Skip to contentSkip to conversation

media: imx: imx7-media-csi: Fix applying format constraints

PATCH In review Fabio Estevam 4 participants 7 messages Jul 20, 2023 – Jul 25, 2023

Subsystems (3)
lore
Share & tools
lore

Fabio Estevam

author reviewer
#0 lore raw

PATCH v2 1/1 media: imx: imx7-media-csi: Fix applying format constraints

From: Fabio Estevam <[email protected]>

v4l_bound_align_image() aligns to a multiple power of 2 of walign, but the result only needs to be a multiple of walign.

This causes a 640x480 sensor that used to report:

	Width/Height      : 640/480

to incorrectly report:

	Width/Height      : 768/480

Fix this problem by doing the correct alignment via clamp_roundup().

Fixes 6f482c4729d9 ("media: imx: imx7-media-csi: Get rid of superfluous call to imx7_csi_mbus_fmt_to_pix_fmt")

---
Changes since v1:
- Export clamp_roundup().

 drivers/media/platform/nxp/imx7-media-csi.c | 4 ++--
 drivers/media/v4l2-core/v4l2-common.c       | 5 +++--
 include/media/v4l2-common.h                 | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)
drivers/media/platform/nxp/imx7-media-csi.c +2 -2
index 0bd2613b9320..f3c506fc19c4 100644
@@ -1137,8 +1137,8 @@ __imx7_csi_video_try_fmt(struct v4l2_pix_format *pixfmt,
 	 * TODO: Implement configurable stride support.
 	 */
 	walign = 8 * 8 / cc->bpp;
-	v4l_bound_align_image(&pixfmt->width, 1, 0xffff, walign,
-			      &pixfmt->height, 1, 0xffff, 1, 0);
+	pixfmt->width = clamp_roundup(pixfmt->width, 1, 0xffff, walign);
+	pixfmt->height = clamp_roundup(pixfmt->height, 1, 0xffff, 1);
 
 	pixfmt->bytesperline = pixfmt->width * cc->bpp / 8;
 	pixfmt->sizeimage = pixfmt->bytesperline * pixfmt->height;
drivers/media/v4l2-core/v4l2-common.c +3 -2
index bee1535b04d3..3e8c16bcb0f6 100644
@@ -106,8 +106,8 @@ static unsigned int clamp_align(unsigned int x, unsigned int min,
 	return x;
 }
 
-static unsigned int clamp_roundup(unsigned int x, unsigned int min,
-				   unsigned int max, unsigned int alignment)
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment)
 {
 	x = clamp(x, min, max);
 	if (alignment)
@@ -115,6 +115,7 @@ static unsigned int clamp_roundup(unsigned int x, unsigned int min,
 
 	return x;
 }
+EXPORT_SYMBOL(clamp_roundup);
 
 void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
 			   unsigned int walign,
include/media/v4l2-common.h +2 -0
index d278836fd9cb..7059b99f4afa 100644
@@ -521,6 +521,8 @@ int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 pixelformat,
 		     u32 width, u32 height);
 int v4l2_fill_pixfmt_mp(struct v4l2_pix_format_mplane *pixfmt, u32 pixelformat,
 			u32 width, u32 height);
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment);
 
 /**
  * v4l2_get_link_freq - Get link rate from transmitter
Signature · 1 line
2.34.1
Acked-by on v2 1/1
#1 lore raw
Quoted 82 lines · Fabio Estevam

From: Fabio Estevam <[email protected]>

v4l_bound_align_image() aligns to a multiple power of 2 of walign, but the result only needs to be a multiple of walign.

This causes a 640x480 sensor that used to report:

        Width/Height      : 640/480

to incorrectly report:

        Width/Height      : 768/480

Fix this problem by doing the correct alignment via clamp_roundup().

