[android-porting] Re: Audio stops after some time in android from video/Music player
I've done following change in android_audio_output.cpp. It is better
after enabling RESET mode instead of STOP. Is there any fix for PV
core which mentioned in the comments of code snippet below??
# if 0 //SAMPATH commented
// FIXME: Per PV, we should be exiting thread in Reset, not Stop.
However, doing so
// causes app death, so presumably PV is asserting somewhere and
causing an abort.
// When this gets fixed, remove this Stop function and bring back the
Reset function.
PVMFCommandId AndroidAudioOutput::Stop(const OsclAny* aContext)
{
// request output thread to exit
LOGV(Stop (%p), aContext);
RequestAndWaitForThreadExit();
return AndroidAudioMIO::Stop(aContext);
}
#endif
#if 1 //SAMPATH enabled
PVMFCommandId AndroidAudioOutput::Reset(const OsclAny* aContext)
{
// request output thread to exit
LOGV(Reset (%p), aContext);
RequestAndWaitForThreadExit();
return AndroidAudioMIO::Reset(aContext);
}
#endif
On Feb 17, 10:40 am, getandroid sampath...@gmail.com wrote:
I'm working on android 1.0 release
On Feb 16, 9:28 pm, Dave Sparks davidspa...@android.com wrote:
Which branch are you working on?
On Feb 16, 4:15 am, getandroid sampath...@gmail.com wrote:
Hi,
I'm facing a problem with Video/Music player where audio stops
coming after a random number of times. After some debugging I found
that:
In android_audio_output.cpp-audout_thread_func, there is
iAudioThreadSem-Wait(); before the while(1) loop and it is stuck
there indefinitely. I understand it is waiting for a signal from MIO
saying that the buffer is written and ready to be dumped to audio
flinger. For some reason it is not getting the signal. Is this
correct?
Can anybody throw some light on this?
--~--~-~--~~~---~--~~
unsubscribe: android-porting+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-porting
-~--~~~~--~~--~--~---
[android-porting] Re: Removing external components in Android Build
This is not supposed to happen. What kind of change are you talking about ? Sounds like a build system bug. You should not have to rebuild Webkit if you change something that is unrelated. For the record, you can still use standard GNU Make -B option to mm or mmm to force a rebuild without having to do a 'make clean' On Tue, Feb 17, 2009 at 8:50 AM, AJAY ajay...@gmail.com wrote: Hi, Is there anyway I can exclude certain external libraries/components which I am not testing like webkit during the Android build. I want to speed up my build process during development. I see that making changes in build/target/product/core.mk also has no effect. This is because I see that after making certain changes to the code, say PV omx components , the change doesnt get reflected unless I do a make clean. --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: timer_create problem
On Mon, Feb 16, 2009 at 8:48 PM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: That's great. I found similar problem for SIGEV_SIGNAL. Could you please let me know when the fix is released? I think I nailed that one too. All fixes are in bionic/libc/bionic/pthread-timers.c, essentially: - in timer_settime() replace the call to __timer_gettime() to __timer_settime() with proper parameters (silly typo) - in the worker thread implementation, a subtraction reversal when computing the timeout (should be expires - now instead of now - expires) Thanks! On Feb 16, 1:49 am, David Turner di...@android.com wrote: ok, I found the bug in the timer implementation, a fix is coming soon. On Mon, Feb 16, 2009 at 10:27 AM, David Turner di...@android.com wrote: this one's for me :-) Can you send me a small compilable test program to reproduce this (it'd be much better for me since I don't know what your timer_handler function is doing there). It might be a bug in the SIGEV_THREAD timer implementation, I checked and tested it for correctness but didn't look at CPU usage, I admit. By the way, sigev_signo isn't used with SIGEV_THREAD timers. On Mon, Feb 16, 2009 at 4:13 AM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: hi, i played with timer_create routine with the following code. it seems the timer is working, but the cpu utilization is really high(over 90%) if i run top. Any idea what's going on on G1? Thanks! - se.sigev_notify = SIGEV_THREAD; se.sigev_notify_function = timer_handler; se.sigev_signo = SIGUSR1; timer_create(CLOCK_REALTIME, se, tid); memset(ts, 0, sizeof(struct itimerspec)); ts.it_value.tv_sec = 0; ts.it_value.tv_nsec = 3000; ts.it_interval.tv_sec = 0; ts.it_interval.tv_nsec = 3000; result = timer_settime(tid, 0, ts, 0); --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] How does arm11_clock_init() works
I am analyzing the bootloader code of Android. My understanding is
that the int _main(void) function of nandwrite.c is the bootloader
running in flash mode. Is that right?
