changeset 3c38d3e74980 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=3c38d3e74980
description:
ruby: fix round robin arbiter in garnet2.0
The rr arbiter pointer in garnet was getting updated on every request,
even if there is no grant. This was
lly generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3809/#review9418
---
On Feb. 12, 2017, 4:36 a.m., Tushar Krishna wrote:
>
> ---
> This is an automatically ge
it to update upon a grant.
Diffs
-
src/mem/ruby/network/garnet2.0/SwitchAllocator.cc 79c3f6a60392
Diff: http://reviews.gem5.org/r/3809/diff/
Testing
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
gem5-dev@gem5.org
http://m5sim.org
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3753/#review9246
---
Ship it!
Ship It!
- Tushar Krishna
On Jan. 10, 2017, 9:42 p.m
> On Jan. 10, 2017, 4:43 p.m., Tushar Krishna wrote:
> > src/mem/ruby/network/garnet2.0/NetworkInterface.cc, line 150
> > <http://reviews.gem5.org/r/3753/diff/4/?file=64266#file64266line150>
> >
> > Hey Matt,
> > There is a minor change I hav
The additional 1-cycle in NIC0 is going to be part of the
src_queueing_delay so it is being accounted for.
Does this make sense?
If yes, can you update network delay to
dequeue_time - enqueue_time - 1 ?
And then ship it.
Thanks,
Tushar
- Tushar
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3753/#review9242
---
Ship it!
Ship It!
- Tushar Krishna
On Jan. 6, 2017, 11:59 p.m
> On Dec. 9, 2016, 5:03 a.m., Tushar Krishna wrote:
> > src/mem/ruby/network/garnet2.0/NetworkInterface.cc, line 139
> > <http://reviews.gem5.org/r/3753/diff/2/?file=63961#file63961line139>
> >
> > If a flit is waiting for the output buffer to become free,
nstalled, then this variable becomes true.
But if the stall queue is empty, unstalledMessage is returned false, which
means stalledMessage is true .. which is not quite right.
- Tushar Krishna
On Dec. 9, 2016, 1:58 a.m., Matthew Poremba wrote:
>
> ---
e called from here or when
the stall queue is read.
...
The extLinkAvail piece of code is not needed. It gets subsumed here.
- Tushar Krishna
On Dec. 8, 2016, 11:34 p.m., Matthew Poremba wrote:
>
> ---
> This is an automatically ge
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3751/#review9194
---
Ship it!
Ship It!
- Tushar Krishna
On Dec. 8, 2016, 11:32 p.m
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3711/#review9180
---
Ship it!
Ship It!
- Tushar Krishna
On Dec. 1, 2016, 11:37 p.m
711/#comment7888>
500,000 seems like a pretty large threshold value in my opinion, whatever
be the size of the network.
It means no packet left the router connected to the NI for 500K cycles.
I think we can make it 50,000.
- Tushar Krishna
On Nov. 17, 2016, 6:37 p.m., Jieming Yin
changeset 220fa4099b9a in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=220fa4099b9a
description:
config: fix typo in cluster topology.
diffstat:
configs/topologies/Cluster.py | 4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diffs (15 lines):
diff -r
changeset ebf2acd02fc5 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=ebf2acd02fc5
description:
ruby: Add M5_VAR_USED before variables used only inside assert in
garnet2.0.
This removes errors when building gem5.fast
diffstat:
changeset 10d59d546ea2 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=10d59d546ea2
description:
ruby: garnet2.0
Revamped version of garnet with more optimized single-cycle routers,
more configurability, and cleaner code.
diffstat:
changeset db895719c482 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=db895719c482
description:
ruby: remove the original garnet code.
Only garnet2.0 will be supported henceforth.
diffstat:
src/mem/ruby/network/garnet/BaseGarnetNetwork.cc
LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+# Authors: Brad Beckmann
+# Tushar Krishna
+
+import m5
+from m5.objects import *
+from
OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+# Authors: Tush
R BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+# Author: Tushar Krishna
+
+import m
changeset 2365e9e396f7 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=2365e9e396f7
description:
config: add port directions and per-router delay in topology.
