943bf197fc9af24f3b4ed45a80646f7b51fbd2b8
22 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Sudhir Sharma
|
bea6dbbe98 |
Revert "Camera Bring-up MR1."
This reverts commit 5cf67b1a1f82e20e9659a4a120c67abe4f918893. Change-Id: If4e182553f46918e5b7a84e9f915039a52541ebb Signed-off-by: Sudhir Sharma <sudsha@codeaurora.org> |
||
Azam Sadiq Pasha Kapatrala Syed
|
ded2688a9d |
Camera Bring-up MR1.
Change-Id: I84d3b4d7ff4147506fdfc81cd26b0aa9cfb72467 Signed-off-by: Sudhir Sharma <sudsha@codeaurora.org> |
||
David Brown
|
a01a8cabfd |
msm: video: vcap: Add video capture driver
commit 6a744e339817d68e2b371fa35b0040382b38d044 Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Dec 7 15:50:32 2012 -0500 media: video: Allow NR threshold to be updated by motion est VCAP did not allow for client to update threshold values based on the motion estimation. By enabling an option for this threshold to be updated dynamically by hardware, this allows for a more aggressive noise reduction to be applied at the request of the client. Change-Id: I0ac590ef826ce4567c048745efafbeb05011f31e Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 06550627e1f08b3956e9bf190999cd14123acf9d Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Nov 30 14:35:16 2012 -0500 media: video: Prevent memory leak in VCAP If a client were to do the following, VCAP would not free the ion handle: -Opened VCAP -Start initial configuration -Queues buffers -Close VCAP without ever calling stream on Since VCAP was not freeing ion this would cause a memory leak CRs-Fixed: 423881 Change-Id: I594924d85884aa5e0a647bc52fa9e89e720f4ff6 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 14600e5b94992cebb0cdc1b541293c8f6c952715 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Nov 7 17:16:27 2012 -0500 media: video: vp_enabled state incorrect after a forced reset After a forced reset occurred on VP, during a stream_off call, vp_enabled state was incorrect causing one extra call to disable_irq. This would result in VP isr to be called whenever interrupt was triggered. CRs-Fixed: 418792 Change-Id: Ia66b2d70506d4b06782bc5c25c83587c78f8e9c8 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit df1d4deafb7931eed7ae102440aea76647aab234 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Nov 5 17:06:02 2012 -0500 media: video: Stride alignment configurable by client Prior to this patch VCAP assumed that the line stride should be aligned to 16 bytes. It is possible that that the client needs a line stride to be 32 bytes aligned. Line stride is now a configurable parameter. Change-Id: Id05aa862813582c23b18c79bd9edf95b2cdfffbe Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit b0c58b1c28f95b653808a32e1069f88e474bfd0d Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Oct 30 17:56:36 2012 -0400 media: video: Prevent buffer ion resource from being freed twice It was possible for the client to have two threads running where one would be blocked in dq_buf that would wake up while the other thread was in the process of turning the stream off. This cause two thread to try freeing the ion resource which would result in a crash. Change-Id: Ibbbf34c7770eed99e1a89d2a4c41ad48f5b6a13a Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit aa64d3a2c70286485c37ae640bf929c451ac6188 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Oct 29 13:05:37 2012 -0400 media: video: VC ISR restructure to make code more readable As new features were added to VCAP the ISR for VC has become increasingly complex and difficult to read. It was determined that the VC ISR code should be simplified and cleaned up in order to make the code flow more easier to read. Change-Id: If7bc0a41e14c40f67120dacde9a1544e8f1ef7b1 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 37b1ecc312d7fcc3baa4423dfde6a107d85c2500 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Oct 29 17:53:39 2012 -0400 media: video: Make VCAP properly call regulator_get VCAP was not providing a pointer to the device when calling regulator_get. Although the call was able to get the regulator, there was always an error message regarding not being able to create a debugfs node. The problem was because the device provided to the regulator_get call was NULL. Change-Id: I6b344cafd96db50e95abfbf46021a361ac0de638 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 1f0f274818326a871184fa3e5597d0d2d82002b1 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Oct 24 17:43:47 2012 -0400 media: video: VC to use data polarity given by client VC would ignore the polarity given by the client and hardcode the value to zero. Now client is properly reporting value, VC now takes the value provided by