Skip to contentSkip to conversation

Enable video engine

PATCH Superseded Eddie James 3 participants 15 messages Mar 21, 2019 – Mar 28, 2019

Subsystems (4)

v1 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

cover Enable video engine

This series adds the necessary clocks for the video engine to the Aspeed clock driver. It also adds the video engine node to the AST2500 dts and enables that node in the witherspoon and romulus dts.

Eddie James (6):
clk: aspeed: Add video engine reset index definition
clk: aspeed: Setup video engine clocking
ARM: dts: aspeed-g5: Add video engine
ARM: dts: witherspoon: Enable video engine
ARM: dts: witherspoon: Enable vhub
ARM: dts: romulus: Enable video engine

Wei Yongjun (1):
media: platform: Fix missing spin_lock_init()

 arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts     |  7 +++-
 arch/arm/boot/dts/aspeed-bmc-opp-witherspoon.dts | 11 ++++++-
 arch/arm/boot/dts/aspeed-g5.dtsi                 | 11 +++++++
 drivers/clk/clk-aspeed.c                         | 41 ++++++++++++++++++++++--
 drivers/media/platform/aspeed-video.c            |  1 +
 include/dt-bindings/clock/aspeed-clock.h         |  1 +
 6 files changed, 68 insertions(+), 4 deletions(-)
Signature · 1 line
1.8.3.1
#1 lore raw

PATCH 1/7 media: platform: Fix missing spin_lock_init()

From: Wei Yongjun <[email protected]>

The driver allocates the spinlock but not initialize it. Use spin_lock_init() on it to initialize it correctly.

This is detected by Coccinelle semantic patch.

Fixes d2b4387f3bdf ("media: platform: Add Aspeed Video Engine driver")
---
 drivers/media/platform/aspeed-video.c | 1 +
 1 file changed, 1 insertion(+)
drivers/media/platform/aspeed-video.c +1 -0
index dfec813..692e08e 100644
@@ -1661,6 +1661,7 @@ static int aspeed_video_probe(struct platform_device *pdev)
 
 	video->frame_rate = 30;
 	video->dev = &pdev->dev;
+	spin_lock_init(&video->lock);
 	mutex_init(&video->video_lock);
 	init_waitqueue_head(&video->wait);
 	INIT_DELAYED_WORK(&video->res_work, aspeed_video_resolution_work);
Signature · 1 line
1.8.3.1
maintainer
#2 lore raw

Re: [PATCH dev-5.0 1/7] media: platform: Fix missing spin_lock_init()

Quoted 6 lines · Eddie James

From: Wei Yongjun <[email protected]>

The driver allocates the spinlock but not initialize it. Use spin_lock_init() on it to initialize it correctly.

Quoted 6 lines

This is detected by Coccinelle semantic patch.

Fixes d2b4387f3bdf ("media: platform: Add Aspeed Video Engine driver")

This lacks your sign off.

I have found the upstream commit and cherry picked that instead.

Cheers,

Joel

#3 lore raw

PATCH 2/7 clk: aspeed: Add video engine reset index definition

In reply to Eddie James

Add an index into the array of resets kept in the Aspeed clock driver for the video engine. This isn't a HW bit definition.

---
 include/dt-bindings/clock/aspeed-clock.h | 1 +
 1 file changed, 1 insertion(+)
include/dt-bindings/clock/aspeed-clock.h +1 -0
index f437386..15a9059 100644
@@ -50,5 +50,6 @@
 #define ASPEED_RESET_I2C		7
 #define ASPEED_RESET_AHB		8
 #define ASPEED_RESET_CRT1		9
+#define ASPEED_RESET_VIDEO		10
 
 #endif
Signature · 1 line
1.8.3.1
#4 lore raw

PATCH 3/7 clk: aspeed: Setup video engine clocking

In reply to Eddie James

Add the video engine reset bit. Add eclk mux and clock divider table.

