imx-drm: cleanups and YUV 4:2:0 memory read/write reduction support

- Remove counter load enable form PRE, which has no effect.
 - Add support for setting the double read/write reduction flag in channel
   parameter memory. This can be used to save some memory bandwidth when
   capturing in YUV 4:2:0 chroma subsampled formats.
 - Allocate DMA channel structures as needed, most of the 64 channels are
   unused or even reserved.
 - Remove unused interrupt busy waiting routine.
 - Set VDIC field order for both AUTO and MAN inputs simultaneously as
   both can't be active at the same time.
 -----BEGIN PGP SIGNATURE-----
 
 iQJLBAABCAA1FiEEBsBxhV1FaKwXuCOBUMKIHHCeYOsFAlk49zAXHHAuemFiZWxA
 cGVuZ3V0cm9uaXguZGUACgkQUMKIHHCeYOtX+g/+M5DPkGhcP8fN9m47MOAEa+nP
 oDciYC76VrQev/Qys/zLM3/6sWF9h82USJ52H+zw41RuKKkYlcOzVWnSPQd2yN6Y
 hIl0fCsFzGxOMnIAhmi6BHFnvJKP1jsfeBdXSHyxI0y5kGoufG7BEHiJ7TTSgy/I
 JhccDKTRV9NzAfwpD37EI3a/Nc53DRpw3jrnHPnAaBJ6hYVPZ9YCSrBYbQQbIrDr
 x6NB8E1Ga3KRGZMTw45bTBiOs4AbZKSunzrqWQFnTRjbE+aTDs9W5n5wcdr7AoWi
 gqnx+b6TkiarNK3taHffjYioYvvn2nbGhuoAtg7hpS0CeUup0gitQquKM0kqsWnk
 yBykkP+Z7udASRgXdK6Gtzo6hzdhPeFPmmMbKmSBdIvT26t0ikf9RN1UlEhE6nY3
 A68jKC4+gNTu8kF6imzWCfwM9KB4pWn0N0qTY5U9Y7/gWFky6IEDn3V5OM3XXqUa
 c/iglYyzO+B7vVu6ZajlH+shemO1mVaxGjVFrfX29syVooZrmo0NVJPdoKwiYx0r
 E08FCOdUIEhzFS6h1/FII+mZ6YzAmNkXVz+l+MWaoOW2tIWWX4xSsFTKBka2hDfq
 daU3CDcKcg8LuSybHg6lzj8Hw+/CWkqgtv6ESVnvHEUYCHoZsvArJTNoQrx/zJbE
 EwgZIM4daufEuv6I5zo=
 =m3BY
 -----END PGP SIGNATURE-----

Merge tag 'imx-drm-next-2017-06-08' of git://git.pengutronix.de/git/pza/linux into drm-next

imx-drm: cleanups and YUV 4:2:0 memory read/write reduction support

- Remove counter load enable form PRE, which has no effect.
- Add support for setting the double read/write reduction flag in channel
  parameter memory. This can be used to save some memory bandwidth when
  capturing in YUV 4:2:0 chroma subsampled formats.
- Allocate DMA channel structures as needed, most of the 64 channels are
  unused or even reserved.
- Remove unused interrupt busy waiting routine.
- Set VDIC field order for both AUTO and MAN inputs simultaneously as
  both can't be active at the same time.