Fixes 6f482c4729d9 ("media: imx: imx7-media-csi: Get rid of superfluous call to imx7_csi_mbus_fmt_to_pix_fmt")

---
Changes since v1:
- Export clamp_roundup().

 drivers/media/platform/nxp/imx7-media-csi.c | 4 ++--
 drivers/media/v4l2-core/v4l2-common.c       | 5 +++--
 include/media/v4l2-common.h                 | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)
drivers/media/platform/nxp/imx7-media-csi.c +2 -2
index 0bd2613b9320..f3c506fc19c4 100644
@@ -1137,8 +1137,8 @@ __imx7_csi_video_try_fmt(struct v4l2_pix_format *pixfmt,
         * TODO: Implement configurable stride support.
         */
        walign = 8 * 8 / cc->bpp;
-       v4l_bound_align_image(&pixfmt->width, 1, 0xffff, walign,
-                             &pixfmt->height, 1, 0xffff, 1, 0);
+       pixfmt->width = clamp_roundup(pixfmt->width, 1, 0xffff, walign);
+       pixfmt->height = clamp_roundup(pixfmt->height, 1, 0xffff, 1);
 
        pixfmt->bytesperline = pixfmt->width * cc->bpp / 8;
        pixfmt->sizeimage = pixfmt->bytesperline * pixfmt->height;
drivers/media/v4l2-core/v4l2-common.c +3 -2
index bee1535b04d3..3e8c16bcb0f6 100644
@@ -106,8 +106,8 @@ static unsigned int clamp_align(unsigned int x, unsigned int min,
        return x;
 }
 
-static unsigned int clamp_roundup(unsigned int x, unsigned int min,
-                                  unsigned int max, unsigned int alignment)
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+                          unsigned int max, unsigned int alignment)
 {
        x = clamp(x, min, max);
        if (alignment)
@@ -115,6 +115,7 @@ static unsigned int clamp_roundup(unsigned int x, unsigned int min,
 
        return x;
 }
+EXPORT_SYMBOL(clamp_roundup);
 
 void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
                           unsigned int walign,
include/media/v4l2-common.h +2 -1
index d278836fd9cb..7059b99f4afa 100644
@@ -521,6 +521,8 @@ int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 pixelformat,
                     u32 width, u32 height);
 int v4l2_fill_pixfmt_mp(struct v4l2_pix_format_mplane *pixfmt, u32 pixelformat,
                        u32 width, u32 height);
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+                          unsigned int max, unsigned int alignment);
 
 /**
  * v4l2_get_link_freq - Get link rate from transmitter
--
Quoted 2 lines

2.34.1

Thanks Alexander and Fabio!

tested on a imx8mm-gw72xx-0x with imx8mm-gw72xx-0x-imx219 overlay

Best regards,

Tim

#2 lore raw
In reply to Fabio Estevam

Am Freitag, 21. Juli 2023, 00:25:43 CEST schrieb Fabio Estevam:

Quoted 52 lines

From: Fabio Estevam <[email protected]>

v4l_bound_align_image() aligns to a multiple power of 2 of walign, but the result only needs to be a multiple of walign.

This causes a 640x480 sensor that used to report:

	Width/Height      : 640/480

to incorrectly report:

	Width/Height      : 768/480

Fix this problem by doing the correct alignment via clamp_roundup().

Fixes 6f482c4729d9 ("media: imx: imx7-media-csi: Get rid of superfluous

call to imx7_csi_mbus_fmt_to_pix_fmt") Co-developed-by: Alexander Stein <[email protected]> Signed-off-by: Alexander Stein <[email protected]>

---
Changes since v1:
- Export clamp_roundup().

 drivers/media/platform/nxp/imx7-media-csi.c | 4 ++--
 drivers/media/v4l2-core/v4l2-common.c       | 5 +++--
 include/media/v4l2-common.h                 | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)
(quoted diff) +0 -0

b/drivers/media/platform/nxp/imx7-media-csi.c index 0bd2613b9320..f3c506fc19c4 100644

drivers/media/platform/nxp/imx7-media-csi.c +2 -2
@@ -1137,8 +1137,8 @@ __imx7_csi_video_try_fmt(struct v4l2_pix_format
 *pixfmt, * TODO: Implement configurable stride support.
 	 */
 	walign = 8 * 8 / cc->bpp;
-	v4l_bound_align_image(&pixfmt->width, 1, 0xffff, walign,
-			      &pixfmt->height, 1, 0xffff, 1, 0);
+	pixfmt->width = clamp_roundup(pixfmt->width, 1, 0xffff, walign);
+	pixfmt->height = clamp_roundup(pixfmt->height, 1, 0xffff, 1);
 
 	pixfmt->bytesperline = pixfmt->width * cc->bpp / 8;
 	pixfmt->sizeimage = pixfmt->bytesperline * pixfmt->height;
(quoted diff) +0 -0

b/drivers/media/v4l2-core/v4l2-common.c index bee1535b04d3..3e8c16bcb0f6 100644

drivers/media/v4l2-core/v4l2-common.c static unsigned int clamp_align(unsigned int x, unsigned · line 110

