On May 1, 2014, 11:05 p.m., Nilay Vaish wrote:
Why is the clock period associated with each clocked object not enough?
There is a difference between the controller clock (which is what the DRAMCtrl
clock domain would currently communicate), and the DRAM clock (tCK in DRAM
speak). The other
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2254/
---
(Updated May 2, 2014, 8 a.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2247/
---
(Updated May 7, 2014, 9:44 p.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2266/
---
(Updated May 7, 2014, 9:49 p.m.)
Review request for Default.
Summary (updated)
changeset 39eb5d4c400a in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=39eb5d4c400a
description:
mem: Add precharge all (PREA) to the DRAM controller
This patch adds the basic ingredients for a precharge all operation,
to be used in conjunction with
changeset 793e5ff26e0b in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=793e5ff26e0b
description:
mem: Add tWR to DRAM activate and precharge constraints
This patch adds the write recovery time to the DRAM timing
constraints, and changes the current
changeset e084db2b1527 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=e084db2b1527
description:
mem: Merge DRAM latency calculation and bank state update
This patch merges the two control paths used to estimate the latency
and update the bank state.
changeset acc1131e01d6 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=acc1131e01d6
description:
mem: Add tRTP to the DRAM controller
This patch adds the tRTP timing constraint, governing the minimum time
between a read command and a precharge.
changeset 5a45f124a2f7 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=5a45f124a2f7
description:
tests: Reflect name change in DRAM tests
This patch reflects the recent name change in the DRAM TrafficGen
tests and also tidies up the test directory.
changeset 4161cfba9658 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=4161cfba9658
description:
config: Bump DRAM sweep bus speed to match DDR4 config
This patch bumps the bus clock speed such that the interconnect does
not become a bottleneck with a
changeset 823f7fd1a82f in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=823f7fd1a82f
description:
mem: Make DRAM read/write switching less conservative
This patch changes the read/write event loop to use a single event
(nextReqEvent), along with a
changeset 3112b31596f0 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=3112b31596f0
description:
mem: Ensure DRAM refresh respects timings
This patch adds a state machine for the refresh scheduling to
ensure that no accesses are allowed while the
changeset 2e630c6c2042 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=2e630c6c2042
description:
mem: Remove printing of DRAM params
This patch removes the redundant printing of DRAM params.
diffstat:
src/mem/dram_ctrl.cc | 57
changeset 52d46098c1b6 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=52d46098c1b6
description:
mem: Simplify DRAM response scheduling
This patch simplifies the DRAM response scheduling based on the
assumption that they are always returned in order.
changeset ac71c857e1e1 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=ac71c857e1e1
description:
mem: Merge DRAM page-management calculations
This patch treats the closed page policy as yet another case of
auto-precharging, and thus merges the code
changeset c249f7660eb7 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=c249f7660eb7
description:
mem: Add DRAM power states to the controller
This patch adds power states to the controller. These states and the
transitions can be used together with
changeset 52c869140fc2 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=52c869140fc2
description:
mem: Add DRAM cycle time
This patch extends the current timing parameters with the DRAM cycle
time. This is needed as the DRAMPower tool expects
changeset baf8754fd5be in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=baf8754fd5be
description:
mem: Update DDR3 and DDR4 based on datasheets
This patch makes a more firm connection between the DDR3-1600
configuration and the corresponding datasheet,
changeset 9eab5efc02e8 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=9eab5efc02e8
description:
stats: Bump stats for the fixes, and mostly DRAM controller changes
diffstat:
tests/long/fs/10.linux-boot/ref/alpha/linux/tsunami-o3-dual/stats.txt
Steve, is there any chance you could bump the t1000 test as well?
The pc-switcheroo-full seems to suffering from different stats on
different systems (I havenĀ¹t dug any deeper).
Andreas
On 11/05/2014 11:07, Cron Daemon via gem5-dev gem5-dev@gem5.org wrote:
*
Thanks a million.
On 12/05/2014 14:18, Steve Reinhardt via gem5-dev gem5-dev@gem5.org
wrote:
changeset d51e31eef415 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=d51e31eef415
description:
tests: update t1000 pc-switcheroo-full stats
committed reference
On May 14, 2014, 2:55 p.m., Ali Saidi wrote:
This looks fine to me, but I'm going to guess Andreas will have a comment.
This patch fixes the lovelock you mentioned, correct?
Steve Reinhardt wrote:
Yea, getting rid of the '+ 1' in PacketQueue::scheduleSend is needed to
do two
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2272/#review5102
---
src/arch/x86/pagetable_walker.cc
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2276/#review5104
---
src/mem/ruby/system/Sequencer.cc
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2277/#review5105
---
src/cpu/o3/lsq_unit_impl.hh
http://reviews.gem5.org/r/2277/#comment4623
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2109/#review5106
---
With the last patches that went in, the DRAM controller now has e.g. tWR
Hi Nilay,
I think something went wrong in the last push:
*
build/X86/tests/opt/long/fs/10.linux-boot/x86/linux/pc-switcheroo-full
CHANGED!
