[Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Jani Monoses]

I was confused by these error messages in dmesg when porting to a new
device, thinking some are vendor binaries I need to add to vendor/ .

<3>[1, init] [6.016357] init: cannot find '/system/bin/vold',
disabling 'vold'
<3>[1, init] [6.018890] init: cannot find '/system/bin/netd',
disabling 'netd'
<3>[1, init] [6.020904] init: cannot find
'/system/bin/app_process', disabling 'zygote'
<3>[1, init] [6.029113] init: cannot find '/system/bin/keystore',
disabling 'keystore'

Turns out they are standard Android services which are not built and
used in Ubuntu Touch.

I attach a patch that disables starting these in init.rc to avoid such
error messages.

thanks
Jani


0001-Do-not-try-starting-inexistent-services-in-init.rc.patch
Description: Binary data
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] [Fwd: Image based updates for Ubuntu Touch]

[Oliver Grawert]

--- Begin Message ---
Hello,

I initially wanted to send just a short e-mail on the subject of Image
based updates for our mobile devices, but it evolved into something far
longer. Hopefully I managed to make it readable ;)

== Context ==
Some may remember the session on image based updates from the last vUDS.
We discussed how we'd like the phones and tablets to update and how
different this would have to be from regular APT based updates to meet
some of the new requirements that come with those platforms.

vUDS session:
http://summit.ubuntu.com/uds-1303/meeting/21605/foundations-1303-single-image-update/

== Current work ==
Since UDS, I've been talking with quite a few people about that subject
and we've come up with a rather detailed plan which we are now prototyping.

The details of that plan can be found at:
https://wiki.ubuntu.com/ImageBasedUpgrades/Mobile

This is a rather lengthy spec that tries to cover how we're planning on
updating those devices keeping in mind that some will be updated by the
carrier/OEM and some will be updated by us.

This spec is still evolving as we find new issues so please don't
hesitate to get in touch with me if you feel we overlooked something or
you have questions about some specific bits.

== Presentation ==
Last Friday we had an Hangout On Air with Daniel Holbach and most of the
people involved in the image based update design so far. This covers the
spec itself and some more details on the various prototypes we're
working on.

The recording may be found here:
http://www.youtube.com/watch?v=iWum6Yv9XOU

== Quick intro ==
The overall design can be split into 3 pieces:
 - The update server which contains all the various Ubuntu rootfs images
and Android images for the various supported devices and placed into
update "channels" (that vary in stability and update frequency).
The server code is also responsible for generating the diffs between
images and generating the final zip files that are used by the upgrader.

 - The update client connects to the update server and base on the
current version on the device, the update channel and the amount of
available disk space, resolves the best upgrade path that can fit in the
device's memory. After validating all the files, it reboots the device
into recovery mode.

 - The upgrader sits in the recovery partition and when triggered scans
for any update that was downloaded by the update client, validates the
file signatures, then apply all the updates sequentially directly on top
of the current rootfs. When done, it reboots the device into standard mode.

The updates themselves are currently a squashfs, a list of files to
remove and an update script stored into a zip file. The current server
implementation doesn't use GPG yet, but will in the next couple of days
(the indexes will be signed and so will the content of the zip files).

The rootfs on the device is a read-only ext4 partition on top of which
we basically apply either full updates (wipe everything and unpack a
clean image) or apply a diff (copy a bunch of files, remove some others).
In the current version of the spec, anyone who opts to move to
read/write and directly use apt-get/dpkg won't be able to use image
based updates afterwards and will have to directly use apt-get on the
device.

== Prototype ==
I'm currently working on a prototype server which is available at:
https://phablet.stgraber.org

Barry is working on an initial client that'll get the right data from
the server, calculate an upgrade path, then download and validate the
various bits.

Finally, Ondrej is looking into the actual upgrader which will live in
the recovery ROM and do the final unpacking, validation and application
of the updates.

== Timeline ==
We're hoping to have a fully working prototype of the whole solution in
the next week or so which we'll then try to integrate within the
existing Touch images.

Additional work to support full convergence mode (running Ubuntu desktop
from an Ubuntu touch device) is also on the map but isn't scheduled for
the initial iteration (October) so we still have time to figure out what
to do there.
Some initial thoughts on the topic can be found at:
https://wiki.ubuntu.com/ImageBasedUpgrades/Mobile/LocalPackages

== Getting involved ==
So as I mentioned, the different people involved in the implementation are:
 - Client side: Barry Warsaw
 - Upgrader/recovery ROM: Ondrej Kubik
 - Server side: Stéphane Graber

If you want to help, you can directly get in touch with us.
If you have comments on the spec as a whole, then simply reply to this
e-mail or get in touch with me.

I'm trying to keep the wiki as up to date as I can without getting into
too small details that'd make it even harder to read than it currently
is, however I'm planning on creating a few more wiki pages to detail the
way the client/updater/server each work.

-- 
Stéphane Graber
Ubuntu developer
http://www.ubuntu.com



signature.asc
Description: OpenPGP digital signature
-- 
ubuntu-devel mailing 

[Ubuntu-phone] Choice of included ARM toolchains

[Jani Monoses]

Hi,

I notice there are a four ARM cross compilers for linux-x86 in our tree.
Is it because various hw has been only shown to build and work
correctly with a specific one only?

./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/

thanks
Jani

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Oliver Grawert]

hi
On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:
> Hi,
> 
> I notice there are a four ARM cross compilers for linux-x86 in our tree.
> Is it because various hw has been only shown to build and work
> correctly with a specific one only?
> 
> ./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
> ./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
> ./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
> ./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/
> 
lol, funny that you just looked at the same stuff ...
with teh following:

rm -rf prebuilts/gcc/linux-x86/mips
rm -rf prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6
rm -rf prebuilts/gcc/darwin-x86
rm -rf prebuilt/darwin-x86 # needs change for java version check
(comment out line 126-128 in build/core/main.mk and make sure openjdk-6
is installed on the host)

this still gets me properly built images over here ... and saves about
700M in the tree ...


steve langasek was looking into dropping all toolchains from the tree
yesterday and instaed use the gcc-cross packages we provide in the
archive ... afaik he ran into some linker issues with gold, if these are
resolved we might be able to switch over and save a few gigs ...
additionally the kernels for supported devices are just being moved out
as well ... in the end we should have a much smaller tree.

ciao
oli 


signature.asc
Description: This is a digitally signed message part
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Sergio Schvezov]

On Thu, Apr 18, 2013 at 02:44:30PM +0200, Oliver Grawert wrote:
> hi
> On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:
> > Hi,
> > 
> > I notice there are a four ARM cross compilers for linux-x86 in our tree.
> > Is it because various hw has been only shown to build and work
> > correctly with a specific one only?
> > 
> > ./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
> > ./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
> > ./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
> > ./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/
> > 
> lol, funny that you just looked at the same stuff ...
> with teh following:
> 
> rm -rf prebuilts/gcc/linux-x86/mips
> rm -rf prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6
> rm -rf prebuilts/gcc/darwin-x86
> rm -rf prebuilt/darwin-x86 # needs change for java version check
> (comment out line 126-128 in build/core/main.mk and make sure openjdk-6
> is installed on the host)
> 
> this still gets me properly built images over here ... and saves about
> 700M in the tree ...
> 
> 
> steve langasek was looking into dropping all toolchains from the tree
> yesterday and instaed use the gcc-cross packages we provide in the
> archive ... afaik he ran into some linker issues with gold, if these are
> resolved we might be able to switch over and save a few gigs ...
> additionally the kernels for supported devices are just being moved out
> as well ... in the end we should have a much smaller tree.