Quoted 5 lines
drivers/media/v4l2-core/v4l2-common.c +0 -2
@@ -106,8 +106,8 @@ static unsigned int clamp_align(unsigned int x, unsigned
 int min, return x;
 }
 
-static unsigned int clamp_roundup(unsigned int x, unsigned int min,
-				   unsigned int max, unsigned int 

alignment)

Quoted 19 lines
drivers/media/v4l2-core/v4l2-common.c +3 -0
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment)
 {
 	x = clamp(x, min, max);
 	if (alignment)
@@ -115,6 +115,7 @@ static unsigned int clamp_roundup(unsigned int x,
 unsigned int min,
 
 	return x;
 }
+EXPORT_SYMBOL(clamp_roundup);
 
 void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
drivers/media/v4l2-core/v4l2-common.c +0 -0
 			   unsigned int walign,
include/media/v4l2-common.h +0 -0
index d278836fd9cb..7059b99f4afa 100644

include/media/v4l2-common.h int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 · line 528

Quoted 8 lines
include/media/v4l2-common.h +2 -0
@@ -521,6 +521,8 @@ int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32
 pixelformat, u32 width, u32 height);
 int v4l2_fill_pixfmt_mp(struct v4l2_pix_format_mplane *pixfmt, u32
 pixelformat, u32 width, u32 height);
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment);
 
 /**
  * v4l2_get_link_freq - Get link rate from transmitter

I expect this to work, but I'm not so sure about exporting clamp_roundup(). I would rather use v4l2_apply_frmsize_constraints() instead, but this requires additional work, refer to the TODO in imx7_csi_video_enum_framesizes(). See also my first (broken) try in [1].

As a short term fix, using clamp_roundup() is okay to me. But I'm thinking about using a copy of clamp_roundup() as in v1 instead,so to avoid an eventual unexport of clamp_roundup in the future. Maybe this concern doesn't matter, I guess the maintainers can tell.

Best regards,
Alexander

[1] https://lore.kernel.org/linux-media/[email protected]/T/
#ma88811098cb96f3698a8248d903f0c3455febec7

Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
http://www.tq-group.com/
maintainer
#3 lore raw
Quoted 84 lines · Alexander Stein

Am Freitag, 21. Juli 2023, 00:25:43 CEST schrieb Fabio Estevam:

Quoted 52 lines

From: Fabio Estevam <[email protected]>

v4l_bound_align_image() aligns to a multiple power of 2 of walign, but the result only needs to be a multiple of walign.

This causes a 640x480 sensor that used to report:

	Width/Height      : 640/480

to incorrectly report:

	Width/Height      : 768/480

Fix this problem by doing the correct alignment via clamp_roundup().

Fixes 6f482c4729d9 ("media: imx: imx7-media-csi: Get rid of superfluous

call to imx7_csi_mbus_fmt_to_pix_fmt") Co-developed-by: Alexander Stein <[email protected]> Signed-off-by: Alexander Stein <[email protected]>

---
Changes since v1:
- Export clamp_roundup().

 drivers/media/platform/nxp/imx7-media-csi.c | 4 ++--
 drivers/media/v4l2-core/v4l2-common.c       | 5 +++--
 include/media/v4l2-common.h                 | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)
(quoted diff) +0 -0

b/drivers/media/platform/nxp/imx7-media-csi.c index 0bd2613b9320..f3c506fc19c4 100644

