On 07/01/13 18:23, Lukasz Majewski wrote:

lcd_set_flush_dcache(1) function is called from trats configuration file to enable lcd_sync function.

Tested-at: TRATS (Exynos 4210) Tested-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Kyungmin Park kyungmin.park@samsung.com Cc: Minkyu Kang mk7.kang@samsung.com

---

board/samsung/trats/trats.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/board/samsung/trats/trats.c b/board/samsung/trats/trats.c index 88d193d..5d0fdd0 100644 --- a/board/samsung/trats/trats.c +++ b/board/samsung/trats/trats.c @@ -789,5 +789,8 @@ void init_panel_info(vidinfo_t *vid) s6e8ax0_init(); exynos_set_dsim_platform_data(&s6e8ax0_platform_data);

• /* Enable flushing after LCD writes if requested */
• lcd_set_flush_dcache(1);
• setenv("lcdinfo", "lcd=s6e8ax0");

}

Acked-by: Minkyu Kang mk7.kang@samsung.com

--- Thanks, Minkyu Kang.

Dear Lukasz,

On 07/01/13 18:23, Lukasz Majewski wrote:

lcd_set_flush_dcache(1) function is called from trats configuration file to enable lcd_sync function.

Tested-at: TRATS (Exynos 4210) Tested-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Kyungmin Park kyungmin.park@samsung.com Cc: Minkyu Kang mk7.kang@samsung.com

---

board/samsung/trats/trats.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/board/samsung/trats/trats.c b/board/samsung/trats/trats.c index 88d193d..5d0fdd0 100644 --- a/board/samsung/trats/trats.c +++ b/board/samsung/trats/trats.c @@ -789,5 +789,8 @@ void init_panel_info(vidinfo_t *vid) s6e8ax0_init(); exynos_set_dsim_platform_data(&s6e8ax0_platform_data);

• /* Enable flushing after LCD writes if requested */
• lcd_set_flush_dcache(1);

Please move it into exynos_fb.c (lcd_ctrl_init? or somewhere)

• setenv("lcdinfo", "lcd=s6e8ax0");

}

Thanks, Minkyu Kang.

Hi Simon,

Hi Lukasz,

On Mon, Jan 7, 2013 at 1:23 AM, Lukasz Majewski l.majewski@samsung.com wrote:

...

This commit makes the video subsystem code cache aware. Memory allocated for decompressed BMP memory is now cache line aligned.

Tested-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Lukasz Majewski l.majewski@samsung.com Signed-off-by: Kyungmin Park kyungmin.park@samsung.com Cc: Anatolij Gustschin agust@denx.de

---

common/cmd_bmp.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/common/cmd_bmp.c b/common/cmd_bmp.c index 5a52edd..57f3eb5 100644 --- a/common/cmd_bmp.c +++ b/common/cmd_bmp.c @@ -55,7 +55,7 @@ bmp_image_t *gunzip_bmp(unsigned long addr, unsigned long *lenp) * Decompress bmp image */ len = CONFIG_SYS_VIDEO_LOGO_MAX_SIZE;

•   dst = malloc(CONFIG_SYS_VIDEO_LOGO_MAX_SIZE);
  

•   dst = memalign(CONFIG_SYS_CACHELINE_SIZE, len);
    if (dst == NULL) {
            puts("Error: malloc in gunzip failed!\n");
            return NULL;
  

Sorry, I still have a question. Does this 'dst' address get used as the actual LCD frame buffer on your board, or is it just copied to the frame buffer?

I must admit, that I've misunderstood the LCD code a bit. Reply to this post from Anatolij helped me. The buffer (dst) is only the "internal" buffer from which we are assigning BMP data to actual frame buffer area. Then only the frame buffer (pointed by fb pointer) area needs cache flush (which is done at lcd_sync()).

To sum up - this patch shall be dropped. However the second patch -

[PATCH 2/2] video:cache:fix:trats: Enable dcache flush for TRATS board's LCD subsystem

is crucial to fix the BMP image display distortion.

Regards, Simon

...

-- 1.7.2.3