And I see the following arm11_clock_init() function which is called at
the beginning at the _main() function is very interesting. It looks
like it writes some magic numbers to some memory addresses from the
tbl. What is the function really do? Does it related with the ARM cpu?
Is there any docs about these magic number and memory addresses?
Please help. Thanks!
#define A11S_CLK_CNTL 0xC0100100
#define A11S_CLK_SEL 0xC0100104
#define C A11S_CLK_CNTL
#define S A11S_CLK_SEL
static unsigned tbl[] = {
#if EXPLORE
C, 0x64, S, 2,
C, 0x640001, S, 3,
C, 0x640201, S, 2,
C, 0x640203, S, 3,
C, 0x640403, S, 2,
C, 0x640405, S, 3,
C, 0x640605, S, 2,
C, 0x640607, S, 3,
C, 0x640807, S, 2,
C, 0x640809, S, 3,
C, 0x640A09, S, 2,
C, 0x640A0B, S, 3,
C, 0x640C0B, S, 2,
C, 0x640C0D, S, 3,
C, 0x640E0D, S, 2,
C, 0x640E0F, S, 3,
#endif
C, 0x64, S, 2,
C, 0x64001F, S, 3,
C, 0x64171F, S, 2,
C, 0x641715, S, 5,
C, 0x641315, S, 6,
C, 0x641312, S, 7,
C, 0x641112, S, 6,
0
};
void arm11_clock_init(void)
{
unsigned *x = tbl;
while(*x) {
unsigned *ptr = (unsigned*) *x++;
unsigned val = *x++;
*ptr = val;
}
}
--~--~-~--~~~---~--~~
unsubscribe: android-porting+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-porting
-~--~~~~--~~--~--~---
[android-porting] Re: timer_create problem
Hi David, Can this bug be updated in both cupcake and open source ? Can you send the quick fix for a solution temporarily . May be u can attach ! Regards Girish --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: Android Emulator, Working altogether with ARM and DSP ?
Hi All, As for that Android's Emulator based on QEMU, if somebody want to emulate ARM together with DSP (like TI's TMS320C5000/C6000 series) by Android at the same time, then how should the next step be implemented upon QEMU? I would really appreciate if you can leave some hints about that!! Best Regards, Samuel On Feb 14, 2:04 pm, Samuel samuel.om...@gmail.com wrote: Hi, Guys, The current Android's Emulator residing in QEMU can emulate the basic architectures, such asARM-core based platform, or X86-based, also including of course other architectures!! But, here's a problem that i want to further learn, that whether it has the ability of imitating dual-core platform simultaneously, like based onARMandDSP, by modifying the implementation of the original QEMU architecture!! Or have there been something that supports this way, that is, the dual-core emulation in the open community or proprietary products? Any suggestions are really appreciated!! --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: Android porting on Qemu
Yes David, I was thinking of making Android system running on QEMU. Like you said, HW changes have been made to make Android run. So it seems like there is no way, It can be done on QEMU or any of its derivative platform. On Tue, Feb 17, 2009 at 4:06 PM, David Turner di...@android.com wrote: Indra, I still don't have a clue as to what you really want to do. The terms you're using are rather vague, it would be very helpful if you could describe what you intend to do in more practical terms, as in: - I want to port Android to run on a stock Linux distribution (won't happen without kernel changes) - I want to run the Android system on a stock QEMU (won't happen for lack of proper HW support) - I want to look at the source code and modify it and run my modifications (you can already do that from the open-source tree, after a build, just type emulator to launch your new system image) etc... On Tue, Feb 17, 2009 at 10:14 AM, indra dutt indrad...@gmail.com wrote: Hi David, Thanks for your notes. Definitely that clarifies a lots of things. It means we can not build emulator the way I was taking it, but still I am interested in putting the things together. Do you think that can be done? If that is so, what procedures do I have to follow? Thanks for all your help!! On Tue, Feb 17, 2009 at 1:34 PM, David Turner di...