Hi Dan, On Thu, Jul 24, 2025 at 7:04 PM Dan Scally <dan.scally@xxxxxxxxxxxxxxxx> wrote: > > Hi Prabhakar - thanks for the review > > On 24/07/2025 17:52, Lad, Prabhakar wrote: > > Hi Daniel, > > > > Thank you for the patch. > > > > On Mon, Jul 14, 2025 at 4:25 PM Daniel Scally > > <dan.scally@xxxxxxxxxxxxxxxx> wrote: > >> Add a driver for the Input Video Control block in an RZ/V2H SoC which > >> feeds data into the Arm Mali-C55 ISP. > >> > > s/V2H/V2HP everywhere. > Ack! > >> Signed-off-by: Daniel Scally <dan.scally@xxxxxxxxxxxxxxxx> > >> --- > >> Changes in v5: > >> > >> - Fixed .enum_frame_sizes() to properly check that the > >> given mbus_code matches the source pads format. > >> - Tidy up extra space in Kconfig > >> - Revise Kconfig option message > >> - Don't mark functions inline > >> - Fixup misleading comment > >> - select CONFIG_PM > >> - Use the new pm_sleep_ptr() functionality > >> - Minor formatting > >> > >> Changes in v4: > >> > >> - Update the compatible to renesas,r9a09g057-ivc > >> - Dropped the media jobs / scheduler functionality, and re > >> worked the driver to have its own workqueue pushing frames > >> - Fix .enum_mbus_code() to return 20-bit output for source > >> pad. > >> - Fix some alignment issues > >> - Make the forwarding of sink to source pad format a more > >> explicit operation. > >> - Rename rzv2h_initialise_video_device_and_queue() > >> - Reversed order of v4l2_subdev_init_finalize() and > >> v4l2_async_register_subdev() to make sure everything is > >> finished initialising before registering the subdev. > >> - Change function to MEDIA_ENT_F_PROC_VIDEO_PIXEL_FORMATTER > >> - Use a parametised macro for min vblank > >> - Minor formatting > >> - Use the DEFAULT macros for quantization / ycbcr_enc values > >> - Switch to using the mplane API > >> - Dropped select RESET_CONTROLLER > >> - Used the new helpers for starting a media pipeline > >> - Switch from threaded irq to normal with driver workqueue > >> and revised startup routine > >> > >> Changes in v3: > >> > >> - Account for the renamed CRU pixel formats > >> > >> Changes in v2: > >> > >> - Added selects and depends statements to Kconfig entry > >> - Fixed copyright year > >> - Stopped including in .c files headers already included in .h > >> - Fixed uninitialized variable in iterator > >> - Only check vvalid member in interrupt function and wait > >> unconditionally elsewhere > >> - __maybe_unused for the PM ops > >> - Initialise the subdevice after setting up PM > >> - Fixed the remove function for the driver to actually do > >> something. > >> - Some minor formatting changes > >> - Fixed the quantization member for the format > >> - Changes accounting for the v2 of the media jobs framework > >> - Change min_queued_buffers to 0 > >> --- > >> drivers/media/platform/renesas/Kconfig | 1 + > >> drivers/media/platform/renesas/Makefile | 1 + > >> drivers/media/platform/renesas/rzv2h-ivc/Kconfig | 18 + > >> drivers/media/platform/renesas/rzv2h-ivc/Makefile | 5 + > >> .../platform/renesas/rzv2h-ivc/rzv2h-ivc-dev.c | 229 +++++++++ > >> .../platform/renesas/rzv2h-ivc/rzv2h-ivc-subdev.c | 376 ++++++++++++++ <snip> > > ISP Probe: > > [ 11.600383] mali-c55 16080000.isp: Detected Mali-C55 ISP 9000043.31032022.0 > > [ 11.622062] mali-c55 16080000.isp: Runtime PM usage count underflow! > > > > Logs from IVC: > > root@rzv2h-evk:~/c55# media-ctl -p > > Media controller API version 6.16.0 > > > > Media device information > > ------------------------ > > driver mali-c55 > > model ARM Mali-C55 ISP > > serial > > bus info platform:16080000.isp > > hw revision 0x1d982d6 > > driver version 6.16.0 > > > > Device topology > > - entity 1: mali-c55 tpg (1 pad, 1 link, 0 routes) > > type V4L2 subdev subtype Sensor flags 0 > > device node name /dev/v4l-subdev0 > > pad0: SOURCE > > [stream:0 fmt:SRGGB20_1X20/1920x1080 field:none > > colorspace:raw xfer:none ycbcr:601 quantization:lim-range] > > -> "mali-c55 isp":0 [] > > > > - entity 3: mali-c55 isp (5 pads, 6 links, 0 routes) > > type V4L2 subdev subtype Unknown flags 0 > > device node name /dev/v4l-subdev1 > > pad0: SINK,MUST_CONNECT > > [stream:0 fmt:SGRBG20_1X20/2304x1296 field:none > > colorspace:raw xfer:none ycbcr:601 quantization:lim-range > > crop:(0,0)/2304x1296] > > <- "mali-c55 tpg":0 [] > > <- "rzv2h ivc block":1 [ENABLED] > > pad1: SOURCE > > [stream:0 fmt:RGB121212_1X36/2304x1296 field:none > > colorspace:srgb xfer:none ycbcr:601 quantization:lim-range] > > -> "mali-c55 resizer fr":0 [ENABLED,IMMUTABLE] > > pad2: SOURCE > > [stream:0 fmt:SGRBG20_1X20/2304x1296 field:none > > colorspace:raw xfer:none ycbcr:601 quantization:lim-range] > > -> "mali-c55 resizer fr":2 [ENABLED,IMMUTABLE] > > pad3: SOURCE > > [stream:0 fmt:unknown/0x0 field:none] > > -> "mali-c55 3a stats":0 [] > > pad4: SINK > > [stream:0 fmt:unknown/0x0 field:none] > > <- "mali-c55 3a params":0 [] > > > > - entity 9: mali-c55 resizer fr (3 pads, 3 links, 0 routes) > > type V4L2 subdev subtype Unknown flags 0 > > device node name /dev/v4l-subdev2 > > pad0: SINK > > [stream:0 fmt:RGB121212_1X36/2304x1296 field:none > > colorspace:srgb xfer:srgb ycbcr:601 quantization:lim-range > > crop:(0,0)/2304x1296 > > compose:(0,0)/2304x1296] > > <- "mali-c55 isp":1 [ENABLED,IMMUTABLE] > > pad1: SOURCE > > [stream:0 fmt:RGB121212_1X36/2304x1296 field:none > > colorspace:srgb xfer:srgb ycbcr:601 quantization:lim-range] > > -> "mali-c55 fr":0 [ENABLED] > > pad2: SINK > > <- "mali-c55 isp":2 [ENABLED,IMMUTABLE] > > > > - entity 13: mali-c55 fr (1 pad, 1 link) > > type Node subtype V4L flags 0 > > device node name /dev/video0 > > pad0: SINK > > <- "mali-c55 resizer fr":1 [ENABLED] > > > > - entity 17: mali-c55 3a params (1 pad, 1 link) > > type Node subtype V4L flags 0 > > device node name /dev/video1 > > pad0: SOURCE > > -> "mali-c55 isp":4 [] > > > > - entity 21: mali-c55 3a stats (1 pad, 1 link) > > type Node subtype V4L flags 0 > > device node name /dev/video2 > > pad0: SINK > > <- "mali-c55 isp":3 [] > > > > - entity 37: rzv2h ivc block (2 pads, 2 links, 0 routes) > > type V4L2 subdev subtype Unknown flags 0 > > device node name /dev/v4l-subdev3 > > pad0: SINK > > [stream:0 fmt:SGRBG10_1X10/2304x1296 field:none > > colorspace:raw xfer:none ycbcr:601 quantization:full-range] > > <- "rzv2h-ivc":0 [ENABLED,IMMUTABLE] > > pad1: SOURCE > > [stream:0 fmt:SGRBG20_1X20/2304x1296 field:none > > colorspace:raw xfer:none ycbcr:601 quantization:full-range] > > -> "mali-c55 isp":0 [ENABLED] > > > > - entity 40: rzv2h-ivc (1 pad, 1 link) > > type Node subtype V4L flags 0 > > device node name /dev/video3 > > pad0: SOURCE > > -> "rzv2h ivc block":0 [ENABLED,IMMUTABLE] > > > > root@rzv2h-evk:~/c55# > > root@rzv2h-evk:~# v4l2-ctl -d3 --stream-out-mmap > > --stream-from=/root/c55/frame-15.bin --stream-loop > >>>>> VIDIOC_STREAMON returned -1 (Input/output error) > > root@rzv2h-evk:~# > > > > Logs from ISP: > > root@rzv2h-evk:~/c55# ./isp.sh > > Device /dev/video0 opened. > > Device `ARM Mali-C55 ISP' on `platform:16080000.isp' (driver > > 'mali-c55') supports video, capture, with mplanes. > > Video format set: RGB565 (50424752) 2304x1296 field none, 1 planes: > > * Stride 4608, buffer size 5971968 > > Video format: RGB565 (50424752) 2304x1296 field none, 1 planes: > > * Stride 4608, buffer size 5971968 > > 8 buffers requested. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 0/0 mapped at address 0xffff81f2e000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 1/0 mapped at address 0xffff8197c000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 2/0 mapped at address 0xffff813ca000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 3/0 mapped at address 0xffff80e18000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 4/0 mapped at address 0xffff80866000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 5/0 mapped at address 0xffff802b4000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 6/0 mapped at address 0xffff7fd02000. > > length: 1 offset: 4017363672 timestamp type/source: mono/EoF > > Buffer 7/0 mapped at address 0xffff7f750000. > > [ 92.647719] kauditd_printk_skb: 8 callbacks suppressed > > [ 92.647734] audit: type=1006 audit(1753371566.385:25): pid=407 > > uid=0 old-auid=4294967295 auid=0 tty=(none) old-ses=4294967295 ses=4 > > res=1 > > [ 92.665263] audit: type=1300 audit(1753371566.385:25): > > arch=c00000b7 syscall=64 success=yes exit=1 a0=7 a1=ffffc4ff5740 a2=1 > > a3=1 items=0 ppid=1 pid=407 auid=0 uid=0 gid=0 euid=0 suid=0 fsuid=0 > > egid=0 sgid=0 fsgid=0 tty=(none) ses=4 comm="sshd" > > exe="/usr/sbin/sshd" key=(null) > > [ 92.689604] audit: type=1327 audit(1753371566.385:25): > > proctitle=737368643A20726F6F74205B707269765D > > > > [ 100.932191] rz-dmac 11400000.dma-controller: dma_sync_wait: timeout! > > [ 100.938566] mali-c55 16080000.isp: Failed to DMA xfer ISP config > > [ 100.944702] mali-c55 16080000.isp: failed to write ISP config > > [ 100.950562] mali-c55 16080000.isp: Failed to start ISP > > > Hm, Is this on the EVK? How did you set up devicetree? It's trying to use the DMA write; I've so > far just been using CPU write on the RZ/V2H...I'll try to replicate it and see where I get > > Yes this is on the EVK. I have the changes [0] done to get it working (Ive not enabled the DMA). [0] https://gist.github.com/prabhakarlad/bb1d3649243617d88c9b4e8b386f0803 Cheers, Prabhakar
