[gem5-dev] Re: Proposal for new committing policies

2020-10-19 Thread Jason Lowe-Power via gem5-dev
Hi again gem5 community! Sorry for sending the previous email at the beginning of the weekend! However, now that it's Monday, I want to this back up for discussion. To give a bit more context, this proposal is meant to codify our current culture for code review. Almost everything proposed here

[gem5-dev] Proposal for new committing policies

2020-10-16 Thread Jason Lowe-Power via gem5-dev
Hi all, I have just submitted a changeset to gerrit which updates the CONTRIBUTING documentation with a proposal for updating the committing policies. Issue: https://gem5.atlassian.net/browse/GEM5-799 Changeset: https://gem5-review.googlesource.com/c/public/gem5/+/36256 The gem5 project is

[gem5-dev] A call for maintainers

2020-10-16 Thread Jason Lowe-Power via gem5-dev
Hi everyone! As you all are aware, we're pushing to increase the stability of gem5. However, at the same time, we're seeing a significant increase in the rate of gem5 changes. Between gem5-20.0 and gem5-20.1 we had over 650 commits! And, that doesn't count the many changes on gerrit that never

[gem5-dev] Determining the APIC id for a CPU

2020-10-15 Thread Jason Lowe-Power via gem5-dev
Hi all (and specifically our AMD colleagues), Does anyone know how real hardware assigns APIC IDs in x86? We need to do something more than just use the CPU number if we want to support multiple hardware threads. We have a proposal here:

[gem5-dev] Re: Implementing PSHUFB instruction

2020-10-13 Thread Jason Lowe-Power via gem5-dev
Thanks, Gabe! I think the slow option is probably the right way to go for now. It will match other things like our slow integer division implementation ;). There's a lot in our x86 implementation that needs to be improved if we want to match the actual latencies of instructions in modern systems

[gem5-dev] Re: Build failed in Jenkins: Nightly #91

2020-10-13 Thread Jason Lowe-Power via gem5-dev
Hi all, Yeah, this is 100% a scons problem. I'm not sure what we can do about it. To be more straightforward here are the possible solutions: 1. install python-is-python3 2. pip install scons 3. Edit /usr/bin/scons to use python3 4. Add `alias scons3="/usr/bin/env python3 $(which scons)"` to

[gem5-dev] Re: Implementing PSHUFB instruction

2020-10-13 Thread Jason Lowe-Power via gem5-dev
Hi Gabe, Thanks for the info! This is a bit helpful. Although, I'm still not sure what the next steps would be or how to even start on (1) or (2) that you listed. Is it possible to focus on functional correctness first and then work on the timing correctness? The problem is that modern

[gem5-dev] Re: simple simulator performance check?

2020-10-07 Thread Jason Lowe-Power via gem5-dev
I think Linux boot is pretty reasonable. Or, Linux boot + some multithreaded tests (parsec is available for x86 in gem5-resources). If there isn't much performance impact there, I think that would be strong evidence for little performance impact, generally. Cheers, Jason On Mon, Oct 5, 2020 at

[gem5-dev] Re: MOESI_CMP_directory coherence protocol

2020-10-05 Thread Jason Lowe-Power via gem5-dev
Hello, Without digging deeper into things, I *believe* `Writeback_Ack_Data` has the data with the acknowledgement, but `Writeback_Ack` just has the ack. That's why you have to send the data from the TBE if you receive a Writeback_Ack_Data. Cheers, Jason On Fri, Sep 25, 2020 at 6:26 AM

[gem5-dev] Status of Ruby tester and DMA

2020-10-05 Thread Jason Lowe-Power via gem5-dev
Hi all (specifically Matt P. and other AMDers and Tuan), What's the status of adding DMA to the ruby tester? I see a couple of issues on Jira (e.g., https://gem5.atlassian.net/browse/GEM5-740) and there are some patches on gerrit for the GPU tester (e.g.,

[gem5-dev] Re: PCI memory BARs broken on ARM

2020-10-02 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, I believe the developers at AMD have already implemented a lot of similar things to get their GPU model working. The relevant Jira issue is https://gem5.atlassian.net/browse/GEM5-470 (I think). There may be other relevant issues on this Epic: https://gem5.atlassian.net/browse/GEM5-195.