the client. Change-Id: I511b59ed3e9f6b47d14cd688592d85ab97a5204d Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit f0ab1ee4a19e9616b8726d7974b1fb89baec9086 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Oct 22 12:43:36 2012 -0400 media: video: Force drop 2nd field after VC drops 1st field If a field was dropped by the VC, the VP would not be able to de-interlace the content properly. The problem with the VP is it would have an unbalanced number of top fields and bottom fields. In order to alleviate this problem the subsequent field will also be dropped. That way there would be a balanced number of top and bottom fields in the VP which would prevent jumping behavior observed in the de-interlaced frame. Change-Id: I96e8c0670b5b29baea8d1574b21c259c16a7a32e Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 0857c97b89fdc6316ed3a1029179963a07c785b4 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Oct 17 16:24:21 2012 -0400 media: video: Prevent VCAP from dequeue buffers once turned off VCAP should not report any buffers ready to be dequeued or allow for any buffers to be dq once streamoff has been called. CRs-Fixed: 398896 Change-Id: I2eccf62b7ddff0f7cc0343c8e26bd699b335b9dc Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 8d1c327634466332a93c6d7403758e61de79f728 Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Jul 27 16:36:31 2012 -0400 media: video: Use IOMMU mapping API The ion_phys API is being deprecated. Instead there is a new set of API. Change-Id: I701cc045fe5a0cb0026574dd08016a2bccbc4c72 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit f1b7b584430f6cd33bfee13b916a084fc1b80419 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Oct 16 11:49:24 2012 -0400 media: video: Shutdown VP before freeing VP buffers It was possible for buffers to be freed while VP was still using them. This patch ensures that VP has been stopped before that happens. Change-Id: Ifff689f690cdc43c935ee355dfaca61ade9e9d65 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 9d6f89cb2343ff5adaccb77a177fdc6c3844c65d Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Oct 3 16:24:04 2012 -0400 msm: mpq8064: Correct VCAP stride calucation VCAP was not properly calucating the stride values. This patch allows for line stride to be 16 byte aligned. CRs-Fixed: 390191 Change-Id: Id9eca2fd3ec4a51347f2656271591f7bccf36ad5 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit c4aa4c09475ed3029d010417393a6a6368449772 Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Aug 31 04:49:21 2012 -0400 msm: mpq8064: Adding debugfs support to VCAP In order to assist in debugging various VCAP issue it would be useful to have debugfs. This patch is about making everything to debug VCAP, from message to getting important data, available within debugfs. Change-Id: Ic9373036af6a5eb795da2d89d6f26fcd404264af Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit f3cd9a12838bfe4688e184f90ad16b6c304a8245 Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Sep 21 11:18:02 2012 -0400 msm: mpq8064: Correct VP event type There is a VP error that was being incorrectly reported as a VC error event. This patch allows for the error to properly be reported. Change-Id: Ic2a1c6eb3b149bb7075417d54e2040d2261dda08 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 674f4799a54da7f5c38675308c3471c1b70bba4f Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Sep 21 11:10:22 2012 -0400 msm: mpq8064: VCAP probe properly reports failures It was possible for VCAP probe to fail, but still return 0. Change-Id: I395cecea90475ea07491a57dde12aae1fa67166e Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit f51f6e6d43c8d1d9b5a8ed86e044f19020146407 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Aug 29 11:02:17 2012 -0400 msm: mpq8064: Improve VCAP performance This patch increase bandwidth request and VCAP core clock for 1080 interlace or progressive content. This increase helps VCAP capture and process content in a timely manner. Change-Id: I8cb8d6e87954272b28a8e83351152354ea75a184 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 911b4b702a63fbd241863966cad45909e40e4665 Author: Mitchel Humpherys <mitchelh@codeaurora.org> Date: Wed Sep 12 14:42:50 2012 -0700 ion: change ion kernel map function to not take flags argument Buffer flags are going to be specified at allocation time rather than map time. This removes the flags argument from the ion kernel map function. Change-Id: I91589ca7b1ab142bb5ac8e6b0f4c06fa1305f23a Signed-off-by: Mitchel Humpherys <mitchelh@codeaurora.org> commit 7d72bad55a0e03392d81eee77a9a8c9df6b73a9a Author: Hanumant Singh <hanumant@codeaurora.org> Date: Wed Aug 29 18:39:44 2012 -0700 ion: Port heap mask change to ion Heap mask field passes as argument at allocation time to specify ion heap. Change-Id: I8881dfc491dc0f8f70581ad97b35756a6a33dd6d Signed-off-by: Hanumant Singh <hanumant@codeaurora.org> Signed-off-by: Mitchel Humpherys <mitchelh@codeaurora.org> commit 8ba7ad9492496e12c5555da796bc1f60d52c46f8 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Sep 18 17:01:34 2012 -0400 msm: mpq8064: Improve VP de-interlacing completion time VP time to de-interlacing for 1080i content was over 20ms. This was way too long considering content updating at 60Hz requires latency to be under 16ms. This change allow for de-interlacing to complete within 7ms. Change-Id: Iae2be758ef8d5a6fcecb96df93448226c9ed9109 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 9c4f8a220532407fd890187522b0ddb365b1d660 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Sep 12 10:25:50 2012 -0400 msm: mpq8064: VC report field info that is used by VP VP assumed that the first field provided was a bottom field, if VP was incorrect in this assumption it would not properly de-interlace content. It was also possible for VC to drop a field, when this happened VP would become out of sync again causing de-interlacing to be incorrect. CRs-Fixed: 397534 Change-Id: I7a75a5dcd27149220273284a3d48f407bc5a4f31 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 3af95c07d0ff5be80502734f87ed90dc811face7 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Sep 18 16:43:38 2012 -0400 msm: mpq8064: VP report proper timestamp value There was a problem where de-interlace content was not getting a valid timestamp. Change-Id: Ief9e19c9d20bae3d785a7932b5021e2bba36d238 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit a1e8af4e6cf4bf8a4187a1be844a1462d37affff Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Sep 18 16:41:57 2012 -0400 msm: mpq8064: Timestamp properly populated within VCAP Before this change VCAP was only populating the usec parameter in timeval. Now VCAP populates sec and accounts for timestamp counter overflows. Change-Id: I32b2c49604baba95f56faa483acfe787fdb13b34 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 616762dc78540d8cf82fed1cbbc19cf90956aa4b Author: Terence Hampson <thampson@codeaurora.org> Date: Thu Aug 16 17:47:44 2012 -0400 msm: mpq8064: Allow any client to control noise reduction parms It was requested that any client be allowed to modify the nr parameters. Reason for this is that there will be a separate pc tool that will be controlling NR for demos. Also fixed a bug with that cause s_fmt to failed with EINVAL when arguments were valid. Change-Id: Ie58972a036bb6637a938b48212ef85faf894002e Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 4b88553634efb3b8748a4c2462c71b69bf491947 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Sep 4 03:51:55 2012 -0400 msm: mpq8064: Prevent VC from running after error reported If VC continues to capture frames after receiving an error the frames do not contain a valid frame it is possible for a system crash to occur. This change resets NPL as soon as an error occurs, thus preventing VC from capturing any more frames. VC waits for client to handle error reported by VC. CRs-Fixed: 372564 Change-Id: I72bdbab3251852520801541ed6d5521dc75f7cf1 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 2cba63f9f64c1c5e7d53eeddefc67d075d78d6f8 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Aug 21 10:54:38 2012 -0400 msm: mpq8064: Allow client to select num bufs for VC to hold onto VC part of the VCAP driver was initial implmented for the hardware to 2 store and work with 2 buffers. Since hardware is capable of storing and filling up to 6 buffers, a new API was created to allow client to select how many they want VC to use. Change-Id: I0548f1322c9703d9ccf12700be05511b0e1eb0de Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit b128b98ac802c210dfb0ea0b279271c0c24bc2eb Author: Terence Hampson <thampson@codeaurora.org> Date: Thu Aug 16 14:41:19 2012 -0400 msm: mpq8064: Correct VCAP shutdown procedure VCAP hardware shutdown sequence was incorrect. This would cause the hardware to become unstable, which would result in the system crashing. This patch corrects it so the driver calls the correct shutdown procedure. Change-Id: I78273f2c70d9eab060dcc34aef0b3704ed685f40 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 72452631e2db3ca641414d05d2437144ac0366c6 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Aug 13 12:32:24 2012 -0400 msm: mpq8064: Fix false error reported by VP Logical or where there should have been a bitwise or. Also ignoring cadence change interrupt, because there is nothing for the driver to do cadence change occurs. Change-Id: I8a6214a66b67f6397a7b99ef54969761640abedf Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit a2cba0df366dc571feb8f8a750405f2ad8284cf6 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Aug 8 11:39:53 2012 -0400 msm: mpq8064: Use ION to alloc driver buffers Since VCAP is behind IOMMU, in order to get proper address for VCAP hw to write to driver needs to alloc buffer from ION and not kzalloc. Change-Id: I1f4fcc6834d0ea09ba0fc812b301219d35f40721 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit cc88c1256aaf04ad1ecc1d7ae6b35ae05112a397 Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Jul 16 18:20:02 2012 -0400 msm: mpq8064: Avoid NPL clock overflow Since external components has its own variation there is a need for an additional 2% margin for the npl_clk Change-Id: I27d77ccd2f5cd9a9635975ea286bf3d0f8dba1d2 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 3dff4ef26eb40494eebb17adef80e66a0a469478 Author: Terence Hampson <thampson@codeaurora.org> Date: Wed Jun 13 15:20:59 2012 -0400 msm: mpq8064: Enable separate ioctl call for noise reduction (nr) NR for VCAP is capable of receiving manual tuning parameters. In order for the client to pass in these parameters a separate ioctl was developed. Change-Id: Ia8d7e8de80ee4fd7ccd2019d30b4457f1000e106 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit 54d3a6d39374c5fe3bd22be970717f57480d6ad8 Author: Terence Hampson <thampson@codeaurora.org> Date: Tue Jul 10 14:05:35 2012 -0400 msm: mpq8064: Dummy read to prevent false interrupt from triggering The VBIF is capable of triggering a false bus error the first time it performs a read. This fix creates a dummy read event that happens every time that the VCAP is powered on that will ignore the false positive VBIF interrupt error, so that during normal operation the error can be properly identified. Change-Id: Ie56e60faf2067fab7a05a0f16c2d840408c35035 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit ad33c513b4b9455f0822c125affc39138603459e Author: Terence Hampson <thampson@codeaurora.org> Date: Mon Jul 16 17:50:00 2012 -0400 msm: mpq8064: VCAP corner case bug fixes Closing VCAP devices could be called without calling stream off: This problem would make VCAP resources taken by devices no longer running. Using unsigned variable to store signed result: This problem would allow stream on to continue even when function returned error. Fix problem of reading VP irq status register too early: It was possible for cadence detection bit to be propgated faster by only a few clock cycles then the picture done bit. When the irq status read was performed and by the time a write to the clear register was performed the picture done bit would propgate. Since the irq is rising edge triggered this would cause a problem and the irq would not be triggered until VP was reset. Remove unused code Change-Id: Iba72354c06287efb27178edf64d8d3f459dbf954 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit a6c9648dd8e85e1e1e854347a78e4497578ebf51 Author: Terence Hampson <thampson@codeaurora.org> Date: Fri Jul 6 17:53:09 2012 -0400 msm: mpq8064: Separate q/dq buffer for VCAP de-interlacing mode This change fixes a crash when running VCAP in de-interlace mode. The problem was noticed in the introduction of kernel 3.4. VCAP was improperly using the videobuf2 framework APIs. Since VCAP is already reliant on queue structure provided by videobuf2 framework, we created our own basic version of the qbuf and dqbuf calls that have some similarities to the videobuf2 framework. This addition was necessary since we are calling qbuf from a kernel work thread. Since moving to kernel 3.4 there has been a change in videobuf2 qbuf that dereferences the current process.s mm_struct, from a kernel work thread this struct is NULL, which was causing a crash. When we are de-interlacing a captured video we are cycling the buffers internally in the driver, which is why we needed to create our own q/dq buffer functionality. Change-Id: Id51b775ae7f0a58bbac8a581e07fa02417802a30 Signed-off-by: Terence Hampson <thampson@codeaurora.org> commit b14ed96f74b1258b1f102831a2b0ccd56f991132 Author: Laura Abbott <lauraa@codeaurora.org> Date: Mon Jan 30 14:18:08 2012 -0800 gpu: ion: Pull in patches for 3.4 Pull in Ion patches for 3.4 upgrade. Inclues the following patches from google: commit |
||
Linus Torvalds
|
250f6715a4 |
Merge tag 'device-for-3.4' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux
Pull <linux/device.h> avoidance patches from Paul Gortmaker: "Nearly every subsystem has some kind of header with a proto like: void foo(struct device *dev); and yet there is no reason for most of these guys to care about the sub fields within the device struct. This allows us to significantly reduce the scope of headers including headers. For this instance, a reduction of about 40% is achieved by replacing the include with the simple fact that the device is some kind of a struct. Unlike the much larger module.h cleanup, this one is simply two commits. One to fix the implicit <linux/device.h> users, and then one to delete the device.h includes from the linux/include/ dir wherever possible." * tag 'device-for-3.4' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux: device.h: audit and cleanup users in main include dir device.h: cleanup users outside of linux/include (C files) |
||
Paul Gortmaker
|
313162d0b8 |
device.h: audit and cleanup users in main include dir
The <linux/device.h> header includes a lot of stuff, and it in turn gets a lot of use just for the basic "struct device" which appears so often. Clean up the users as follows: 1) For those headers only needing "struct device" as a pointer in fcn args, replace the include with exactly that. 2) For headers not really using anything from device.h, simply delete the include altogether. 3) For headers relying on getting device.h implicitly before being included themselves, now explicitly include device.h 4) For files in which doing #1 or #2 uncovers an implicit dependency on some other header, fix by explicitly adding the required header(s). Any C files that were implicitly relying on device.h to be present have already been dealt with in advance. Total removals from #1 and #2: 51. Total additions coming from #3: 9. Total other implicit dependencies from #4: 7. As of 3.3-rc1, there were 110, so a net removal of 42 gives about a 38% reduction in device.h presence in include/* Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> |
||
Hans Verkuil
|
a26243b0e3 |
[media] v4l2-ctrls: add helper functions for control events
Many drivers just support control events, and most radio drivers just need to poll for control events. Add some functions to simplify those jobs. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
e2ecb257ee |
[media] v4l2: standardize log start/end message
For drivers that properly use the v4l2 framework (i.e. set v4l2_dev in the video_device struct), the start and end messages of VIDIOC_LOG_STATUS are now generated automatically. People tended to forget these, but the v4l2-ctl tool scans for these messages, and it also makes it easier to read the status output in the kernel log. The cx18, ivtv and bttv drivers were changed since they no longer need to log these start/end messages. In saa7164 two empty log_status functions were removed. Also added a helper function to v4l2-ctrl.c that can be used as the vidioc_log_status callback if all you need to do is to log the current control values. This is now used by pwc and vivi. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Cc: Steven Toth <stoth@kernellabs.com> Cc: Andy Walls <awalls@md.metrocast.net> Cc: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
5626b8c75b |
[media] v4l2-ctrls: implement new volatile autocluster scheme
The problem tackled in this patch is how to handle volatile autoclusters correctly. A volatile autocluster is a cluster of related controls where one control is the control that toggles between manual and auto mode and the other controls are the values for the manual mode. For example autogain and gain, autoexposure and exposure, etc. If the hardware lets you read out the automatically calculated manual values while in automode, then those manual controls should be marked volatile. gain value as calculated by the autogain circuitry, then you would mark the gain control as volatile (i.e. continuously changing). The question in such use cases is what to do when switching from the auto mode to the manual mode. Should we switch to the last set manual values or should the volatile values be copied and used as the initial manual values. For example: suppose the mode is manual gain and gain is set to 5. Then autogain is turned on and the gain is set by the hardware to 2. Finally the user switches back to manual gain. What should the gain be? 2 or 5? After a long discussion the decisions was made to keep the last value as calculated by the auto mode (so 2 in the example above). The reason is that webcams that do such things will adapt themselves to the current light conditions and when you switch back to manual mode you expect that you keep the same picture. If you would switch back to old manual values, then that would give you a suddenly different picture, which is jarring for the user. Additionally, this would be difficult to implement in applications that store and restore the control values at application exit and start. If you want to keep the old manual values when you switch from auto to manual, then there would have to be a way for applications to get hold of those old values while in auto mode, but there isn't. So this patch will do all the heavy lifting in v4l2-ctrls.c: if you go from auto mode to manual mode and the manual controls are volatile, then g_volatile_ctrl will be called to get the current values for the manual controls before switching to manual mode. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
88365105d6 |
[media] v4l2-ctrls: replace is_volatile with V4L2_CTRL_FLAG_VOLATILE
With the new flag there is no need anymore to have a separate is_volatile field. Modify all users to use the new flag. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Acked-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
3f66f0ed31 |
[media] v4l2-ctrls/v4l2-events: small coding style cleanups
Thanks to Laurent Pinchart <laurent.pinchart@ideasonboard.com>. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
eb5b16efb2 |
[media] v4l2-ctrls: improve discovery of controls of the same cluster
The implementation of VIDIOC_G/S/TRY_EXT_CTRLS in the control framework has to figure out which controls in the control list belong to the same cluster. Since controls belonging to the same cluster need to be handled as a unit, this is important information. It did that by going over the controls in the list and for each control that belonged to a multi-control cluster it would walk the remainder of the list to try and find controls that belong to that same cluster. This approach has two disadvantages: 1) it was a potentially quadratic algorithm (although highly unlikely that it would ever be that bad in practice). 2) it took place with the control handler's lock held. Since we want to make it possible in the future to change control values from interrupt context, doing a lot of work while holding a lock is not a good idea. In the new code the algorithm is no longer quadratic but linear in the number of controls in the list. Also, it now can be done beforehand. Another change that was made was to so the try and set at the same time. Before when S_TRY_EXT_CTRLS was called it would 'try' the controls first, and then it would 'set' them. The idea was that any 'try' errors would prevent the 'set' from happening, thus avoiding having partially set control lists. However, this caused more problems than it solved because between the 'try' and the 'set' changes might have happened, so it had to try a second time, and since actual controls with a try_ctrl op are very rare (and those that we have just adjust values and do not return an error), I've decided to drop that two-stage approach and just combine try and set. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
f1e393de38 |
[media] v4l2-event/ctrls/fh: allocate events per fh and per type instead of just per-fh
The driver had to decide how many events to allocate when the v4l2_fh struct was created. It was possible to add more events afterwards, but there was no way to ensure that you wouldn't miss important events if the event queue would fill up for that filehandle. In addition, once there were no more free events, any new events were simply dropped on the floor. For the control event in particular this made life very difficult since control status/value changes could just be missed if the number of allocated events and the speed at which the application read events was too low to keep up with the number of generated events. The application would have no idea what the latest state was for a control since it could have missed the latest control change. So this patch makes some major changes in how events are allocated. Instead of allocating events per-filehandle they are now allocated when subscribing an event. So for that particular event type N events (determined by the driver) are allocated. Those events are reserved for that particular event type. This ensures that you will not miss events for a particular type altogether. In addition, if there are N events in use and a new event is raised, then the oldest event is dropped and the new one is added. So the latest event is always available. This can be further improved by adding the ability to merge the state of two events together, ensuring that no data is lost at all. This will be added in the next patch. This also makes it possible to allow the user to determine the number of events that will be allocated. This is not implemented at the moment, but would be trivial. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
77068d36d8 |
[media] v4l2-ctrls/event: remove struct v4l2_ctrl_fh, instead use v4l2_subscribed_event
The v4l2_ctrl_fh struct connected v4l2_ctrl with v4l2_fh so the control would know which filehandles subscribed to it. However, it is much easier to use struct v4l2_subscribed_event directly for that and get rid of that intermediate struct. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
82a7c04944 |
[media] v4l2-ctrls: make manual_mode_value 8 bits and check against control range
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
ce7275747b |
[media] v4l2-ctrls: simplify event subscription
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
6e239399e5 |
[media] v4l2-ctrls: add control events
Whenever a control changes value or state an event is sent to anyone that subscribed to it. This functionality is useful for control panels but also for applications that need to wait for (usually status) controls to change value. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
ab892bac84 |
[media] v4l2-ctrls: add v4l2_fh pointer to the set control functions
When an application changes a control you want to generate an event. However, you want to avoid sending such an event back to the application (file handle) that caused the change. Add the filehandle to the various set control functions. The filehandle isn't used yet, but the control event patches will need this. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
72d877cac0 |
[media] v4l2-ctrls: add v4l2_ctrl_auto_cluster to simplify autogain/gain scenarios
It is a bit tricky to handle autogain/gain type scenerios correctly. Such controls need to be clustered and the V4L2_CTRL_FLAG_UPDATE should be set on the autofoo controls. In addition, the manual controls should be marked inactive when the automatic mode is on, and active when the manual mode is on. This also requires specialized volatile handling. The chances of drivers doing all these things correctly are pretty remote. So a new v4l2_ctrl_auto_cluster function was added that takes care of these issues. Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
Laurent Pinchart
|
01c40c048b |
[media] v4l: Include linux/videodev2.h in media/v4l2-ctrls.h
The later makes extensive use of structures defined in the former. Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
Hans Verkuil
|
2a863793be |
[media] v4l2-ctrls: v4l2_ctrl_handler_setup must set is_new to 1
Renamed has_new to is_new. Drivers can use the is_new field to determine if a new value was specified for a control. The v4l2_ctrl_handler_setup() must always set this to 1 since the setup has to force a full update of all controls. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
513521eaee |
[media] v4l2-ctrls: use const char * const * for the menu arrays
This prevents checkpatch warnings generated when defining 'static const char *foo[]' arrays. It makes sense to use const char * const * anyway since the pointers in the array are indeed const. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |
||
|
Hans Verkuil
|
0996517cf8 |
V4L/DVB: v4l2: Add new control handling framework
Add a new framework to handle controls which makes life for driver developers much easier. Note that this patch moves some of the control support that used to be in v4l2-common.c to v4l2-ctrls.c. The tables were copied unchanged. The body of v4l2_ctrl_query_fill() was copied to a new v4l2_ctrl_fill() function in v4l2-ctrls.c. This new function doesn't use the v4l2_queryctrl struct anymore, which makes it more general. The remainder of v4l2-ctrls.c is all new. Highlights include: - No need to implement VIDIOC_QUERYCTRL, QUERYMENU, S_CTRL, G_CTRL, S_EXT_CTRLS, G_EXT_CTRLS or TRY_EXT_CTRLS in either bridge drivers or subdevs. New wrapper functions are provided that can just be plugged in. Once everything has been converted these wrapper functions can be removed as well. - When subdevices are added their controls can be automatically merged with the bridge driver's controls. - Most drivers just need to implement s_ctrl to set the controls. The framework handles the locking and tries to be as 'atomic' as possible. - Ready for the subdev device nodes: the same mechanism applies to subdevs and their device nodes as well. Sub-device drivers can make controls local, preventing them from being merged with bridge drivers. - Takes care of backwards compatibility handling of VIDIOC_S_CTRL and VIDIOC_G_CTRL. Handling of V4L2_CID_PRIVATE_BASE is fully transparent. CTRL_CLASS controls are automatically added. Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com> |