drivers/media/platform/nxp/imx7-media-csi.c +2 -2
@@ -1137,8 +1137,8 @@ __imx7_csi_video_try_fmt(struct v4l2_pix_format
 *pixfmt, * TODO: Implement configurable stride support.
 	 */
 	walign = 8 * 8 / cc->bpp;
-	v4l_bound_align_image(&pixfmt->width, 1, 0xffff, walign,
-			      &pixfmt->height, 1, 0xffff, 1, 0);
+	pixfmt->width = clamp_roundup(pixfmt->width, 1, 0xffff, walign);
+	pixfmt->height = clamp_roundup(pixfmt->height, 1, 0xffff, 1);
 
 	pixfmt->bytesperline = pixfmt->width * cc->bpp / 8;
 	pixfmt->sizeimage = pixfmt->bytesperline * pixfmt->height;
(quoted diff) +0 -0

b/drivers/media/v4l2-core/v4l2-common.c index bee1535b04d3..3e8c16bcb0f6 100644

drivers/media/v4l2-core/v4l2-common.c static unsigned int clamp_align(unsigned int x, unsigned · line 110

Quoted 5 lines
drivers/media/v4l2-core/v4l2-common.c +0 -2
@@ -106,8 +106,8 @@ static unsigned int clamp_align(unsigned int x, unsigned
 int min, return x;
 }
 
-static unsigned int clamp_roundup(unsigned int x, unsigned int min,
-				   unsigned int max, unsigned int 

alignment)

Quoted 25 lines
drivers/media/v4l2-core/v4l2-common.c +3 -0
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment)
 {
 	x = clamp(x, min, max);
 	if (alignment)
@@ -115,6 +115,7 @@ static unsigned int clamp_roundup(unsigned int x,
 unsigned int min,
 
 	return x;
 }
+EXPORT_SYMBOL(clamp_roundup);
 
 void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
drivers/media/v4l2-core/v4l2-common.c +0 -0
 			   unsigned int walign,
include/media/v4l2-common.h +2 -0
index d278836fd9cb..7059b99f4afa 100644
@@ -521,6 +521,8 @@ int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32
 pixelformat, u32 width, u32 height);
 int v4l2_fill_pixfmt_mp(struct v4l2_pix_format_mplane *pixfmt, u32
 pixelformat, u32 width, u32 height);
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment);
 
Quoted 4 lines

include/media/v4l2-common.h int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 pixelformat, · line 528

Quoted 2 lines
include/media/v4l2-common.h +0 -0
 /**
  * v4l2_get_link_freq - Get link rate from transmitter

I expect this to work, but I'm not so sure about exporting clamp_roundup().

I agree. The function name is too generic to be exported by V4L2. It should be moved to include/linux/minmax.h (and the implementation to a corresponding source file if it's too large to be inlined) if we want to export it.

Quoted 2 lines

I would rather use v4l2_apply_frmsize_constraints() instead, but this requires additional work, refer to the TODO in imx7_csi_video_enum_framesizes().

That TODO is definitely outdated, it would be nice to address it. I'll submit a patch.

Quoted 12 lines

See also my first (broken) try in [1].

As a short term fix, using clamp_roundup() is okay to me. But I'm thinking about using a copy of clamp_roundup() as in v1 instead,so to avoid an eventual unexport of clamp_roundup in the future. Maybe this concern doesn't matter, I guess the maintainers can tell.

Best regards,
Alexander

[1] https://lore.kernel.org/linux-media/[email protected]/T/
#ma88811098cb96f3698a8248d903f0c3455febec7

Signature · 3 lines
Regards,

Laurent Pinchart
maintainer
#4 lore raw
In reply to Fabio Estevam

Hi Fabio,

Thank you for the patch.

Quoted 72 lines · Fabio Estevam

From: Fabio Estevam <[email protected]>

v4l_bound_align_image() aligns to a multiple power of 2 of walign, but the result only needs to be a multiple of walign.

This causes a 640x480 sensor that used to report:

	Width/Height      : 640/480

to incorrectly report:

	Width/Height      : 768/480

Fix this problem by doing the correct alignment via clamp_roundup().