[gem5-dev] gem5-20.1 release imminent

2020-09-29 Thread Jason Lowe-Power via gem5-dev
Hi all, We're planning on making the gem5-20.1 release official tomorrow. Please let us know if there's any last minute blocking issues. There are a couple of small things we're still waiting on, but we're confident they'll be resolved by tomorrow morning (US west coast time). In the meantime,

[gem5-dev] Re: MMU object vs. DTB and ITB

2020-09-29 Thread Jason Lowe-Power via gem5-dev
g the TLB hierarchy to the rest of the cpu code. > > The remaining thing after this patches would be to move most TLB methods > to the MMU class and to make the TLB a simple passive translation cache. > That will allow us to stack them in any way we want > > > > Giacomo >

[gem5-dev] Re: Build failed in Jenkins: Nightly #75

2020-09-22 Thread Jason Lowe-Power via gem5-dev
I think the problem is that google perftools isn't installed in this image. I guess we could skip building perf for the unittests? On Mon, Sep 21, 2020 at 11:26 PM jenkins-no-reply--- via gem5-dev < gem5-dev@gem5.org> wrote: > See

[gem5-dev] Re: MMU object vs. DTB and ITB

2020-09-21 Thread Jason Lowe-Power via gem5-dev
We (well, mostly Ayaz) have also been looking at this interface. We've been thinking more about x86 and RISC-V, but would also like to be kept up to date! We were also thinking that many of the TLB/MMU concepts are shared between ISAs (or are microarchitecture details). So, it would be nice to be

[gem5-dev] Re: gem5 python version?

2020-09-20 Thread Jason Lowe-Power via gem5-dev
Hi Mike, I believe we've decided to drop 2.7 support after the 20.1 release. This hasn't been checked into develop, yet, but I think it's a safe assumption that develop now only supports python3. Cheers, Jason On Sun, Sep 20, 2020 at 10:52 AM mike upton via gem5-dev wrote: > > Do we still

[gem5-dev] Re: A few quick thoughts

2020-09-19 Thread Jason Lowe-Power via gem5-dev
t;>> Cheers, >>> Jason >>> >>> On Fri, Sep 18, 2020 at 8:29 AM Gutierrez, Anthony < >>> anthony.gutier...@amd.com> wrote: >>> >>>> [AMD Public Use] >>>> >>>> >>>> >>>> Hey, Jason per

[gem5-dev] Re: A few quick thoughts

2020-09-18 Thread Jason Lowe-Power via gem5-dev
afelinks.protection.outlook.com/?url=https%3A%2F%2Fgem5.googlesource.com%2Fpublic%2Fgem5%2F%2B%2Frefs%2Fheads%2Fdevelop%2Fsrc%2Fbase%2Fcompiler.hh%2393=02%7C01%7Canthony.gutierrez%40amd.com%7C8dda55bf862b42f3f16408d85be45700%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637360383571770101=ufibvsE%2FQ7NY7I

[gem5-dev] Re: A few quick thoughts

2020-09-18 Thread Jason Lowe-Power via gem5-dev
this somewhere but what is the reason for > such a strong aversion to the template approach? It seems to solve the > issue nicely with what seems to be a minor change in syntax. gem5 is C++, > so we should allow users to write C++. > > > > Tony > > > > *From:* Jason

[gem5-dev] Re: A few quick thoughts

2020-09-18 Thread Jason Lowe-Power via gem5-dev
ute (see >> https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/src/base/compiler.hh#55 >> ). >> >> We should be able to get rid of this: >> https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/src/base/compiler.hh#93 >> And maybe this: >

[gem5-dev] Re: A few quick thoughts

2020-09-17 Thread Jason Lowe-Power via gem5-dev
https://gem5.googlesource.com/public/gem5/+/refs/heads/develop/src/base/compiler.hh#69 > >> -- >> Dr. Bobby R. Bruce >> Room 2235, >> Kemper Hall, UC Davis >> Davis, >> CA, 95616 >> >> web: https://www.bobbybruce.net >> >> >&