* build/X86/tests/opt/long/se/10.mcf/x86/linux/o3-timing CHANGED!
* build/X86/tests/opt/long/fs/10.linux-boot/x86/linux/pc-o3-timing
CHANGED!
Hi all,
What are your thoughts on bumping the gcc requirement from 4.4 to 4.6? From a
syntactic sugar point of view this means e.g. range-based for loops and lambdas
(and minor things like constant nullexpr). Most importantly, from a
functionality point of view it means fully functional
On May 30, 2014, 9:19 a.m., Ahmad Hassan wrote:
Hi Andreas,
According to the literature, both tCL and tWR should be independent of the
memory technology because they are buffer constraint timings and they
reduce the bandwidth (Source: http://dl.acm.org/citation.cfm?id=1555758).
On May 30, 2014, 9:19 a.m., Ahmad Hassan wrote:
Hi Andreas,
According to the literature, both tCL and tWR should be independent of the
memory technology because they are buffer constraint timings and they
reduce the bandwidth (Source: http://dl.acm.org/citation.cfm?id=1555758).
On May 30, 2014, 9:19 a.m., Ahmad Hassan wrote:
Hi Andreas,
According to the literature, both tCL and tWR should be independent of the
memory technology because they are buffer constraint timings and they
reduce the bandwidth (Source: http://dl.acm.org/citation.cfm?id=1555758).
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2281/#review5117
---
Ship it!
Ship It!
- Andreas Hansson
On June 1, 2014, 12:47 a.m.,
On June 1, 2014, 8:22 p.m., Andreas Hansson wrote:
Ship It!
Perhaps we should add style to the official keywords on
http://www.gem5.org/Commit_Access
- Andreas
---
This is an automatically generated e-mail. To reply, visit:
On May 30, 2014, 9:19 a.m., Ahmad Hassan wrote:
Hi Andreas,
According to the literature, both tCL and tWR should be independent of the
memory technology because they are buffer constraint timings and they
reduce the bandwidth (Source: http://dl.acm.org/citation.cfm?id=1555758).
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2282/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2284/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2285/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2286/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2287/#review5133
---
Ship it!
Looks great. Thanks for the additions.
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2275/#review5134
---
Ship it!
Seems like a good first step.
- Andreas Hansson
On May 14,
Hi Pavlos,
It is a crossbar implemented by muxes.
I am in the process of renaming the class XBar (from Bus) to avoid any
confusion.
Andreas
On 10/06/2014 15:50, Pavlos Maniotis via gem5-dev gem5-dev@gem5.org
wrote:
Hello everyone,
I have read the following thread and I feel that I have not
is
it like the configuration in the following image with three layers (1 for
requests, 1 for snoops and 1 for responses)?
http://www.doulos.com/knowhow/arm/Migrating_from_AHB_to_AXI/multilayer_bus
_simple.png
Thanks again,
Pavlos
On Tue, Jun 10, 2014 at 5:55 PM, Andreas Hansson via gem5-dev
changeset b21b3aad6bd1 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=b21b3aad6bd1
description:
scons: Bump the compiler version to gcc 4.6 and clang 3.0
This patch bumps the supported version of gcc from 4.4 to 4.6, and
clang from 2.9 to 3.0. This
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2289/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
Hi Alex,
That sounds rather odd. The only event that should happen when there are
no request going through the memory system (the KVM CPU operates directly
on the underlying backing store), is the refresh, which happens roughly
every 64 ms. Could you elaborate on what event it is you are
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2301/#review5136
---
configs/common/FSConfig.py
http://reviews.gem5.org/r/2301/#comment4657
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2286/#review5138
---
It would be great to get some feedback if anyone has actually used this.
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2301/#review5139
---
configs/common/FSConfig.py
http://reviews.gem5.org/r/2301/#comment4659
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2302/#review5156
---
Perhaps I am missing something, but why would Ruby forward the packet to
On June 25, 2014, 2 p.m., Andreas Hansson wrote:
Perhaps I am missing something, but why would Ruby forward the packet to
the iobus in the first place if the address is not valid?
