Re: [gem5-dev] Use of exceptions in constexpr constructs

The comma operator is totally valid C++. In fact you can overload operator,. Before C++ 11, cprintf and all of it's kin were implemented by a gross hack that involved overloading operator, If you're having a problem, it may be simply because assert is probably some fancy builtin these days and al

Re: [gem5-dev] Review Request 2826: multi-gem5: add support for multi gem5 runs

Hi Steve, Shared-memory MultiIface would be synchronising the gem5 instances on a coarse granularity, corresponding to a link delay (it is a number of non-coherent machine instances). The multi-threaded event-queue implementation has to synchronise all event queues on every single memory access (a

Re: [gem5-dev] pd-gem5: simulating a parallel/distributed system on multiple physical hosts

Hi all, Thank you Steve for you insightful comments. I do agree with Andreas's first two points, too. Just add one comment to his first point, even if the violation for packet arrival happens, which is unlikely, you can abort simulation (or continue but do not expect determinism). Regarding 2, p

Re: [gem5-dev] Review Request 2874: sim: Decouple draining from the SimObject hierarchy

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2874/#review6695 --- Ship it! Ship It! - Nilay Vaish On June 8, 2015, 12:13 p.m., Andreas

Re: [gem5-dev] Review Request 2908: ruby: Fix checkpointing and restore

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2908/#review6693 --- Ship it! Ship It! - Nilay Vaish On July 1, 2015, 7:45 p.m., Timothy J

Re: [gem5-dev] Help with adding a simple regression test

Thanks, Jason! This looks like a valuable addition. My main comment is that the first two steps you have in your new "execution description" section basically overlap what's already there in the "running regressions" section. I'd prefer to see you expand the "running regressions" section with ad

Re: [gem5-dev] pd-gem5: simulating a parallel/distributed system on multiple physical hosts

Hi everyone, Sorry for taking so long to engage. This is a great development and I think both these patches are terrific contributions. Thanks to Mohammad, Gabor, and everyone else involved. I agree with Andreas that we should start with some top-level goals & assumptions, agree on those, and the

Re: [gem5-dev] Review Request 2826: multi-gem5: add support for multi gem5 runs

> On June 29, 2015, 12:55 p.m., Tony Gutierrez wrote: > > Firstly, thanks for this patch, this is really nice work. > > > > I have only done a cursory review of this patch so I'm still looking over > > the code in more detail, but I thought I'd share some of my initial > > thoughts to get the

Re: [gem5-dev] Review Request 2865: multi-gem5: Add checkpoint support for multi gem5 runs

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2865/#review6690 --- Ship it! Ship It! - Tony Gutierrez On June 1, 2015, 1:05 p.m., Curtis

Re: [gem5-dev] Review Request 2902: ruby: removes g_system_ptr and replaces with object based references

On Thu, 2 Jul 2015, Jason Power wrote: On July 2, 2015, 5:31 a.m., Nilay Vaish wrote: Brandon, overall I would suggest, if it is possible, that you spend sometime on making RubySystem a sim object. That way you would not require to copy the pointer to so many places. I think objects that

Re: [gem5-dev] Review Request 2773: ruby: slicc: Dynamically find+declare all MachineTypes

> On June 18, 2015, 9:38 p.m., Sooraj Puthoor wrote: > > Hi Joel, > > I have some minor concerns and some major concerns about this patch..:). > > Let me start with the minor concerns first: > > > > 1) If this patch is applied on top of Nilay's > > http://reviews.gem5.org/r/2550/ patch, gem5 w

Re: [gem5-dev] Review Request 2826: multi-gem5: add support for multi gem5 runs

> On July 2, 2015, 9:40 a.m., Tony Gutierrez wrote: > > Ship It! Assuming the style problems are addressed, it's a ship it. - Tony --- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2826/#review668

Re: [gem5-dev] pd-gem5: simulating a parallel/distributed system on multiple physical hosts

I agree, and I think points 1 and 3 are also non-negotiable. Given that, I think the multi-gem5 design is more robust and fits in with the overall gem5 design philosophy. I've been slowly going over the code and see no major problems - certainly nothing to warrant keeping it out of the code base

Re: [gem5-dev] Review Request 2826: multi-gem5: add support for multi gem5 runs

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2826/#review6687 --- Ship it! Ship It! - Tony Gutierrez On June 24, 2015, 4:56 p.m., Curti

Re: [gem5-dev] Help with adding a simple regression test

Thanks, Andreas. This was exactly what I was missing. I've updated the wiki page on regressions with everything that I learned. If someone that knows more about the regression tester could check it out and make sure I didn't miss anything and I'm not getting anything wrong, that would be great. ht

Re: [gem5-dev] Review Request 2884: mem: Add ReadCleanReq and ReadSharedReq packets

