Hello,
I was looking for ways to read in QEMU memory dump files/ QEMU snapshot
files.
I currently have two files with me - one of them is a snapshot of the
memory which I created using the command (virsh snapshot-create-as
--memspec ... ) and the other is a dump of the QEMU guest memory, obtained
Hello,
I was trying to understand the ELF file generated by the virsh dump
(--memory-only) command. I have successfully generated a dump of the QEMU
VM memory using this command.
I specifically am trying to understand the loadable segments of this ELF
file.
I ran readelf -a to get the informati
Hello,
I was trying to understand the ELF file generated by the virsh dump
(--memory-only) command. I have successfully generated a dump of the QEMU
VM memory using this command.
I specifically am trying to understand the loadable segments of this ELF
file.
I ran readelf -a to get the informa
Hello,
I am repeatedly facing an issue while loading a saved snapshot into QEMU.
I will detail the steps which I follow when I take a snapshot:
1. I start QEMU with *-enable-kvm* option. I then run a workload(a program
from Spec2006) on QEMU.
2. Immediately after I start running the workload, I
QEMU version 2.10 as well. I will be grateful if
you can provide me with some more details on the root cause of this issue.
Best Regards,
Arnab
On Mon, Sep 4, 2017 at 4:58 AM, Peter Maydell
wrote:
> On 1 September 2017 at 18:33, Arnabjyoti Kalita
> wrote:
> > 1. I start QEMU w
Hi all,
I have been trying to modify the execution flow in QEMU. Briefly, at
certain points in my program, I have tried to reverse the direction of the
branch i.e. if the branch was taken, I have forced the branch to not be
taken or if the branch was not-taken, I have forced the branch to be taken
Thanks Peter for guiding me through this information. I was able to find
out the source of this problem when I added the -nochain feature to the
QEMU command line.
On Thu, Mar 15, 2018 at 6:06 AM, Peter Maydell
wrote:
> On 15 March 2018 at 00:51, Arnabjyoti Kalita
> wrote:
>
Hi all,
I have been trying to understand how the -icount shift=N | auto flags work
in QEMU. There are small descriptions as to how it works already, but I
just wanted a clearer perspective.
I obtained a trace of the target instructions using "-d in_asm" switch
while having the icount flags in the
March 2018 at 20:22, Arnabjyoti Kalita
> wrote:
> > I see that in the trace file, some of the TCG blocks seem to be
> translated
> > more than once -
> >
> > ( NOTE: I am not using the TB Cache/Hash Table and have managed to
> disable
> > it in the Q
.
That helped me to atleast narrow down the sources of this cause.
Best Regards,
Arnab
On Fri, Mar 23, 2018 at 6:03 AM, Peter Maydell
wrote:
> On 22 March 2018 at 22:34, Arnabjyoti Kalita
> wrote:
> > From what I can see from the logs, it is quite hard to tell why this
> occurs.
12:27 PM, Peter Maydell
wrote:
> On 23 March 2018 at 19:47, Arnabjyoti Kalita
> wrote:
> > Hi Peter,
> >
> > Yes only I could see that. Sorry for dumping out incomplete logs :)
> >
> > Well I think I figured out the source of this representation of
> rep
Hi all,
I have been trying to modify the execution flow in QEMU. Briefly, at
certain points in my program, I have tried to reverse the direction of the
branch i.e. if the branch was taken, I have forced the branch to not be
taken or if the branch was not-taken, I have forced the branch to be taken
Hi all,
I have been trying to simulate the occurrence of an interrupt during the
execution of an application in TCG mode. Specifically the interrupt number
is 239 (local apic timer interrupt).
Briefly, I am using the -loadvm command to load a previously saved state of
QEMU. When I load the saved
it hard to
understand what it returns from the source code.
Basically, I am finding it hard to determine if this function returns the
current TB that was executed or that it returns the next TB that is to be
executed.
Thanks and Regards,
Arnabjyoti Kalita
Hi Peter,
Thanks for highlighting this.
Regards,
Arnabjyoti Kalita
On Wed, May 16, 2018 at 9:14 AM, Peter Maydell
wrote:
> On 15 May 2018 at 22:09, Arnabjyoti Kalita
> wrote:
> > I was trying to understand what the function tcg_qemu_tb_exec() returns.