Steve Reinhardt wrote:
This is an FS thing, where you misspeculate in the kernel and generate a
changeset e0e3efe3b1d5 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=e0e3efe3b1d5
description:
mem: Add bank and rank indices as fields to the DRAM bank
This patch adds the index of the bank and rank as a field so that we can
determine the identity
changeset 72277952d444 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=72277952d444
description:
mem: DRAMPower trace formatting script
This patch adds a first version of a script that processes the debug
output and generates a command trace for
changeset 0ad233f0a77d in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=0ad233f0a77d
description:
mem: DRAMPower trace output
This patch adds a DRAMPower flag to enable off-line DRAM power
analysis using the DRAMPower tool. A new DRAMPower flag is
changeset 70333502b9b5 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=70333502b9b5
description:
mem: Extend DRAM row bits from 16 to 32 for larger densities
This patch extends the DRAM row bits to 32 to support larger density
memories. Additional
Hi all,
Apologies for the hiccup. The checkpoint upgrade script is now reflecting
the change.
Andreas
On 01/07/2014 13:49, Stephan Diestelhorst via gem5-dev
gem5-dev@gem5.org wrote:
On Tuesday 01 July 2014 13:04:52 Nilay Vaish via gem5-dev wrote:
The following changeset breaks existing
Hi Steve,
Just to chime in here. There are definitely objects that are clocked and
not ticked. Thus, as Andrew already suggested, the most sensible thing to
do is derive TickedObject from ClockedObject.
Andreas
On 01/07/2014 18:44, Steve Reinhardt via gem5-dev gem5-dev@gem5.org
wrote:
Yes,
Hi all,
It seems that the regression
build/ALPHA/tests/opt/quick/fs/80.netperf-stream/alpha/linux/twosys-tsunami-simple-atomic
livelocks with the latest changeset:
changeset: 10251:878f2f30b12d
tag: tip
user:Anthony Gutierrez atgut...@umich.edu
date:Wed Jul 02
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2310/#review5180
---
Ship it!
Thanks for chasing this down
- Andreas Hansson
On July 7,
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2302/#review5181
---
Ship it!
As I mentioned before, I'm not opposed this going in as is if
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2305/#review5182
---
src/dev/Ethernet.py
http://reviews.gem5.org/r/2305/#comment4695
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2315/#review5198
---
configs/common/FSConfig.py
http://reviews.gem5.org/r/2315/#comment4752
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2314/#review5209
---
Ship it!
I assume this changes the X86 FS regressions? If so, I've got
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2315/#review5210
---
Ship it!
Ship It!
- Andreas Hansson
On July 17, 2014, 9:40 a.m.,
On July 17, 2014, 12:25 p.m., Andreas Hansson wrote:
Ship It!
I think this should be combined with http://reviews.gem5.org/r/2302/ such that
any address that is now forwarded to the iobus, but not actually used is sent
to a badaddr responder.
- Andreas
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2167/#review5218
---
src/mem/cache/tags/base_set_assoc.cc
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2305/#review5223
---
src/dev/etherswitch.cc
http://reviews.gem5.org/r/2305/#comment4773
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2316/#review5228
---
Thanks for the input Amin. One high-level question: what is the main aim
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2305/#review5229
---
Ship it!
Ship It!
- Andreas Hansson
On July 24, 2014, 7:44 p.m.,
changeset dc198e224a85 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=dc198e224a85
description:
stats: Bump stats for the regressions using the minor CPU
Updating the stats to match the current behaviour.
diffstat:
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2246/
---
(Updated July 28, 2014, 5:49 a.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2317/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2318/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
On July 28, 2014, 12:34 a.m., Andreas Hansson wrote:
Thanks for the input Amin. One high-level question: what is the main aim of
the patch? Until now we have tried to keep the timing constraints to a
minimum (without sacrificing fidelity). In general you could argue that
tBURST can
changeset d4090f0cab30 in /z/repo/gem5
details: http://repo.gem5.org/gem5?cmd=changeset;node=d4090f0cab30
description:
cpu: Ensure the traffic generator suppresses non-memory packets
This patch adds a check to ensure that packets which are not going to
a memory range are
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2324/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2327/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2325/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2331/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2330/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2333/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2336/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2332/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2341/
---
(Updated Aug. 14, 2014, 7:54 a.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2340/
---
(Updated Aug. 15, 2014, 2:24 p.m.)
Review request for Default.
Repository: gem5
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.
Obviously I need to work on my
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2351/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2337/
---
(Updated Aug. 17, 2014, 10:48 a.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2352/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2353/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2325/
---
(Updated Aug. 18, 2014, 12:27 p.m.)
Review request for Default.
Repository: gem5
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 one
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2327/
---
(Updated Aug. 18, 2014, 9:55 p.m.)
Review request for Default.
Repository: gem5
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2354/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
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 one
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2351/
---
(Updated Aug. 19, 2014, 9:22 p.m.)
Review request for Default.
Summary (updated)
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2357/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2356/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2355/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2359/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2360/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
1 - 100 of 633 matches
Mail list logo