Yea, devices are the issue... from one perspective that's the main problem
that the KVM cpu model solves, which is how to use KVM with gem5's device
models so that their state can be maintained properly.
The problem with going the other way and leveraging qemu devices is really
qemu's gpl
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2849/#review6494
---
Ship it!
I just got an explanation that I'm happy with.
- Andreas
Hi Tim,
I have not been completely following this thread, but I can answer your
question about unserializing cache contents.
The benefit for creating at trace, rather than just inserting data into the
cache, is two-fold. First, by creating a trace from a very large cache system,
one can
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2890/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2889/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2891/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
* 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.
*
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2893/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
Thanks Nilay and Joel for the information.
I've been playing around with this over the past few days and I can't
work out what the point of the flush is. The CacheRecorder already has
a copy of all the data blocks in the trace before the flush starts.
Removing the flush event and subsequent
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2892/
---
Review request for Default.
Repository: gem5
Description
---
Changeset
Mike,
I agree with Andreas ‹ devices are going to be the pain. You might be able
to get the CPU state to transfer, but it will be extremely hard and
fragile.
Someone at ISCA (I think it was in the SST workshop) mentioned a project
called qsim which builds qemu as a library that can be used for
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2855/#review6492
---
src/cpu/o3/fetch_impl.hh (lines 331 - 355)
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2851/#review6493
---
Ship it!
I'm happy with the explanation you just gave. Could you add
It seems to me that it should be possible to directly import kvm or qemu
generated checkpoints into the simulator.
Does the community agree, or am I missing something.
Thanks,
Mike
___
gem5-dev mailing list
gem5-dev@gem5.org
Hi Mike,
I think the main challenge is device state. You will need the same
devices, and a coherent view of their state.
The CPU’s should be easy in comparison. There it’s just a matter of
someone sitting down and doing the work.
Andreas
On 17/06/2015 09:31, mike upton michaelup...@gmail.com
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2860/#review6490
---
Ship it!
- mike upton
On May 28, 2015, 3:41 p.m., Alexandru Dutu
---
This is an automatically generated e-mail. To reply, visit:
http://reviews.gem5.org/r/2782/#review6491
---
Ship it!
- mike upton
On May 11, 2015, 10:18 p.m., Tony Gutierrez
17 matches
Mail list logo