> I
> > can underst
Hello all,
I am trying to implement a 'minimal' record-replay mechanism for KVM, which
is similar to the one existing for TCG via -icount. I am trying to record
I/O events only (specifically disk and network events) when KVM does a
VMEXIT. This has led me to the function kvm_cpu_exec where I can c
l network packets
without fail.
Regards,
Arnabjyoti Kalita
size);*
given this scenario, when we do *savevm* and then a *loadvm*, will the
memory addresses like *nc*, *buf* in the function above still retain
correct values, or will they be garbage?
Thank you for answering.
Best Regards,
Arnabjyoti Kalita
Hello all,
I am running QEMU-3.0.0 in TCG mode, and my QEMU as well as TCG target is
x86_64 architecture.
What I am trying to do is inject an I/O interrupt in the middle of a
translation block.
I have started TCG mode using the following debug flags
*-d in_asm,cpu,exec,nochain*
Let's say, I hav
all kinds of
interrupts, hardware/software/faults/traps/vmexits etc. ?
Best Regards,
Arnab
On Sat, Aug 22, 2020 at 7:01 PM Peter Maydell
wrote:
> On Sat, 22 Aug 2020 at 09:42, Arnabjyoti Kalita
> wrote:
> > I am running QEMU-3.0.0 in TCG mode, and my QEMU as well as TCG target
Thank you for the detailed explanation, Peter. It makes a lot of things
clear.
Arnab
On Tue, Aug 25, 2020 at 4:00 PM Peter Maydell
wrote:
> On Tue, 25 Aug 2020 at 06:53, Arnabjyoti Kalita
> wrote:
> > This makes sense. In this scenario, when QEMU takes an interrupt at the
>
Hello all,
I am running the QEMU stable version of 3.0 in KVM mode with the host as
well as target architecture being x86_64. I am using a virtio-net device
using the command line -
-device virtio-net-pci,netdev=tap1,mac=00:00:00:00:00:00
I have been trying to record all the virtqueue kicks tha
Hello all,
I am running QEMU with KVM using the below command line -
*sudo ./qemu-system-x86_64 -m 1024 -machine pc-i440fx-3.0 *
*-cpu qemu64,-kvmclock -accel kvm -netdev
tap,id=tap1,ifname=tap0,script=no,downscript=no *
*-device virtio-net-pci,netdev=tap1,mac=00:00:00:00:00:00 *
*-drive file=ubu
I am running an experiment with QEMU as follows -
1. Start the QEMU process with "accel=kvm" using the default qemu64 CPU,
the command line I use is -
*sudo ./qemu-system-x86_64 -m 1024 -machine pc-i440fx-2.5 -cpu
qemu64,-kvmclock -accel kvm -netdev
tap,id=tap1,ifname=tap0,script=no,downscript=n
Hello all,
Is there a way to mask *all* interrupts in the TCG mode of QEMU execution?
How should I go about doing that in code?
Scenario:
I start the QEMU process in KVM mode and then do a savevm. I load the saved
QEMU state in TCG mode. I'd like all the interrupts to be disabled/masked
when I d
different numbers in the
KVM mode and in TCG mode? Like say, an interrupt is 0x15 in KVM, but is
0x75 in TCG.
Best Regards,
Arnab
On Fri, Oct 2, 2020 at 4:05 PM Peter Maydell
wrote:
> On Fri, 2 Oct 2020 at 04:17, Arnabjyoti Kalita
> wrote:
> > Is there a way to mask all interrupts in
On Fri, Oct 2, 2020 at 11:39 PM Peter Maydell
wrote:
> On Fri, 2 Oct 2020 at 17:50, Arnabjyoti Kalita
> wrote:
> > What I can understand is that when we restore the VM State, the same
> interrupt state will be restored which was before we did a savevm. And then
> these i
Hello all,
I am doing a KVM-TCG migration of VMState as follows.
I start the QEMU process in KVM mode using the below command line -
*sudo ./qemu-system-x86_64 -m 1024 --machine pc-i440fx-2.5 -netdev
tap,id=tap1,ifname=tap0,script=no,downscript=no --accel kvm -device
virtio-net-pci,netdev=tap1,m
Hello all,
I am trying to mimic the CPU configuration that QEMU starts up with in both
TCG and KVM modes. I do not see a mechanism to change the CPU model in TCG,
while there is a way in KVM with the help of the -cpu flag.
As a starting point, I used the *qemu64 *CPU when I start QEMU with KVM
us
No, it won't work with KVM.
