Hi Hiremath Vaibhav,
I have fixed this bugs in omap_vout.c :
vidioc_streamoff()
{
...
- return ovl->set_overlay_info(ovl, &info);
+ ovl->set_overlay_info(ovl, &info);
...
}
because it should not return ovl->set_overlay_info() here, in this way,
the below code will not be executed:
videobuf_streamoff() -> videobuf_queue_cancel()
{
...
q->streaming = 0
...
}
so, vidioc_reqbufs->videobuf_reqbufs() will always fail.
Do you agree with me?
Thanks & Best Regards,
George.Qiao
＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝＝
-----Original Message-----
From: George.Qiao [mailto:qiao_shanshan@xxxxxxxxxxxx]
Sent: Friday, May 08, 2009 11:57 AM
To: Hiremath, Vaibhav
Cc: Syed Mohammed, Khasim; Shah, Hardik; wan_jinxue@xxxxxxxxxxxx;
Bill Lin; shawnofarrell@xxxxxxxxxxxx; Chatterjee, Amit; Andrews,
Gerard; Kridner, Jason; 'Bangfei Jin'; 'Xuejun Dong'; 'YangCai';
linux-omap@xxxxxxxxxxxxxxx
Subject: Re: Status of Beagle Board
Hi Vaibhav Hiremath,
My application can work in OMAP3 EVM and other v4l2 platform, but
there
is another issue in beagle board. as follow:
[Hiremath, Vaibhav] Are you saying, same application works on OMAP3EVM but fails on beagle board?
1. NormalSize() -> RequestBuffer() -> mmap()
2. playing...
3. FullScreen() -> VIDIOC_STREAMOFF
-> munmap()
-> VIDIOC_REQBUFS
-> VIDIOC_QUERYBUF
-> RequestBuffer() -> mmap()
-> VIDIOC_S_CROP
-> VIDIOC_S_FMT
-> VIDIOC_QBUF
-> VIDIOC_STREAMON
When I want to call VIDIOC_REQBUFS in FullScreen(), it always fail.
[Hiremath, Vaibhav] Is it possible for you to share your application, so that I can give try here at my end?
Thanks,
Vaibhav Hiremath
Best Regards,
George.Qiao
============================
Great, please see my comments in-lined below -
Thanks,
Vaibhav Hiremath
Platform Support Products
Texas Instruments Inc
Ph: +91-80-25099927
-----Original Message-----
From: George.Qiao [mailto:qiao_shanshan@xxxxxxxxxxxx]
Sent: Thursday, May 07, 2009 3:51 PM
To: Hiremath, Vaibhav
Cc: Syed Mohammed, Khasim; Shah, Hardik; wan_jinxue@xxxxxxxxxxxx;
Bill Lin; shawnofarrell@xxxxxxxxxxxx; Chatterjee, Amit; Andrews,
Gerard; Kridner, Jason; 'Bangfei Jin'; 'Xuejun Dong'; 'YangCai';
linux-omap-open-source@xxxxxxxxxxxxxx
Subject: Re: Status of Beagle Board
Dear Vaibhav Hiremath,
I can play video by v4l2 now! Thank you!
I have add code in board-omap3beagle.c :
#ifdef CONFIG_FB_OMAP2
static struct resource omap3beagle_vout_resource[3 -
CONFIG_FB_OMAP2_NUM_FBS] = {
};
#else
static struct resource omap3beagle_vout_resource[2] = {
};
#endif
static struct platform_device omap3beagle_vout_device = {
.name = "omap_vout",
.num_resources = ARRAY_SIZE(omap3beagle_vout_resource),
.resource = &omap3beagle_vout_resource[0],
.id = -1,
};
static struct platform_device *omap3_beagle_devices[] __initdata
= {
&beagle_dss_device,
&leds_gpio,
&keys_gpio,
&omap3beagle_vout_device,
};
I have got some omapdss error and voutBuffer Size. as follow:
omapdss: Could not find exact pixel clock. Requested 23500 kHz,
got
24000 kHz
omapdss error: display already enabled
omap_voutDisplay already enabled
omapdss error: display already enabled
omap_voutDisplay already enabled
[Hiremath, Vaibhav] This is not an error as such, it warning
message. Actually here we are trying to enable the display which has
already been enabled by Fbdev.
We can suppress this message, atleast during init.
omap_voutBuffer Size = 3686400
omap_vout: registered and initialized video device 0 [v4l2]
omap_voutBuffer Size = 3686400
omap_vout: registered and initialized video device 1 [v4l2]
[Hiremath, Vaibhav] This is just a debug massage; it is neither a
error nor a warning. So don't worry.
Can you share the code-base or submit the patches to the list, so
that interested people can use it.
Could I change 'voutBuffer Size'? How to fix it?
Best Regards,
George.Qiao
=======================
Yes Definitely this is the issue. If the platform_device.name
doesn't match with the platform_driver.driver.name then your
probe
function will not be called at all.
Can you just copy the board-omap3evm.c changes related to
V4L2/DSS2 and give a shot? I think it should work straight away.
Please let me know if you need any further clarification or
help.
Thanks,
Vaibhav Hiremath
Platform Support Products
Texas Instruments Inc
Ph: +91-80-25099927
-----Original Message-----
From: George.Qiao [mailto:qiao_shanshan@xxxxxxxxxxxx]
Sent: Thursday, May 07, 2009 12:55 PM
To: Hiremath, Vaibhav
Cc: Syed Mohammed, Khasim; Shah, Hardik;
wan_jinxue@xxxxxxxxxxxx;
Bill Lin; shawnofarrell@xxxxxxxxxxxx; Chatterjee, Amit;
Andrews,
Gerard; Kridner, Jason; 'Bangfei Jin'; 'Xuejun Dong';
'YangCai';
linux-omap-open-source@xxxxxxxxxxxxxx
Subject: Re: Status of Beagle Board
Hi Hiremath, Vaibhav,
Thank you very much for your instant response!
I've checked the file 'board-omap3beagle.c' and found nothing
looking
like omap_vout in it, no platform_device, either. And I've also
compared
it with board-omap3evm.c.
Below is some difference between them:
board-omap3evm.c has:
*+static struct platform_device omap3evm_vout_device = {
**+ .name = "omap_vout",
...
}
*
But 'board-omap3beagle.c' does not have anything looking like:
*+static struct platform_device omap3beagle_vout_device = {*
*+ .name = "omap_vout",
...
}
*
So, I think there's no patch on V4L2 for beagle done and this
is
the
root cause of this issue. Do you agree with me? Thanks a lot.
Best Regards,
George.Qiao
===================================
Hi George,
I have looked into your config file and it looks ok to me.
Can you conform that, you have added "platform_device"
definitions
in board-omap3beagle.c?
Can you please create complete patch on top of your baseline
(O-
L
and Tomi's tree), so that I can review that directly? I will be
seeing complete changes done for beagle.
Thanks,
Vaibhav Hiremath
Platform Support Products
Texas Instruments Inc
Ph: +91-80-25099927
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