Ricardo was working on a patch yesterday to remove all these. Should happen
soon.

I'm working on getting the need for java out of the way and that last email
ogra forwarded about updates sort of clears out a doubt we had last night.
But if anywhere, I'll follow up on that thread.

Cheers
Sergio

PS: frameworks base decreased in 400MB 6 hours ago. More on the way.


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Sergio Schvezov]

On Thu, Apr 18, 2013 at 11:59:23AM +0300, Jani Monoses wrote:
> I attach a patch that disables starting these in init.rc to avoid such
> error messages.

Any reason to comment instead of just remove? I would prefer removal myself.

Cheers
Sergio

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Ricardo Salveti de Araujo]

On 04/18/2013 09:44 AM, Oliver Grawert wrote:

hi
On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:

Hi,

I notice there are a four ARM cross compilers for linux-x86 in our tree.
Is it because various hw has been only shown to build and work
correctly with a specific one only?

./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/


We have multiple toolchains but I believe only the linaro one is being 
actually used at the build, that's why Oliver pointed that the build 
worked when removing the others.


And they are still there simply because there was no explicit need to 
remove them (which is kind of the case now).



steve langasek was looking into dropping all toolchains from the tree
yesterday and instaed use the gcc-cross packages we provide in the
archive ... afaik he ran into some linker issues with gold, if these are
resolved we might be able to switch over and save a few gigs ...
additionally the kernels for supported devices are just being moved out
as well ... in the end we should have a much smaller tree.


I'm working on removing the not needed ones at least, but I'm not yet 
convinced that we should remove all and just reuse whatever is available 
in the archive (at least not at this point).


The huge benefit of having the prebuilt toolchain as part of the build 
environment is that people don't need to care about which specific 
version of Ubuntu they are using at the host side.


Cheers,
--
Ricardo Salveti de Araujo

--
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Ricardo Salveti de Araujo]

On 04/18/2013 10:55 AM, Sergio Schvezov wrote:

On Thu, Apr 18, 2013 at 11:59:23AM +0300, Jani Monoses wrote:

I attach a patch that disables starting these in init.rc to avoid such
error messages.


Any reason to comment instead of just remove? I would prefer removal myself.


+1 to remove.

I'm just not so sure about vold, as I remember people saying that it's 
needed for a few devices, need to investigate further.


Cheers,
--
Ricardo Salveti de Araujo

--
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Jani Monoses]

> I'm working on removing the not needed ones at least, but I'm not yet
> convinced that we should remove all and just reuse whatever is available in
> the archive (at least not at this point).

I agree. I found the raring default cross-toolchain failed to build
mako's kernel a while ago, and none but gcc-4.6 managed to build a
bootable 3.0.8 based kernel for a Huawei device.

Jani

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Jani Monoses]

On Thu, Apr 18, 2013 at 4:59 PM, Ricardo Salveti de Araujo
 wrote:
> On 04/18/2013 10:55 AM, Sergio Schvezov wrote:
>>
>> On Thu, Apr 18, 2013 at 11:59:23AM +0300, Jani Monoses wrote:
>>>
>>> I attach a patch that disables starting these in init.rc to avoid such
>>> error messages.
>>
>>
>> Any reason to comment instead of just remove? I would prefer removal
>> myself.
>
>
> +1 to remove.
>

I just followed what was done for bootanim, which is commented out in
the same file, and for keeping
the context handy in case there's a merge with upstream.
I am fine with removing.

> I'm just not so sure about vold, as I remember people saying that it's
> needed for a few devices, need to investigate further.

I just saw it is not built for mako, did not know we build different
Android bits depending on target.

Jani

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Sergio Schvezov]

On Thu, Apr 18, 2013 at 10:57:23AM -0300, Ricardo Salveti de Araujo wrote:
> On 04/18/2013 09:44 AM, Oliver Grawert wrote:
> >hi
> >On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:
> >>Hi,
> >>
> >>I notice there are a four ARM cross compilers for linux-x86 in our tree.
> >>Is it because various hw has been only shown to build and work
> >>correctly with a specific one only?
> >>
> >>./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
> >>./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
> >>./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
> >>./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/
> 
> We have multiple toolchains but I believe only the linaro one is
> being actually used at the build, that's why Oliver pointed that the
> build worked when removing the others.

Wait, don't forget about that kernel rule to use 4.6 (IIRC). That should
stay even if we get all the nexus kernels in from packages or think of a 
way for porters to get it working, or some extra download in the 
cm.dependencies for the device.

Well, the IIRC comes from [1].

Cheers
Sergio

[1] 
http://phablet.ubuntu.com/gitweb?p=CyanogenMod/android_build.git;a=commitdiff;h=0828186ad8d02876bc00c73285d9c341ee76bab2

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Sergio Schvezov]

On Thu, Apr 18, 2013 at 05:02:55PM +0300, Jani Monoses wrote:
> On Thu, Apr 18, 2013 at 4:59 PM, Ricardo Salveti de Araujo
>  wrote:
> > On 04/18/2013 10:55 AM, Sergio Schvezov wrote:
> >>
> >> On Thu, Apr 18, 2013 at 11:59:23AM +0300, Jani Monoses wrote:
> >>>
> >>> I attach a patch that disables starting these in init.rc to avoid such
> >>> error messages.
> >>
> >>
> >> Any reason to comment instead of just remove? I would prefer removal
> >> myself.
> >
> >
> > +1 to remove.
> >
> 
> I just followed what was done for bootanim, which is commented out in
> the same file, and for keeping
> the context handy in case there's a merge with upstream.
> I am fine with removing.

That's a leftover from when we didn't have a repo and just some mega patches.
Feel free to remove that one too if you like.

> > I'm just not so sure about vold, as I remember people saying that it's
> > needed for a few devices, need to investigate further.
> 
> I just saw it is not built for mako, did not know we build different
> Android bits depending on target.

Well if someone needs it, they can add it to their devices init.rc, right?


-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Huawei Ascend P1 WIP port

[Jani Monoses]

Hi,

I added a new page for the Huawei Ascend P1.
For now it only boots into the Phone UI and does not do much else, but
I'll keep improving it hopefully.
This is a device with a SoC similar to the one in the Galaxy Nexus,
but ships only with Android 4.0.3 and is not supported by CyanogenMod.

https://wiki.ubuntu.com/Touch/Devices/u9200

Jani

-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Oliver Grawert]

hi,
Am Donnerstag, den 18.04.2013, 10:57 -0300 schrieb Ricardo Salveti de
Araujo:
> On 04/18/2013 09:44 AM, Oliver Grawert wrote:
> > hi
> > On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:
> >> Hi,
> >>
> >> I notice there are a four ARM cross compilers for linux-x86 in our tree.
> >> Is it because various hw has been only shown to build and work
> >> correctly with a specific one only?
> >>
> >> ./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
> >> ./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
> >> ./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
> >> ./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/
> 
> We have multiple toolchains but I believe only the linaro one is being 
> actually used at the build, that's why Oliver pointed that the build 
> worked when removing the others.
> 
you still need prebuilts/gcc/linux-x86/host unless you hack around it in
the build system (it is referenced in many places it seems)

and prebuilts/gcc/linux-x86/arm/arm-eabi-4.6 was needed by the mako
kernel for me, it failed badly with teh linaro chain (i guess that
should easily be solvable with the gcc-cross from the archive or just by
using the binaries with your last kernel removal patch)

ciao
oli


signature.asc
Description: This is a digitally signed message part
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] [PATCH] Do not try starting vold, netd, keystore and app_process from init.rc

[Jani Monoses]

Revised patch that removes those lines and bootanim too.

cheers
Jani

On Thu, Apr 18, 2013 at 5:24 PM, Sergio Schvezov
 wrote:
> On Thu, Apr 18, 2013 at 05:02:55PM +0300, Jani Monoses wrote:
>> On Thu, Apr 18, 2013 at 4:59 PM, Ricardo Salveti de Araujo
>>  wrote:
>> > On 04/18/2013 10:55 AM, Sergio Schvezov wrote:
>> >>
>> >> On Thu, Apr 18, 2013 at 11:59:23AM +0300, Jani Monoses wrote:
>> >>>
>> >>> I attach a patch that disables starting these in init.rc to avoid such
>> >>> error messages.
>> >>
>> >>
>> >> Any reason to comment instead of just remove? I would prefer removal
>> >> myself.
>> >
>> >
>> > +1 to remove.
>> >
>>
>> I just followed what was done for bootanim, which is commented out in
>> the same file, and for keeping
>> the context handy in case there's a merge with upstream.
>> I am fine with removing.
>
> That's a leftover from when we didn't have a repo and just some mega patches.
> Feel free to remove that one too if you like.
>
>> > I'm just not so sure about vold, as I remember people saying that it's
>> > needed for a few devices, need to investigate further.
>>
>> I just saw it is not built for mako, did not know we build different
>> Android bits depending on target.
>
> Well if someone needs it, they can add it to their devices init.rc, right?
>
>
> --
> Mailing list: https://launchpad.net/~ubuntu-phone
> Post to : ubuntu-phone@lists.launchpad.net
> Unsubscribe : https://launchpad.net/~ubuntu-phone
> More help   : https://help.launchpad.net/ListHelp


0001-Do-not-try-starting-inexistent-services-in-init.rc.patch
Description: Binary data
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Choice of included ARM toolchains

[Ricardo Salveti de Araujo]

On 04/18/2013 11:07 AM, Sergio Schvezov wrote:

On Thu, Apr 18, 2013 at 10:57:23AM -0300, Ricardo Salveti de Araujo wrote:

On 04/18/2013 09:44 AM, Oliver Grawert wrote:

hi
On Do, 2013-04-18 at 15:32 +0300, Jani Monoses wrote:

Hi,

I notice there are a four ARM cross compilers for linux-x86 in our tree.
Is it because various hw has been only shown to build and work
correctly with a specific one only?

./prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/
./prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6/
./prebuilts/gcc/linux-x86/arm/linaro-4.7.2/
./prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/


We have multiple toolchains but I believe only the linaro one is
being actually used at the build, that's why Oliver pointed that the
build worked when removing the others.


Wait, don't forget about that kernel rule to use 4.6 (IIRC). That should
stay even if we get all the nexus kernels in from packages or think of a
way for porters to get it working, or some extra download in the
cm.dependencies for the device.


Right, indeed, the 4.6 based toolchain is only used by the kernel currently.

Should probably stay as well.

Cheers,
--
Ricardo Salveti de Araujo

--
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Николай Шатохин]

I did this:

Grid

{

id: grid

anchors.centerIn: parent

rows: 4

columns: 4

spacing: __buttonRightMargin

Repeater

{

id: buttons


GameButton

{

row: index % __gameFieldWidth

column: index / __gameFieldWidth

width: __buttonWidth

height: __buttonHeight

}

}

}


onCreateField:

{

grid.rows = fieldHeight;

grid.columns = fieldWidth;

buttons.model = fieldWidth * fieldHeight;


fieldCreated(fieldWidth, fieldHeight);

}


Now, I don't see buttons from old field if I start new game with
smaller field, but it still so slow.