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2884/#review6686 --- Ship it! I said no need to repost ;) --- looks good though, thanks - St

Re: [gem5-dev] pd-gem5: simulating a parallel/distributed system on multiple physical hosts

Hi all, I think we need to up-level this a bit. From our perspective (and I suspect in general): 1. Robustness is important. Having a design that _may_ break, however unlikely is simply not an option. 2. Performance and scaling is important. We can compare actual numbers here, and I am fairly su

Re: [gem5-dev] Review Request 2902: ruby: removes g_system_ptr and replaces with object based references

> On July 2, 2015, 5:31 a.m., Nilay Vaish wrote: > > Brandon, overall I would suggest, if it is possible, that you spend > > sometime on making > > RubySystem a sim object. That way you would not require to copy the > > pointer to so > > many places. I think objects that need clock really sh

Re: [gem5-dev] Review Request 2908: ruby: Fix checkpointing and restore

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2908/#review6684 --- Ship it! What did you use to test this? It would be wonderful if we had

Re: [gem5-dev] Review Request 2932: mem: minor fixes in the HMC vault model

> On July 1, 2015, 3:07 p.m., Andreas Hansson wrote: > > src/mem/DRAMCtrl.py, line 460 > > > > > > Really? I was under the impression that the vault controllers were > > simple closed page. See for example > > http://ieeex

Re: [gem5-dev] Review Request 2746: mem: Avoid DRAM write queue iteration for merging and read lookup

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2746/ --- (Updated July 2, 2015, 3:04 p.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] Review Request 2885: mem: Split WriteInvalidateReq into write and invalidate

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2885/ --- (Updated July 2, 2015, 2:56 p.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] Review Request 2884: mem: Add ReadCleanReq and ReadSharedReq packets

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2884/ --- (Updated July 2, 2015, 2:51 p.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] Review Request 2882: mem: Add clean evicts to improve snoop filter tracking

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2882/ --- (Updated July 2, 2015, 2:48 p.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] Review Request 2908: ruby: Fix checkpointing and restore

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2908/#review6682 --- Ship it! Looks good to me. Thanks for tackling this! - Joel Hestness

Re: [gem5-dev] Review Request 2885: mem: Split WriteInvalidateReq into write and invalidate

> On July 2, 2015, 4:50 a.m., Nilay Vaish wrote: > > src/mem/cache/cache_impl.hh, line 1039 > > > > > > Why is this true? Suppose we have two cores with private caches and > > both of them have some block in shared state

Re: [gem5-dev] Review Request 2746: mem: Avoid DRAM write queue iteration for merging and read lookup

> On July 2, 2015, 5:55 a.m., Nilay Vaish wrote: > > src/mem/dram_ctrl.hh, line 647 > > > > > > I am guessing that somewhere you make sure that burstSize is a power of > > two. Yes. I will add a check in the constructor.

Re: [gem5-dev] Review Request 2885: mem: Split WriteInvalidateReq into write and invalidate

> On July 2, 2015, 4:50 a.m., Nilay Vaish wrote: > > src/mem/cache/cache_impl.hh, line 1039 > > > > > > Why is this true? Suppose we have two cores with private caches and > > both of them have some block in shared state

Re: [gem5-dev] Review Request 2886: mem: Remove redundant is_top_level cache parameter

> On June 28, 2015, 5:31 p.m., Nilay Vaish wrote: > > src/mem/cache/cache_impl.hh, line 2145 > > > > > > Same question as previous one? > > Andreas Hansson wrote: > Arguably, the new check is actually more sensible tha

Re: [gem5-dev] Review Request 2932: mem: minor fixes in the HMC vault model

> On July 1, 2015, 3:07 p.m., Andreas Hansson wrote: > > src/mem/DRAMCtrl.py, line 460 > > > > > > Really? I was under the impression that the vault controllers were > > simple closed page. See for example > > http://ieeex

Re: [gem5-dev] Review Request 2908: ruby: Fix checkpointing and restore

> On July 1, 2015, 7:51 p.m., Andreas Hansson wrote: > > Thanks for getting this in shape Tim! Much nicer. Thanks Andreas. I'd just like to get comments from a ruby person too before I push. Nilay, Brad, Joel, any thoughts? - Timothy

[gem5-dev] Cron /z/m5/regression/do-regression quick

* build/ALPHA/tests/opt/quick/se/00.hello/alpha/linux/minor-timing passed. * build/ALPHA/tests/opt/quick/se/00.hello/alpha/linux/o3-timing passed. * build/ALPHA/tests/opt/quick/se/00.hello/alpha/linux/simple-atomic passed. * build/ALPHA/tests/opt/quick/se/00.hello/alpha/linux/simple