[PATCH 1/3] media: atomisp: don't print errors for ignored MBUS formats

public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed

* [PATCH 1/3] media: atomisp: don't print errors for ignored MBUS formats
@ 2021-10-31 11:25 Mauro Carvalho Chehab
  2021-10-31 11:25 ` [PATCH 2/3] media: atomisp: report the visible resolution Mauro Carvalho Chehab
  2021-10-31 11:25 ` [PATCH 3/3] media: atomisp: comment-out JPEG format Mauro Carvalho Chehab
  0 siblings, 2 replies; 5+ messages in thread
From: Mauro Carvalho Chehab @ 2021-10-31 11:25 UTC (permalink / raw)
  Cc: linuxarm, mauro.chehab, Mauro Carvalho Chehab, Arnd Bergmann,
	Dan Carpenter, Greg Kroah-Hartman, Kaixu Xia,
	Mauro Carvalho Chehab, Peter Zijlstra, Sakari Ailus,
	Thomas Gleixner, Tsuchiya Yuto, linux-kernel, linux-media,
	linux-staging

The bayer formats aren't currently available for userspace
to select: those are marked as IA_CSS_FRAME_FORMAT_RAW and
trying to get them result on binary firmware load errors:

	[74625.258097] atomisp-isp2 0000:00:03.0: Using binary isp_preview_var_isp2 (id 22), type 0, mode 1, continuous true
	[74625.258146] atomisp-isp2 0000:00:03.0: Seeking for binary failed at:
	[74625.258161] CPU: 3 PID: 2792 Comm: v4l2grab Tainted: G         C        5.15.0-rc4+ #77
	[74625.258190] Hardware name: ASUSTeK COMPUTER INC. T101HA/T101HA, BIOS T101HA.306 04/23/2019
	[74625.258208] Call Trace:
	[74625.258231]  dump_stack_lvl+0x46/0x5a
	[74625.258272]  ia_css_binary_find+0xa7d/0xcf0 [atomisp]
	[74625.258570]  load_preview_binaries+0x323/0x3c0 [atomisp]
	...
	[74625.265892] atomisp-isp2 0000:00:03.0: can't create streams
	[74625.265937] atomisp-isp2 0000:00:03.0: __get_frame_info 1604x1200 (padded to 0) returned -22
	[74625.265962] atomisp-isp2 0000:00:03.0: Can't set format on ISP. Error -22

As those formats are ignored by purpose, it doesn't make any sense
to print a message like:

	atomisp_enum_fmt_cap(): format for code 3007 not found.

for those.

Yet, some day it would be interesting to also report the bayer
formats, letting userspace to decode them on a different way.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
---
 drivers/staging/media/atomisp/pci/atomisp_ioctl.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
index f082d7a67875..8f0a9a69f075 100644
--- a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
+++ b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
@@ -829,8 +829,6 @@ static int atomisp_enum_fmt_cap(struct file *file, void *fh,
 		f->pixelformat = format->pixelformat;
 		return 0;
 	}
-	dev_err(isp->dev, "%s(): format for code %x not found.\n",
-		__func__, code.code);
 
 	return -EINVAL;
 }
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH 2/3] media: atomisp: report the visible resolution
  2021-10-31 11:25 [PATCH 1/3] media: atomisp: don't print errors for ignored MBUS formats Mauro Carvalho Chehab
@ 2021-10-31 11:25 ` Mauro Carvalho Chehab
  2021-10-31 11:25 ` [PATCH 3/3] media: atomisp: comment-out JPEG format Mauro Carvalho Chehab
  1 sibling, 0 replies; 5+ messages in thread
From: Mauro Carvalho Chehab @ 2021-10-31 11:25 UTC (permalink / raw)
  Cc: linuxarm, mauro.chehab, Mauro Carvalho Chehab, Arnd Bergmann,
	Dan Carpenter, Greg Kroah-Hartman, Ingo Molnar, Kaixu Xia,
	Mauro Carvalho Chehab, Peter Zijlstra, Sakari Ailus,
	Tsuchiya Yuto, linux-kernel, linux-media, linux-staging

The atomisp sensors and logic adds an extra pad lines/columns,
called "dvs envelope". It also uses an extra 12 lines/columns
at the sensor for BYT.