---
 drivers/clk/clk-aspeed.c | 41 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 39 insertions(+), 2 deletions(-)
drivers/clk/clk-aspeed.c +39 -2
index 5961367..f16ce7d 100644
@@ -87,7 +87,7 @@ struct aspeed_clk_gate {
 /* TODO: ask Aspeed about the actual parent data */
 static const struct aspeed_gate_data aspeed_gates[] = {
 	/*				 clk rst   name			parent	flags */
-	[ASPEED_CLK_GATE_ECLK] =	{  0, -1, "eclk-gate",		"eclk",	0 }, /* Video Engine */
+	[ASPEED_CLK_GATE_ECLK] =	{  0,  6, "eclk-gate",		"eclk",	0 }, /* Video Engine */
 	[ASPEED_CLK_GATE_GCLK] =	{  1,  7, "gclk-gate",		NULL,	0 }, /* 2D engine */
 	[ASPEED_CLK_GATE_MCLK] =	{  2, -1, "mclk-gate",		"mpll",	CLK_IS_CRITICAL }, /* SDRAM */
 	[ASPEED_CLK_GATE_VCLK] =	{  3,  6, "vclk-gate",		NULL,	0 }, /* Video Capture */
@@ -113,6 +113,24 @@ struct aspeed_clk_gate {
 	[ASPEED_CLK_GATE_LHCCLK] =	{ 28, -1, "lhclk-gate",		"lhclk", 0 }, /* LPC master/LPC+ */
 };
 
+static const char * const eclk_parent_names[] = {
+	"mpll",
+	"hpll",
+	"dpll",
+};
+
+static const struct clk_div_table ast2500_eclk_div_table[] = {
+	{ 0x0, 2 },
+	{ 0x1, 2 },
+	{ 0x2, 3 },
+	{ 0x3, 4 },
+	{ 0x4, 5 },
+	{ 0x5, 6 },
+	{ 0x6, 7 },
+	{ 0x7, 8 },
+	{ 0 }
+};
+
 static const struct clk_div_table ast2500_mac_div_table[] = {
 	{ 0x0, 4 }, /* Yep, really. Aspeed confirmed this is correct */
 	{ 0x1, 4 },
@@ -192,18 +210,21 @@ static struct clk_hw *aspeed_ast2500_calc_pll(const char *name, u32 val)
 
 struct aspeed_clk_soc_data {
 	const struct clk_div_table *div_table;
+	const struct clk_div_table *eclk_div_table;
 	const struct clk_div_table *mac_div_table;
 	struct clk_hw *(*calc_pll)(const char *name, u32 val);
 };
 
 static const struct aspeed_clk_soc_data ast2500_data = {
 	.div_table = ast2500_div_table,
+	.eclk_div_table = ast2500_eclk_div_table,
 	.mac_div_table = ast2500_mac_div_table,
 	.calc_pll = aspeed_ast2500_calc_pll,
 };
 
 static const struct aspeed_clk_soc_data ast2400_data = {
 	.div_table = ast2400_div_table,
+	.eclk_div_table = ast2400_div_table,
 	.mac_div_table = ast2400_div_table,
 	.calc_pll = aspeed_ast2400_calc_pll,
 };
@@ -317,6 +338,7 @@ struct aspeed_reset {
 	[ASPEED_RESET_PECI]	= 10,
 	[ASPEED_RESET_I2C]	=  2,
 	[ASPEED_RESET_AHB]	=  1,
+	[ASPEED_RESET_VIDEO]	=  6,
 
 	/*
 	 * SCUD4 resets start at an offset to separate them from
@@ -522,6 +544,22 @@ static int aspeed_clk_probe(struct platform_device *pdev)
 		return PTR_ERR(hw);
 	aspeed_clk_data->hws[ASPEED_CLK_24M] = hw;
 
