[gem5-dev] changeset in gem5: ruby: eliminate type Time

changeset 0e9e99e6369a in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=0e9e99e6369a description: ruby: eliminate type Time There is another type Time in src/base class which results in a conflict. diffstat: src/cpu/testers/rubytest/Check.cc

[gem5-dev] changeset in gem5: ruby: remove typedef of Index as int64

changeset 94b6b28fc968 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=94b6b28fc968 description: ruby: remove typedef of Index as int64 The Index type defined as typedef int64 does not really provide any help since in most places we use primitive types

[gem5-dev] changeset in gem5: ruby: network: move getNumNodes() to base class

changeset 71e0934af9f1 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=71e0934af9f1 description: ruby: network: move getNumNodes() to base class All the implementations were doing the same things. diffstat: src/mem/ruby/network/Network.cc

[gem5-dev] changeset in gem5: ruby: move files from ruby/system to ruby/str...

changeset 44839e8febbd in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=44839e8febbd description: ruby: move files from ruby/system to ruby/structures The directory ruby/system is crowded and unorganized. Hence, the files the hold actual physical struc

Re: [gem5-dev] Review Request 2348: ruby: message buffers: significant changes

> On Aug. 24, 2014, 2:25 p.m., Andreas Hansson wrote: > > src/cpu/testers/traffic_gen/generators.cc, line 304 > > > > > > This probably breaks a bunch of python scripts as well? > > Nilay Vaish wrote: > Well, the code c

[gem5-dev] Review Request 2372: style: add .clang-format file

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2372/ --- Review request for Default. Repository: gem5 Description --- Changeset 10318

[gem5-dev] Ruby regression tests and null isa

Andreas I was thinking of editing the regression tests for ruby coherence protocols (memtest and rubytest) so that they use null isa, instead of alpha. But then I thought more about it and it seems that there is nothing to be gained over here. We will still compile alpha isa with different

Re: [gem5-dev] Review Request 2348: ruby: message buffers: significant changes

> On Aug. 24, 2014, 2:25 p.m., Andreas Hansson wrote: > > build_opts/NULL, line 3 > > > > > > Could you elaborate on this one? > > > > Are we also transitioning the Ruby tests to use the NULL ISA? > > Nilay Vaish wro

[gem5-dev] Review Request 2371: build opts: add MI_example to NULL ISA

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2371/ --- Review request for Default. Repository: gem5 Description --- Changeset 10310

[gem5-dev] Review Request 2370: mem: change the namespace Message to ProtoMessage

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2370/ --- Review request for Default. Repository: gem5 Description --- Changeset 10309

Re: [gem5-dev] Updating gem5-stable

I have updated gem5-stable to the following changeset from gem5: --- changeset: 10231:cb2e6950956d user: Steve Reinhardt date: Sat May 31 18:00:23 2014 -0700 summary: style: eliminate equality tests with true and false --

[gem5-dev] changeset in gem5: Added tag stable_2014_08_26 for changeset cb2...

changeset 2a1d75864ad2 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=2a1d75864ad2 description: Added tag stable_2014_08_26 for changeset cb2e6950956d diffstat: .hgtags | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diffs (8 lines): diff -r 35738ad3c7c6

Re: [gem5-dev] Review Request 2348: ruby: message buffers: significant changes

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2348/ --- (Updated Aug. 25, 2014, 8:01 p.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] Review Request 2348: ruby: message buffers: significant changes

> On Aug. 24, 2014, 2:25 p.m., Andreas Hansson wrote: > > build_opts/NULL, line 3 > > > > > > Could you elaborate on this one? > > > > Are we also transitioning the Ruby tests to use the NULL ISA? No. I don't recall

Re: [gem5-dev] Review Request 2350: ruby: remove typedef of Index as int64

> On Aug. 24, 2014, 2:08 p.m., Andreas Hansson wrote: > > src/mem/ruby/common/TypeDefines.hh, line 33 > > > > > > It would be nice to prune these as well and rely in the established > > uint64_t int64_t etc. I'll drop these

Re: [gem5-dev] Review Request 2343: ruby: eliminate type Time

> On Aug. 24, 2014, 2:06 p.m., Andreas Hansson wrote: > > Could you elaborate on why Time is Cycles in some cases and Ticks in > > others? Is this also fixing up bugs? No, this patch is not fixing any bugs. I changed the type of the variable depending on how it is being used in the rest of the

Re: [gem5-dev] Review Request 2353: base: Use STL C++11 random number generation

You are assuming that each object is associated with a particular thread and only that thread would call on the rng of the object. I don't think this assumption holds. I understand that experiments would not be repeatable in the strictest sense. But if some change in the order of events cause

Re: [gem5-dev] Review Request 2332: cpu: Fix cached block load behavior in o3 cpu

> On Aug. 19, 2014, 5:11 p.m., Nilay Vaish wrote: > > src/cpu/o3/inst_queue_impl.hh, line 1116 > > > > > > We should use nullptr now that we have gcc minimum dependency at 4.6. > > Mitch Hayenga wrote: > Hmm, at one po

Re: [gem5-dev] Review Request 2218: ruby: Fixes clock domains in configuration files.

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2218/#review5294 --- tests/configs/rubytest-ruby.py

Re: [gem5-dev] Review Request 2331: style: Add support for a style ignore list and ignore ext/

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2331/#review5293 --- Ship it! Overall I am fine with the patch. Somehow I do not like the op

Re: [gem5-dev] Review Request 2330: style: Fixup strange semantics in hg m5style

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2330/#review5292 --- Ship it! - Nilay Vaish On Aug. 13, 2014, 12:51 p.m., Andreas Hansson w

Re: [gem5-dev] Review Request 2351: misc: README direct to website for dependencies

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2351/#review5289 --- Ship it! Ship It! - Nilay Vaish On Aug. 19, 2014, 9:22 p.m., Andreas

Re: [gem5-dev] Review Request 2353: base: Use STL C++11 random number generation

On Wed, 20 Aug 2014, Andreas Hansson wrote: On Aug. 20, 2014, 5:06 p.m., Nilay Vaish wrote: The patch seems fine. I think we should declare rngs to be thread local to begin with and have different seeds for each of these rng to avoid any correlation. Agreed. Finding the thread boundaries

Re: [gem5-dev] Review Request 2327: arm: support 16kb vm granules

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2327/#review5285 --- While I am ok with the changes made, do you think it is possible to reta

Re: [gem5-dev] Review Request 2325: mem: Fix address interleaving bug in DRAM controller

> On Aug. 16, 2014, 2:13 a.m., Nilay Vaish wrote: > > While I don't understand why there are column bits at the lower end of the > > address, > > I suggest we change the name of the addressing policy to reflect that > > column > > address bits have two subsets. > > Andreas Hansson wrote: >

Re: [gem5-dev] Review Request 2353: base: Use STL C++11 random number generation

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2353/#review5281 --- Ship it! The patch seems fine. I think we should declare rngs to be thr

Re: [gem5-dev] Review Request 2352: base: Use the global Mersenne twister throughout

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2352/#review5280 --- Ship it! This patch seems fine. Is it possible to prevent random() from

Re: [gem5-dev] Review Request 2341: mem: write streaming support via WriteInvalidate promotion

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2341/#review5279 --- I think we need better explanation in the opening comment. While I sort

Re: [gem5-dev] Updating gem5-stable

10231 there are a number of compiler and bug fixes along with a few big changes. I think the big changes are quite contained, so I'd be in favor for making changeset 10283 79fde1c67ed8 the new stable if others were ok with that. Ali On 19.08.2014 09:15, Nilay Vaish via gem5-dev wrote:

Re: [gem5-dev] Review Request 2351: misc: Update dependencies in README and fit in 80 char

> On Aug. 18, 2014, 5:03 p.m., Nilay Vaish wrote: > > I am fine with the patch as such. I think we should either maintain the > > dependencies > > here in this file or on the website. I don't see why we would like > > maintain them in > > two different places unless we can link them so that o

Re: [gem5-dev] Review Request 2218: ruby: Fixes clock domains in configuration files.

