[Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Stefan Hajnoczi]

QEMU will be applying to Google Summer of Code again this year.

Do you want to help some of the most talented university students
begin contributing to open source by working on our codebase and
joining our community?

We need to put together a list of project ideas by February 9:
http://qemu-project.org/Google_Summer_of_Code_2017#Project_Ideas

Active contributors are invited to post project ideas.  The project
should be doable in 12 weeks by someone fluent in C programming but
not familiar with the codebase.  Good project ideas have a clear
scope, few dependencies, a high chance of being merged, and can be
broken down into smaller steps.

For background on why QEMU participates in GSoC and how it works,
check out my KVM Forum presentation:
https://www.youtube.com/watch?v=xNVCX7YMUL8
https://vmsplice.net/~stefan/stefanha-kvm-forum-2016.pdf

KVM and Jailhouse are invited to participate under the QEMU umbrella
organization.

Please let me know if you have any questions.

Stefan

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Alex Bennée]

Stefan Hajnoczi <stefanha@gmail.com> writes:

> QEMU will be applying to Google Summer of Code again this year.
>
> Do you want to help some of the most talented university students
> begin contributing to open source by working on our codebase and
> joining our community?
>
> We need to put together a list of project ideas by February 9:
> http://qemu-project.org/Google_Summer_of_Code_2017#Project_Ideas

Will Outreachy point here as well or are we going to be duplicating
ideas? I seem to recall a lot of copy and pasting last time.

Could we not write a page-per-project and use categories with
transclusion on the wiki?

> Active contributors are invited to post project ideas.  The project
> should be doable in 12 weeks by someone fluent in C programming but
> not familiar with the codebase.  Good project ideas have a clear
> scope, few dependencies, a high chance of being merged, and can be
> broken down into smaller steps.
>
> For background on why QEMU participates in GSoC and how it works,
> check out my KVM Forum presentation:
> https://www.youtube.com/watch?v=xNVCX7YMUL8
> https://vmsplice.net/~stefan/stefanha-kvm-forum-2016.pdf
>
> KVM and Jailhouse are invited to participate under the QEMU umbrella
> organization.
>
> Please let me know if you have any questions.
>
> Stefan


--
Alex Bennée

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Stefan Hajnoczi]

On Wed, Jan 18, 2017 at 11:12 AM, Alex Bennée <alex.bennee@linaro.org> wrote:
>
> Stefan Hajnoczi <stefanha@gmail.com> writes:
>
>> QEMU will be applying to Google Summer of Code again this year.
>>
>> Do you want to help some of the most talented university students
>> begin contributing to open source by working on our codebase and
>> joining our community?
>>
>> We need to put together a list of project ideas by February 9:
>> http://qemu-project.org/Google_Summer_of_Code_2017#Project_Ideas
>
> Will Outreachy point here as well or are we going to be duplicating
> ideas? I seem to recall a lot of copy and pasting last time.
>
> Could we not write a page-per-project and use categories with
> transclusion on the wiki?

Sure, feel free to create a page for the project idea.

Stefan

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Peter Maydell]

On 18 January 2017 at 10:52, Stefan Hajnoczi <stefanha@gmail.com> wrote:
> QEMU will be applying to Google Summer of Code again this year.
>
> Do you want to help some of the most talented university students
> begin contributing to open source by working on our codebase and
> joining our community?
>
> We need to put together a list of project ideas by February 9:
> http://qemu-project.org/Google_Summer_of_Code_2017#Project_Ideas

Slightly late, but I had an idea last night: would "get
Raspbian booting on our raspi2 board model" be a good
project? Would involve some mix of bug fixing, cleaning
up and adding device models from the raspi2 tree on github,
and implementing missing devices. Bit of a "how long is a
piece of string" project, but on the other hand breaks
down easily into small parts that can all go upstream
individually. Would suit student who likes debugging :-)

thanks
-- PMM

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[John Snow]

On 02/09/2017 07:46 AM, Peter Maydell wrote:
> On 18 January 2017 at 10:52, Stefan Hajnoczi <stefanha@gmail.com> wrote:
>> QEMU will be applying to Google Summer of Code again this year.
>>
>> Do you want to help some of the most talented university students
>> begin contributing to open source by working on our codebase and
>> joining our community?
>>
>> We need to put together a list of project ideas by February 9:
>> http://qemu-project.org/Google_Summer_of_Code_2017#Project_Ideas
> 
> Slightly late, but I had an idea last night: would "get
> Raspbian booting on our raspi2 board model" be a good
> project? Would involve some mix of bug fixing, cleaning
> up and adding device models from the raspi2 tree on github,
> and implementing missing devices. Bit of a "how long is a
> piece of string" project, but on the other hand breaks
> down easily into small parts that can all go upstream
> individually. Would suit student who likes debugging :-)
> 
> thanks
> -- PMM
> 

Who was working on this most recently? There was someone submitting
patches pretty frequently for the raspi2 board within the last year,
wasn't there?

