Patchwork [0/4] qemuppc: replace emulation of qemuppc

login
register
mail settings
Submitter Liming Wang
Date Jan. 5, 2012, 8:37 a.m.
Message ID <1325752666-28560-1-git-send-email-liming.wang@windriver.com>
Download mbox
Permalink /patch/18425/
State New
Headers show

Pull-request

git://128.224.162.211/poky.git qemuppc-new

Comments

Liming Wang - Jan. 5, 2012, 8:37 a.m.
Hi Richard,

These patches are aimed to replace emulation of qemuppc from prep to mac99.
With this new emulation, qemuppc now supports framebuffer and the network
works well.

The kernel config also needs modification and I will let Bruce to review the
patches.

The following changes since commit 0ff1d667dc23b0cc903c18f1b5e032c2b6630aae:

  distro: Add poky-tiny distro definition

are available in the git repository at:

  git://128.224.162.211/poky.git qemuppc-new

Liming Wang (4):
      qemuppc: replace emulation of qemuppc from prep to mac99
      qemuppc: add pointercal file
      qemuppc: add machconfig
      tslib: enable big endian support for qemuppc

 meta/conf/machine/qemuppc.conf                     |    2 +-
 .../formfactor/files/qemuppc/machconfig            |   10 ++++++++++
 .../pointercal/files/qemuppc/pointercal            |  Bin 0 -> 36 bytes
 meta/recipes-graphics/tslib/tslib_1.0.bb           |    1 +
 scripts/runqemu                                    |    7 +------
 scripts/runqemu-internal                           |   16 ++++++++--------
 6 files changed, 21 insertions(+), 15 deletions(-)
 create mode 100755 meta/recipes-bsp/formfactor/files/qemuppc/machconfig
 create mode 100644 meta/recipes-bsp/pointercal/files/qemuppc/pointercal
Bruce Ashfield - Jan. 6, 2012, 2:29 p.m.
On Thu, Jan 5, 2012 at 3:37 AM, Liming Wang <liming.wang@windriver.com> wrote:
> Hi Richard,
>
> These patches are aimed to replace emulation of qemuppc from prep to mac99.
> With this new emulation, qemuppc now supports framebuffer and the network
> works well.
>
> The kernel config also needs modification and I will let Bruce to review the
> patches.

I have the patches in my inbox now, I'm reviewing and merging them shortly.
I'll pull these changes in to do some boot and runtime testing as well.

Richard/Saul: if it works better for coordination purposes, I can do the kernel
part, take this into my tree as well and then send a single coordinated pull
request when everything is known to work together.

Cheers,

Bruce

>
> The following changes since commit 0ff1d667dc23b0cc903c18f1b5e032c2b6630aae:
>
>  distro: Add poky-tiny distro definition
>
> are available in the git repository at:
>
>  git://128.224.162.211/poky.git qemuppc-new
>
> Liming Wang (4):
>      qemuppc: replace emulation of qemuppc from prep to mac99
>      qemuppc: add pointercal file
>      qemuppc: add machconfig
>      tslib: enable big endian support for qemuppc
>
>  meta/conf/machine/qemuppc.conf                     |    2 +-
>  .../formfactor/files/qemuppc/machconfig            |   10 ++++++++++
>  .../pointercal/files/qemuppc/pointercal            |  Bin 0 -> 36 bytes
>  meta/recipes-graphics/tslib/tslib_1.0.bb           |    1 +
>  scripts/runqemu                                    |    7 +------
>  scripts/runqemu-internal                           |   16 ++++++++--------
>  6 files changed, 21 insertions(+), 15 deletions(-)
>  create mode 100755 meta/recipes-bsp/formfactor/files/qemuppc/machconfig
>  create mode 100644 meta/recipes-bsp/pointercal/files/qemuppc/pointercal
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Richard Purdie - Jan. 6, 2012, 2:33 p.m.
On Fri, 2012-01-06 at 09:29 -0500, Bruce Ashfield wrote:
> On Thu, Jan 5, 2012 at 3:37 AM, Liming Wang <liming.wang@windriver.com> wrote:
> > Hi Richard,
> >
> > These patches are aimed to replace emulation of qemuppc from prep to mac99.
> > With this new emulation, qemuppc now supports framebuffer and the network
> > works well.
> >
> > The kernel config also needs modification and I will let Bruce to review the
> > patches.
> 
> I have the patches in my inbox now, I'm reviewing and merging them shortly.
> I'll pull these changes in to do some boot and runtime testing as well.
> 
> Richard/Saul: if it works better for coordination purposes, I can do the kernel
> part, take this into my tree as well and then send a single coordinated pull
> request when everything is known to work together.

Yes please.

The userspace bits look reasonable to me but they need to be in step
with the kernel changes. I'm looking forward to working framebuffer
support :)

Cheers,

Richard
Bruce Ashfield - Jan. 9, 2012, 4:07 p.m.
On Fri, Jan 6, 2012 at 9:33 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Fri, 2012-01-06 at 09:29 -0500, Bruce Ashfield wrote:
>> On Thu, Jan 5, 2012 at 3:37 AM, Liming Wang <liming.wang@windriver.com> wrote:
>> > Hi Richard,
>> >
>> > These patches are aimed to replace emulation of qemuppc from prep to mac99.
>> > With this new emulation, qemuppc now supports framebuffer and the network
>> > works well.
>> >
>> > The kernel config also needs modification and I will let Bruce to review the
>> > patches.
>>
>> I have the patches in my inbox now, I'm reviewing and merging them shortly.
>> I'll pull these changes in to do some boot and runtime testing as well.
>>
>> Richard/Saul: if it works better for coordination purposes, I can do the kernel
>> part, take this into my tree as well and then send a single coordinated pull
>> request when everything is known to work together.
>
> Yes please.
>
> The userspace bits look reasonable to me but they need to be in step
> with the kernel changes. I'm looking forward to working framebuffer
> support :)

In case anyone is wondering, I merged and started testing this on Friday. I'm
working with Liming to hunt down a mouse grab/cusor and keyboard issue that I'm
seeing. But other than that, everything worked nicely. If I can't
resolve the two
minor items quickly, I'll send the queue anyway and more eyes can be put on
the problem.

Cheers,

Bruce

>
> Cheers,
>
> Richard
>
Saul Wold - Jan. 21, 2012, 3:50 a.m.
On 01/05/2012 12:37 AM, Liming Wang wrote:
> Hi Richard,
>
> These patches are aimed to replace emulation of qemuppc from prep to mac99.
> With this new emulation, qemuppc now supports framebuffer and the network
> works well.
>
Is there anything that needs to be setup differently for the runqemu 
scripts?  Our sanity tests have failed after changing to the new ppc 
code and I wonder if there is something for the command line or ???

THanks
	Sau!

> The kernel config also needs modification and I will let Bruce to review the
> patches.
>
> The following changes since commit 0ff1d667dc23b0cc903c18f1b5e032c2b6630aae:
>
>    distro: Add poky-tiny distro definition
>
> are available in the git repository at:
>
>    git://128.224.162.211/poky.git qemuppc-new
>
> Liming Wang (4):
>        qemuppc: replace emulation of qemuppc from prep to mac99
>        qemuppc: add pointercal file
>        qemuppc: add machconfig
>        tslib: enable big endian support for qemuppc
>
>   meta/conf/machine/qemuppc.conf                     |    2 +-
>   .../formfactor/files/qemuppc/machconfig            |   10 ++++++++++
>   .../pointercal/files/qemuppc/pointercal            |  Bin 0 ->  36 bytes
>   meta/recipes-graphics/tslib/tslib_1.0.bb           |    1 +
>   scripts/runqemu                                    |    7 +------
>   scripts/runqemu-internal                           |   16 ++++++++--------
>   6 files changed, 21 insertions(+), 15 deletions(-)
>   create mode 100755 meta/recipes-bsp/formfactor/files/qemuppc/machconfig
>   create mode 100644 meta/recipes-bsp/pointercal/files/qemuppc/pointercal
>
Bruce Ashfield - Jan. 21, 2012, 4:12 a.m.
On Fri, Jan 20, 2012 at 10:50 PM, Saul Wold <sgw@linux.intel.com> wrote:
> On 01/05/2012 12:37 AM, Liming Wang wrote:
>>
>> Hi Richard,
>>
>> These patches are aimed to replace emulation of qemuppc from prep to
>> mac99.
>> With this new emulation, qemuppc now supports framebuffer and the network
>> works well.
>>
> Is there anything that needs to be setup differently for the runqemu
> scripts?  Our sanity tests have failed after changing to the new ppc code
> and I wonder if there is something for the command line or ???

What type of failure ? boot ? I'm building and booting it all the time, so I can
help there. Or are you talking about something in particular like the network ?
disk ?

Bruce

>
> THanks
>        Sau!
>
>
>> The kernel config also needs modification and I will let Bruce to review
>> the
>> patches.
>>
>> The following changes since commit
>> 0ff1d667dc23b0cc903c18f1b5e032c2b6630aae:
>>
>>   distro: Add poky-tiny distro definition
>>
>> are available in the git repository at:
>>
>>   git://128.224.162.211/poky.git qemuppc-new
>>
>> Liming Wang (4):
>>       qemuppc: replace emulation of qemuppc from prep to mac99
>>       qemuppc: add pointercal file
>>       qemuppc: add machconfig
>>       tslib: enable big endian support for qemuppc
>>
>>  meta/conf/machine/qemuppc.conf                     |    2 +-
>>  .../formfactor/files/qemuppc/machconfig            |   10 ++++++++++
>>  .../pointercal/files/qemuppc/pointercal            |  Bin 0 ->  36 bytes
>>  meta/recipes-graphics/tslib/tslib_1.0.bb           |    1 +
>>  scripts/runqemu                                    |    7 +------
>>  scripts/runqemu-internal                           |   16
>> ++++++++--------
>>  6 files changed, 21 insertions(+), 15 deletions(-)
>>  create mode 100755 meta/recipes-bsp/formfactor/files/qemuppc/machconfig
>>  create mode 100644 meta/recipes-bsp/pointercal/files/qemuppc/pointercal
>>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Richard Purdie - Jan. 21, 2012, 11:48 a.m.
On Fri, 2012-01-20 at 23:12 -0500, Bruce Ashfield wrote:
> On Fri, Jan 20, 2012 at 10:50 PM, Saul Wold <sgw@linux.intel.com> wrote:
> > On 01/05/2012 12:37 AM, Liming Wang wrote:
> >>
> >> Hi Richard,
> >>
> >> These patches are aimed to replace emulation of qemuppc from prep to
> >> mac99.
> >> With this new emulation, qemuppc now supports framebuffer and the network
> >> works well.
> >>
> > Is there anything that needs to be setup differently for the runqemu
> > scripts?  Our sanity tests have failed after changing to the new ppc code
> > and I wonder if there is something for the command line or ???
> 
> What type of failure ? boot ? I'm building and booting it all the time, so I can
> help there. Or are you talking about something in particular like the network ?
> disk ?

I ran a test build and can confirm that whilst qemuppc can boot, there
is no network. We need to fix the network ASAP as our automated testing
depends on working network access...

Cheers,

Richard
Bruce Ashfield - Jan. 21, 2012, 2:29 p.m.
On Sat, Jan 21, 2012 at 6:48 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Fri, 2012-01-20 at 23:12 -0500, Bruce Ashfield wrote:
>> On Fri, Jan 20, 2012 at 10:50 PM, Saul Wold <sgw@linux.intel.com> wrote:
>> > On 01/05/2012 12:37 AM, Liming Wang wrote:
>> >>
>> >> Hi Richard,
>> >>
>> >> These patches are aimed to replace emulation of qemuppc from prep to
>> >> mac99.
>> >> With this new emulation, qemuppc now supports framebuffer and the network
>> >> works well.
>> >>
>> > Is there anything that needs to be setup differently for the runqemu
>> > scripts?  Our sanity tests have failed after changing to the new ppc code
>> > and I wonder if there is something for the command line or ???
>>
>> What type of failure ? boot ? I'm building and booting it all the time, so I can
>> help there. Or are you talking about something in particular like the network ?
>> disk ?
>
> I ran a test build and can confirm that whilst qemuppc can boot, there
> is no network. We need to fix the network ASAP as our automated testing
> depends on working network access...

Liming is out for the next week or so, but I'll see what I can do on Monday.
He did report a properly working networking with his initial submission, so it
should just be a configuration issue.

Cheers,

Bruce

>
> Cheers,
>
> Richard
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
Bruce Ashfield - Jan. 23, 2012, 3:44 p.m.
On Sat, Jan 21, 2012 at 9:29 AM, Bruce Ashfield
<bruce.ashfield@gmail.com> wrote:
> On Sat, Jan 21, 2012 at 6:48 AM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
>> On Fri, 2012-01-20 at 23:12 -0500, Bruce Ashfield wrote:
>>> On Fri, Jan 20, 2012 at 10:50 PM, Saul Wold <sgw@linux.intel.com> wrote:
>>> > On 01/05/2012 12:37 AM, Liming Wang wrote:
>>> >>
>>> >> Hi Richard,
>>> >>
>>> >> These patches are aimed to replace emulation of qemuppc from prep to
>>> >> mac99.
>>> >> With this new emulation, qemuppc now supports framebuffer and the network
>>> >> works well.
>>> >>
>>> > Is there anything that needs to be setup differently for the runqemu
>>> > scripts?  Our sanity tests have failed after changing to the new ppc code
>>> > and I wonder if there is something for the command line or ???
>>>
>>> What type of failure ? boot ? I'm building and booting it all the time, so I can
>>> help there. Or are you talking about something in particular like the network ?
>>> disk ?
>>
>> I ran a test build and can confirm that whilst qemuppc can boot, there
>> is no network. We need to fix the network ASAP as our automated testing
>> depends on working network access...
>
> Liming is out for the next week or so, but I'll see what I can do on Monday.
> He did report a properly working networking with his initial submission, so it
> should just be a configuration issue.

Does anyone have more details on this for me ? I just did a build and boot into
sato and everything worked here. Mouse was fine, opened the terminal, ifconfig
reports and up and running ethernet, and I can ping and ssh out of qemppc to
my host machine.

So I've got nothing to test here, since it all worked.

Note: there is one small change to runqemu that I can send along to enable the
keyboard in sato, but that's all that I have to go on.

Cheers,

Bruce

>
> Cheers,
>
> Bruce
>
>>
>> Cheers,
>>
>> Richard
>>
>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
>
>
> --
> "Thou shalt not follow the NULL pointer, for chaos and madness await
> thee at its end"
Bruce Ashfield - Jan. 23, 2012, 5:57 p.m.
On Mon, Jan 23, 2012 at 10:44 AM, Bruce Ashfield
<bruce.ashfield@gmail.com> wrote:
> On Sat, Jan 21, 2012 at 9:29 AM, Bruce Ashfield
> <bruce.ashfield@gmail.com> wrote:
>> On Sat, Jan 21, 2012 at 6:48 AM, Richard Purdie
>> <richard.purdie@linuxfoundation.org> wrote:
>>> On Fri, 2012-01-20 at 23:12 -0500, Bruce Ashfield wrote:
>>>> On Fri, Jan 20, 2012 at 10:50 PM, Saul Wold <sgw@linux.intel.com> wrote:
>>>> > On 01/05/2012 12:37 AM, Liming Wang wrote:
>>>> >>
>>>> >> Hi Richard,
>>>> >>
>>>> >> These patches are aimed to replace emulation of qemuppc from prep to
>>>> >> mac99.
>>>> >> With this new emulation, qemuppc now supports framebuffer and the network
>>>> >> works well.
>>>> >>
>>>> > Is there anything that needs to be setup differently for the runqemu
>>>> > scripts?  Our sanity tests have failed after changing to the new ppc code
>>>> > and I wonder if there is something for the command line or ???
>>>>
>>>> What type of failure ? boot ? I'm building and booting it all the time, so I can
>>>> help there. Or are you talking about something in particular like the network ?
>>>> disk ?
>>>
>>> I ran a test build and can confirm that whilst qemuppc can boot, there
>>> is no network. We need to fix the network ASAP as our automated testing
>>> depends on working network access...
>>
>> Liming is out for the next week or so, but I'll see what I can do on Monday.
>> He did report a properly working networking with his initial submission, so it
>> should just be a configuration issue.
>
> Does anyone have more details on this for me ? I just did a build and boot into
> sato and everything worked here. Mouse was fine, opened the terminal, ifconfig
> reports and up and running ethernet, and I can ping and ssh out of qemppc to
> my host machine.
>
> So I've got nothing to test here, since it all worked.
>
> Note: there is one small change to runqemu that I can send along to enable the
> keyboard in sato, but that's all that I have to go on.

As was discussed on IRC, I was booting my 3.2 kernel .. where everything worked.
I've fixed this on 3.0 and sent a pull request.

Bruce

>
> Cheers,
>
> Bruce
>
>>
>> Cheers,
>>
>> Bruce
>>
>>>
>>> Cheers,
>>>
>>> Richard
>>>
>>>
>>> _______________________________________________
>>> Openembedded-core mailing list
>>> Openembedded-core@lists.openembedded.org
>>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>>
>>
>>
>> --
>> "Thou shalt not follow the NULL pointer, for chaos and madness await
>> thee at its end"
>
>
>
> --
> "Thou shalt not follow the NULL pointer, for chaos and madness await
> thee at its end"