+	hw = clk_hw_register_mux(dev, "eclk-mux", eclk_parent_names,
+				 ARRAY_SIZE(eclk_parent_names), 0,
+				 scu_base + ASPEED_CLK_SELECTION, 2, 0x3, 0,
+				 &aspeed_clk_lock);
+	if (IS_ERR(hw))
+		return PTR_ERR(hw);
+	aspeed_clk_data->hws[ASPEED_CLK_ECLK_MUX] = hw;
+
+	hw = clk_hw_register_divider_table(dev, "eclk", "eclk-mux", 0,
+					   scu_base + ASPEED_CLK_SELECTION, 28,
+					   3, 0, soc_data->eclk_div_table,
+					   &aspeed_clk_lock);
+	if (IS_ERR(hw))
+		return PTR_ERR(hw);
+	aspeed_clk_data->hws[ASPEED_CLK_ECLK] = hw;
+
 	/*
 	 * TODO: There are a number of clocks that not included in this driver
 	 * as more information is required:
@@ -531,7 +569,6 @@ static int aspeed_clk_probe(struct platform_device *pdev)
 	 *   RGMII
 	 *   RMII
 	 *   UART[1..5] clock source mux
-	 *   Video Engine (ECLK) mux and clock divider
 	 */
 
 	for (i = 0; i < ARRAY_SIZE(aspeed_gates); i++) {
Signature · 1 line
1.8.3.1
maintainer
#5 lore raw

Re: [PATCH dev-5.0 3/7] clk: aspeed: Setup video engine clocking

Quoted 14 lines · Eddie James

Add the video engine reset bit. Add eclk mux and clock divider table.

---
 drivers/clk/clk-aspeed.c | 41 +++++++++++++++++++++++++++++++++++++++--
 1 file changed, 39 insertions(+), 2 deletions(-)
drivers/clk/clk-aspeed.c +0 -0
index 5961367..f16ce7d 100644

drivers/clk/clk-aspeed.c struct aspeed_clk_gate { · line 90

Quoted 5 lines
drivers/clk/clk-aspeed.c +1 -1
@@ -87,7 +87,7 @@ struct aspeed_clk_gate {
 /* TODO: ask Aspeed about the actual parent data */
 static const struct aspeed_gate_data aspeed_gates[] = {
        /*                               clk rst   name                 parent  flags */
-       [ASPEED_CLK_GATE_ECLK] =        {  0, -1, "eclk-gate",          "eclk", 0 }, /* Video Engine */
+       [ASPEED_CLK_GATE_ECLK] =        {  0,  6, "eclk-gate",          "eclk", 0 }, /* Video Engine */

As I've said on previous reviews, it's incorrect to add the reset line to the clock, and have it be a separate reset device.

When you add the reset bit to the clk device, we control both the clock gating (bit 0, SCU0C) and the reset line (bit 6, SCU04) with the clk device. You can see this happen in aspeed_clk_enable if gate->reset_idx >= 0.

Assuming you are doing this on purpose, can you explain why you want to have the reset device and the clk device both control the one bit?

Cheers,

Joel

Quoted 92 lines
drivers/clk/clk-aspeed.c +38 -2
        [ASPEED_CLK_GATE_GCLK] =        {  1,  7, "gclk-gate",          NULL,   0 }, /* 2D engine */
        [ASPEED_CLK_GATE_MCLK] =        {  2, -1, "mclk-gate",          "mpll", CLK_IS_CRITICAL }, /* SDRAM */
        [ASPEED_CLK_GATE_VCLK] =        {  3,  6, "vclk-gate",          NULL,   0 }, /* Video Capture */
@@ -113,6 +113,24 @@ struct aspeed_clk_gate {
        [ASPEED_CLK_GATE_LHCCLK] =      { 28, -1, "lhclk-gate",         "lhclk", 0 }, /* LPC master/LPC+ */
 };
 
+static const char * const eclk_parent_names[] = {
+       "mpll",
+       "hpll",
+       "dpll",
+};
+
+static const struct clk_div_table ast2500_eclk_div_table[] = {
+       { 0x0, 2 },
+       { 0x1, 2 },
+       { 0x2, 3 },
+       { 0x3, 4 },
+       { 0x4, 5 },
+       { 0x5, 6 },
+       { 0x6, 7 },
+       { 0x7, 8 },
+       { 0 }
+};
+
 static const struct clk_div_table ast2500_mac_div_table[] = {
        { 0x0, 4 }, /* Yep, really. Aspeed confirmed this is correct */
        { 0x1, 4 },
@@ -192,18 +210,21 @@ static struct clk_hw *aspeed_ast2500_calc_pll(const char *name, u32 val)
 