2013/4/16 Николай Шатохин 

> Nope, but very similar by interface
>
>
> 2013/4/16 Roman Shchekin 
>
>> MineSweeper?
>>
>>
>> 2013/4/16 Michael Zanetti 
>>
>>> On Tuesday 16 April 2013 18:54:01 Michael Zanetti wrote:
>>> > As a rule of thumb, you should be able to write more than 90% of your
>>> > application without the need to the work "function" (game logic
>>> excluded).
>>>
>>> That should have been: ... without the need to use the word "function"...
>>>
>>> Regards,
>>> Michael
>>>
>>> > >
>>> > > 2013/4/16 Rick Spencer 
>>> > >
>>> > > > I suppose you would use a loader for the Repeater. After you make
>>> your
>>> > > > repeater work for your first level, you can use QtCreator to wrap
>>> the
>>> > > > Repeater in a loader for you. Then I think for each level you would
>>> > > > destroy the whole Repeater and then use the loader to create a new
>>> one
>>> > > > with the new model for the next level.
>>> > > >
>>> > > > At least that is the approach I would take starting out.
>>> > > >
>>> > > > HTH
>>> > > >
>>> > > > Cheers, Rick
>>> > > >
>>> > > > On Tue, Apr 16, 2013 at 7:31 AM, Николай Шатохин <
>>> n.shatok...@gmail.com>
>>> > > >
>>> > > > wrote:
>>> > > > > When I create button I assign destroy signal to each. So, when I
>>> start
>>> > > >
>>> > > > new
>>> > > >
>>> > > > > game, I destroying buttons from old game and see old field in
>>> this
>>> > > >
>>> > > > moment -
>>> > > >
>>> > > > > destroing slow too. Is quickly removing possible? Can I destroy
>>> > > > > repeater?
>>> > > > >
>>> > > > >
>>> > > > > 2013/4/16 Николай Шатохин 
>>> > > > >
>>> > > > >> Oh, I didn't know that. I'll try.
>>> > > > >>
>>> > > > >>
>>> > > > >> 2013/4/16 Rick Spencer 
>>> > > > >>
>>> > > > >>> Hi.
>>> > > > >>>
>>> > > > >>> I'm not sure how your code flows overall, but if you are
>>> creating
>>> > > > >>> all
>>> > > >
>>> > > > the
>>> > > >
>>> > > > >>> buttons at once, you might consider using a Repeater instead of
>>> > > > >>> doing
>>> > > >
>>> > > > it all
>>> > > >
>>> > > > >>> in javascript.
>>> > > > >>>
>>> > > > >>> It looks like you might have some kind of list that fires a
>>> > > >
>>> > > > createButton
>>> > > >
>>> > > > >>> signal. You could potentially use that list as a model for a
>>> > > > >>> Repeater
>>> > > >
>>> > > > and
>>> > > >
>>> > > > >>> build your buttons in the delegate of the Repeater.
>>> > > > >>>
>>> > > > >>> Cheers, Rick
>>> > > > >>>
>>> > > > >>>
>>> > > > >>>
>>> > > > >>> On Tue, Apr 16, 2013 at 5:11 AM, Николай Шатохин <
>>> > > >
>>> > > > n.shatok...@gmail.com>
>>> > > >
>>> > > > >>> wrote:
>>> > > >  This is my code that creates buttons field:
>>> > > > 
>>> > > > 
>>> > > >  onCreateField:
>>> > > >  {
>>> > > > 
>>> > > >  for(var i=0;i>> > > >  {
>>> > > > 
>>> > > >  for(var j=0;j>> > > >  {
>>> > > > 
>>> > > >  createButton(i, j);
>>> > > > 
>>> > > >  }
>>> > > > 
>>> > > >  }
>>> > > > 
>>> > > >  }
>>> > > > 
>>> > > >  function createButton(x, y)
>>> > > >  {
>>> > > > 
>>> > > >  __buttonX = x;
>>> > > >  __buttonY = y;
>>> > > > 
>>> > > >  __component = Qt.createComponent("GameButton.qml");
>>> > > > 
>>> > > >  if(__component != null)
>>> > > > 
>>> > > >  continueButtonCreation();
>>> > > > 
>>> > > >  else
>>> > > > 
>>> > > >  __component.ready.connect(continueButtonCreation);
>>> > > > 
>>> > > >  }
>>> > > > 
>>> > > >  function continueButtonCreation()
>>> > > >  {
>>> > > > 
>>> > > >  var button = __component.createObject(field, {"row":
>>> __buttonY,
>>> > > > 
>>> > > >  "column": __buttonX});
>>> > > > 
>>> > > >   if (button == null) {
>>> > > > 
>>> > > >   // Error Handling
>>> > > >   console.log("Error creating object");
>>> > > > 
>>> > > >   return;
>>> > > > 
>>> > > >   }
>>> > > > 
>>> > > >   updateValveState.connect(button.stateUpdated);
>>> > > >   button.buttonClicked.connect(buttonClicked);
>>> > > > 

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Michael Zanetti]

Doesn't look too bad... so the slowness must be either in the GameButton 
itself or something connected to the the fieldCreated() signal.

Can you push the whole code to some repository? I would be willing to give it 
a shot and try to find out whats going wrong.

On Thursday 18 April 2013 18:32:53 Николай Шатохин wrote:
> I did this:
> 
> Grid
> 
> {
> 
> id: grid
> 
> anchors.centerIn: parent
> 
> rows: 4
> 
> columns: 4
> 
> spacing: __buttonRightMargin
> 
> Repeater
> 
> {
> 
> id: buttons
> 
> 
> GameButton
> 
> {
> 
> row: index % __gameFieldWidth
> 
> column: index / __gameFieldWidth
> 
> width: __buttonWidth
> 
> height: __buttonHeight
> 
> }
> 
> }
> 
> }
> 
> 
> onCreateField:
> 
> {
> 
> grid.rows = fieldHeight;
> 
> grid.columns = fieldWidth;
> 
> buttons.model = fieldWidth * fieldHeight;
> 
> 
> fieldCreated(fieldWidth, fieldHeight);
> 
> }
> 
> 
> Now, I don't see buttons from old field if I start new game with
> smaller field, but it still so slow.
> 
> 
> 
> 2013/4/16 Николай Шатохин 
> 
> > Nope, but very similar by interface
> > 
> > 
> > 2013/4/16 Roman Shchekin 
> > 
> >> MineSweeper?
> >> 
> >> 
> >> 2013/4/16 Michael Zanetti 
> >> 
> >>> On Tuesday 16 April 2013 18:54:01 Michael Zanetti wrote:
> >>> > As a rule of thumb, you should be able to write more than 90% of your
> >>> > application without the need to the work "function" (game logic
> >>> 
> >>> excluded).
> >>> 
> >>> That should have been: ... without the need to use the word
> >>> "function"...
> >>> 
> >>> Regards,
> >>> Michael
> >>> 
> >>> > > 2013/4/16 Rick Spencer 
> >>> > > 
> >>> > > > I suppose you would use a loader for the Repeater. After you make
> >>> 
> >>> your
> >>> 
> >>> > > > repeater work for your first level, you can use QtCreator to wrap
> >>> 
> >>> the
> >>> 
> >>> > > > Repeater in a loader for you. Then I think for each level you
> >>> > > > would
> >>> > > > destroy the whole Repeater and then use the loader to create a new
> >>> 
> >>> one
> >>> 
> >>> > > > with the new model for the next level.
> >>> > > > 
> >>> > > > At least that is the approach I would take starting out.
> >>> > > > 
> >>> > > > HTH
> >>> > > > 
> >>> > > > Cheers, Rick
> >>> > > > 
> >>> > > > On Tue, Apr 16, 2013 at 7:31 AM, Николай Шатохин <
> >>> 
> >>> n.shatok...@gmail.com>
> >>> 
> >>> > > > wrote:
> >>> > > > > When I create button I assign destroy signal to each. So, when I
> >>> 
> >>> start
> >>> 
> >>> > > > new
> >>> > > > 
> >>> > > > > game, I destroying buttons from old game and see old field in
> >>> 
> >>> this
> >>> 
> >>> > > > moment -
> >>> > > > 
> >>> > > > > destroing slow too. Is quickly removing possible? Can I destroy
> >>> > > > > repeater?
> >>> > > > > 
> >>> > > > > 
> >>> > > > > 2013/4/16 Николай Шатохин 
> >>> > > > > 
> >>> > > > >> Oh, I didn't know that. I'll try.
> >>> > > > >> 
> >>> > > > >> 
> >>> > > > >> 2013/4/16 Rick Spencer 
> >>> > > > >> 
> >>> > > > >>> Hi.
> >>> > > > >>> 
> >>> > > > >>> I'm not sure how your code flows overall, but if you are
> >>> 
> >>> creating
> >>> 
> >>> > > > >>> all
> >>> > > > 
> >>> > > > the
> >>> > > > 
> >>> > > > >>> buttons at once, you might consider using a Repeater instead
> >>> > > > >>> of
> >>> > > > >>> doing
> >>> > > > 
> >>> > > > it all
> >>> > > > 
> >>> > > > >>> in javascript.
> >>> > > > >>> 
> >>> > > > >>> It looks like you might have some kind of list that fires a
> >>> > > > 
> >>> > > > createButton
> >>> > > > 
> >>> > > > >>> signal. You could potentially use that list as a model for a
> >>> > > > >>> Repeater
> >>> > > > 
> >>> > > > and
> >>> > > > 
> >>> > > > >>> build your buttons in the delegate of the Repeater.
> >>> > > > >>> 
> >>> > > > >>> Cheers, Rick
> >>> > > > >>> 
> >>> > > > >>> 
> >>> > > > >>> 
> >>> > > > >>> On Tue, Apr 16, 2013 at 5:11 AM, Николай Шатохин <
> >>> > > > 
> >>> > > > n.shatok...@gmail.com>
> >>> > > > 
> >>> > > > >>> wrote:
> >>> > > >  This is my code that creates buttons field:
> >>> > > >  
> >>> > > >  
> >>> > > >  onCreateField:
> >>> > > >  {
> >>> > > >  
> >>> > > >  for(var i=0;i >>> > > >  {
> >>> > > >  
> >>> > > >  for(var j=0;j >>> > > >  {
> >>> > > >  
> >>> > > >  createButton(i, j);
> >>> > > >  
> >>> > > >  }
> >>> > > >  
> >>> > > >  }
> >>> > > >  
> >>> > > >  }
> >>> > > >  
> >>> > > >  function createButton(x, y)
> >>> > > >  {
> >>> > > >  
> >>> > > >  __buttonX = x;
> >>> > > >  __buttonY = y;
> >>> > > >  
> >>> > > >  __component = Qt.createComponent("GameButton.qml");
> >>> > > >  
> >>> > > >  if(__component != n

Re: [Ubuntu-phone] Building for Encore

[Joey Carlini]

Remember when I said it was alive? It's a little less then alive it turns
out. Flashing it works via CWM Recovery, but upon boot, it turns the
backlight on then off, then kills power to the device. More work is
required...

*%&$#&@!!!*


On Wed, Apr 17, 2013 at 4:53 PM, Joey Carlini  wrote:

> Hey guys, the build is done... :)
>
>
>  
> cm-10.1-20130417-UNOFFICIAL-encore.zip
> Want to flash it on my nook first, and it may make things explode, but
> it's alive!
>
>
> On Wed, Apr 17, 2013 at 3:04 PM, Joey Carlini wrote:
>
>> Mo' Makes, Mo' Problems.
>>
>> http://paste.ubuntu.com/5716685/
>>
>> I think I'm starting to get close though.
>>
>>
>> On Tue, Apr 16, 2013 at 11:00 PM, Joey Carlini wrote:
>>
>>> New Build Error time!
>>>
>>> http://paste.ubuntu.com/5714902/
>>>
>>>
>>> On Tue, Apr 16, 2013 at 10:14 PM, Joey Carlini wrote:
>>>
 Success, sudo apt-get install u-boot-tools adds it to my computer.
 Building again.


 On Tue, Apr 16, 2013 at 9:52 PM, Joey Carlini wrote:

> What worked for me was going into (myworkspace)/build/core/main.mk with
> gedit and putting "system/extras/mkimage" inbetween 
> system/media/audio_utils
> and system/extras/ext4_utils with the backslash thing, and brunch encore
> seemed to work okay until that one bit.  I guess I should dive into the
> Nook Source, find that mkimage file, and just put it where it needs to go.
>
>
> On Tue, Apr 16, 2013 at 9:40 PM, Sergio Schvezov <
> sergio.schve...@canonical.com> wrote:
>
>> On Tue, Apr 16, 2013 at 09:27:10PM -0400, Joey Carlini wrote:
>> > http://paste.ubuntu.com/5714772/
>> >
>> > Uh oh. Ran into this about an hour or so in, cannot make heads or
>> tails of
>> > it. Anyone care to help?
>>
>> So how did you get the build going? If you did get that mkimage from
>> somewhere
>> at least add it to the PATH.
>>
>> If you got mkimage from the actual source tree released from the
>> Nook, add
>> it to the build rules so it gets build and lands in the original
>> out//host/.../mkimage
>>
>> Cheers
>> Sergio
>>
>> --
>> Mailing list: https://launchpad.net/~ubuntu-phone
>> Post to : ubuntu-phone@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>> More help   : https://help.launchpad.net/ListHelp
>>
>
>

>>>
>>
>
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Building for Encore

[Mika Baumeister]

-
Mika Baumeister
Amselweg 6
45721 Haltern am See
01578/5415104
Gesendet von meinem Windows Phone.

Von: Joey Carlini
Gesendet: ‎18.‎04.‎2013 17:58
An: Sergio Schvezov
Cc: ubuntu-phone@lists.launchpad.net
Betreff: Re: [Ubuntu-phone] Building for Encore

Remember when I said it was alive? It's a little less then alive it turns
out. Flashing it works via CWM Recovery, but upon boot, it turns the
backlight on then off, then kills power to the device. More work is
required...