Thanks,
Arnab
On Mon, Jan 18, 2021 at 7:03 AM Duo Zhang wrote:
> Hi developers,
>
> I am trying to capture input instructions through QEMU, I see on the
> manual: "-d in_asm" shows input assembly code for each compiled TB. Does
> the option still work when enabling K
Hello all,
I am trying to understand how the clock values are recorded and replayed in
QEMU (when it runs in TCG mode). I have been specifically following the
document that has been provided here -
https://github.com/qemu/qemu/blob/master/docs/replay.txt
I have the following 4 questions to ask-
Thank you Lukas for your answer.
I was only looking to record and replay host clock values (not any other
events), i.e. record in KVM mode and replay in TCG mode.
Best Regards,
Arnab
On Thu, Feb 25, 2021 at 10:00 PM Lukas Straub wrote:
> On Thu, 25 Feb 2021 20:39:20 +0530
> Arnabjyoti
for a way to ensure that the timer activity is synchronized with
the KVM mode, when I load the guest in TCG mode. Will this be possible to
achieve ?
Thank you very much again.
Best Regards,
Arnab
On Thu, Feb 25, 2021 at 10:30 PM Alex Bennée wrote:
>
> Arnabjyoti Kalita writes:
>
> >
t; > On 2/25/21 4:09 PM, Arnabjyoti Kalita wrote:
> >> Hello all,
> >>
> >> I am trying to understand how the clock values are recorded and replayed
> >> in QEMU (when it runs in TCG mode). I have been specifically following
> >> the document that has
the host may
* undergo (e.g. due to NTP).
```
Thanks.
Best Regards,
Arnab
On Fri, Feb 26, 2021 at 3:25 PM Alex Bennée wrote:
>
> Pavel Dovgalyuk writes:
>
> > On 25.02.2021 18:43, Philippe Mathieu-Daudé wrote:
> >> Cc'ing Pavel/Alex.
> >>
> >&
Hello all,
I am really thankful for the wonderful answers in my last post linked below-
https://lists.nongnu.org/archive/html/qemu-discuss/2021-02/msg00131.html
In continuation with the last post, I have a few more questions to ask -
My experiment is still, mostly the same. I record clock value
Thank you Pavel.
Your answers make the clock record-replay process much clearer to me now.
Best Regards,
Arnab
On Tue, Mar 2, 2021 at 12:49 PM Pavel Dovgalyuk
wrote:
> On 01.03.2021 20:16, Arnabjyoti Kalita wrote:
> > Hello all,
> >
> > I am really thankful for the wo
Hello all,
This is a continuation of some of the questions I had about the clock
record-replay handling process in QEMU. My previous post is here -
https://www.mail-archive.com/qemu-discuss@nongnu.org/msg06231.html
My experiment involves two steps. In the first step, I start QEMU in KVM
mode, s
e of rdtsc. This requires me to know which functions get called
when the rdtsc emulation happens within QEMU.
How do I go about determining whether a VMEXIT happened because of the
rdtsc instruction? Which function does QEMU execute to emulate rdtsc?
Thank you very much.
Best Regards,
Arnabjyoti Kalita
values are being replayed correctly.
Best Regards,
Arnabjyoti Kalita
o you see any other potential causes of concern?
Best Regards,
Arnabjyoti Kalita
On Mon, May 17, 2021 at 1:05 PM Pavel Dovgalyuk
wrote:
>
> On 14.05.2021 06:30, Arnabjyoti Kalita wrote:
> > Hello all,
> >
> > I am recording and replaying HPET timer values for my guest.
>
will start with the same
disk state? Can I get rid of the extra snapshot that is created here?
Thank you very much for all your help so far.
Best Regards,
Arnabjyoti Kalita
do I have to make to ensure synchronous processing of network
packets happens in TCG?
Thanks for all your help so far.
Best regards,
Arnabjyoti Kalita
is x86-64.
Thank you very much for all your help so far.
Best regards,
Arnabjyoti Kalita
Hello all,
Is it possible to disable scsi disk requests being generated by the
guest when using a virtio-blk device?
I am starting the guest in KVM mode using the below command line-
sudo ./qemu-system-x86_64 -m 1024 --machine pc-i440fx-2.5 -cpu
qemu64,-kvmclock,-kvm_pv_eoi
-enable-kvm -netdev t
t as well. Are these interrupts independent or related?