This patch adds port direction names to the links during topology
creation, which can be used
T, STRICT LIABILITY, OR TORT
+# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+# Authors: Tushar Krishna
+
+import math
+import m5
+from m5.objects import *
+from m5.defines import buildEnv
+from m
Thanks,
Tushar
- Tushar
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3558/#review8734
---
On Aug. 24, 2016, 3:24 p.m., Tushar Krishna wrote:
>
> ---
tically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/3555/#review8504
-----------
On Aug. 24, 2016, 3:22 p.m., Tushar Krishna wrote:
>
> ---
> This is an automatically generated e-mail. To reply, visit:
> http://rev
-CREATION
src/mem/ruby/network/garnet2.0/InputUnit.cc PRE-CREATION
src/mem/ruby/network/garnet2.0/NetworkInterface.hh PRE-CREATION
src/mem/ruby/network/garnet2.0/NetworkInterface.cc PRE-CREATION
Diff: http://reviews.gem5.org/r/3558/diff/
Testing
---
Thanks,
Tushar Krishna
/GarnetNetwork_d.cc d726d0cea027
src/mem/ruby/network/garnet/fixed-pipeline/GarnetNetwork_d.py d726d0cea027
src/mem/ruby/network/garnet/fixed-pipeline/InputUnit_d.hh d726d0cea027
Diff: http://reviews.gem5.org/r/3557/diff/
Testing
---
Thanks,
Tushar Krishna
configs/topologies/Torus.py d726d0cea027
src/mem/ruby/network/BasicLink.hh d726d0cea027
src/mem/ruby/network/BasicLink.cc d726d0cea027
Diff: http://reviews.gem5.org/r/3555/diff/
Testing
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
/3556/diff/
Testing
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
gem5-dev@gem5.org
http://m5sim.org/mailman/listinfo/gem5-dev
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
gem5-dev@gem5.org
http://m5sim.org/mailman/listinfo/gem5-dev
-CREATION
src/cpu/testers/networktest/NetworkTest.py d726d0cea027
Diff: http://reviews.gem5.org/r/3550/diff/
Testing
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
gem5-dev@gem5.org
http://m5sim.org/mailman/listinfo/gem5-dev
/Garnet_standalone.py PRE-CREATION
configs/ruby/Network_test.py d726d0cea027
src/mem/protocol/Garnet_standalone-cache.sm PRE-CREATION
src/mem/protocol/Garnet_standalone-dir.sm PRE-CREATION
Diff: http://reviews.gem5.org/r/3549/diff/
Testing
---
Thanks,
Tushar Krishna
default value of mesh_rows to sqrt(num_routers)
Diffs
-
configs/ruby/Ruby.py b4d0bdb52694
configs/topologies/Mesh.py b4d0bdb52694
configs/topologies/MeshDirCorners.py b4d0bdb52694
Diff: http://reviews.gem5.org/r/1337/diff/
Testing
---
Thanks,
Tushar Krishna
: remove unnecessary memory allocation
Diffs
-
src/cpu/testers/networktest/networktest.cc 4148f9af0b70
Diff: http://reviews.gem5.org/r/1136/diff/
Testing
---
Thanks,
Tushar Krishna
___
gem5-dev mailing list
gem5-dev@gem5.org
http://m5sim.org
changeset c739a3a829f5 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=c739a3a829f5
description:
Garnet: Stats at vnet granularity + code cleanup
This patch
(1) Moves redundant code from fixed and flexible networks to
BaseGarnetNetwork.
(2)
changeset df3b7a1e883f in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=df3b7a1e883f
description:
GARNET: adding a fault model for resilient on-chip network research.
This patch adds a fault model, which provides the probability of a
number of
These discussions might be better on the gem5-users list ...
Can you re-send this email on that? I'll explain how you can separate
out the power for different routers and ports ...
On 10/25/2011 06:21 AM, lyh wrote:
In a NoC, there are many routers. In general, some with more ports, some
They affect the capacitance.
larger width -- larger cap
larger spacing -- smaller cap
What configuration did you run?
I'll take a look ...
On 10/24/2011 02:45 AM, lyh wrote:
Thks for your quick answer.
In the code, i have found that WIRE_WIDTH_SPACING has no effect on the static or
dynamic
Hi Lyh,
Garnet requires the ni_flit_size to be same as bandwidth_factor which
means you cannot use different bandwidth factors for different links ..
However, if you specify different link latencies, you can mimic
different bandwidths since flits will take more cycles on some links and
fewer
wrote:
This is an automatically generated e-mail. To reply, visit:
http://reviews.m5sim.org/r/776/
Review request for Default, Ali Saidi, Gabe Black, Steve Reinhardt, Nathan
Binkert, Brad Beckmann, and Tushar Krishna.
By Konstantinos Aisopos.
Description
GARNET: adding a fault model
39 matches
Mail list logo