On Fri, Aug 04, 2023 at 04:50:05PM +0200, Julien Stephan wrote:

On Mon, Jul 03, 2023 at 01:02:02PM +0200, AngeloGioacchino Del Regno wrote:

quoted

Il 30/06/23 12:01, Julien Stephan ha scritto:

..snip..

quoted

quoted

+
+static const struct mtk_seninf_format_info mtk_seninf_formats[] = {
+	{
+		.code = MEDIA_BUS_FMT_SBGGR8_1X8,
+		.flags = MTK_SENINF_FORMAT_BAYER,

Each entry fits in one line.

	{ .code = MEDIA_BUS_FMT_SBGGR8_1X8, .flags = MTK_SENINF_FORMAT_BAYER },

Hi Angelo,

Actually not all entries fit in one line. The last 4 ones don't:

       { .code = MEDIA_BUS_FMT_SGRBG10_DPCM8_1X8, .flags = MTK_SENINF_FORMAT_DPCM | MTK_SENINF_FORMAT_INPUT_ONLY },
which is 115 chars..

so what is the best? put all in one line except the last 4 one? or keep
them all as is?

I'd keep them all as is.

quoted

quoted

+	}, {

..snip..

quoted

quoted

+	udelay(1);
+	mtk_seninf_input_update(input, SENINF_CTRL, CSI2_SW_RST, 0);

Is there any way to check if the CSI port did reset, or is it *guaranteed* to get
out of reset in a microsecond after deasserting SW_RST?

I will double check this

quoted

quoted

+}
+

..snip..

quoted

quoted

+
+	val = mtk_seninf_mux_read(mux, SENINF_MUX_CTRL);

rst_mask = SENINF_MUX_CTRL_SENINF_IRQ_SW_RST | SENINF_MUX_CTRL_SENINF_MUX_SW_RST;

writel(mux->base + SENINF_MUX_CTRL, val | rst_mask);
writel(mux->base + SENINFMUX_CTRL, val & ~rst_mask);

that's better, right? :-)

With mtk_seninf_mux_write() instead of writel(), yes :-)

right :)

quoted

quoted

+	mtk_seninf_mux_write(mux, SENINF_MUX_CTRL, val |
+			     SENINF_MUX_CTRL_SENINF_IRQ_SW_RST |
+			     SENINF_MUX_CTRL_SENINF_MUX_SW_RST);

-- 
Regards,

Laurent Pinchart

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel