-dev] Review Request 3185: mem: support for scoped atomics in GPU ruby protocols

. - Andreas Hansson On Nov. 2, 2015, 10:43 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

Re: [gem5-dev] Review Request 3172: mem: Use the packet delays and do not just zero them out

This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3172/#review7471 --- On Oct. 30, 2015, 10:35 a.m., Andreas Hansson wrote: > > -

Re: [gem5-dev] Review Request 3172: mem: Use the packet delays and do not just zero them out

This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3172/#review7471 ------- On Oct. 30, 2015, 10:35 a.m., Andreas Hansson wrote: > >

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

patch that changes all instances in this file from NULL -> nullptr. - Andreas --- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/3156/#review7441 ------------

[gem5-dev] Patches ready to go

Hi all, There is a bunch of patches related to the classic memory system that I’d like to get out. The big changes (support for mostly inclusive and mostly exclusive caches, and the addition of clean write backs) have been on the review board for two weeks now. Please let me know if you have

Re: [gem5-dev] Review Request 3158: mem: Add an option to perform clean writebacks from caches

/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org/mailman/listinfo/gem5-dev

Re: [gem5-dev] Review Request 3156: mem: Add cache clusivity

/cache/cache.cc 4daf60db14d7 src/mem/cache/mshr.hh 4daf60db14d7 src/mem/cache/mshr.cc 4daf60db14d7 Diff: http://reviews.gem5.org/r/3156/diff/ Testing --- Thanks, Andreas Hansson ___ gem5-dev mailing list gem5-dev@gem5.org http://m5sim.org

Re: [gem5-dev] Review Request 3190: ruby: fixed latency memory controller

that? - Andreas Hansson On Oct. 30, 2015, 9:51 p.m., Tony Gutierrez wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

Re: [gem5-dev] Review Request 3185: mem: support for scoped atomics in GPU ruby protocols

what the patch is accomplishing, and the concept/idea behind the implementation. From a quick glance it seems _very_ invasive on the memory system, and seems to mix a number of different pieces of functionality. - Andreas Hansson On Oct. 30, 2015, 9:52 p.m., Tony Gutierrez wrote

Re: [gem5-dev] Review Request 3183: mem: increase arbitrary limit on the packet queue

check. Perhaps we can call it "ludicrous mode" :-) If you want to set that from Ruby and remove the safety net that's fine. - Andreas Hansson On Oct. 30, 2015, 9:52 p.m., Tony Gutierrez wrote: > > --- > This is a

Re: [gem5-dev] Review Request 3189: gpu: AMD's baseline GPU model

, could you please make it a subdirectory in the gpu dir (as from above). There are more GPUs in this world...in fact there is already the NoMali gpu in src/dev/arm (we could move that if needed). If it's x86 specific, then perhaps src/dev/x86/gpu? - Andreas Hansson On Oct. 30, 2015, 9:52 p.m

Re: [gem5-dev] Review Request 3171: mem: Do not treat CleanEvict as a write operation

converged on going with your suggestion. - Andreas Hansson On Oct. 30, 2015, 10:33 a.m., Andreas Hansson wrote: > > --- > This is an automatically generated e-mail. To reply, visit: > http://reviews.gem

<    1   2   3   4   5   6   7   8   9   10   >