*%&$#&@!!!*


On Wed, Apr 17, 2013 at 4:53 PM, Joey Carlini  wrote:

> Hey guys, the build is done... :)
>
>
>  
> cm-10.1-20130417-UNOFFICIAL-encore.zip
> Want to flash it on my nook first, and it may make things explode, but
> it's alive!
>
>
> On Wed, Apr 17, 2013 at 3:04 PM, Joey Carlini wrote:
>
>> Mo' Makes, Mo' Problems.
>>
>> http://paste.ubuntu.com/5716685/
>>
>> I think I'm starting to get close though.
>>
>>
>> On Tue, Apr 16, 2013 at 11:00 PM, Joey Carlini wrote:
>>
>>> New Build Error time!
>>>
>>> http://paste.ubuntu.com/5714902/
>>>
>>>
>>> On Tue, Apr 16, 2013 at 10:14 PM, Joey Carlini wrote:
>>>
 Success, sudo apt-get install u-boot-tools adds it to my computer.
 Building again.


 On Tue, Apr 16, 2013 at 9:52 PM, Joey Carlini wrote:

> What worked for me was going into (myworkspace)/build/core/main.mk with
> gedit and putting "system/extras/mkimage" inbetween 
> system/media/audio_utils
> and system/extras/ext4_utils with the backslash thing, and brunch encore
> seemed to work okay until that one bit.  I guess I should dive into the
> Nook Source, find that mkimage file, and just put it where it needs to go.
>
>
> On Tue, Apr 16, 2013 at 9:40 PM, Sergio Schvezov <
> sergio.schve...@canonical.com> wrote:
>
>> On Tue, Apr 16, 2013 at 09:27:10PM -0400, Joey Carlini wrote:
>> > http://paste.ubuntu.com/5714772/
>> >
>> > Uh oh. Ran into this about an hour or so in, cannot make heads or
>> tails of
>> > it. Anyone care to help?
>>
>> So how did you get the build going? If you did get that mkimage from
>> somewhere
>> at least add it to the PATH.
>>
>> If you got mkimage from the actual source tree released from the
>> Nook, add
>> it to the build rules so it gets build and lands in the original
>> out//host/.../mkimage
>>
>> Cheers
>> Sergio
>>
>> --
>> Mailing list: https://launchpad.net/~ubuntu-phone
>> Post to : ubuntu-phone@lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>> More help   : https://help.launchpad.net/ListHelp
>>
>
>

>>>
>>
>
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Николай Шатохин]

This is my git repository:

https://bitbucket.org/nshatokhin/valves.git

And how to connect signals to buttons created by Repeater?


2013/4/18 Michael Zanetti 

> Doesn't look too bad... so the slowness must be either in the GameButton
> itself or something connected to the the fieldCreated() signal.
>
> Can you push the whole code to some repository? I would be willing to give
> it
> a shot and try to find out whats going wrong.
>
> On Thursday 18 April 2013 18:32:53 Николай Шатохин wrote:
> > I did this:
> >
> > Grid
> >
> > {
> >
> > id: grid
> >
> > anchors.centerIn: parent
> >
> > rows: 4
> >
> > columns: 4
> >
> > spacing: __buttonRightMargin
> >
> > Repeater
> >
> > {
> >
> > id: buttons
> >
> >
> > GameButton
> >
> > {
> >
> > row: index % __gameFieldWidth
> >
> > column: index / __gameFieldWidth
> >
> > width: __buttonWidth
> >
> > height: __buttonHeight
> >
> > }
> >
> > }
> >
> > }
> >
> >
> > onCreateField:
> >
> > {
> >
> > grid.rows = fieldHeight;
> >
> > grid.columns = fieldWidth;
> >
> > buttons.model = fieldWidth * fieldHeight;
> >
> >
> > fieldCreated(fieldWidth, fieldHeight);
> >
> > }
> >
> >
> > Now, I don't see buttons from old field if I start new game with
> > smaller field, but it still so slow.
> >
> >
> >
> > 2013/4/16 Николай Шатохин 
> >
> > > Nope, but very similar by interface
> > >
> > >
> > > 2013/4/16 Roman Shchekin 
> > >
> > >> MineSweeper?
> > >>
> > >>
> > >> 2013/4/16 Michael Zanetti 
> > >>
> > >>> On Tuesday 16 April 2013 18:54:01 Michael Zanetti wrote:
> > >>> > As a rule of thumb, you should be able to write more than 90% of
> your
> > >>> > application without the need to the work "function" (game logic
> > >>>
> > >>> excluded).
> > >>>
> > >>> That should have been: ... without the need to use the word
> > >>> "function"...
> > >>>
> > >>> Regards,
> > >>> Michael
> > >>>
> > >>> > > 2013/4/16 Rick Spencer 
> > >>> > >
> > >>> > > > I suppose you would use a loader for the Repeater. After you
> make
> > >>>
> > >>> your
> > >>>
> > >>> > > > repeater work for your first level, you can use QtCreator to
> wrap
> > >>>
> > >>> the
> > >>>
> > >>> > > > Repeater in a loader for you. Then I think for each level you
> > >>> > > > would
> > >>> > > > destroy the whole Repeater and then use the loader to create a
> new
> > >>>
> > >>> one
> > >>>
> > >>> > > > with the new model for the next level.
> > >>> > > >
> > >>> > > > At least that is the approach I would take starting out.
> > >>> > > >
> > >>> > > > HTH
> > >>> > > >
> > >>> > > > Cheers, Rick
> > >>> > > >
> > >>> > > > On Tue, Apr 16, 2013 at 7:31 AM, Николай Шатохин <
> > >>>
> > >>> n.shatok...@gmail.com>
> > >>>
> > >>> > > > wrote:
> > >>> > > > > When I create button I assign destroy signal to each. So,
> when I
> > >>>
> > >>> start
> > >>>
> > >>> > > > new
> > >>> > > >
> > >>> > > > > game, I destroying buttons from old game and see old field in
> > >>>
> > >>> this
> > >>>
> > >>> > > > moment -
> > >>> > > >
> > >>> > > > > destroing slow too. Is quickly removing possible? Can I
> destroy
> > >>> > > > > repeater?
> > >>> > > > >
> > >>> > > > >
> > >>> > > > > 2013/4/16 Николай Шатохин 
> > >>> > > > >
> > >>> > > > >> Oh, I didn't know that. I'll try.
> > >>> > > > >>
> > >>> > > > >>
> > >>> > > > >> 2013/4/16 Rick Spencer 
> > >>> > > > >>
> > >>> > > > >>> Hi.
> > >>> > > > >>>
> > >>> > > > >>> I'm not sure how your code flows overall, but if you are
> > >>>
> > >>> creating
> > >>>
> > >>> > > > >>> all
> > >>> > > >
> > >>> > > > the
> > >>> > > >
> > >>> > > > >>> buttons at once, you might consider using a Repeater
> instead
> > >>> > > > >>> of
> > >>> > > > >>> doing
> > >>> > > >
> > >>> > > > it all
> > >>> > > >
> > >>> > > > >>> in javascript.
> > >>> > > > >>>
> > >>> > > > >>> It looks like you might have some kind of list that fires a
> > >>> > > >
> > >>> > > > createButton
> > >>> > > >
> > >>> > > > >>> signal. You could potentially use that list as a model for
> a
> > >>> > > > >>> Repeater
> > >>> > > >
> > >>> > > > and
> > >>> > > >
> > >>> > > > >>> build your buttons in the delegate of the Repeater.
> > >>> > > > >>>
> > >>> > > > >>> Cheers, Rick
> > >>> > > > >>>
> > >>> > > > >>>
> > >>> > > > >>>
> > >>> > > > >>> On Tue, Apr 16, 2013 at 5:11 AM, Николай Шатохин <
> > >>> > > >
> > >>> > > > n.shatok...@gmail.com>
> > >>> > > >
> > >>> > > > >>> wrote:
> > >>> > > >  This is my code that creates buttons field:
> > >>> > > > 
> > >>> > > > 
> > >>> > > >  onCreateField:
> > >>> > > >  {
> > >>> > > > 
> > >>> > > >  for(var i=0;i > >>> > > >  {
> > >>> > > > 
> > >>> > > >  for(var j=0;j > >>> > > >  {
> > >>> > > > 
> > >>> > > >  createButton(i, j);