* tag 'imx-drm-next-2017-06-08' of git://git.pengutronix.de/git/pza/linux:
  gpu: ipu-v3: vdic: include AUTO field order bit in ipu_vdi_set_field_order
  gpu: ipu-v3: remove interrupt busy waiting routine
  gpu: ipu-v3: allocate ipuv3_channels as needed
  gpu: ipu-v3: Add support for double read/write reduction
  gpu: ipu-v3: prg: remove counter load enable

drivers/gpu/ipu-v3/ipu-common.c

@@ -274,15 +274,22 @@ struct ipuv3_channel *ipu_idmac_get(struct ipu_soc *ipu, unsigned num)
 
 	mutex_lock(&ipu->channel_lock);
 
-	channel = &ipu->channel[num];
+	list_for_each_entry(channel, &ipu->channels, list) {
+		if (channel->num == num) {
+			channel = ERR_PTR(-EBUSY);
+			goto out;
+		}
+	}
 
-	if (channel->busy) {
+	channel = kzalloc(sizeof(*channel), GFP_KERNEL);
-		channel = ERR_PTR(-EBUSY);
+	if (!channel) {
+		channel = ERR_PTR(-ENOMEM);
 		goto out;
 	}
 
-	channel->busy = true;
 	channel->num = num;
+	channel->ipu = ipu;
+	list_add(&channel->list, &ipu->channels);
 
 out:
 	mutex_unlock(&ipu->channel_lock);
@@ -299,7 +306,8 @@ void ipu_idmac_put(struct ipuv3_channel *channel)
 
 	mutex_lock(&ipu->channel_lock);
 
-	channel->busy = false;
+	list_del(&channel->list);
+	kfree(channel);
 
 	mutex_unlock(&ipu->channel_lock);
 }
@@ -589,22 +597,6 @@ int ipu_idmac_wait_busy(struct ipuv3_channel *channel, int ms)
 }
 EXPORT_SYMBOL_GPL(ipu_idmac_wait_busy);
 
-int ipu_wait_interrupt(struct ipu_soc *ipu, int irq, int ms)
-{
-	unsigned long timeout;
-
-	timeout = jiffies + msecs_to_jiffies(ms);
-	ipu_cm_write(ipu, BIT(irq % 32), IPU_INT_STAT(irq / 32));
-	while (!(ipu_cm_read(ipu, IPU_INT_STAT(irq / 32) & BIT(irq % 32)))) {
-		if (time_after(jiffies, timeout))
-			return -ETIMEDOUT;
-		cpu_relax();
-	}
-
-	return 0;
-}
-EXPORT_SYMBOL_GPL(ipu_wait_interrupt);
-
 int ipu_idmac_disable_channel(struct ipuv3_channel *channel)
 {
 	struct ipu_soc *ipu = channel->ipu;
@@ -1376,7 +1368,7 @@ static int ipu_probe(struct platform_device *pdev)
 	struct ipu_soc *ipu;
 	struct resource *res;
 	unsigned long ipu_base;
-	int i, ret, irq_sync, irq_err;
+	int ret, irq_sync, irq_err;
 	const struct ipu_devtype *devtype;
 
 	devtype = of_device_get_match_data(&pdev->dev);
@@ -1409,13 +1401,12 @@ static int ipu_probe(struct platform_device *pdev)
 			return -EPROBE_DEFER;
 	}
 
-	for (i = 0; i < 64; i++)
-		ipu->channel[i].ipu = ipu;
 	ipu->devtype = devtype;
 	ipu->ipu_type = devtype->type;
 
 	spin_lock_init(&ipu->lock);
 	mutex_init(&ipu->channel_lock);
+	INIT_LIST_HEAD(&ipu->channels);
 
 	dev_dbg(&pdev->dev, "cm_reg:   0x%08lx\n",
 			ipu_base + devtype->cm_ofs);

drivers/gpu/ipu-v3/ipu-cpmem.c

@@ -224,6 +224,12 @@ void ipu_cpmem_set_resolution(struct ipuv3_channel *ch, int xres, int yres)
 }
 EXPORT_SYMBOL_GPL(ipu_cpmem_set_resolution);
 