 struct aspeed_clk_soc_data {
        const struct clk_div_table *div_table;
+       const struct clk_div_table *eclk_div_table;
        const struct clk_div_table *mac_div_table;
        struct clk_hw *(*calc_pll)(const char *name, u32 val);
 };
 
 static const struct aspeed_clk_soc_data ast2500_data = {
        .div_table = ast2500_div_table,
+       .eclk_div_table = ast2500_eclk_div_table,
        .mac_div_table = ast2500_mac_div_table,
        .calc_pll = aspeed_ast2500_calc_pll,
 };
 
 static const struct aspeed_clk_soc_data ast2400_data = {
        .div_table = ast2400_div_table,
+       .eclk_div_table = ast2400_div_table,
        .mac_div_table = ast2400_div_table,
        .calc_pll = aspeed_ast2400_calc_pll,
 };
@@ -317,6 +338,7 @@ struct aspeed_reset {
        [ASPEED_RESET_PECI]     = 10,
        [ASPEED_RESET_I2C]      =  2,
        [ASPEED_RESET_AHB]      =  1,
+       [ASPEED_RESET_VIDEO]    =  6,
 
        /*
         * SCUD4 resets start at an offset to separate them from
@@ -522,6 +544,22 @@ static int aspeed_clk_probe(struct platform_device *pdev)
                return PTR_ERR(hw);
        aspeed_clk_data->hws[ASPEED_CLK_24M] = hw;
 
+       hw = clk_hw_register_mux(dev, "eclk-mux", eclk_parent_names,
+                                ARRAY_SIZE(eclk_parent_names), 0,
+                                scu_base + ASPEED_CLK_SELECTION, 2, 0x3, 0,
+                                &aspeed_clk_lock);
+       if (IS_ERR(hw))
+               return PTR_ERR(hw);
+       aspeed_clk_data->hws[ASPEED_CLK_ECLK_MUX] = hw;
+
+       hw = clk_hw_register_divider_table(dev, "eclk", "eclk-mux", 0,
+                                          scu_base + ASPEED_CLK_SELECTION, 28,
+                                          3, 0, soc_data->eclk_div_table,
+                                          &aspeed_clk_lock);
+       if (IS_ERR(hw))
+               return PTR_ERR(hw);
+       aspeed_clk_data->hws[ASPEED_CLK_ECLK] = hw;
+
        /*
         * TODO: There are a number of clocks that not included in this driver
         * as more information is required:
@@ -531,7 +569,6 @@ static int aspeed_clk_probe(struct platform_device *pdev)
         *   RGMII
         *   RMII
         *   UART[1..5] clock source mux
-        *   Video Engine (ECLK) mux and clock divider
         */
 
        for (i = 0; i < ARRAY_SIZE(aspeed_gates); i++) {
--

1.8.3.1

#6 lore raw

Re: [PATCH dev-5.0 3/7] clk: aspeed: Setup video engine clocking

Quoted 22 lines · Joel Stanley
Quoted 13 lines · Eddie James

Add the video engine reset bit. Add eclk mux and clock divider table.

---
  drivers/clk/clk-aspeed.c | 41 +++++++++++++++++++++++++++++++++++++++--
  1 file changed, 39 insertions(+), 2 deletions(-)
drivers/clk/clk-aspeed.c +0 -0
index 5961367..f16ce7d 100644

drivers/clk/clk-aspeed.c struct aspeed_clk_gate { · line 90

Quoted 5 lines
drivers/clk/clk-aspeed.c +1 -1
@@ -87,7 +87,7 @@ struct aspeed_clk_gate {
  /* TODO: ask Aspeed about the actual parent data */
  static const struct aspeed_gate_data aspeed_gates[] = {
         /*                               clk rst   name                 parent  flags */
-       [ASPEED_CLK_GATE_ECLK] =        {  0, -1, "eclk-gate",          "eclk", 0 }, /* Video Engine */
+       [ASPEED_CLK_GATE_ECLK] =        {  0,  6, "eclk-gate",          "eclk", 0 }, /* Video Engine */

As I've said on previous reviews, it's incorrect to add the reset line to the clock, and have it be a separate reset device.

Quoted 7 lines

When you add the reset bit to the clk device, we control both the clock gating (bit 0, SCU0C) and the reset line (bit 6, SCU04) with the clk device. You can see this happen in aspeed_clk_enable if gate->reset_idx >= 0.

Assuming you are doing this on purpose, can you explain why you want to have the reset device and the clk device both control the one bit?

Yes. The reset line must be added to the clock device because it's an important part of the clock enable/disable operation, as required in the AST2500 spec.

The video driver has been written around the idea that the reset device is available by itself as well. I ran into a number of issues trying to remove that dependency. Firstly its much slower to toggle the clocks than to simply toggle the reset, and there are cases where we need to quickly reset the engine and reacquire the video signal. But mainly I saw issues with interrupts being unstoppable, since the clock driver doesn't enable the reset before the clocks turn off, and then the video engine registers can no longer be changed once the clocks are off.

I don't really understand what's so incorrect about having both things control the reset. Maybe we can chat tonight to resolve this.

Thanks,

Eddie

Quoted 98 lines

Cheers,

Joel

Quoted 92 lines
drivers/clk/clk-aspeed.c +38 -2
         [ASPEED_CLK_GATE_GCLK] =        {  1,  7, "gclk-gate",          NULL,   0 }, /* 2D engine */
         [ASPEED_CLK_GATE_MCLK] =        {  2, -1, "mclk-gate",          "mpll", CLK_IS_CRITICAL }, /* SDRAM */
         [ASPEED_CLK_GATE_VCLK] =        {  3,  6, "vclk-gate",          NULL,   0 }, /* Video Capture */
@@ -113,6 +113,24 @@ struct aspeed_clk_gate {
         [ASPEED_CLK_GATE_LHCCLK] =      { 28, -1, "lhclk-gate",         "lhclk", 0 }, /* LPC master/LPC+ */
  };
 
+static const char * const eclk_parent_names[] = {
+       "mpll",
+       "hpll",
+       "dpll",
+};
+
+static const struct clk_div_table ast2500_eclk_div_table[] = {
+       { 0x0, 2 },
+       { 0x1, 2 },
+       { 0x2, 3 },
+       { 0x3, 4 },
+       { 0x4, 5 },
+       { 0x5, 6 },
+       { 0x6, 7 },
+       { 0x7, 8 },
+       { 0 }
+};
+
  static const struct clk_div_table ast2500_mac_div_table[] = {
         { 0x0, 4 }, /* Yep, really. Aspeed confirmed this is correct */
         { 0x1, 4 },
@@ -192,18 +210,21 @@ static struct clk_hw *aspeed_ast2500_calc_pll(const char *name, u32 val)
 
  struct aspeed_clk_soc_data {
         const struct clk_div_table *div_table;
+       const struct clk_div_table *eclk_div_table;
         const struct clk_div_table *mac_div_table;
         struct clk_hw *(*calc_pll)(const char *name, u32 val);
  };
 
  static const struct aspeed_clk_soc_data ast2500_data = {
         .div_table = ast2500_div_table,
+       .eclk_div_table = ast2500_eclk_div_table,
         .mac_div_table = ast2500_mac_div_table,
         .calc_pll = aspeed_ast2500_calc_pll,
  };
 
  static const struct aspeed_clk_soc_data ast2400_data = {
         .div_table = ast2400_div_table,
+       .eclk_div_table = ast2400_div_table,
         .mac_div_table = ast2400_div_table,
         .calc_pll = aspeed_ast2400_calc_pll,
  };
@@ -317,6 +338,7 @@ struct aspeed_reset {
         [ASPEED_RESET_PECI]     = 10,
         [ASPEED_RESET_I2C]      =  2,
         [ASPEED_RESET_AHB]      =  1,
+       [ASPEED_RESET_VIDEO]    =  6,
 