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Michał Sawicz]

W dniu 18.04.2013 18:28, Николай Шатохин pisze:
> https://bitbucket.org/nshatokhin/valves.git

404

> And how to connect signals to buttons created by Repeater?

Usually:

  onClicked: someObject.someHandler(index)

-- 
Michał Sawicz 
Canonical Services Ltd.



signature.asc
Description: OpenPGP digital signature
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Michał Sawicz]

W dniu 18.04.2013 19:05, Michał Sawicz pisze:
>> https://bitbucket.org/nshatokhin/valves.git
> 404

https://bitbucket.org/nshatokhin/valves

Seems to work.
-- 
Michał Sawicz 
Canonical Services Ltd.



signature.asc
Description: OpenPGP digital signature
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Running QML functions in parallel thread

[Michael Zanetti]

Wow, the slowness really comes from using the Button element from 
Ubuntu.Components.

Inside GameButton.qml, replace Button {} with this:

Image
{
id: button
source: "closed.png"
width: buttonHolder.width
height: buttonHolder.height

MouseArea {
anchors.fill: parent
onClicked:
{
console.log("(" + row + ", " + column + ")")
buttonClicked(row, column)
}
}
}


and it'll be _much_ faster. I tend to say this is an issue in the Button 
component. On the other hand its probably not supposed to be used 256 times on 
one page as it requires lots of theming stuff which you don't really need in 
your game.

What do the SDK people think of this?

Br,
Michael

On Thursday 18 April 2013 19:28:29 Николай Шатохин wrote:
> This is my git repository:
> 
> https://bitbucket.org/nshatokhin/valves.git
> 
> And how to connect signals to buttons created by Repeater?
> 
> 
> 2013/4/18 Michael Zanetti 
> 
> > Doesn't look too bad... so the slowness must be either in the GameButton
> > itself or something connected to the the fieldCreated() signal.
> > 
> > Can you push the whole code to some repository? I would be willing to give
> > it
> > a shot and try to find out whats going wrong.
> > 
> > On Thursday 18 April 2013 18:32:53 Николай Шатохин wrote:
> > > I did this:
> > > Grid
> > > 
> > > {
> > > 
> > > id: grid
> > > 
> > > anchors.centerIn: parent
> > > 
> > > rows: 4
> > > 
> > > columns: 4
> > > 
> > > spacing: __buttonRightMargin
> > > 
> > > Repeater
> > > 
> > > {
> > > 
> > > id: buttons
> > > 
> > > 
> > > GameButton
> > > 
> > > {
> > > 
> > > row: index % __gameFieldWidth
> > > 
> > > column: index / __gameFieldWidth
> > > 
> > > width: __buttonWidth
> > > 
> > > height: __buttonHeight
> > > 
> > > }
> > > 
> > > }
> > > 
> > > }
> > > 
> > > 
> > > onCreateField:
> > > 
> > > {
> > > 
> > > grid.rows = fieldHeight;
> > > 
> > > grid.columns = fieldWidth;
> > > 
> > > buttons.model = fieldWidth * fieldHeight;
> > > 
> > > 
> > > fieldCreated(fieldWidth, fieldHeight);
> > > 
> > > }
> > > 
> > > Now, I don't see buttons from old field if I start new game with
> > > smaller field, but it still so slow.
> > > 
> > > 
> > > 
> > > 2013/4/16 Николай Шатохин 
> > > 
> > > > Nope, but very similar by interface
> > > > 
> > > > 
> > > > 2013/4/16 Roman Shchekin 
> > > > 
> > > >> MineSweeper?
> > > >> 
> > > >> 
> > > >> 2013/4/16 Michael Zanetti 
> > > >> 
> > > >>> On Tuesday 16 April 2013 18:54:01 Michael Zanetti wrote:
> > > >>> > As a rule of thumb, you should be able to write more than 90% of
> > 
> > your
> > 
> > > >>> > application without the need to the work "function" (game logic
> > > >>> 
> > > >>> excluded).
> > > >>> 
> > > >>> That should have been: ... without the need to use the word
> > > >>> "function"...
> > > >>> 
> > > >>> Regards,
> > > >>> Michael
> > > >>> 
> > > >>> > > 2013/4/16 Rick Spencer 
> > > >>> > > 
> > > >>> > > > I suppose you would use a loader for the Repeater. After you
> > 
> > make
> > 
> > > >>> your
> > > >>> 
> > > >>> > > > repeater work for your first level, you can use QtCreator to
> > 
> > wrap
> > 
> > > >>> the
> > > >>> 
> > > >>> > > > Repeater in a loader for you. Then I think for each level you
> > > >>> > > > would
> > > >>> > > > destroy the whole Repeater and then use the loader to create a
> > 
> > new
> > 
> > > >>> one
> > > >>> 
> > > >>> > > > with the new model for the next level.
> > > >>> > > > 
> > > >>> > > > At least that is the approach I would take starting out.
> > > >>> > > > 
> > > >>> > > > HTH
> > > >>> > > > 
> > > >>> > > > Cheers, Rick
> > > >>> > > > 
> > > >>> > > > On Tue, Apr 16, 2013 at 7:31 AM, Николай Шатохин <
> > > >>> 
> > > >>> n.shatok...@gmail.com>
> > > >>> 
> > > >>> > > > wrote:
> > > >>> > > > > When I create button I assign destroy signal to each. So,
> > 
> > when I
> > 
> > > >>> start
> > > >>> 
> > > >>> > > > new
> > > >>> > > > 
> > > >>> > > > > game, I destroying buttons from old game and see old field
> > > >>> > > > > in
> > > >>> 
> > > >>> this
> > > >>> 
> > > >>> > > > moment -
> > > >>> > > > 
> > > >>> > > > > destroing slow too. Is quickly removing possible? Can I
> > 
> > destroy
> > 
> > > >>> > > > > repeater?
> > > >>> > > > > 
> > > >>> > > > > 
> > > >>> > > > > 2013/4/16 Николай Шатохин 
> > > >>> > > > > 
> > > >>> > > > >> Oh, I didn't know that. I'll try.
> > > >>> > > > >> 
> > > >>> > > > >> 
> > > >>

[Ubuntu-phone] Remove TI/panda support

[Jani Monoses]

device/ti/panda seems not to be needed by our official devices, this
patch removes it from the default manifest


0001-Remove-device-ti-panda.patch
Description: Binary data
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Fwd: Remove TI/panda support

[Mathieu Trudel-Lapierre]

That was meant for the mailing list...