[gem5-dev] Re: A few quick thoughts

2020-09-17 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, On Thu, Sep 17, 2020 at 4:46 AM Gabe Black via gem5-dev wrote: > 1. Use __builtin_expect() for panic, fatal, etc. Preexisting library > functions like assert probably already have this, but our versions don't > and have similar behavior patterns. This should improve performance. >

[gem5-dev] Re: MessageBuffer double counting delay between arrival and dequeue ticks?

2020-09-15 Thread Jason Lowe-Power via gem5-dev
;> >> In any case, it would be great if you can post a fix. >> >> Thanks, >> Srikant >> >> On Mon, Sep 14, 2020 at 10:48 AM Jason Lowe-Power via gem5-dev < >> gem5-dev@gem5.org> wrote: >> >>> Hey Ryan, >>> >>> Sorry

[gem5-dev] Re: MessageBuffer double counting delay between arrival and dequeue ticks?

2020-09-14 Thread Jason Lowe-Power via gem5-dev
Hey Ryan, Sorry for the slow reply. Yes, it looks like delayedTicks may be double counting in some cases. I wonder if a little microbenchmark might be able to confirm more clearly? Assuming it is being double counted, we'd welcome a fix! @Bharadwaj, Srikant might be able to help as well.

[gem5-dev] Re: Build failed in Jenkins: Nightly #65

2020-09-11 Thread Jason Lowe-Power via gem5-dev
> Error setting param MemCtrl.dram to > > > > Do you know which commits are responsible for this? > > > > Thanks in advance > > > > -- > > Timothy Hayes > > Senior Research Engineer > > Arm Research > > Phone: +44-1223 645690 > > t

[gem5-dev] Re: Build failed in Jenkins: Nightly #65

2020-09-11 Thread Jason Lowe-Power via gem5-dev
(I think this message might have bounced. If not, sorry for the duplication) It looks like there are two (or more) problems here: 1. Something wrong with MIPS (we'll look into this ASAP, probably on Monday) 2. ARM FS + Ruby I'd like to discuss the latter for a moment. Here's the backtrace:

[gem5-dev] Re: Build failed in Jenkins: Nightly #65

2020-09-11 Thread Jason Lowe-Power via gem5-dev
It looks like there are two (or more) problems here: 1. Something wrong with MIPS (we'll look into this ASAP, probably on Monday) 2. ARM FS + Ruby I'd like to discuss the latter for a moment. Here's the backtrace: Traceback (most recent call last): File "", line 1, in File

[gem5-dev] Re: Scons minimum version breaks GCN-gpu dockerfile build

2020-09-03 Thread Jason Lowe-Power via gem5-dev
hat completely in hopes that it > only finds the pip version. > > > > > > -Matt > > > > *From:* Jason Lowe-Power via gem5-dev > *Sent:* Wednesday, September 2, 2020 8:03 AM > *To:* Gabe Black > *Cc:* gem5 Developer List ; Kyle Roarty < > kroa...@wisc

[gem5-dev] Re: Scons minimum version breaks GCN-gpu dockerfile build

2020-09-02 Thread Jason Lowe-Power via gem5-dev
ions. Cheers, jason On Wed, Sep 2, 2020 at 1:46 AM Gabe Black wrote: > What's a ROCM? > > Gabe > > On Tue, Sep 1, 2020 at 11:43 AM Jason Lowe-Power via gem5-dev < > gem5-dev@gem5.org> wrote: > >> Hi Dan, >> >> :facepalm: At least this is on develop and

[gem5-dev] Re: Scons minimum version breaks GCN-gpu dockerfile build

2020-09-01 Thread Jason Lowe-Power via gem5-dev
Hi Dan, :facepalm: At least this is on develop and not the stable branch! I think we need to get input from Kyle on this since he developed that docker image. TBH, I think pip isn't a bad solution given that 16.04 is more than four years old and we're stuck with that to support the older version

[gem5-dev] Re: For gem5 20.1: Remove "master" and replace with new "stable" default branch (?)

2020-08-28 Thread Jason Lowe-Power via gem5-dev
+1 Thanks for getting the ball rolling on this, Bobby! On Fri, Aug 28, 2020 at 5:26 PM Bobby Bruce via gem5-dev wrote: > Dear all, > > Back when we moved from the master branch being under constant development > to a dual "master-as-stable" and "develop" branch setup, there was talk of >

[gem5-dev] Re: Change in gem5/gem5[develop]: mem-cache: Create Compressor namespace

2020-08-26 Thread Jason Lowe-Power via gem5-dev
It's a good idea to send an email to gem5-users. However, with our new release model, it should affect many fewer people. It's probably not true today, but the vision is that the only people who will care if develop breaks something are people subscribed to gem5-dev :). Cheers, Jason On Wed, Aug

[gem5-dev] Re: switching to python 3 for config scripts?

2020-08-26 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, We've had many discussions on this in the past couple of months. The underlying problem is that each distro has decided to do something different with /usr/bin/python. Some of them allow it to point to python3 others force it to point to python2. This screws up Scons, our scripts, and

[gem5-dev] Re: Problem with libelf

2020-08-24 Thread Jason Lowe-Power via gem5-dev
l apply RPATH >> # and be able to find them at run time. If a shared libary >> includes >> # a path, the dynamic linker will apparently ignore RPATH when >> looking >> # for it. >> >> Or in other words, .../build/ARM/libfoo.a and -lfoo for li

[gem5-dev] Re: Remove CP annotation support?

2020-08-19 Thread Jason Lowe-Power via gem5-dev
Hi all, Generally, I think that any code that doesn't have tests and we are not actively supporting should be removed. We can always resurrect it with about the same amount of work as getting it up to date today. Plus, we can now say "the code is available in gem5-20." We have an extremely lean

[gem5-dev] Problem with libelf

2020-08-11 Thread Jason Lowe-Power via gem5-dev
Hi all, Does anyone remember why we have our own libelf in ext? Gabe, you were recently changing the elf loader, do you have any idea? I'm having a huge frustration trying to get gem5 to work with openSUSE. For some reason, gem5 is picking up the libelf.so file in /usr/lib instead of the

[gem5-dev] Re: Build failed in Jenkins: Nightly #27

2020-08-07 Thread Jason Lowe-Power via gem5-dev
thought I’d > give Gabe a chance to take a look. > > > > > > -Matt > > > > *From:* Jason Lowe-Power via gem5-dev > *Sent:* Thursday, August 6, 2020 7:36 AM > *To:* gem5 Developer List ; Bobby Bruce < > bbr...@ucdavis.edu> > *Cc:* Jason Lowe-Po

[gem5-dev] Re: Build failed in Jenkins: Nightly #27

2020-08-06 Thread Jason Lowe-Power via gem5-dev
Cool! The nightly builds are "working"! (At least in the sense that they let us know when something failed :D) It looks like the change that caused this issue is https://gem5-review.googlesource.com/c/public/gem5/+/29403. @Bobby Bruce , it might be nice if this message could give us links to the

[gem5-dev] Re: Ticks to Cycles

2020-07-24 Thread Jason Lowe-Power via gem5-dev
Hi John, Yeah, that would work! In the c++ code you can also use the Cycle() object. Cheers, Jason On Fri, Jul 24, 2020 at 3:17 PM John Domingo wrote: > Hi Jason, > > If I look at the config.json, I see "clk_domain->clock" under it. > To manually calculate the number of cycles, can I divide

[gem5-dev] Re: Ticks to Cycles

2020-07-24 Thread Jason Lowe-Power via gem5-dev
Hi John, The clock domain object should do this conversion for you. If you set a clock to 1 GHz, then when you use "Cycle" in that object it will be converted to 1000 ticks (assuming then normal tick rate of 1ps). Most objects are a "ClockedObject" Cheers, Jason On Fri, Jul 24, 2020 at 2:41 PM

[gem5-dev] Call for Ruby reviewers

2020-07-17 Thread Jason Lowe-Power via gem5-dev
Hi all, Tiago has recently pushed the largest number of Ruby changes we've seen for quite a while! I'm going to work to get through these in the next week or so, but it would be good if others with Ruby expertise and that use Ruby frequently would also give these a look. I believe this is the

[gem5-dev] gem5-20 paper on arXiv

2020-07-08 Thread Jason Lowe-Power via gem5-dev
Hi everyone! I'm excited to announce that we've published a new gem5 paper! Right now, it's available on arXiv at this URL: https://arxiv.org/abs/2007.03152 I tried to reach out to everyone who has been involved in gem5 development since its inception in 2011. However, I'm certain to have missed

[gem5-dev] gem5-20.1 roadmap

2020-07-06 Thread Jason Lowe-Power via gem5-dev
Hi all, We have just posted the gem5-20.1 roadmap to the gem5 website: http://www.gem5.org/project/2020/07/01/gem5-20-1-roadmap.html After our successful release of gem5-20,

[gem5-dev] Re: [Suggestion] Replace gem5-users mailing-list with Discourse

2020-07-06 Thread Jason Lowe-Power via gem5-dev
g some help as time passes; >>> there >>> >> is a short window and you really have to hope someone flagged your >>> email or >>> >> got some time to address your problem. >>> >> >>> >> I wouldn't use JIRA to be honest for a

[gem5-dev] Re: Feedback on gem5-20.1 roadmap

2020-07-03 Thread Jason Lowe-Power via gem5-dev
Hi all, Let me know if there are any other comments on the roadmap. Otherwise, I'm going to post this early next week. This roadmap isn't "binding." Of course, we--as a community--can add/remove anything we want from the gem5-20.1 release. However, it's useful for planning purposes to have an

[gem5-dev] Feedback on gem5-20.1 roadmap

2020-06-26 Thread Jason Lowe-Power via gem5-dev
Hi all! After the gem5 workshop and town hall we held a couple of weeks ago, I have put together a draft of the gem5-20.1 roadmap. You can find the draft on gerrit: https://gem5-review.googlesource.com/c/public/gem5-website/+/30715. Please let me know what feedback you have! One of the main

[gem5-dev] Re: [Suggestion] Replace gem5-users mailing-list with Discourse

2020-06-09 Thread Jason Lowe-Power via gem5-dev
+1 for Discourse :). Just to give a bit more context: I'm also trying to find a good forum for community engagement during my online Learning gem5 class this summer. I would like to find a platform that could be used generally for my class this summer, future iterations of the class, and general

[gem5-dev] Re: Ruby Prefetcher Segfaulting

2020-06-05 Thread Jason Lowe-Power via gem5-dev
We're working on the hotfix right now. Sorry about that! Cheers, Jason On Fri, Jun 5, 2020 at 3:36 AM Timothy Hayes via gem5-dev wrote: > Thanks Daniel - this patch fixes the problem for me. > > On 05/06/2020, 11:09, "Daniel Carvalho via gem5-dev" > wrote: > > Hello Timothy, > I

[gem5-dev] Call for topics at gem5 town hall

2020-06-01 Thread Jason Lowe-Power via gem5-dev
Hi everyone! As you may be aware, we're going to have a gem5 virtual town hall event on Thursday with the gem5 workshop. I'm super excited for the opportunity to bring the community together and hear feedback! Details: http://www.gem5.org/events/isca-2020#workshop-schedule-june-4th Registration

[gem5-dev] Re: Add AVX512 Support?

2020-06-01 Thread Jason Lowe-Power via gem5-dev
Hey Zhengrong, Thanks for getting started on this! I've also cc'd Miquel at BSC who has implemented many of the x86 vector instructions. Miquel, it would be great to get your input here! As far as getting input from AMD folks... I think this is going to be a tough thing for them to weigh in on

[gem5-dev] Re: New ARM patches on staging branch

2020-05-28 Thread Jason Lowe-Power via gem5-dev
Hi Giacomo, Thanks for letting us know. I've gone ahead and hit maintainer +1 on them so we can get kokoro's results ASAP. Can you find someone to review them so we can merge them today and release this afternoon (they look fine to me, but I have never looked at this code before ;)). Thanks!

[gem5-dev] New ARM patches on staging branch

2020-05-28 Thread Jason Lowe-Power via gem5-dev
Hey Giacomo, I noticed you just pushed two change to the staging branch: https://gem5-review.googlesource.com/c/public/gem5/+/29613 https://gem5-review.googlesource.com/c/public/gem5/+/29612 We were going to merge the staging into develop and release this morning (i.e., it a couple of hours).

[gem5-dev] gem5 workshop details: June 3rd and 4th

2020-05-25 Thread Jason Lowe-Power via gem5-dev
Hi everyone! I'm excited to (finally) announce details for the 3rd gem5 Users’ Workshop. (The first virtual version!) Details also found at http://www.gem5.org/events/isca-2020. Workshop registration (free): https://www.eventbrite.com/e/gem5-workshop-with-isca-2020-tickets-105893911540. Zoom

[gem5-dev] Re: c64 simulation

2020-05-12 Thread Jason Lowe-Power via gem5-dev
This is interesting, Gabe! I agree that keeping it in EXTRAS makes sense. IMO, since we probably don't need code review on it, a github repo or something similar would probably be best. We should link to it on the website, though! Cheers, Jason On Mon, May 11, 2020 at 12:03 AM Gabe Black via

[gem5-dev] Re: gem5 Staging Branch created

2020-05-02 Thread Jason Lowe-Power via gem5-dev
Thanks, Bobby! And thanks to everyone who worked extra hard last week! We're planning on posting a retrospective after this release to try to improve our practices for next time. Expect to see that in the coming weeks. Have a great weekend! Cheers, Jason On Sat, May 2, 2020 at 10:23 AM Bobby

[gem5-dev] Re: m5 utility patch series

2020-05-01 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, Quick follow up: It looks like the most controversial part of this change (at least to me) the conversion to scons from make has already been committed. So, I guess it makes sense to merge this as is and just focus on documentation over the next couple of weeks. The chain is still

[gem5-dev] Re: m5 utility patch series

2020-05-01 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, Sorry for not getting to this until now. To be honest, I was a little intimidated by the size of the changes, and it's hard to find the block of time needed to review all of this. Next time, I think it would be easier for everyone if we had a discussion about design first, and then

[gem5-dev] Re: [UPDATE] gem5 Workshop with ISCA 2020

2020-04-29 Thread Jason Lowe-Power via gem5-dev
One last reminder... The deadline for abstracts for the gem5 workshop is on Friday. We're going to have both asynchronous contributions (e.g., videos and blog posts) and some synchronous time on something like zoom for authors and the community to interact. We'll have at least two synchronous

[gem5-dev] Re: review flurry

2020-04-28 Thread Jason Lowe-Power via gem5-dev
Hey Gabe, While I agree that we need to have as many eyes on changesets as possible, not having enough reviews is not a new phenomenon. IIRC, this has happened with a number of your commits as well ;). It's just going to happen sometimes :). Actually, I want to thank everyone that's working so

[gem5-dev] Call for help: gem5-20 release

2020-04-23 Thread Jason Lowe-Power via gem5-dev
Hi everyone! I'm super excited to say that we're getting close to our first new release: gem5-20! However, there's a ton of work left for us to have a successful release. Anything you (anyone and everyone!) can do to help would be greatly appreciated! As some of you may have noticed, our Gerrit

[gem5-dev] Re: [gem5-users] [NOTICE] Mailing lists successfully migrated!

2020-04-22 Thread Jason Lowe-Power via gem5-dev
Thanks, Bobby! To everyone: In the past few months, we've had many reports of "missing" emails where emails sent to gem5-users or gem5-dev were only received by some people on the list. If you notice this happening again, please let us know ASAP so we can look into it on the new mail server.