--js

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Peter Maydell]

On 9 February 2017 at 20:46, John Snow <jsnow@redhat.com> wrote:
> On 02/09/2017 07:46 AM, Peter Maydell wrote:
>> Slightly late, but I had an idea last night: would "get
>> Raspbian booting on our raspi2 board model" be a good
>> project? Would involve some mix of bug fixing, cleaning
>> up and adding device models from the raspi2 tree on github,
>> and implementing missing devices. Bit of a "how long is a
>> piece of string" project, but on the other hand breaks
>> down easily into small parts that can all go upstream
>> individually. Would suit student who likes debugging :-)

> Who was working on this most recently? There was someone submitting
> patches pretty frequently for the raspi2 board within the last year,
> wasn't there?

That was Andrew Baumann, but I think his use case was
getting Windows 10 to boot on it, which exercises
different bits of the hardware.

thanks
-- PMM

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Andrew Baumann]

> From: Peter Maydell [mailto:peter.maydell@linaro.org]
> Sent: Thursday, 9 February 2017 14:02
> On 9 February 2017 at 20:46, John Snow <jsnow@redhat.com> wrote:
> > On 02/09/2017 07:46 AM, Peter Maydell wrote:
> >> Slightly late, but I had an idea last night: would "get
> >> Raspbian booting on our raspi2 board model" be a good
> >> project? Would involve some mix of bug fixing, cleaning
> >> up and adding device models from the raspi2 tree on github,
> >> and implementing missing devices. Bit of a "how long is a
> >> piece of string" project, but on the other hand breaks
> >> down easily into small parts that can all go upstream
> >> individually. Would suit student who likes debugging :-)
> 
> > Who was working on this most recently? There was someone submitting
> > patches pretty frequently for the raspi2 board within the last year,
> > wasn't there?
> 
> That was Andrew Baumann, but I think his use case was
> getting Windows 10 to boot on it, which exercises
> different bits of the hardware.

That's right, but I thought we also had Raspbian working after Peter C's implementation of SETEND. (I wouldn't be surprised if it had regressed, however.)

I've dropped the ball on this, and don't forsee finding time for it in the immediate future, but if anyone wants to work on it there are still some outstanding bits of emulation lurking in my github tree (github.com/0xabu/qemu) mostly from the original implementation by Gregory Estrade: pi1 model, system timers, stub power management control, and USB controller. The last one (USB) is essential to having a usable device, but also sadly dubious in terms of code quality and completeness.

Cheers,
Andrew

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Peter Maydell]

On 9 February 2017 at 22:12, Andrew Baumann
<Andrew.Baumann@microsoft.com> wrote:
>> From: Peter Maydell [mailto:peter.maydell@linaro.org]
>> Sent: Thursday, 9 February 2017 14:02
>> On 9 February 2017 at 20:46, John Snow <jsnow@redhat.com> wrote:
>> > On 02/09/2017 07:46 AM, Peter Maydell wrote:
>> >> Slightly late, but I had an idea last night: would "get
>> >> Raspbian booting on our raspi2 board model" be a good
>> >> project? Would involve some mix of bug fixing, cleaning
>> >> up and adding device models from the raspi2 tree on github,
>> >> and implementing missing devices. Bit of a "how long is a
>> >> piece of string" project, but on the other hand breaks
>> >> down easily into small parts that can all go upstream
>> >> individually. Would suit student who likes debugging :-)
>>
>> > Who was working on this most recently? There was someone submitting
>> > patches pretty frequently for the raspi2 board within the last year,
>> > wasn't there?
>>
>> That was Andrew Baumann, but I think his use case was
>> getting Windows 10 to boot on it, which exercises
>> different bits of the hardware.
>
> That's right, but I thought we also had Raspbian working after
> Peter C's implementation of SETEND. (I wouldn't be surprised if
> it had regressed, however.)

For the raspbian kernel image I had it never worked -- I
think the kernel gained some new functionality that means
it now touches some device we don't yet implement (vague
recollection it was somethnig message-box related?).

> I've dropped the ball on this, and don't forsee finding time
> for it in the immediate future, but if anyone wants to work
> on it there are still some outstanding bits of emulation
> lurking in my github tree (github.com/0xabu/qemu)

Yeah, I figured that was probably the situation, and I
don't personally have the time to do it either; but
it seemed like maybe it would make a GSoC project.

thanks
-- PMM

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Stefan Hajnoczi]

[-- Attachment #1: Type: text/plain, Size: 2337 bytes --]

On Thu, Feb 09, 2017 at 10:17:45PM +0000, Peter Maydell wrote:
> On 9 February 2017 at 22:12, Andrew Baumann
> <Andrew.Baumann@microsoft.com> wrote:
> >> From: Peter Maydell [mailto:peter.maydell@linaro.org]
> >> Sent: Thursday, 9 February 2017 14:02
> >> On 9 February 2017 at 20:46, John Snow <jsnow@redhat.com> wrote:
> >> > On 02/09/2017 07:46 AM, Peter Maydell wrote:
> >> >> Slightly late, but I had an idea last night: would "get
> >> >> Raspbian booting on our raspi2 board model" be a good
> >> >> project? Would involve some mix of bug fixing, cleaning
> >> >> up and adding device models from the raspi2 tree on github,
> >> >> and implementing missing devices. Bit of a "how long is a
> >> >> piece of string" project, but on the other hand breaks
> >> >> down easily into small parts that can all go upstream
> >> >> individually. Would suit student who likes debugging :-)
> >>
> >> > Who was working on this most recently? There was someone submitting
> >> > patches pretty frequently for the raspi2 board within the last year,
> >> > wasn't there?
> >>
> >> That was Andrew Baumann, but I think his use case was
> >> getting Windows 10 to boot on it, which exercises
> >> different bits of the hardware.
> >
> > That's right, but I thought we also had Raspbian working after
> > Peter C's implementation of SETEND. (I wouldn't be surprised if
> > it had regressed, however.)
> 
> For the raspbian kernel image I had it never worked -- I
> think the kernel gained some new functionality that means
> it now touches some device we don't yet implement (vague
> recollection it was somethnig message-box related?).
> 
> > I've dropped the ball on this, and don't forsee finding time
> > for it in the immediate future, but if anyone wants to work
> > on it there are still some outstanding bits of emulation
> > lurking in my github tree (github.com/0xabu/qemu)
> 
> Yeah, I figured that was probably the situation, and I
> don't personally have the time to do it either; but
> it seemed like maybe it would make a GSoC project.

It sounds like this can be broken down into tasks by a mentor willing to
participate and knowledgable (or willing to do the homework) on device
support status.

Are there any volunteers?

This would be a great project to have on board.

Stefan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 455 bytes --]

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Alex Bennée]

Stefan Hajnoczi <stefanha@gmail.com> writes:

> On Thu, Feb 09, 2017 at 10:17:45PM +0000, Peter Maydell wrote:
>> On 9 February 2017 at 22:12, Andrew Baumann
>> <Andrew.Baumann@microsoft.com> wrote:
>> >> From: Peter Maydell [mailto:peter.maydell@linaro.org]
>> >> Sent: Thursday, 9 February 2017 14:02
>> >> On 9 February 2017 at 20:46, John Snow <jsnow@redhat.com> wrote:
>> >> > On 02/09/2017 07:46 AM, Peter Maydell wrote:
>> >> >> Slightly late, but I had an idea last night: would "get
>> >> >> Raspbian booting on our raspi2 board model" be a good
>> >> >> project? Would involve some mix of bug fixing, cleaning
>> >> >> up and adding device models from the raspi2 tree on github,
>> >> >> and implementing missing devices. Bit of a "how long is a
>> >> >> piece of string" project, but on the other hand breaks
>> >> >> down easily into small parts that can all go upstream
>> >> >> individually. Would suit student who likes debugging :-)
>> >>
>> >> > Who was working on this most recently? There was someone submitting
>> >> > patches pretty frequently for the raspi2 board within the last year,
>> >> > wasn't there?
>> >>
>> >> That was Andrew Baumann, but I think his use case was
>> >> getting Windows 10 to boot on it, which exercises
>> >> different bits of the hardware.
>> >
>> > That's right, but I thought we also had Raspbian working after
>> > Peter C's implementation of SETEND. (I wouldn't be surprised if
>> > it had regressed, however.)
>>
>> For the raspbian kernel image I had it never worked -- I
>> think the kernel gained some new functionality that means
>> it now touches some device we don't yet implement (vague
>> recollection it was somethnig message-box related?).

There are RNG related patches on the list at the moment.

>>
>> > I've dropped the ball on this, and don't forsee finding time
>> > for it in the immediate future, but if anyone wants to work
>> > on it there are still some outstanding bits of emulation
>> > lurking in my github tree (github.com/0xabu/qemu)
>>
>> Yeah, I figured that was probably the situation, and I
>> don't personally have the time to do it either; but
>> it seemed like maybe it would make a GSoC project.
>
> It sounds like this can be broken down into tasks by a mentor willing to
> participate and knowledgable (or willing to do the homework) on device
> support status.
>
> Are there any volunteers?

I'd be happy to help mentor if needed. I guess it will depend of anyone
picks the project up?

>
> This would be a great project to have on board.
>
> Stefan


--
Alex Bennée

Re: [Qemu-devel] Call for GSoC 2017 mentors & project ideas

[Peter Maydell]

On 13 February 2017 at 14:29, Alex Bennée <alex.bennee@linaro.org> wrote:
> There are RNG related patches on the list at the moment.

Yes. We should check whether that's all that's needed, because
if it is then there's no more project left here :-)

thanks
-- PMM