Fixes 6f482c4729d9 ("media: imx: imx7-media-csi: Get rid of superfluous call to imx7_csi_mbus_fmt_to_pix_fmt")

---
Changes since v1:
- Export clamp_roundup().

 drivers/media/platform/nxp/imx7-media-csi.c | 4 ++--
 drivers/media/v4l2-core/v4l2-common.c       | 5 +++--
 include/media/v4l2-common.h                 | 2 ++
 3 files changed, 7 insertions(+), 4 deletions(-)
drivers/media/platform/nxp/imx7-media-csi.c +2 -2
index 0bd2613b9320..f3c506fc19c4 100644
@@ -1137,8 +1137,8 @@ __imx7_csi_video_try_fmt(struct v4l2_pix_format *pixfmt,
 	 * TODO: Implement configurable stride support.
 	 */framesize
 	walign = 8 * 8 / cc->bpp;
-	v4l_bound_align_image(&pixfmt->width, 1, 0xffff, walign,
-			      &pixfmt->height, 1, 0xffff, 1, 0);
+	pixfmt->width = clamp_roundup(pixfmt->width, 1, 0xffff, walign);
+	pixfmt->height = clamp_roundup(pixfmt->height, 1, 0xffff, 1);
 
 	pixfmt->bytesperline = pixfmt->width * cc->bpp / 8;
 	pixfmt->sizeimage = pixfmt->bytesperline * pixfmt->height;
drivers/media/v4l2-core/v4l2-common.c +3 -2
index bee1535b04d3..3e8c16bcb0f6 100644
@@ -106,8 +106,8 @@ static unsigned int clamp_align(unsigned int x, unsigned int min,
 	return x;
 }
 
-static unsigned int clamp_roundup(unsigned int x, unsigned int min,
-				   unsigned int max, unsigned int alignment)
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment)
 {
 	x = clamp(x, min, max);
 	if (alignment)
@@ -115,6 +115,7 @@ static unsigned int clamp_roundup(unsigned int x, unsigned int min,
 
 	return x;
 }
+EXPORT_SYMBOL(clamp_roundup);
 
 void v4l_bound_align_image(u32 *w, unsigned int wmin, unsigned int wmax,
 			   unsigned int walign,
include/media/v4l2-common.h +0 -0
index d278836fd9cb..7059b99f4afa 100644

include/media/v4l2-common.h int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 pixelformat, · line 525

Quoted 5 lines
include/media/v4l2-common.h +2 -0
@@ -521,6 +521,8 @@ int v4l2_fill_pixfmt(struct v4l2_pix_format *pixfmt, u32 pixelformat,
 		     u32 width, u32 height);
 int v4l2_fill_pixfmt_mp(struct v4l2_pix_format_mplane *pixfmt, u32 pixelformat,
 			u32 width, u32 height);
+unsigned int clamp_roundup(unsigned int x, unsigned int min,
+			   unsigned int max, unsigned int alignment);

The function name is too generic to be exported by V4L2. It should move to include/linux/minmax.h (and the implementation to a corresponding source file if it's too large to be inlined) if you want to export it.

I've submitted a v3 of this patch, along with two other patches, and CC'ed you. The series retains your authorship on this patch. Could you please review it ?

Quoted 3 lines
include/media/v4l2-common.h +0 -0
 /**
  * v4l2_get_link_freq - Get link rate from transmitter
Signature · 3 lines
Regards,

Laurent Pinchart
author reviewer
#5 lore raw

Hi Laurent,

Quoted 3 lines · Laurent Pinchart

I've submitted a v3 of this patch, along with two other patches, and CC'ed you. The series retains your authorship on this patch. Could you please review it ?

v3 looks good to me, thanks.

Tim said on another thread he would be out on vacation for two weeks, so he would not be able to
test it soon.

I manually hardcoded Tim's value on the driver so that I could reproduce the original 480 to 768 width miscalculation.

With v3, the width is reported correctly as 480.

Thanks

author reviewer
#6 lore raw
Quoted 12 lines · Fabio Estevam

Hi Laurent,

Quoted 3 lines · Laurent Pinchart

I've submitted a v3 of this patch, along with two other patches, and CC'ed you. The series retains your authorship on this patch. Could you please review it ?

v3 looks good to me, thanks.

Quoted 8 lines

Tim said on another thread he would be out on vacation for two weeks, so he would not be able to
test it soon.

I manually hardcoded Tim's value on the driver so that I could reproduce the original 480 to 768 width miscalculation.

With v3, the width is reported correctly as 480.

Sorry, I meant the width is correctly calculated as 640 here.

So all is good with v3.