+void ipu_cpmem_skip_odd_chroma_rows(struct ipuv3_channel *ch)
+{
+	ipu_ch_param_write_field(ch, IPU_FIELD_RDRW, 1);
+}
+EXPORT_SYMBOL_GPL(ipu_cpmem_skip_odd_chroma_rows);
+
 void ipu_cpmem_set_stride(struct ipuv3_channel *ch, int stride)
 {
 	ipu_ch_param_write_field(ch, IPU_FIELD_SLY, stride - 1);

drivers/gpu/ipu-v3/ipu-prg.c

@@ -318,8 +318,6 @@ int ipu_prg_channel_configure(struct ipuv3_channel *ipu_chan,
 	writel(val, prg->regs + IPU_PRG_BADDR(prg_chan));
 
 	val = readl(prg->regs + IPU_PRG_CTL);
-	/* counter load enable */
-	val |= IPU_PRG_CTL_CNT_LOAD_EN(prg_chan);
 	/* config AXI ID */
 	val &= ~(IPU_PRG_CTL_SOFT_ARID_MASK <<
 		 IPU_PRG_CTL_SOFT_ARID_SHIFT(prg_chan));

drivers/gpu/ipu-v3/ipu-prv.h

@@ -157,11 +157,8 @@ enum ipu_modules {
 
 struct ipuv3_channel {
 	unsigned int num;
-
-	bool enabled;
-	bool busy;
-
 	struct ipu_soc *ipu;
+	struct list_head list;
 };
 
 struct ipu_cpmem;
@@ -184,6 +181,7 @@ struct ipu_soc {
 	enum ipuv3_type		ipu_type;
 	spinlock_t		lock;
 	struct mutex		channel_lock;
+	struct list_head	channels;
 
 	void __iomem		*cm_reg;
 	void __iomem		*idmac_reg;
@@ -193,8 +191,6 @@ struct ipu_soc {
 
 	struct clk		*clk;
 
-	struct ipuv3_channel	channel[64];
-
 	int			irq_sync;
 	int			irq_err;
 	struct irq_domain	*domain;
@@ -229,7 +225,6 @@ int ipu_module_enable(struct ipu_soc *ipu, u32 mask);
 int ipu_module_disable(struct ipu_soc *ipu, u32 mask);
 
 bool ipu_idmac_channel_busy(struct ipu_soc *ipu, unsigned int chno);
-int ipu_wait_interrupt(struct ipu_soc *ipu, int irq, int ms);
 
 int ipu_csi_init(struct ipu_soc *ipu, struct device *dev, int id,
 		 unsigned long base, u32 module, struct clk *clk_ipu);

drivers/gpu/ipu-v3/ipu-vdi.c

@@ -88,9 +88,9 @@ void ipu_vdi_set_field_order(struct ipu_vdi *vdi, v4l2_std_id std, u32 field)
 
 	reg = ipu_vdi_read(vdi, VDI_C);
 	if (top_field_0)
-		reg &= ~VDI_C_TOP_FIELD_MAN_1;
+		reg &= ~(VDI_C_TOP_FIELD_MAN_1 | VDI_C_TOP_FIELD_AUTO_1);
 	else
-		reg |= VDI_C_TOP_FIELD_MAN_1;
+		reg |= VDI_C_TOP_FIELD_MAN_1 | VDI_C_TOP_FIELD_AUTO_1;
 	ipu_vdi_write(vdi, reg, VDI_C);
 
 	spin_unlock_irqrestore(&vdi->lock, flags);

include/video/imx-ipu-v3.h

@@ -250,6 +250,7 @@ struct ipu_image {
 
 void ipu_cpmem_zero(struct ipuv3_channel *ch);
 void ipu_cpmem_set_resolution(struct ipuv3_channel *ch, int xres, int yres);
+void ipu_cpmem_skip_odd_chroma_rows(struct ipuv3_channel *ch);
 void ipu_cpmem_set_stride(struct ipuv3_channel *ch, int stride);
 void ipu_cpmem_set_high_priority(struct ipuv3_channel *ch);
 void ipu_cpmem_set_buffer(struct ipuv3_channel *ch, int bufnum, dma_addr_t buf);