> On May 3, 2014, 6:19 a.m., Nilay Vaish wrote: > > I tried running the regression tests. They fail because some clock > > domain you assume is missing in the test configurations. Here is output > > from one of the regression tests: > > > > Traceback (most recent call last): > > File "", lin

Re: [gem5-dev] Review Request 2332: cpu: Fix cached block load behavior in o3 cpu

> On Aug. 16, 2014, 4:01 p.m., Nilay Vaish wrote: > > Two points that I would like to make: > > * The opening comment in the patch states that it is trying to do two > > things. I would suggest that we split the patch. > > > > * I think we should not drop the original behaviour. Firstly, it wa

Re: [gem5-dev] Review Request 2332: cpu: Fix cached block load behavior in o3 cpu

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2332/#review5272 --- I'll take Mitch and Stephen's word that we need this patch. I'll still

[gem5-dev] Updating gem5-stable

Hello everyone As you might recall that we missed updating gem5-stable around June 15 as I was busy with other commitments. Now I am planning to update the repo next Monday. I propose we move up to the following changeset from gem5-dev: --

Re: [gem5-dev] Review Request 2298: cpu: Fix incorrect speculative branch predictor behavior

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2298/#review5269 --- Ship it! - Nilay Vaish On June 24, 2014, 4:28 p.m., Ali Saidi wrote: >

Re: [gem5-dev] Review Request 2351: misc: Update dependencies in README and fit in 80 char

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2351/#review5268 --- I am fine with the patch as such. I think we should either maintain the

Re: [gem5-dev] Review Request 2308: Adding TSX support to gem5

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2308/#review5265 --- Pradip, several comments. * If possible, split this patch into smaller o

Re: [gem5-dev] Review Request 2338: arm: Fix v8 neon latency issue for loads/stores

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2338/#review5264 --- Two questions: * What are interleave/deinterleave microops? * Why should

Re: [gem5-dev] Review Request 2337: arm: use condition code registers for ARM ISA

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2337/#review5263 --- Ship it! src/arch/arm/ccregs.hh

Re: [gem5-dev] Review Request 2334: tests: Use O3_ARM_v7a config for full-system ARM regressions

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2334/#review5262 --- Ship it! Ship It! - Nilay Vaish On Aug. 13, 2014, 2:07 p.m., Andreas

Re: [gem5-dev] Review Request 2332: cpu: Fix cached block load behavior in o3 cpu

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2332/#review5261 --- Two points that I would like to make: * The opening comment in the patch

Re: [gem5-dev] Review Request 2339: mem: Refactor assignment of Packet types

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2339/#review5260 --- Ship it! Ship It! - Nilay Vaish On Aug. 13, 2014, 2:07 p.m., Andreas

Re: [gem5-dev] Review Request 2329: base: Replace the internal varargs stuff with C++11 constructs

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2329/#review5259 --- Ship it! Ship It! - Nilay Vaish On Aug. 13, 2014, 12:51 p.m., Andreas

Re: [gem5-dev] Review Request 2328: base: Add compiler macros for C++11 final/override

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2328/#review5258 --- Ship it! Ship It! - Nilay Vaish On Aug. 13, 2014, 12:50 p.m., Andreas

Re: [gem5-dev] Review Request 2327: arm: support 16kb vm granules

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2327/#review5257 --- src/arch/arm/table_walker.hh

Re: [gem5-dev] Review Request 2340: cpu, mem: Make software prefetches non-blocking

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2340/#review5256 --- Ship it! - Nilay Vaish On Aug. 15, 2014, 2:24 p.m., Andreas Hansson wr

Re: [gem5-dev] Review Request 2325: mem: Fix address interleaving bug in DRAM controller

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2325/#review5255 --- While I don't understand why there are column bits at the lower end of th

Re: [gem5-dev] Review Request 2324: sim: bump checkpoint version for multiple event queues

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2324/#review5254 --- Ship it! Mea culpa. - Nilay Vaish On Aug. 13, 2014, 12:49 p.m., Andre

[gem5-dev] Review Request 2350: ruby: remove typedef of Index as int64

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2350/ --- Review request for Default. Repository: gem5 Description --- Changeset 10294

[gem5-dev] Review Request 2349: x86: set op class of two fp instructions

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2349/ --- Review request for Default. Repository: gem5 Description --- Changeset 10293