@android.com wrote: I fear you don't understand much how Android works, but I'll try to give a few hints: Android requires its own kernel, because it includes a few drivers that are specific to the platform (e.g. the Binder driver used to implement inter-process communication) or specific configuration settings. You thus can't run Android properly on a stock Linux kernel, even one compiled for ARM. For the record, Android currently uses 2.6.27 and there are chances that trying to use 2.6.24 is not going to work at all, even if you try to integrate Android-specific changes into it. The Android kernel sources are available from android.kernel.org. This includes a virtual platform named goldfish corresponding to an ARM-based virtual machine that can be run in the Android emulator, which is a derivative of QEMU that includes goldfish hardware emulation. In other words you cannot use a stock QEMU to run this kernel because it doesn't include the necessary hw support. Apart from that, the SDK contains the emulator plus some files in the tools/lib/images directory that are: - kernel-qemu: a prebuilt image of the Android kernel built for the goldfish platform (ARM-based) - ramdisk.img: the ramdisk image used to boot the system (which includes Android-specific /init and config files) - system.img: a YAFFS2 image mounted as / when the emulated system starts - userdata.img: another YAFFS2 image mounted as /data when the emulated system starts. Actually, system.img and userdata.img are not mounted directly, they are copied into either a temporary file or one in ~/.android/userdata-qemu.img when the emulator starts up, so should only be considered as initial version of the corresponding filesystems. These already contain an ARM-based Android system, including an ARM-based Dalvik. You can download the Android SDK from this page: http://developer.android.com/sdk/1.1_r1/index.html The emulator sources are available here too: http://code.google.com/p/android/downloads/list (note: the emulator in the 1.1 SDK is the same one than in the 1.0r2 SDK) The sources of most of the Android platform (i.e. everything that has been open-sourced) are available from: source.android.com I still don't understand what you mean by reverse-engineering, but so be it. Hope this helps On Tue, Feb 17, 2009 at 7:35 AM, indra dutt indrad...@gmail.com wrote: Ok. perhaps let me put it like, Suppose I do not have android emulator and I want to make that, We know on Qemu there is Linux 2.6.23 and on top of that there are android images and Goldfish FS, then there is Dalvik VM ported. Does that make sense? Any help?? On Tue, Feb 17, 2009 at 11:01 AM, Avtar Singh s.av...@gmail.comwrote: On my windows environment, I am installing 2.6.24 kernel, and on that I plan to port Middleware ( Android file-system, ARM file-simulation), on then Dalvik VM, finally I wish to run any android application on that. I am sure you now have clue what I intend to do. Absolutely no clue. I am planning to make my own android emulator and to play with it later. I am seeking help on that. Do you plan to use/build on QEMU source or not? Have you tried installing Android source code and looking at its emulator code? -- Thanks -Indra -- Thanks -Indra -- Thanks -Indra --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: logcat can't display all log?
Ah, I forgot that you are doing the increment inside the call. This
could be a classical macro expansion issue. I think all your outputs are
happening, but the increment can be expanded in several places inside
the macro. Try taking the ++c outside of the call:
++c;
LOGI(...);
forest wrote:
add \n has no effect
On 2月17日, 上午11时36分, Sean McNeil seanmcne...@gmail.com wrote:
Maybe try a \n at the end of the string?
forest wrote:
for example:
void fun()
{
static int c = 0;
LOGI(call func() %d, ++c);
}
the log sometimes display
call func() 1
call func() 5
call func() 8
..
when debug PVMFStatus AndroidAudioInput::DoRead() int the file
external/opencore/android/author/android_audio_input.cpp
static int AndroidAudioInputDoRead = 0;
LOGI(--AndroidAudioInputDoRead=%d, +
+AndroidAudioInputDoRead);
find above case- 隐藏被引用文字 -
- 显示引用的文字 -
--~--~-~--~~~---~--~~
unsubscribe: android-porting+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-porting
-~--~~~~--~~--~--~---
[android-porting] Re: logcat can't display all log?
You can try to launch logcat like below:adb logcat -r 8000 -f
/data/log.txt
Perhaps it works.
On Tue, Feb 17, 2009 at 9:25 PM, Sean McNeil seanmcne...@gmail.com wrote:
Ah, I forgot that you are doing the increment inside the call. This
could be a classical macro expansion issue. I think all your outputs are
happening, but the increment can be expanded in several places inside
the macro. Try taking the ++c outside of the call:
++c;
LOGI(...);
forest wrote:
add \n has no effect
On 2月17日, 上午11时36分, Sean McNeil seanmcne...@gmail.com wrote:
Maybe try a \n at the end of the string?
forest wrote:
for example:
void fun()
{
static int c = 0;
LOGI(call func() %d, ++c);
}
the log sometimes display
call func() 1
call func() 5
call func() 8
..
when debug PVMFStatus AndroidAudioInput::DoRead() int the file
external/opencore/android/author/android_audio_input.cpp
static int AndroidAudioInputDoRead = 0;
LOGI(--AndroidAudioInputDoRead=%d, +
+AndroidAudioInputDoRead);
find above case- 隐藏被引用文字 -
- 显示引用的文字 -
--~--~-~--~~~---~--~~
unsubscribe: android-porting+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-porting
-~--~~~~--~~--~--~---
[android-porting] Re: Camera Object in Video Capture Case
We cannot allow the application to make any changes to the camera that could potentially violate the contract between the camera and the media recorder. For example, let's say that the video frame size is set to CIF, and the application changes it to QCIF in the middle of a recording. This will cause the encoder to read beyond the end of the frame and probably crash the media server process. In a future release, we will enable a subset of safe API's that the application can use while video recording is in progress. On Feb 17, 6:20 am, steve2641 steve2...@gmail.com wrote: Hi, In the cupcake baseline and beyond, the potential exists for an application to create and configure a Camera object and then pass that object to the media recorder to be used to deliver preview frames to be encoded into a video file. This is fine except for the fact that the media recorder does a reconnect to the Camera object, thus seemingly disconnecting the application from the camera object. Does this reconnect prevent the application from any further interaction with the camera device? I'm pretty sure, but please correct me if I'm wrong, that at a minimum the reconnect will prevent the application from being able to receive any callbacks that may be of interest. For instance, say an implementation adds a new capability to the camera interface which includes a callback to denote completion of a requested task. When the camera object is passed off to the media recorder the Application seems to be losing access to these expanded features that maybe desired during the capture of the video. We have such an enhancement planned, but I don't feel comfortable disclosing the details in a public forum. It would seem more appropriate, at least in my case, that the media recorder reconnect be changed to a very specific request frames call which would not take over all of the potential callbacks that could occur, but simply tell Camera Services that the video engine requests the preview frames. With this Camera Services could continue to service the application as it does when no video record use case is running. The addition would be that Camera Services would also have an additional client to pass frames to, the video engine. When Camera Services receives a preview frame from HAL it would pass it to the Surface Flinger, potentially pass it to the application (when requested), and potentially send it to the video engine (also, when requested). In the case where the application does not provide a Camera object the media recorder would still have to create it's own camera object. That's a hook that seems very clean and clearly needed. Comments? Thanks, Steve. --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] How to change the screen resolution.
