- 20 Apr, 2016 10 commits
-
-
Ezequiel Garcia authored
Set vb2_queue.gfp_flags to GFP_DMA32. Otherwise it will start to create bounce buffers which is something you want to avoid since those are in limited supply. Without this patch, DMA scatter-gather may not work because machines can ran out of buffers easily. Signed-off-by:
Ezequiel Garcia <ezequiel@vanguardiasur.com.ar> Signed-off-by:
Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
Mauro Carvalho Chehab <mchehab@osg.samsung.com>
-
Hans Verkuil authored
Signed-off-by:
-
Hans Verkuil authored
Signed-off-by:
-
Hans Verkuil authored
The mainline tw686x driver also supports audio, that's missing here as well. Signed-off-by:
-
Hans Verkuil authored
Set vb2_queue.gfp_flags to GFP_DMA32. Otherwise it will start to create bounce buffers which is something you want to avoid since those are in limited supply. Signed-off-by:
-
Mauro Carvalho Chehab authored
the dma_requests field is cached, but cache is not used: drivers/staging/media/tw686x-kh/tw686x-kh-video.c: In function 'tw686x_video_irq': drivers/staging/media/tw686x-kh/tw686x-kh-video.c:622:6: warning: variable 'requests' set but not used [-Wunused-but-set-variable] u32 requests; ^ Use the cache instead, as it seems reading it needs to be done with spin lock taken. Signed-off-by: -
Ezequiel Garcia authored
This commit introduces the support for the Techwell TW686x video capture IC. This hardware supports a few DMA modes, including scatter-gather and frame (contiguous). This commit makes little use of the DMA engine and instead has a memcpy based implementation. DMA frame and scatter-gather modes support may be added in the future. Currently supported chips: - TW6864 (4 video channels), - TW6865 (4 video channels, not tested, second generation chip), - TW6868 (8 video channels but only 4 first channels using built-in video decoder are supported, not tested), - TW6869 (8 video channels, second generation chip). [mchehab@osg.samsung.com: make checkpatch happy by using "unsigned int" instead of just "unsigned"] Cc: Krzysztof Halasa <khalasa@piap.pl> Signed-off-by: -
Krzysztof Hałasa authored
A driver for Intersil/Techwell TW686x-based PCIe frame grabbers. [hans.verkuil@cisco.com: renamed staging tw686x to tw686x-kh to prevent naming conflicts] [hans.verkuil@cisco.com: don't build tw686x-kh if tw686x is already selected to prevent conflicts] [mchehab@osg.samsung.com: use "unsigned int" instead of just "unsigned" and add some whitespaces to make checkpatch happier] Signed-off-by:
Krzysztof Halasa <khalasa@piap.pl> Signed-off-by:
-
Mauro Carvalho Chehab authored
Those callbacks are called with the media_device.graph_mutex held. Add a note about that, as the code called by those notifiers should not be touching in the mutex. Acked-by:
Sakari Ailus <sakari.ailus@linux.intel.com> Acked-by:
-
Mauro Carvalho Chehab authored
Right now, the lock schema for media_device struct is messy, since sometimes, it is protected via a spin lock, while, for media graph traversal, it is protected by a mutex. Solve this conflict by always using a mutex. As a side effect, this prevents a bug when the media notifiers is called at atomic context, while running the notifier callback: BUG: sleeping function called from invalid context at mm/slub.c:1289 in_atomic(): 1, irqs_disabled(): 0, pid: 3479, name: modprobe 4 locks held by modprobe/3479: #0: (&dev->mutex){......}, at: [<ffffffff81ce8933>] __driver_attach+0xa3/0x160 #1: (&dev->mutex){......}, at: [<ffffffff81ce8941>] __driver_attach+0xb1/0x160 #2: (register_mutex#5){+.+.+.}, at: [<ffffffffa10596c7>] usb_audio_probe+0x257/0x1c90 [snd_usb_audio] #3: (&(&mdev->lock)->rlock){+.+.+.}, at: [<ffffffffa0e6051b>] media_device_register_entity+0x1cb/0x700 [media] CPU: 2 PID: 3479 Comm: modprobe Not tainted 4.5.0-rc3+ #49 Hardware name: /NUC5i7RYB, BIOS RYBDWi35.86A.0350.2015.0812.1722 08/12/2015 0000000000000000 ffff8803b3f6f288 ffffffff81933901 ffff8803c4bae000 ffff8803c4bae5c8 ffff8803b3f6f2b0 ffffffff811c6af5 ffff8803c4bae000 ffffffff8285d7f6 0000000000000509 ffff8803b3f6f2f0 ffffffff811c6ce5 Call Trace: [<ffffffff81933901>] dump_stack+0x85/0xc4 [<ffffffff811c6af5>] ___might_sleep+0x245/0x3a0 [<ffffffff811c6ce5>] __might_sleep+0x95/0x1a0 [<ffffffff8155aade>] kmem_cache_alloc_trace+0x20e/0x300 [<ffffffffa0e66e3d>] ? media_add_link+0x4d/0x140 [media] [<ffffffffa0e66e3d>] media_add_link+0x4d/0x140 [media] [<ffffffffa0e69931>] media_create_pad_link+0xa1/0x600 [media] [<ffffffffa0fe11b3>] au0828_media_graph_notify+0x173/0x360 [au0828] [<ffffffffa0e68a6a>] ? media_gobj_create+0x1ba/0x480 [media] [<ffffffffa0e606fb>] media_device_register_entity+0x3ab/0x700 [media] Reviewed-by:Javier Martinez Canillas <javier@osg.samsung.com> Acked-by:
-
- 13 Apr, 2016 30 commits
-
-
Peter Rosin authored
s32tmp in the below code may be negative, and dev->mclk_khz is an unsigned type. s32tmp = 0x10000 * (tuner_frequency - c->frequency); s32tmp = DIV_ROUND_CLOSEST(s32tmp, dev->mclk_khz); This is undefined, as DIV_ROUND_CLOSEST is undefined for negative dividends when the divisor is of unsigned type. So, change mclk_khz to be signed (s32). Signed-off-by:
Peter Rosin <peda@axentia.se> Signed-off-by:
Antti Palosaari <crope@iki.fi> Signed-off-by:
-
Simon Horman authored
Make use of ARCH_RENESAS in place of ARCH_SHMOBILE. This is part of an ongoing process to migrate from ARCH_SHMOBILE to ARCH_RENESAS the motivation for which being that RENESAS seems to be a more appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs. Acked-by:
Geert Uytterhoeven <geert+renesas@glider.be> Acked-by:
Simon Horman <horms+renesas@verge.net.au> Signed-off-by:
Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com> Signed-off-by:
-
Laurent Pinchart authored
The maximum image size supported by the WPF is 2048x2048 on Gen2 and 8190x8190 on Gen3. Update the code accordingly, and fix the maximum LIF size for both Gen2 and Gen3. Signed-off-by:
-
Laurent Pinchart authored
The format is erroneously defined with an alpha channel. Fix it. Signed-off-by:
-
Laurent Pinchart authored
Make the global alpha multiplier of DRM planes configurable. All the necessary infrastructure is there, we just need to store the alpha value passed through the DRM API. Signed-off-by:
-
Laurent Pinchart authored
Make the Z-order of planes configurable by assigning RPFs to BRU inputs dynamically based on the Z-order position. Signed-off-by:
-
Laurent Pinchart authored
The Gen3 RPF includes an alpha multiplier that can both multiply the alpha channel by a fixed global alpha value, and multiply the pixel components to convert the input to premultiplied alpha. As alpha premultiplication is available in the BRU for both Gen2 and Gen3 we handle it there and use the Gen3 alpha multiplier for global alpha multiplication only. This prevents conversion to premultiplied alpha if no BRU is present in the pipeline, that use case will be implemented later if needed. Signed-off-by:
-
Laurent Pinchart authored
Synchronize the userspace LUT setup with the pipeline operation by using a display list fragment to store LUT data. Signed-off-by:
-
Laurent Pinchart authored
Display lists support up to 8 bodies but we currently use a single one. To support preparing display lists for large look-up tables, add support for multi-body display lists. Signed-off-by:
-
Laurent Pinchart authored
The vsp1_dl_list_put() function expects to be called with the display list manager lock held. This assumption is correct for calls from within the vsp1_dl.c file, but not for the external calls. Fix it by taking the lock inside the function and providing an unlocked version for the internal callers. Signed-off-by:
-
Laurent Pinchart authored
The field takes positive values only, make it unsigned. Signed-off-by:
-
Laurent Pinchart authored