[gem5-dev] Review Request 2348: ruby: message buffers: significant changes

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2348/ --- Review request for Default. Repository: gem5 Description --- Changeset 10292

[gem5-dev] Review Request 2347: ruby: slicc: change the way configurable members are specified

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2347/ --- Review request for Default. Repository: gem5 Description --- Changeset 10291

[gem5-dev] Review Request 2346: ruby: slicc: improve the grammar

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2346/ --- Review request for Default. Repository: gem5 Description --- Changeset 10290

[gem5-dev] Review Request 2345: ruby: slicc: donot prefix machine name to variables

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2345/ --- Review request for Default. Repository: gem5 Description --- Changeset 10288

[gem5-dev] Review Request 2344: ruby: network: move getNumNodes() to base class

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2344/ --- Review request for Default. Repository: gem5 Description --- Changeset 10286

[gem5-dev] Review Request 2343: ruby: eliminate type Time

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2343/ --- Review request for Default. Repository: gem5 Description --- Changeset 10285

Re: [gem5-dev] Review Request 1544: ruby: move files from ruby/system to ruby/structures

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/1544/ --- (Updated Aug. 16, 2014, 12:34 a.m.) Review request for Default. Repository: gem5

Re: [gem5-dev] [PATCH] Implementation of exit_group on X86

Severin, thanks for the patch. Please post it on reviews.gem5.org, the website we use for reviewing patches. -- Nilay On Mon, 7 Jul 2014, Severin Wischmann via gem5-dev wrote: # HG changeset patch # User Severin Wischmann # Date 1404719147 -7200 # Mon Jul 07 09:45:47 2014 +0200 # Node

[gem5-dev] About changeset: 6bbb7ae309ac

The following changeset breaks existing checkpoints. Please update the script for upgrading checkpoints and the checkpoint version number as required. changeset: 10249:6bbb7ae309ac tag: qparent user:Stephan Diestelhorst date:Mon Jun 30 13:56:06 2014 -0400 summary:

Re: [gem5-dev] changeset in gem5: ruby: some small changes

On Mon, 9 Jun 2014, Joel Hestness via gem5-dev wrote: On Mon, Jun 9, 2014 at 7:04 AM, Nilay Vaish via gem5-dev wrote: I prefer simulations that fail visibly to simulations that silently do the wrong thing. I definitely agree that it is better to outright fail than proceed silently with

[gem5-dev] Updating gem5-stable

Hello everyone As per schedule, gem5-stable needs to be updated this month. I would not be able to do so for the time being. We can choose to delay it and I'll perform the update in late August. If someone else is interested in doing it, he/she can do it. Thanks Nilay

Re: [gem5-dev] changeset in gem5: ruby: some small changes

On Sun, 8 Jun 2014, Joel Hestness via gem5-dev wrote: Sure, but this fails with non-power-of-2 directories. The m_num_directories_bits is used in mapAddressToDirectoryVersion(), where m_num_directories_bits of the address are masked out and that value is returned. Suppose there are 3 directori

Re: [gem5-dev] changeset in gem5: ruby: some small changes

If the number of directories is a power of 2, then both ceil and floor would give the same output. But if the number of directories is not a power of 2, then floor will give 1 less bit than required to represent the number of directories. -- Nilay On Sat, 7 Jun 2014, Joel Hestness via gem5-d

[gem5-dev] changeset in gem5: stats: changes due to recent o3 patch.

changeset a2bb75a474fd in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=a2bb75a474fd description: stats: changes due to recent o3 patch. diffstat: tests/long/fs/10.linux-boot/ref/x86/linux/pc-o3-timing/stats.txt | 1 - tests/long/fs/10.linux-boot/ref/x86/linux

[gem5-dev] changeset in gem5: ruby: message buffer: drop dequeue_getDelayCy...

changeset 056363356d15 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=056363356d15 description: ruby: message buffer: drop dequeue_getDelayCycles() The functionality of updating and returning the delay cycles would now be performed by the dequeue() fu

[gem5-dev] changeset in gem5: ruby: remove old protocol documentation

changeset 3ffc86fefc49 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=3ffc86fefc49 description: ruby: remove old protocol documentation diffstat: src/mem/protocol/doc/MSI_MOSI_CMP_directory_2level-protocol-description.txt | 48 -- src/mem/protocol/doc/