Best Regards,
Arnab
On Mon, Jul 19, 2021 at 11:55 AM Thomas Huth wrote:
>
> On 18/07/2021 15.37, Arnabjyoti Kalita wrote:
> > Hello all,
> >
> > Is it possible to disable scsi disk requests being generated by the
> &g
eterministic event, like say, a
disk interrupt?
3) How do we separate disk requests that are actually generated by the
guest from the ones generated by QEMU (for eg. with operations like
loadvm/savevm)?
Thank you very much.
Best Regards,
Arnabjyoti Kalita
Dear Pavel,
Thank you very much for your answers.
So if I understand correctly, the blkreplay layer would only record
disk events generated by the virtual device, so this solves the
problem of recording only the relevant disk events that change the
guest state.
Best Regards,
Arnabjyoti Kalita
p so far.
Best Regards,
Arnabjyoti Kalita
Dear Pavel,
Thank you again for the answers.
I had another question -
How is it possible to just use the request id while recording and use
it during replay? Don't we have to record other parameters of the disk
I/O request (like offset, bytes, iov buffer etc.) ?
Best Regards,
Arnabjyoti K
ping does not happen?
If there are good articles that describe this process of DMA address space
mapping/unmapping, feel free to point me to them and I would love to read
them too.
Thank you very much.
Best Regards,
Arnabjyoti Kalita
quot; virtqueue buffers be written to the guest
memory, I can possibly just move forward without doing an unmap() in this
case.
Best Regards,
Arnabjyoti Kalita
int in recording them.
During replay, all I would do is call virtqueue_push and pass in the
members of the "VirtQueueElement" structure that I recorded.
I hope this makes it clear. Thank you again for helping out, Peter.
Best Regards,
Arnabjyoti Kalita
On Sun, Aug 29, 2021 at 5:49 PM P
hieve something else. And directly using the
upstream record-and-replay feature is a slight impediment for me since
I record in KVM mode and replay in TCG mode. But I get your point. I
might try a more generic design once I have a working system.
Best Regards,
Arnabjyoti Kalita
On Mon, Aug 30,
memory_unmap() happens, the contents of the buffer that
is pointed to by the host virtual address, will be finally written to
the guest physical memory which is pointed to by "pa" in the code
above?
Best Regards,
Arnabjyoti Kalita
On Mon, Aug 30, 2021 at 3:16 PM Arnabjyoti Kalita
wrote:
Thank you Peter. This makes it very clear.
Best Regards,
Arnabjyoti Kalita
On Tue, Aug 31, 2021 at 3:32 PM Peter Maydell wrote:
>
> On Tue, 31 Aug 2021 at 05:48, Arnabjyoti Kalita
> wrote:
> >
> > Dear Peter,
> >
> > I had a few additional questions wit
much.
Best Regards,
Arnabjyoti Kalita
chitecture and is running Linux kernel 4.4.0.
Thank you very much.
Best Regards,
Arnabjyoti Kalita
?
For context, I am using QEMU version 5.0.1 and am looking for page
fault handling mechanisms for the x86_64 architecture.
Thank you very much.
Best Regards,
Arnabjyoti Kalita
TB instruction fetch)? How can this
happen? Will guest execution fail if this happens?
- Can it happen that a correct host PC map to an incorrect guest PC ?
Best Regards,
Arnabjyoti Kalita
On Tue, Nov 9, 2021 at 4:42 PM Peter Maydell wrote:
>
> On Tue, 9 Nov 2021 at 05:44, Arnabjyoti
need to
raise an interrupt at the end of the execution of every TB ? Where in
code should I start making changes ? I do not want the functionality
of the original TCG execution driver to change.
Best Regards,
Arnabjyoti Kalita
robin implementation going.
Whenever the rdtsc instruction executes for a CPU, I read the next
recorded TSC values for each CPU and make the next scheduling
decision. I try to ensure that the TSC values are read in sequence.
Best Regards,
Arnabjyoti Kalita
On Tue, Jan 11, 2022 at 11:03 PM Alex B
is raised?
The guest and host architecture is both x86_64. My host runs Linux 5.8.0.
Thank you very much for your reply.
Best Regards,
Arnabjyoti Kalita
this? Any
pointers would be appreciated.
Thank you very much for all your help.
Best Regards,
Arnabjyoti Kalita
interrupt number assigned to virtio
devices.
Additionally, are there other programmatic ways in QEMU for me to
identify the interrupt numbers?
Thanks,
Arnabjyoti Kalita
65 matches
Mail list logo