On 21/08/2024 18:12, Dave Stevenson wrote: > Hi Hans > > This is a very useful little series - thanks. > > On Wed, 21 Aug 2024 at 15:16, Hans Verkuil <hverkuil-cisco@xxxxxxxxx> wrote: >> >> Export InfoFrames to debugfs. > > I had a tc358743 to hand, so thought this warranted a quick test. > I think you have an off-by-one on the length that this exposes. > > log_status is giving me state > [ 428.600874] tc358743 10-000f: -----HDMI status----- > [ 428.600881] tc358743 10-000f: HDCP encrypted content: no > [ 428.600887] tc358743 10-000f: Input color space: RGB limited range > [ 428.601404] tc358743 10-000f: AV Mute: off > [ 428.601921] tc358743 10-000f: Deep color mode: 8-bits per channel > [ 428.604407] tc358743 10-000f: HDMI infoframe: Auxiliary Video > Information (AVI), version 2, length 13 > [ 428.604425] tc358743 10-000f: colorspace: RGB > [ 428.604433] tc358743 10-000f: scan mode: Underscan > [ 428.604441] tc358743 10-000f: colorimetry: No Data > [ 428.604449] tc358743 10-000f: picture aspect: 16:9 > [ 428.604456] tc358743 10-000f: active aspect: Same as Picture > [ 428.604463] tc358743 10-000f: itc: No Data > [ 428.604469] tc358743 10-000f: extended colorimetry: xvYCC 601 > [ 428.604476] tc358743 10-000f: quantization range: Limited > [ 428.604483] tc358743 10-000f: nups: Unknown Non-uniform Scaling > [ 428.604490] tc358743 10-000f: video code: 4 > [ 428.604497] tc358743 10-000f: ycc quantization range: Limited > [ 428.604505] tc358743 10-000f: hdmi content type: Graphics > [ 428.604511] tc358743 10-000f: pixel repeat: 0 > [ 428.604519] tc358743 10-000f: bar top 0, bottom 0, left 0, right 0 > > pi@bookworm64:~/edid-decode $ xxd /sys/kernel/debug/v4l2/tc358743\ > 10-000f/infoframes/avi > 00000000: 8202 0d2d 1228 0404 0000 0000 0000 0000 ...-.(.......... > > At the transmitting end I've got > pi@bookworm64:~/edid-decode $ sudo xxd > /sys/kernel/debug/dri/1/HDMI-A-1/infoframes/avi > 00000000: 8202 0d2d 1228 0404 0000 0000 0000 0000 ...-.(.......... > 00000010: 00 > > edid-decode -I decodes the latter fine, but aborts on the former. > Oddly the "fail" message from parse_if_hdr [1] doesn't get printed, it > just stops with I fixed this in edid-decode. > pi@bookworm64:~/edid-decode $ ./build/edid-decode -I > /sys/kernel/debug/v4l2/tc358743\ 10-000f/infoframes/avi > edid-decode InfoFrame (hex): > > 82 02 0d 2d 12 28 04 04 00 00 00 00 00 00 00 00 > > ---------------- > > HDMI InfoFrame Checksum: 0x2d > > AVI InfoFrame > Version: 2 > Length: 13 > > > [1] https://git.linuxtv.org/edid-decode.git/tree/parse-if.cpp#n21 > >> Signed-off-by: Hans Verkuil <hverkuil-cisco@xxxxxxxxx> >> --- >> drivers/media/i2c/tc358743.c | 36 +++++++++++++++++++++++++++++++++++- >> 1 file changed, 35 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/media/i2c/tc358743.c b/drivers/media/i2c/tc358743.c >> index 65d58ddf0287..c7652c0dbaeb 100644 >> --- a/drivers/media/i2c/tc358743.c >> +++ b/drivers/media/i2c/tc358743.c >> @@ -87,6 +87,10 @@ struct tc358743_state { >> struct timer_list timer; >> struct work_struct work_i2c_poll; >> >> + /* debugfs */ >> + struct dentry *debugfs_dir; >> + struct v4l2_debugfs_if *infoframes; >> + >> /* edid */ >> u8 edid_blocks_written; >> >> @@ -430,12 +434,35 @@ static void tc358743_erase_bksv(struct v4l2_subdev *sd) >> >> /* --------------- AVI infoframe --------------- */ >> >> +static ssize_t >> +tc358743_debugfs_if_read(u32 type, void *priv, struct file *filp, >> + char __user *ubuf, size_t count, loff_t *ppos) >> +{ >> + u8 buf[V4L2_DEBUGFS_IF_MAX_LEN] = {}; >> + struct v4l2_subdev *sd = priv; >> + int len; >> + >> + if (!is_hdmi(sd)) >> + return 0; >> + >> + if (type != V4L2_DEBUGFS_IF_AVI) >> + return 0; >> + >> + i2c_rd(sd, PK_AVI_0HEAD, buf, PK_AVI_16BYTE - PK_AVI_0HEAD + 1); >> + len = buf[2] + 3; > > tda1997x has len = buffer[2] + 4; > adv7842 and adv7604 have len = buf[2] + 1; and then return len + 3; > adv7511 has len = buffer[2]; and return len + 4; > > So I think this should be len = buf[2] + 4; Yes, that's correct. I forgot about the extra checksum byte that HDMI inserts in InfoFrames. Thank you, I've updated the patch and added a Tested-by with your email. Regards, Hans > > Doing so makes edid-decode happy. > pi@bookworm64:~/edid-decode $ sudo ./build/edid-decode -I > /sys/kernel/debug/v4l2/tc358743\ 10-000f/infoframes/avi > edid-decode InfoFrame (hex): > > 82 02 0d 2d 12 28 04 04 00 00 00 00 00 00 00 00 > 00 > > ---------------- > > HDMI InfoFrame Checksum: 0x2d > > AVI InfoFrame > Version: 2 > Length: 13 > VIC 4: 1280x720 60.000000 Hz 16:9 45.000 kHz 74.250000 MHz > Y: Color Component Sample Format: RGB > A: Active Format Information Present: Yes > B: Bar Data Present: Bar Data not present > S: Scan Information: Composed for an underscanned display > C: Colorimetry: No Data > M: Picture Aspect Ratio: 16:9 > R: Active Portion Aspect Ratio: 8 > ITC: IT Content: No Data > EC: Extended Colorimetry: xvYCC601 > Q: RGB Quantization Range: Limited Range > SC: Non-Uniform Picture Scaling: No Known non-uniform scaling > YQ: YCC Quantization Range: Limited Range > CN: IT Content Type: Graphics > PR: Pixel Data Repetition Count: 0 > Line Number of End of Top Bar: 0 > Line Number of Start of Bottom Bar: 0 > Pixel Number of End of Left Bar: 0 > Pixel Number of Start of Right Bar: 0 > > I haven't double checked the maths to ensure that we have read that > extra byte via i2c_rd though. > > Dave > >> + if (len > V4L2_DEBUGFS_IF_MAX_LEN) >> + len = -ENOENT; >> + if (len > 0) >> + len = simple_read_from_buffer(ubuf, count, ppos, buf, len); >> + return len < 0 ? 0 : len; >> +} >> + >> static void print_avi_infoframe(struct v4l2_subdev *sd) >> { >> struct i2c_client *client = v4l2_get_subdevdata(sd); >> struct device *dev = &client->dev; >> union hdmi_infoframe frame; >> - u8 buffer[HDMI_INFOFRAME_SIZE(AVI)]; >> + u8 buffer[HDMI_INFOFRAME_SIZE(AVI)] = {}; >> >> if (!is_hdmi(sd)) { >> v4l2_info(sd, "DVI-D signal - AVI infoframe not supported\n"); >> @@ -2161,6 +2188,11 @@ static int tc358743_probe(struct i2c_client *client) >> if (err < 0) >> goto err_work_queues; >> >> + state->debugfs_dir = debugfs_create_dir(sd->name, v4l2_debugfs_root()); >> + state->infoframes = v4l2_debugfs_if_alloc(state->debugfs_dir, >> + V4L2_DEBUGFS_IF_AVI, sd, >> + tc358743_debugfs_if_read); >> + >> v4l2_info(sd, "%s found @ 0x%x (%s)\n", client->name, >> client->addr << 1, client->adapter->name); >> >> @@ -2188,6 +2220,8 @@ static void tc358743_remove(struct i2c_client *client) >> flush_work(&state->work_i2c_poll); >> } >> cancel_delayed_work_sync(&state->delayed_work_enable_hotplug); >> + v4l2_debugfs_if_free(state->infoframes); >> + debugfs_remove_recursive(state->debugfs_dir); >> cec_unregister_adapter(state->cec_adap); >> v4l2_async_unregister_subdev(sd); >> v4l2_device_unregister_subdev(sd); >> -- >> 2.43.0 >> >