[gem5-dev] changeset in gem5: ruby: slicc: remove unused ids DNUCA*

changeset 1a85c4fc805c in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=1a85c4fc805c description: ruby: slicc: remove unused ids DNUCA* diffstat: src/mem/slicc/symbols/StateMachine.py | 3 +-- src/mem/slicc/symbols/Transition.py | 11 --- src/mem/slicc/s

[gem5-dev] changeset in gem5: stats: changes due to o3 cpu and ruby message...

changeset aae7735450a9 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=aae7735450a9 description: stats: changes due to o3 cpu and ruby message buffer patches diffstat: tests/long/fs/10.linux-boot/ref/alpha/linux/tsunami-o3-dual/stats.txt | 2 -

[gem5-dev] changeset in gem5: cpu: o3: remove stat totalCommittedInsts

changeset 01df075d9f93 in /z/repo/gem5 details: http://repo.gem5.org/gem5?cmd=changeset;node=01df075d9f93 description: cpu: o3: remove stat totalCommittedInsts This patch removes the stat totalCommittedInsts. This variable was used for recording the total number of instruc

Re: [gem5-dev] Review Request 2276: ruby: don't make O3 CPU squash on loads that hit outstanding requests

diff -r 9eab5efc02e8 tests/configs/pc-simple-timing-ruby.py --- a/tests/configs/pc-simple-timing-ruby.py Fri May 09 18:58:50 2014 -0400 +++ b/tests/configs/pc-simple-timing-ruby.py Tue May 20 20:19:57 2014 -0500 @@ -65,7 +65,7 @@ voltage_domain = system.volt

Re: [gem5-dev] Review Request 2276: ruby: don't make O3 CPU squash on loads that hit outstanding requests

On Wed, 14 May 2014, Steve Reinhardt wrote: On May 14, 2014, 6:41 a.m., Nilay Vaish wrote: gem5 aborts due to segmentation fault with this patch applied. Program received signal SIGSEGV, Segmentation fault. Sequencer::writeCallback (this=0x3c9d800, address=..., data=..., externalHit=false,

Re: [gem5-dev] Review Request 2276: ruby: don't make O3 CPU squash on loads that hit outstanding requests

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2276/#review5094 --- gem5 aborts due to segmentation fault with this patch applied. Program r

Re: [gem5-dev] Review Request 2259: mem: Make DRAM read/write switching less conservative

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2259/#review5081 --- Looks fine to me. - Nilay Vaish On April 23, 2014, 12:27 p.m., Andreas

Re: [gem5-dev] Review Request 2218: ruby: Fixes clock domains in configuration files.

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2218/#review5080 --- I tried running the regression tests. They fail because some clock domai

Re: [gem5-dev] Review Request 2269: mem: Add DRAM cycle time

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2269/#review5079 --- Ship it! Ship It! - Nilay Vaish On April 23, 2014, 12:37 p.m., Andrea

Re: [gem5-dev] Review Request 2257: arm: allow DC instructions by default so SE mode works

--- This is an automatically generated e-mail. To reply, visit: http://reviews.gem5.org/r/2257/#review5078 --- Ship it! Ship It! - Nilay Vaish On April 23, 2014, 12:27 p.m., Andrea

Re: [gem5-dev] Review Request 2266: mem: Print all DRAM params

On Fri, 2 May 2014, Andreas Hansson wrote: On May 1, 2014, 11:08 p.m., Nilay Vaish wrote: Is this information not available in config.ini? It is indeed. This debug printout is merely to make life easier when using the DRAM debug flag, and this patch merely adds the missing bits. I'm not

Re: [gem5-dev] Review Request 2257: arm: allow DC instructions by default so SE mode works

Yes, you are right. -- Nilay On Fri, 2 May 2014, Ali Saidi wrote: On May 1, 2014, 11:20 p.m., Nilay Vaish wrote: Where are this two bits? I could not find them in src/arch/arm/miscregs.hh. Hi Nilay, They are there, perhaps you looked at an older version of gem5 before the armv8 code

<    1   2   3