Hi Sir, I'm porting Android to our hardware with 320x240 LCD. The Android resolution is too high, so that the icons and messages on the desktop is too small to be recognized. How can I configure the resolution of screen to proper size? Thank you for your great help. Best Regards, YJ --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Non-Ti Wifi driver
Hi I'm porting Android to the Freescale iMX31 PDK. On it there is a CSR Unifi-based APM 6628 wifi module. This module / chip's driver is propriety, and you need a few legal documents and NDA's to have the source of it. But hey, I've got mine running! Then there's the question, how does Android handle non-standard Wifi drivers? If you look as /sys/class/net the Unifi chip registers it as eth0 (in the source its behaviour is explained that that Red Hat 9's - eek! - network config doesn't pick up wlan0 devices but only eth0s). The device node /dev/eth0 is not created (nor /dev/wlan0). The char device /dev/unifi0 and /dev/unifudi0 is created. I've enabled WPA_SUPPLICANT by adding HAVE_CUSTOM_WIFI_DRIVER_0 := true to it. Any pointers as how to get Android pick this device up? --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: How to change the screen resolution.
320x240 screens are not supported in the current platform, thought this is on the roadmap. You can play around with such resolutions by adjusting the density appropriately... however, if your problems is that the UI shown on that screen is too -small- rather than being too large, then I don't think we will be supporting such a screen officially in the platform (with application compatibility and such) any time soon. Both reducing the screen resolution and increasing the screen density is going to leave so little space that most existing apps simply won't work. :} On Tue, Feb 17, 2009 at 8:32 AM, cack dearc...@hotmail.com wrote: Hi Sir, I'm porting Android to our hardware with 320x240 LCD. The Android resolution is too high, so that the icons and messages on the desktop is too small to be recognized. How can I configure the resolution of screen to proper size? Thank you for your great help. Best Regards, YJ -- Dianne Hackborn Android framework engineer hack...@android.com Note: please don't send private questions to me, as I don't have time to provide private support. All such questions should be posted on public forums, where I and others can see and answer them. --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: Porting to PPC (MPC7xxx)
I am also interested to stat porting on ppc. On Tue, Feb 17, 2009 at 5:43 AM, Androidphan niels.kee...@tass.nl wrote: I am very interested as well. On Feb 16, 12:50 am, Karthik manian.kart...@gmail.com wrote: Hi, I'm trying to port the android to PPC (sayMPC7xxx). Please guide me in proceeding. Also let me know any effort is spent on porting the same. Detailed approach for the porting is very much appriciated. Regards, Karthik.S -- Thanks Rizavan --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] bluetooth cpu utilization
When using bluetooth (bluez) to play audio mp3 file on G1, i observed that cpu utilization(from top command) is very high. The mediaserver process takes 30% of cpu. Bluez should also consume significant amount of cpu since the l2cap is in the kernel. The cpu frequency is 384Mhz, powerful enough. I'm wondering if this 30% is accurate or not. Is it too high? Also can anybody clarify any better tool to test cpu utilization? and how to measure the cpu utilization for l2cap kernel socket? Thanks! --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: bluetooth cpu utilization
Yes we have CPU usage of around 30-40% by mediaserver when playing back MP3 over Bluetooth. Note that we do SBC encoding in the mediaserver. What kind of numbers were you expecting? The cpu usage in the kernel side shouldn't be too bad. It just slaps on some L2CAP headers, and the uart transfer uses DMA. Nick On Tue, Feb 17, 2009 at 2:57 PM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: When using bluetooth (bluez) to play audio mp3 file on G1, i observed that cpu utilization(from top command) is very high. The mediaserver process takes 30% of cpu. Bluez should also consume significant amount of cpu since the l2cap is in the kernel. The cpu frequency is 384Mhz, powerful enough. I'm wondering if this 30% is accurate or not. Is it too high? Also can anybody clarify any better tool to test cpu utilization? and how to measure the cpu utilization for l2cap kernel socket? Thanks! --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: a1200 porting issue, mismatch uid, binder cannot be permitted, pls help