Mathieu Trudel-Lapierre 
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93


-- Forwarded message --
From: Mathieu Trudel-Lapierre 
Date: Thu, Apr 18, 2013 at 4:27 PM
Subject: Re: [Ubuntu-phone] Remove TI/panda support


On Thu, Apr 18, 2013 at 2:15 PM, Jani Monoses  wrote:

> device/ti/panda seems not to be needed by our official devices, this
> patch removes it from the default manifest
>

What do we benefit from removing it?

Seems to me like it doesn't impact anything by being there, but rather
allows users to build for that device should they choose to.

Mathieu Trudel-Lapierre 
Freenode: cyphermox, Jabber: mathieu...@gmail.com
4096R/EE018C93 1967 8F7D 03A1 8F38 732E  FF82 C126 33E1 EE01 8C93
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

[Ubuntu-phone] Testing "raring" Ubuntu Touch Images

[Nicholas Skaggs]

Raring based ubuntu touch images can now be found on the isotracker 
(http://iso.qa.ubuntu.com) for each of the 4 official builds.
You can find them under the raring daily milestone; 
http://iso.qa.ubuntu.com/qatracker/milestones/243/builds


Ubuntu Touch Preinstalled grouper
Ubuntu Touch Preinstalled maguro
Ubuntu Touch Preinstalled mako
Ubuntu Touch Preinstalled manta

These images are intended to replace the current quantal based images. 
As such, your help in testing and reporting any issues found (especially 
between the current quantal based images and the new ones) is most 
appreciated! If you want to try out the new images;you can do the following;


/sudo add-apt-repository ppa:phablet-team/tools//
//sudo apt-get update && sudo apt-get install phablet-tools//
//wget http://people.canonical.com/~sergiusens/raring_settings.py//
//phablet-flash --alternate-settings raring_settings.py

/Then, go to the isotracker and execute the testcases for the image. 
With a ubuntu sso account, you can login and add your pass or fail 
result as well. If you run into any issues, report a bug and note it in 
your result. The wiki has help on reporting bugs; 
https://wiki.ubuntu.com/Touch/Contribute#How_to_report_bugs if needed.


If you've never used the qatracker, this link will help you understand 
how to report results:

https://wiki.ubuntu.com/Testing/QATracker

More details here: 
http://www.theorangenotebook.com/2013/04/testing-ubunutu-touch-images-now-with.html


Thanks everyone!

Nicholas
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp

Re: [Ubuntu-phone] Building for Encore

[Joey Carlini]

Anyway, here's my latest brunch records if you want to take a look.

http://paste.ubuntu.com/5719982/


On Thu, Apr 18, 2013 at 11:58 AM, Joey Carlini  wrote:

> Remember when I said it was alive? It's a little less then alive it turns
> out. Flashing it works via CWM Recovery, but upon boot, it turns the
> backlight on then off, then kills power to the device. More work is
> required...
>
> *%&$#&@!!!*
>
>
> On Wed, Apr 17, 2013 at 4:53 PM, Joey Carlini wrote:
>
>> Hey guys, the build is done... :)
>>
>>
>>  
>> cm-10.1-20130417-UNOFFICIAL-encore.zip
>> Want to flash it on my nook first, and it may make things explode, but
>> it's alive!
>>
>>
>> On Wed, Apr 17, 2013 at 3:04 PM, Joey Carlini wrote:
>>
>>> Mo' Makes, Mo' Problems.
>>>
>>> http://paste.ubuntu.com/5716685/
>>>
>>> I think I'm starting to get close though.
>>>
>>>
>>> On Tue, Apr 16, 2013 at 11:00 PM, Joey Carlini wrote:
>>>
 New Build Error time!

 http://paste.ubuntu.com/5714902/


 On Tue, Apr 16, 2013 at 10:14 PM, Joey Carlini wrote:

> Success, sudo apt-get install u-boot-tools adds it to my computer.
> Building again.
>
>
> On Tue, Apr 16, 2013 at 9:52 PM, Joey Carlini wrote:
>
>> What worked for me was going into (myworkspace)/build/core/main.mk with
>> gedit and putting "system/extras/mkimage" inbetween 
>> system/media/audio_utils
>> and system/extras/ext4_utils with the backslash thing, and brunch encore
>> seemed to work okay until that one bit.  I guess I should dive into the
>> Nook Source, find that mkimage file, and just put it where it needs to 
>> go.
>>
>>
>> On Tue, Apr 16, 2013 at 9:40 PM, Sergio Schvezov <
>> sergio.schve...@canonical.com> wrote:
>>
>>> On Tue, Apr 16, 2013 at 09:27:10PM -0400, Joey Carlini wrote:
>>> > http://paste.ubuntu.com/5714772/
>>> >
>>> > Uh oh. Ran into this about an hour or so in, cannot make heads or
>>> tails of
>>> > it. Anyone care to help?
>>>
>>> So how did you get the build going? If you did get that mkimage from
>>> somewhere
>>> at least add it to the PATH.
>>>
>>> If you got mkimage from the actual source tree released from the
>>> Nook, add
>>> it to the build rules so it gets build and lands in the original
>>> out//host/.../mkimage
>>>
>>> Cheers
>>> Sergio
>>>
>>> --
>>> Mailing list: https://launchpad.net/~ubuntu-phone
>>> Post to : ubuntu-phone@lists.launchpad.net
>>> Unsubscribe : https://launchpad.net/~ubuntu-phone
>>> More help   : https://help.launchpad.net/ListHelp
>>>
>>
>>
>

>>>
>>
>
-- 
Mailing list: https://launchpad.net/~ubuntu-phone
Post to : ubuntu-phone@lists.launchpad.net
Unsubscribe : https://launchpad.net/~ubuntu-phone
More help   : https://help.launchpad.net/ListHelp