         /*
          * SCUD4 resets start at an offset to separate them from
@@ -522,6 +544,22 @@ static int aspeed_clk_probe(struct platform_device *pdev)
                 return PTR_ERR(hw);
         aspeed_clk_data->hws[ASPEED_CLK_24M] = hw;
 
+       hw = clk_hw_register_mux(dev, "eclk-mux", eclk_parent_names,
+                                ARRAY_SIZE(eclk_parent_names), 0,
+                                scu_base + ASPEED_CLK_SELECTION, 2, 0x3, 0,
+                                &aspeed_clk_lock);
+       if (IS_ERR(hw))
+               return PTR_ERR(hw);
+       aspeed_clk_data->hws[ASPEED_CLK_ECLK_MUX] = hw;
+
+       hw = clk_hw_register_divider_table(dev, "eclk", "eclk-mux", 0,
+                                          scu_base + ASPEED_CLK_SELECTION, 28,
+                                          3, 0, soc_data->eclk_div_table,
+                                          &aspeed_clk_lock);
+       if (IS_ERR(hw))
+               return PTR_ERR(hw);
+       aspeed_clk_data->hws[ASPEED_CLK_ECLK] = hw;
+
         /*
          * TODO: There are a number of clocks that not included in this driver
          * as more information is required:
@@ -531,7 +569,6 @@ static int aspeed_clk_probe(struct platform_device *pdev)
          *   RGMII
          *   RMII
          *   UART[1..5] clock source mux
-        *   Video Engine (ECLK) mux and clock divider
          */
 
         for (i = 0; i < ARRAY_SIZE(aspeed_gates); i++) {
--

1.8.3.1

#7 lore raw

PATCH 4/7 ARM: dts: aspeed-g5: Add video engine

In reply to Eddie James

Add a node to describe the video engine on the AST2500.

---
 arch/arm/boot/dts/aspeed-g5.dtsi | 11 +++++++++++
 1 file changed, 11 insertions(+)
arch/arm/boot/dts/aspeed-g5.dtsi +11 -0
index a79e01f..0cb143f 100644
@@ -252,6 +252,17 @@
 				status = "disabled";
 			};
 
+			video: video@1e700000 {
+				compatible = "aspeed,ast2500-video-engine";
+				reg = <0x1e700000 0x1000>;
+				clocks = <&syscon ASPEED_CLK_GATE_VCLK>,
+					 <&syscon ASPEED_CLK_GATE_ECLK>;
+				clock-names = "vclk", "eclk";
+				resets = <&syscon ASPEED_RESET_VIDEO>;
+				interrupts = <7>;
+				status = "disabled";
+			};
+
 			sram: sram@1e720000 {
 				compatible = "mmio-sram";
 				reg = <0x1e720000 0x9000>;	// 36K
Signature · 1 line
1.8.3.1
#8 lore raw

PATCH 5/7 ARM: dts: witherspoon: Enable video engine

In reply to Eddie James

Enable the video engine and set it's required memory-region property.

---
 arch/arm/boot/dts/aspeed-bmc-opp-witherspoon.dts | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
arch/arm/boot/dts/aspeed-bmc-opp-witherspoon.dts +6 -1
index b933ced..9da8ffa 100644
@@ -28,7 +28,7 @@
 		};
 
 		gfx_memory: framebuffer {
-			size = <0x01000000>;
+			size = <0x04000000>;
 			alignment = <0x01000000>;
 			compatible = "shared-dma-pool";
 			reusable;
@@ -660,4 +660,9 @@
 	status = "okay";
 };
 
+&video {
+	status = "okay";
+	memory-region = <&gfx_memory>;
+};
+
 #include "ibm-power9-dual.dtsi"
Signature · 1 line
1.8.3.1
#9 lore raw

PATCH 6/7 ARM: dts: witherspoon: Enable vhub

In reply to Eddie James

Enable the virtual USB hub.