could you analysis my startup log, now i am porting to a1200 phone. now i could see the desktop, but the windowsapplication error: com.google.googleappls show, so could you tell me what happen. another question: why there are so many permission problems during nfs root file system. --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: bluetooth cpu utilization
Just wondering if the 30~40% cpu utilization was considered in the android design. This is a large number if other applications could run simultaneously. Do you have tools(more accurate than top) that could be used to test cpu utilization? Thanks! On Feb 17, 3:38 pm, Nick Pelly npe...@google.com wrote: Yes we have CPU usage of around 30-40% by mediaserver when playing back MP3 over Bluetooth. Note that we do SBC encoding in the mediaserver. What kind of numbers were you expecting? The cpu usage in the kernel side shouldn't be too bad. It just slaps on some L2CAP headers, and the uart transfer uses DMA. Nick On Tue, Feb 17, 2009 at 2:57 PM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: When using bluetooth (bluez) to play audio mp3 file on G1, i observed that cpu utilization(from top command) is very high. The mediaserver process takes 30% of cpu. Bluez should also consume significant amount of cpu since the l2cap is in the kernel. The cpu frequency is 384Mhz, powerful enough. I'm wondering if this 30% is accurate or not. Is it too high? Also can anybody clarify any better tool to test cpu utilization? and how to measure the cpu utilization for l2cap kernel socket? Thanks! --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: bluetooth cpu utilization
On Tue, Feb 17, 2009 at 4:59 PM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: Just wondering if the 30~40% cpu utilization was considered in the android design. This is a large number if other applications could run simultaneously. Do you have tools(more accurate than top) that could be used to test cpu utilization? schedtop Thanks! On Feb 17, 3:38 pm, Nick Pelly npe...@google.com wrote: Yes we have CPU usage of around 30-40% by mediaserver when playing back MP3 over Bluetooth. Note that we do SBC encoding in the mediaserver. What kind of numbers were you expecting? The cpu usage in the kernel side shouldn't be too bad. It just slaps on some L2CAP headers, and the uart transfer uses DMA. Nick On Tue, Feb 17, 2009 at 2:57 PM, henry.lon...@gmail.com henry.lon...@gmail.com wrote: When using bluetooth (bluez) to play audio mp3 file on G1, i observed that cpu utilization(from top command) is very high. The mediaserver process takes 30% of cpu. Bluez should also consume significant amount of cpu since the l2cap is in the kernel. The cpu frequency is 384Mhz, powerful enough. I'm wondering if this 30% is accurate or not. Is it too high? Also can anybody clarify any better tool to test cpu utilization? and how to measure the cpu utilization for l2cap kernel socket? Thanks! --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: logcat can't display all log?
logcat -r 8000 -f /data/log.txt
can record all log,Freepine,thank you very much,
hi,Sean McNeil,thank you too,
++c;
LOGI(...);
has no effect
On 2月17日, 下午10时02分, Freepine freep...@gmail.com wrote:
You can try to launch logcat like below:adb logcat -r 8000 -f
/data/log.txt
Perhaps it works.
On Tue, Feb 17, 2009 at 9:25 PM, Sean McNeil seanmcne...@gmail.com wrote:
Ah, I forgot that you are doing the increment inside the call. This
could be a classical macro expansion issue. I think all your outputs are
happening, but the increment can be expanded in several places inside
the macro. Try taking the ++c outside of the call:
++c;
LOGI(...);
forest wrote:
add \n has no effect
On 2月17日, 上午11时36分, Sean McNeil seanmcne...@gmail.com wrote:
Maybe try a \n at the end of the string?
forest wrote:
for example:
void fun()
{
static int c = 0;
LOGI(call func() %d, ++c);
}
the log sometimes display
call func() 1
call func() 5
call func() 8
..
when debug PVMFStatus AndroidAudioInput::DoRead() int the file
external/opencore/android/author/android_audio_input.cpp
static int AndroidAudioInputDoRead = 0;
LOGI(--AndroidAudioInputDoRead=%d, +
+AndroidAudioInputDoRead);
find above case- 隐藏被引用文字 -
- 显示引用的文字 -- 隐藏被引用文字 -
- 显示引用的文字 -
--~--~-~--~~~---~--~~
unsubscribe: android-porting+unsubscr...@googlegroups.com
website: http://groups.google.com/group/android-porting
-~--~~~~--~~--~--~---
[android-porting] Re: How to develop more than 20 timer parallely..
Hi One more Q for you Yes, it is, but frankly you should be using a single SIGEV_THREAD timer with your own sorted timer list Do that means SIGEV_THREAD functionality will put the timers in one event list and sort them all together ? and timeouts will happen similar to POSIX signals ? or user has to sort them in order ? Regards Girish On Feb 12, 6:39 pm, David Turner di...@android.com wrote: Yes, it is, but frankly you should be using a single SIGEV_THREAD timer with your own sorted timer list On Thu, Feb 12, 2009 at 2:14 PM, Girish htgir...@gmail.com wrote: Hi, Is SIGEV_THREAD is part of open source and cupcake code now ? With SIGEV_THREAD also i guess we should be able to create 32 timers per- process right ? Can any one point me out using threads over signals for timer operation. Regards Girish --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: Android porting on Qemu
Thanks Avtar, It would be really great if you can forward any e-book on that. I thought I asked a very specific question as you can see my subject line too Android porting on Qemu. Since I was not aware, what it takes to achieve that, I posted this query. David really gave a good insight on that. Thanks a ton for that. On Tue, Feb 17, 2009 at 6:21 PM, Avtar Singh s.av...@gmail.com wrote: Yes David, I was thinking of making Android system running on QEMU. Like you said, HW changes have been made to make Android run. So it seems like there is no way, It can be done on QEMU or any of its derivative platform. I guess you should try to setup Android and run it in emulator. You will also become more aware of how Android is structured and would be able to ask specific questions then. -- Thanks -Indra --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: How to change the screen resolution.
If the issue is that the icons are tiny (say 3 pixels by 3 pixels wide) then you need to adjust the density/dpi returned by your framebuffer kernel driver (/dev/fb0) to the Android lower layer. Regards, Peter On Feb 18, 2009, at 12:32 AM, cack wrote: Hi Sir, I'm porting Android to our hardware with 320x240 LCD. The Android resolution is too high, so that the icons and messages on the desktop is too small to be recognized. How can I configure the resolution of screen to proper size? Thank you for your great help. Best Regards, YJ --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
[android-porting] Re: How to determine requirements for porting Android?
Plz check my reply.. On Mon, Feb 16, 2009 at 5:30 AM, ppmoore polom...@gmail.com wrote: [ I have also cross-posted this to androidcommunity.com ] Hello, I work for a portable entertainment products manufacturer, and we are looking at porting Android to one of our handheld audio-visual devices. My initial task is to determine if Android can be ported to the device. For this , you have to check your device hardware it in terms of CPU, if its other than ARM, i.e MIPS etc, you have to start from scratch to port Android on it. I've been browsing through the various Google-sponsored Android sites, looking for hardware porting and requirements information, and haven't been able to find anything. From reading the postings on this group, the lack of a Porting Guide document seems to be accepted. Our device is primarily an entertainment device, so we will need to support the various audio, still-picture and video applications and codecs normally found on such a device. The device is not a mobile phone, so we will not be supporting GSM/3G. The device supports connectivity through USB/Ethernet/Wi-fi, and the related applications such as web browsing, VOIP, and possibly IPTV. If you have Linux running on your device with driver like USB/Ethernet/wifi, its not big to make them work same on Anroid. Regarding codecs audio/video, I dont have idea. At this stage, I need to determine if Android can be successfully ported to the device. I guess that this should include an evaluation of whether the hardware is already supported. Does Android already include device drivers for various low-level hardware? Or is it expected that each manufacturer has to implement the necessary device drivers and codecs? I apologise because I don't have much experience of embedded Linux, so I'm learning on the fly here. As per my understanding, if you have linux ( not android) up and running with device driver i.e touchscreen, sound, keypad.. and your device based on ARM variant you just need customization in device driver to make them work same as on Android. Can someone please point me in the right direction? I've been looking at the documentation included in the kernel/common (Common Android Kernel Tree) section of the code, but haven't been able to find what I'm looking for. Check this: http://www.linuxdevices.com/articles/AT2892720865.html Many thanks, Paul -- Thanks Rizavan --~--~-~--~~~---~--~~ unsubscribe: android-porting+unsubscr...@googlegroups.com website: http://groups.google.com/group/android-porting -~--~~~~--~~--~--~---
