Re: [PATCH] staging: fbtft: fb_st7789v: support setting offset

[Yuguo Pei]

On Tue, 9 Apr 2024 17:49:25 +0200, Greg KH wrote:
> So you are now hard-coding the size?

Yes, the change just helps developers who want to adapt to different screen 
sizes.

> Is this always going to be 0, if so, why need it at all?

Not sure. For example, when HEIGHT and WIDTH are 240 and 280, LEFT_OFFSET is 20 
and TOP_OFFSET is 0. Different screens may have different offsets.

In addition, do I need to resend the patch to modify the Signed-off-by?

thanks,
purofle

Re: [PATCH] staging: fbtft: fb_st7789v: support setting offset

[purofle]

Re: [PATCH] staging: fbtft: fb_st7789v: support setting offset

[Greg KH]

On Wed, Apr 10, 2024 at 01:28:06AM +0800, Yuguo Pei wrote:
> On Tue, 9 Apr 2024 17:49:25 +0200, Greg KH wrote:
> > So you are now hard-coding the size?
> 
> Yes, the change just helps developers who want to adapt to different screen 
> sizes.

So there is no change?  I don't understand.

> > Is this always going to be 0, if so, why need it at all?
> 
> Not sure. For example, when HEIGHT and WIDTH are 240 and 280, LEFT_OFFSET is 
> 20 and TOP_OFFSET is 0. Different screens may have different offsets.

Then why is this needed?

Don't make changes that are not needed by anyone, especially for code in
drivers/staging/

> 
> In addition, do I need to resend the patch to modify the Signed-off-by?

Of course you do :)

Re: [PATCH] staging: fbtft: fb_st7789v: support setting offset

[Greg KH]

On Sat, Apr 06, 2024 at 12:57:47AM +0800, purofle wrote:
> Some screen sizes using st7789v chips are different from 240x320,
> and offsets need to be set to display all images properly.
> 
> Signed-off-by: purofle 

We need a semi-real name here please.

> ---
>  drivers/staging/fbtft/fb_st7789v.c | 22 ++
>  1 file changed, 22 insertions(+)
> 
> diff --git a/drivers/staging/fbtft/fb_st7789v.c 
> b/drivers/staging/fbtft/fb_st7789v.c
> index 861a15414..d47ab4262 100644
> --- a/drivers/staging/fbtft/fb_st7789v.c
> +++ b/drivers/staging/fbtft/fb_st7789v.c
> @@ -30,6 +30,12 @@
>  
>  #define HSD20_IPS 1
>  
> +#define WIDTH 240
> +#define HEIGHT 320

So you are now hard-coding the size?

> +
> +#define LEFT_OFFSET 0
> +#define TOP_OFFSET 0

Is this always going to be 0, if so, why need it at all?


thanks,

greg k-h

[PATCH] staging: fbtft: fb_st7789v: support setting offset

[purofle]

Some screen sizes using st7789v chips are different from 240x320,
and offsets need to be set to display all images properly.

Signed-off-by: purofle 
---
 drivers/staging/fbtft/fb_st7789v.c | 22 ++
 1 file changed, 22 insertions(+)

diff --git a/drivers/staging/fbtft/fb_st7789v.c 
b/drivers/staging/fbtft/fb_st7789v.c
index 861a15414..d47ab4262 100644
--- a/drivers/staging/fbtft/fb_st7789v.c
+++ b/drivers/staging/fbtft/fb_st7789v.c
@@ -30,6 +30,12 @@
 
 #define HSD20_IPS 1
 
+#define WIDTH 240
+#define HEIGHT 320
+
+#define LEFT_OFFSET 0
+#define TOP_OFFSET 0
+
 /**
  * enum st7789v_command - ST7789V display controller commands
  *
@@ -349,6 +355,21 @@ static int set_gamma(struct fbtft_par *par, u32 *curves)
return 0;
 }
 
+static void set_addr_win(struct fbtft_par *par, int xs, int ys, int xe, int ye)
+{
+   unsigned int x = xs + TOP_OFFSET, y = xe + TOP_OFFSET;
+
+   write_reg(par, MIPI_DCS_SET_COLUMN_ADDRESS, (x >> 8) & 0xFF, xs & 0xFF,
+ (y >> 8) & 0xFF, xe & 0xFF);
+
+   x = ys + LEFT_OFFSET, y = ye + LEFT_OFFSET;
+
+   write_reg(par, MIPI_DCS_SET_PAGE_ADDRESS, (x >> 8) & 0xFF, ys & 0xFF,
+ (y >> 8) & 0xFF, ye & 0xFF);
+
+   write_reg(par, MIPI_DCS_WRITE_MEMORY_START);
+}
+
 /**
  * blank() - blank the display
  *
@@ -379,6 +400,7 @@ static struct fbtft_display display = {
.set_var = set_var,
.set_gamma = set_gamma,
.blank = blank,
+   .set_addr_win = set_addr_win,
},
 };
 
-- 
2.44.0