The LUT set format handler overrides the requested format by mistake. Fix it. Signed-off-by:
-
Laurent Pinchart authored
Most of the entities can't perform scaling and implement the same frame size enumeration function. Factorize the code into a single implementation. Signed-off-by:
-
Mauro Carvalho Chehab authored
There are some issues rised on this patch during patch review. I ended by merging this one by mistake. So, let's revert it. This reverts commit 54ace1cf. Cc: Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
-
Laurent Pinchart authored
Most of the entities can't perform format conversion and implement the same media bus enumeration function. Factorize the code into a single implementation. Signed-off-by:
-
Laurent Pinchart authored
All entities implement the same get pad format handler, factorize it into a common function. Signed-off-by:
-
Laurent Pinchart authored
The RPF entities are located at the very beginning of pipelines, they can't be target nodes in the Data Path Router matrix. Remove their input ID from the routing table. Signed-off-by:
-
Laurent Pinchart authored
Instead of embedding pipelines in the vsp1_video objects allocate them on demand when they are needed. This fixes the streamon race condition where pipelines objects from different video nodes could be used for the same pipeline. Signed-off-by:
-
Laurent Pinchart authored
Move the pipeline initialization and cleanup functions to prepare for the next commit. No functional code change is performed here. Signed-off-by:
-
Laurent Pinchart authored
This prepares for dynamic pipeline allocation by providing a field that can be used to store the pipeline pointer atomically under driver control. Signed-off-by:
-
Laurent Pinchart authored
Pass the pipe explicitly instead of retrieving it through media entities. This decouples device state stored in the pipeline from the active state stored in entities, preparing for dynamic pipeline creation. Signed-off-by:
-
Laurent Pinchart authored
The primary purpose of those functions is to build the pipeline, rename them to make this clearer. Signed-off-by:
-
Laurent Pinchart authored
Modules write register values to the active display list pointed to by the pipeline. In order to support preparing display lists ahead of time, pass them explicitly to all configuration functions. Signed-off-by:
-
Laurent Pinchart authored
Reuse the runtime logic to initialize the default format instead of open-coding it. This ensures coherency between intialization and runtime. Signed-off-by:
-
Laurent Pinchart authored
The two structures are identical, merge them and move the result to vsp1_rwpf.c. All rwpf pad operations can now be declared static. Signed-off-by:
-
Laurent Pinchart authored
The subdev s_stream operation is abused as a generic way to setup modules at every frame. Move the code out to a new VSP1 entity configure operation. Most modules now have an empty s_stream operation that can be removed. The only exception is the WPF module that needs to perform hardware configuration when stopping the stream. The code can be simplified accordingly as we know that that operation never fails. Signed-off-by:
-
Laurent Pinchart authored
Use the pad config structure part of the vsp1_entity to store all active pad selection rectangles. This generalizes the code to operate on pad config structures. Signed-off-by:
-
Laurent Pinchart authored
Add a pad config structure field to the vsp1_entity structure and use it to store all active pad formats. This generalizes the code to operate on pad config structures. Signed-off-by:
-
Laurent Pinchart authored
Turn the custom formats initialization function into a standard pad::init_cfg handler and use it in subdevs instead of initializing formats in the subdev open handler. This makes the subdev open handler empty, so remove it. Signed-off-by:
-
Laurent Pinchart authored
Signed-off-by:
-