As those are not visible to userspace, the V4L2 API should
decrement such values when reporting the current resolution.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
---
 drivers/staging/media/atomisp/pci/atomisp_ioctl.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
index 8f0a9a69f075..d92d41dd0ade 100644
--- a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
+++ b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
@@ -876,6 +876,20 @@ static int atomisp_g_fmt_cap(struct file *file, void *fh,
 		v4l2_fill_pix_format(&f->fmt.pix, &fmt.format);
 
 		f->fmt.pix.pixelformat = fmtdesc.pixelformat;
+
+		/*
+		 * HACK: The atomisp does something different here, as it
+		 * seems to set the sensor to a slightly higher resolution than
+		 * the visible ones. That seems to be needed by atomisp's ISP
+		 * in order to properly handle the frames. So, s_fmt adds 16
+		 * extra columns/lines. See atomisp_subdev_set_selection().
+		 *
+		 * Yet, the V4L2 userspace API doesn't expect those, so it
+		 * needs to be decremented when reporting the visible
+		 * resolution to userspace.
+		 */
+		f->fmt.pix.width -= pad_w;
+		f->fmt.pix.height -= pad_h;
 	}
 
 	depth = atomisp_get_pixel_depth(f->fmt.pix.pixelformat);
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH 3/3] media: atomisp: comment-out JPEG format
  2021-10-31 11:25 [PATCH 1/3] media: atomisp: don't print errors for ignored MBUS formats Mauro Carvalho Chehab
  2021-10-31 11:25 ` [PATCH 2/3] media: atomisp: report the visible resolution Mauro Carvalho Chehab
@ 2021-10-31 11:25 ` Mauro Carvalho Chehab
  2021-10-31 13:03   ` Andy Shevchenko
  1 sibling, 1 reply; 5+ messages in thread
From: Mauro Carvalho Chehab @ 2021-10-31 11:25 UTC (permalink / raw)
  Cc: linuxarm, mauro.chehab, Mauro Carvalho Chehab, Alex Dewar,
	Aline Santana Cordeiro, Andy Shevchenko, Arnd Bergmann,
	Dan Carpenter, Greg Kroah-Hartman, Hans Verkuil, Ingo Molnar,
	Kaixu Xia, Mauro Carvalho Chehab, Peter Zijlstra, Sakari Ailus,
	Tomi Valkeinen, Tsuchiya Yuto, linux-kernel, linux-media,
	linux-staging

This is already disabled on some parts of the code, and trying
to use it with current firmware causes an error:

[   53.799946] atomisp-isp2 0000:00:03.0: can't create streams
[   53.799962] atomisp-isp2 0000:00:03.0: __get_frame_info 1600x900 (padded to 0) returned -22

So, completely disable reporting it.

Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
---
 drivers/staging/media/atomisp/pci/atomisp_cmd.c   | 2 ++
 drivers/staging/media/atomisp/pci/atomisp_ioctl.c | 7 +++----
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/staging/media/atomisp/pci/atomisp_cmd.c b/drivers/staging/media/atomisp/pci/atomisp_cmd.c
index ff8640314f65..feb75491a273 100644
--- a/drivers/staging/media/atomisp/pci/atomisp_cmd.c
+++ b/drivers/staging/media/atomisp/pci/atomisp_cmd.c
@@ -1980,9 +1980,11 @@ v4l2_fmt_to_sh_fmt(u32 fmt)
 		return IA_CSS_FRAME_FORMAT_RGBA888;
 	case V4L2_PIX_FMT_RGB565:
 		return IA_CSS_FRAME_FORMAT_RGB565;
+#if 0
 	case V4L2_PIX_FMT_JPEG:
 	case V4L2_PIX_FMT_CUSTOM_M10MO_RAW:
 		return IA_CSS_FRAME_FORMAT_BINARY_8;
+#endif
 	case V4L2_PIX_FMT_SBGGR16:
 	case V4L2_PIX_FMT_SBGGR10:
 	case V4L2_PIX_FMT_SGBRG10:
diff --git a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
index d92d41dd0ade..936516daec7d 100644
--- a/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
+++ b/drivers/staging/media/atomisp/pci/atomisp_ioctl.c
@@ -492,23 +492,22 @@ const struct atomisp_format_bridge atomisp_output_fmts[] = {
 		.mbus_code = MEDIA_BUS_FMT_BGR565_2X8_LE,
 		.sh_fmt = IA_CSS_FRAME_FORMAT_RGB565,
 		.description = "16 RGB 5-6-5"
+#if 0
 	}, {
 		.pixelformat = V4L2_PIX_FMT_JPEG,
 		.depth = 8,
 		.mbus_code = MEDIA_BUS_FMT_JPEG_1X8,
 		.sh_fmt = IA_CSS_FRAME_FORMAT_BINARY_8,
 		.description = "JPEG"
-	},
-#if 0
-	{
+	}, {
 		/* This is a custom format being used by M10MO to send the RAW data */
 		.pixelformat = V4L2_PIX_FMT_CUSTOM_M10MO_RAW,
 		.depth = 8,
 		.mbus_code = V4L2_MBUS_FMT_CUSTOM_M10MO_RAW,
 		.sh_fmt = IA_CSS_FRAME_FORMAT_BINARY_8,
 		.description = "Custom RAW for M10MO"
-	},
 #endif
+	},
 };
 
 const struct atomisp_format_bridge *
-- 
2.31.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [PATCH 3/3] media: atomisp: comment-out JPEG format
  2021-10-31 11:25 ` [PATCH 3/3] media: atomisp: comment-out JPEG format Mauro Carvalho Chehab
@ 2021-10-31 13:03   ` Andy Shevchenko
  2021-11-01  8:11     ` Mauro Carvalho Chehab
  0 siblings, 1 reply; 5+ messages in thread
From: Andy Shevchenko @ 2021-10-31 13:03 UTC (permalink / raw)
  To: Mauro Carvalho Chehab
  Cc: Linuxarm, mauro.chehab, Alex Dewar, Aline Santana Cordeiro,
	Andy Shevchenko, Arnd Bergmann, Dan Carpenter, Greg Kroah-Hartman,
	Hans Verkuil, Ingo Molnar, Kaixu Xia, Mauro Carvalho Chehab,
	Peter Zijlstra, Sakari Ailus, Tomi Valkeinen, Tsuchiya Yuto,
	Linux Kernel Mailing List, Linux Media Mailing List,
	linux-staging

On Sun, Oct 31, 2021 at 1:30 PM Mauro Carvalho Chehab
<mchehab+huawei@kernel.org> wrote:
>
> This is already disabled on some parts of the code, and trying
> to use it with current firmware causes an error:
>
> [   53.799946] atomisp-isp2 0000:00:03.0: can't create streams
> [   53.799962] atomisp-isp2 0000:00:03.0: __get_frame_info 1600x900 (padded to 0) returned -22
>
> So, completely disable reporting it.


Can't we simply remove all these pieces altogether?
The rationale:
 - code is still kept in the index of the Git
 - avoid robots and humans of sending patches against dead code
 - leaving dead code (even in staging) is not what we want in the long term

-- 
With Best Regards,
Andy Shevchenko

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH 3/3] media: atomisp: comment-out JPEG format
  2021-10-31 13:03   ` Andy Shevchenko
@ 2021-11-01  8:11     ` Mauro Carvalho Chehab
  0 siblings, 0 replies; 5+ messages in thread
From: Mauro Carvalho Chehab @ 2021-11-01  8:11 UTC (permalink / raw)
  To: Andy Shevchenko
  Cc: Linuxarm, mauro.chehab, Alex Dewar, Aline Santana Cordeiro,
	Andy Shevchenko, Arnd Bergmann, Dan Carpenter, Greg Kroah-Hartman,
	Hans Verkuil, Ingo Molnar, Kaixu Xia, Mauro Carvalho Chehab,
	Peter Zijlstra, Sakari Ailus, Tomi Valkeinen, Tsuchiya Yuto,
	Linux Kernel Mailing List, Linux Media Mailing List,
	linux-staging

Hi Andy,

Em Sun, 31 Oct 2021 15:03:39 +0200
Andy Shevchenko <andy.shevchenko@gmail.com> escreveu:

> On Sun, Oct 31, 2021 at 1:30 PM Mauro Carvalho Chehab
> <mchehab+huawei@kernel.org> wrote:
> >
> > This is already disabled on some parts of the code, and trying
> > to use it with current firmware causes an error:
> >
> > [   53.799946] atomisp-isp2 0000:00:03.0: can't create streams
> > [   53.799962] atomisp-isp2 0000:00:03.0: __get_frame_info 1600x900 (padded to 0) returned -22
> >
> > So, completely disable reporting it.  
> 
> 
> Can't we simply remove all these pieces altogether?
> The rationale:
>  - code is still kept in the index of the Git
>  - avoid robots and humans of sending patches against dead code
>  - leaving dead code (even in staging) is not what we want in the long term

Good point.

IMO, it is a little premature to drop those. I mean, clearly at least
some firmware for the ISP2 is capable of encoding in JPEG format,
but the code which creates such pipeline is not selecting the right
binary. 

Btw, the same issue also happens here for BG10 fourcc, which is the
real output of the sensor:

	$ v4l2grab -f BG10 -x 1600 -y 1200 -d /dev/video2 -u -n 1 -R
	Device caps: CAPTURE OUTPUT STREAMING EXT_PIX_FORMAT DEVICE_CAPS STREAMING 
	Driver caps: CAPTURE OUTPUT STREAMING EXT_PIX_FORMAT STREAMING 
	xioctl(5): error 22, Invalid argument

And the relevant logs from firmware load time:

	 atomisp-isp2 0000:00:03.0: binary #0  type SP: sp
	 atomisp-isp2 0000:00:03.0: binary #1  type ISP (Normal), binary id is  0: isp_copy_var
	 atomisp-isp2 0000:00:03.0: binary #2  type ISP (Normal), binary id is  2: isp_vf_pp_full
	 atomisp-isp2 0000:00:03.0: binary #3  type ISP (Normal), binary id is  3: isp_vf_pp_opt
	 atomisp-isp2 0000:00:03.0: binary #4  type ISP (Normal), binary id is 60: isp_capture_pp_var_bli
	 atomisp-isp2 0000:00:03.0: binary #5  type ISP (Normal), binary id is 61: isp_capture_pp_ldc
	 atomisp-isp2 0000:00:03.0: binary #6  type ISP (Normal), binary id is  5: isp_capture_pp_var
	 atomisp-isp2 0000:00:03.0: binary #7  type ISP (Normal), binary id is  4: isp_yuv_scale_var
	 atomisp-isp2 0000:00:03.0: binary #8  type ISP (Normal), binary id is  6: isp_preisp_var
	 atomisp-isp2 0000:00:03.0: binary #9  type ISP (Normal), binary id is  7: isp_preisp_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #10 type ISP (Normal), binary id is 58: isp_pre_de_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #11 type ISP (Normal), binary id is  8: isp_gdc_var
	 atomisp-isp2 0000:00:03.0: binary #12 type ISP (Normal), binary id is 11: isp_anr_var
	 atomisp-isp2 0000:00:03.0: binary #13 type ISP (Normal), binary id is 12: isp_anr_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #14 type ISP (Normal), binary id is  9: isp_postisp_var
	 atomisp-isp2 0000:00:03.0: binary #15 type ISP (Normal), binary id is 10: isp_postisp_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #16 type ISP (Normal), binary id is 15: isp_preview_dec
	 atomisp-isp2 0000:00:03.0: binary #17 type ISP (Normal), binary id is 16: isp_preview_cont_bds125_isp2
	 atomisp-isp2 0000:00:03.0: binary #18 type ISP (Normal), binary id is 17: isp_preview_cont_dpc_bds150_isp2
	 atomisp-isp2 0000:00:03.0: binary #19 type ISP (Normal), binary id is 19: isp_preview_cont_dpc_bds200_isp2
	 atomisp-isp2 0000:00:03.0: binary #20 type ISP (Normal), binary id is 18: isp_preview_cont_bds150_isp2
	 atomisp-isp2 0000:00:03.0: binary #21 type ISP (Normal), binary id is 20: isp_preview_cont_bds200_isp2
	 atomisp-isp2 0000:00:03.0: binary #22 type ISP (Normal), binary id is 21: isp_preview_var
	 atomisp-isp2 0000:00:03.0: binary #23 type ISP (Normal), binary id is 22: isp_preview_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #24 type ISP (Normal), binary id is 24: isp_primary_var
	 atomisp-isp2 0000:00:03.0: binary #25 type ISP (Normal), binary id is 25: isp_primary_var_isp2
	 atomisp-isp2 0000:00:03.0: binary #26 type ISP (Normal), binary id is 26: isp_primary_small
	 atomisp-isp2 0000:00:03.0: binary #27 type ISP (Normal), binary id is 27: isp_primary_striped
	 atomisp-isp2 0000:00:03.0: binary #28 type ISP (Normal), binary id is 28: isp_primary_striped_isp2
	 atomisp-isp2 0000:00:03.0: binary #29 type ISP (Normal), binary id is 29: isp_primary_8mp
	 atomisp-isp2 0000:00:03.0: binary #30 type ISP (Normal), binary id is 30: isp_primary_14mp
	 atomisp-isp2 0000:00:03.0: binary #31 type ISP (Normal), binary id is 31: isp_primary_16mp
	 atomisp-isp2 0000:00:03.0: binary #32 type ISP (Normal), binary id is 33: isp_primary_isp261_stage0
	 atomisp-isp2 0000:00:03.0: binary #33 type ISP (Normal), binary id is 34: isp_primary_isp261_stage1
	 atomisp-isp2 0000:00:03.0: binary #34 type ISP (Normal), binary id is 35: isp_primary_isp261_stage2
	 atomisp-isp2 0000:00:03.0: binary #35 type ISP (Normal), binary id is 36: isp_primary_isp261_stage3
	 atomisp-isp2 0000:00:03.0: binary #36 type ISP (Normal), binary id is 37: isp_primary_isp261_stage4
	 atomisp-isp2 0000:00:03.0: binary #37 type ISP (Normal), binary id is 38: isp_primary_isp261_stage5
	 atomisp-isp2 0000:00:03.0: binary #38 type ISP (Normal), binary id is 42: isp_video_dz
	 atomisp-isp2 0000:00:03.0: binary #39 type ISP (Normal), binary id is 44: isp_video_high
	 atomisp-isp2 0000:00:03.0: binary #40 type ISP (Normal), binary id is 45: isp_video_nodz
	 atomisp-isp2 0000:00:03.0: binary #41 type ISP (Normal), binary id is 46: isp_video_cont_multibds_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #42 type ISP (Normal), binary id is 47: isp_video_cont_bds_300_600_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #43 type ISP (Normal), binary id is 48: isp_video_cont_dpc_bds150_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #44 type ISP (Normal), binary id is 50: isp_video_cont_dpc_bds200_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #45 type ISP (Normal), binary id is 49: isp_video_cont_bds150_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #46 type ISP (Normal), binary id is 51: isp_video_cont_bds200_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #47 type ISP (Normal), binary id is 52: isp_video_cont_nobds_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #48 type ISP (Normal), binary id is 53: isp_video_dz_isp2_min
	 atomisp-isp2 0000:00:03.0: binary #49 type ISP (Normal), binary id is 54: isp_video_dz_isp2
	 atomisp-isp2 0000:00:03.0: binary #50 type ISP (Normal), binary id is 55: isp_video_lp_isp2

And when trying to build the pipeline:

	 ia_css_binary_find() selected = 0000000004885f71, mode = 1 ID = 22
	 ia_css_binary_find() leave: return_err=0
	 atomisp-isp2 0000:00:03.0: Using binary isp_preview_var_isp2 (id 22), type 0, mode 1, continuous true
	 load_preview_binaries() num_invalid_frames=1 dvs_frame_delay=1
	 ia_css_binary_find() enter: descr=00000000d70b101f, (mode=9), binary=00000000bceb2e4a
	 BINARY INFO:
	 0:
	  Name:isp_copy_var Type:0 Cont:0
	 1:
	  Name:isp_preview_var_isp2 Type:0 Cont:1
	  Name:isp_preview_var Type:0 Cont:1
	  Name:isp_preview_cont_bds200_isp2 Type:0 Cont:1
	  Name:isp_preview_cont_bds150_isp2 Type:0 Cont:1
	  Name:isp_preview_cont_dpc_bds200_isp2 Type:0 Cont:1
	  Name:isp_preview_cont_dpc_bds150_isp2 Type:0 Cont:1
	  Name:isp_preview_cont_bds125_isp2 Type:0 Cont:1
	  Name:isp_preview_dec Type:0 Cont:1
	 2:
	  Name:isp_primary_16mp Type:0 Cont:0
	  Name:isp_primary_14mp Type:0 Cont:0
	  Name:isp_primary_8mp Type:0 Cont:0
	  Name:isp_primary_striped_isp2 Type:0 Cont:1
	  Name:isp_primary_striped Type:0 Cont:1
	  Name:isp_primary_small Type:0 Cont:0
	  Name:isp_primary_var_isp2 Type:0 Cont:1
	  Name:isp_primary_var Type:0 Cont:0
	 3:
	  Name:isp_video_lp_isp2 Type:0 Cont:0
	  Name:isp_video_dz_isp2 Type:0 Cont:0
	  Name:isp_video_dz_isp2_min Type:0 Cont:0
	  Name:isp_video_cont_nobds_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_bds200_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_bds150_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_dpc_bds200_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_dpc_bds150_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_bds_300_600_isp2_min Type:0 Cont:1
	  Name:isp_video_cont_multibds_isp2_min Type:0 Cont:1
	  Name:isp_video_nodz Type:0 Cont:0
	  Name:isp_video_high Type:0 Cont:0
	  Name:isp_video_dz Type:0 Cont:0
	 4:
	  Name:isp_preisp_var_isp2 Type:0 Cont:0
	  Name:isp_preisp_var Type:0 Cont:0
	 5:
	  Name:isp_gdc_var Type:0 Cont:0
	 6:
	  Name:isp_postisp_var_isp2 Type:0 Cont:0
	  Name:isp_postisp_var Type:0 Cont:0
	 7:
	  Name:isp_anr_var_isp2 Type:0 Cont:0
	  Name:isp_anr_var Type:0 Cont:0
	 8:
	  Name:isp_yuv_scale_var Type:0 Cont:1
	  Name:isp_capture_pp_var Type:0 Cont:1
	  Name:isp_capture_pp_ldc Type:0 Cont:1
	  Name:isp_capture_pp_var_bli Type:0 Cont:1
	 9:
	  Name:isp_vf_pp_opt Type:0 Cont:1
	  Name:isp_vf_pp_full Type:0 Cont:1
	 10:
	  Name:isp_pre_de_var_isp2 Type:0 Cont:1
	 11:
	  Name:isp_primary_isp261_stage0 Type:0 Cont:1
	 12:
	  Name:isp_primary_isp261_stage1 Type:0 Cont:1
	 13:
	  Name:isp_primary_isp261_stage2 Type:0 Cont:1
	 14:
	  Name:isp_primary_isp261_stage3 Type:0 Cont:1
	 15:
	  Name:isp_primary_isp261_stage4 Type:0 Cont:1
	 16:
	  Name:isp_primary_isp261_stage5 Type:0 Cont:1
	 ia_css_binary_find() candidate = 00000000fc264b22, mode = 9 ID = 3
	 ia_css_binary_find() [1676] continue: !0
	 ia_css_binary_find() candidate = 0000000041fcca04, mode = 9 ID = 2
	 ia_css_binary_find() [1676] continue: !0
	 ia_css_binary_find() selected = 0000000000000000, mode = 0 ID = 0
	 ia_css_binary_find() leave: return_err=-22
	 atomisp-isp2 0000:00:03.0: Seeking for binary failed at:
	 CPU: 0 PID: 1343 Comm: v4l2grab Tainted: G         C        5.15.0-rc4+ #77
	 Hardware name: ASUSTeK COMPUTER INC. T101HA/T101HA, BIOS T101HA.306 04/23/2019
	 Call Trace:
	  dump_stack_lvl+0x46/0x5a
	  ia_css_binary_find+0xa7d/0xcf0 [atomisp]
	  ? vprintk_emit+0x112/0x240
	  load_preview_binaries+0x323/0x3c0 [atomisp]
	  ia_css_stream_create+0xddf/0x18e0 [atomisp]
	  __create_streams+0x255/0xe20 [atomisp]
	  ? __create_pipes+0x2b7/0x4f0 [atomisp]
	  __get_frame_info+0x9d/0x2a0 [atomisp]

See, basically the logic which seeks for a firmware that would allow
to just copy the MBUS format without touching it failed somewhere.

As no binary should be required to convert from BG10, as this is the
native format, I strongly suspect that the logic which sets the pipeline
has some bug.

So, I wouldn't drop support for Bayer and JPEG for now. Instead, the 
best seems to comment out the broken stuff for now, while we're dealing
with more relevant issues. We can return later on to address such
formats.

Regards,
Mauro

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2021-11-01  8:12 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2021-10-31 11:25 [PATCH 1/3] media: atomisp: don't print errors for ignored MBUS formats Mauro Carvalho Chehab
2021-10-31 11:25 ` [PATCH 2/3] media: atomisp: report the visible resolution Mauro Carvalho Chehab
2021-10-31 11:25 ` [PATCH 3/3] media: atomisp: comment-out JPEG format Mauro Carvalho Chehab
2021-10-31 13:03   ` Andy Shevchenko
2021-11-01  8:11     ` Mauro Carvalho Chehab

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox