[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: basic qemu question
From: |
Alex Bennée |
Subject: |
Re: basic qemu question |
Date: |
Fri, 12 Jun 2020 21:14:22 +0100 |
User-agent: |
mu4e 1.5.2; emacs 28.0.50 |
David Beccue <david.qemulist@beccue.com> writes:
> Thanks for your thoughts on this.
>
> Yes, the analysis is just a library without src code that I compile into my
> firmware that runs on a Cortex-M3.
>
> With close to speed parity, I guess I could at least run a bunch in
> parallel on my multicore AMD 1700x more easily than loading the code in 16
> different hardware devices.
It will depend if QEMU models the device you are after. For m3 there are
mps2, stellaris and netduino2 models.
> Plus, I assume that QEMU may have a way that I
> can send debug output to my host machine to store in files there, yes?
Yes - both normal serial ports and semihosting output can be re-directed
to a file using the standard -chardev options.
>
> On Fri, Jun 12, 2020 at 10:29 AM Alex Bennée <alex.bennee@linaro.org> wrote:
>
>> The following message is a courtesy copy of an article
>> that has been posted to gmane.comp.emulators.qemu.user as well.
>>
>> David Beccue <David-UKP4QdZF84nQT0dZR+AlfA@public.gmane.org> writes:
>>
>> > I have a pretty basic question about how qemu works... I have an
>> > analysis library (no source) for ARM Cortex M3 processor that I'd like
>> > to run on many files.
>>
>> The analysis library runs on the Cortex-M3?
>>
>> > My hardware would be very slow doing
>> > this. Will QEMU be able to run this faster (assuming a fast PC, ofc)?
>>
>> There is a fairly big performance penalty for softmmu emulation but
>> given the speed of most modern PCs compared to microcontrollers you
>> would possibly be able to achieve parity.
>>
>> >
>> > Cheers,
>> > David
>> >
>> > - - -
>> > David Beccue
>> >
>>
>> --
>> Alex Bennée
>>
>>
--
Alex Bennée