---
 arch/arm/boot/dts/aspeed-bmc-opp-witherspoon.dts | 4 ++++
 1 file changed, 4 insertions(+)
arch/arm/boot/dts/aspeed-bmc-opp-witherspoon.dts +4 -0
index 9da8ffa..93e4842 100644
@@ -660,6 +660,10 @@
 	status = "okay";
 };
 
+&vhub {
+	status = "okay";
+};
+
 &video {
 	status = "okay";
 	memory-region = <&gfx_memory>;
Signature · 1 line
1.8.3.1
maintainer
#10 lore raw

Re: [PATCH dev-5.0 6/7] ARM: dts: witherspoon: Enable vhub

Quoted 4 lines · Eddie James

Enable the virtual USB hub.

This seems unrelated to the video engine series. I've applied it to dev-5.0 as the commit message is good.

Cheers,

Joel

#11 lore raw

PATCH 7/7 ARM: dts: romulus: Enable video engine

In reply to Eddie James

Enable the video engine and set it's required memory-region property.

---
 arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +6 -1
index ee1a460..6b24608 100644
@@ -37,7 +37,7 @@
 		};
 
 		gfx_memory: framebuffer {
-			size = <0x01000000>;
+			size = <0x04000000>;
 			alignment = <0x01000000>;
 			compatible = "shared-dma-pool";
 			reusable;
@@ -315,4 +315,9 @@
      memory-region = <&gfx_memory>;
 };
 
+&video {
+	status = "okay";
+	memory-region = <&gfx_memory>;
+};
+
 #include "ibm-power9-dual.dtsi"
Signature · 1 line
1.8.3.1
maintainer
#12 lore raw

Re: [PATCH dev-5.0 7/7] ARM: dts: romulus: Enable video engine

Quoted 13 lines · Eddie James

Enable the video engine and set it's required memory-region property.

---
 arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +0 -0
index ee1a460..6b24608 100644

arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts line 40

Quoted 5 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +1 -1
@@ -37,7 +37,7 @@
                };
 
                gfx_memory: framebuffer {
-                       size = <0x01000000>;
+                       size = <0x04000000>;

You're using the memory region we have set aside for the gfx device - the thing that you plug a monitor into the back of a machine with. Are you sure that is what we want to do?

Quoted 16 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +5 -1
                        alignment = <0x01000000>;
                        compatible = "shared-dma-pool";
                        reusable;
@@ -315,4 +315,9 @@
      memory-region = <&gfx_memory>;
 };
 
+&video {
+       status = "okay";
+       memory-region = <&gfx_memory>;
+};
+
 #include "ibm-power9-dual.dtsi"
--

1.8.3.1

#13 lore raw

Re: [PATCH dev-5.0 7/7] ARM: dts: romulus: Enable video engine

Quoted 15 lines · Joel Stanley
Quoted 14 lines · Eddie James

Enable the video engine and set it's required memory-region property.

---
  arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts | 7 ++++++-
  1 file changed, 6 insertions(+), 1 deletion(-)
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +0 -0
index ee1a460..6b24608 100644
@@ -37,7 +37,7 @@
                 };
 
Quoted 6 lines

arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts line 40

Quoted 3 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +1 -1
                 gfx_memory: framebuffer {
-                       size = <0x01000000>;
+                       size = <0x04000000>;

You're using the memory region we have set aside for the gfx device - the thing that you plug a monitor into the back of a machine with. Are you sure that is what we want to do?

It still has to be allocated in the kernel, right? I'm assuming that the kernel won't give memory areas we've grabbed with the gfx driver to the video driver, and same in reverse. I could add a separate memory region too, this just seemed convenient.

Quoted 17 lines
Quoted 17 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +5 -1
                         alignment = <0x01000000>;
                         compatible = "shared-dma-pool";
                         reusable;
@@ -315,4 +315,9 @@
       memory-region = <&gfx_memory>;
  };
 
+&video {
+       status = "okay";
+       memory-region = <&gfx_memory>;
+};
+
  #include "ibm-power9-dual.dtsi"
--

1.8.3.1

#14 lore raw

Re: [PATCH dev-5.0 7/7] ARM: dts: romulus: Enable video engine

Quoted 24 lines · Eddie James
Quoted 15 lines · Joel Stanley
Quoted 14 lines · Eddie James

Enable the video engine and set it's required memory-region property.

---
  arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts | 7 ++++++-
  1 file changed, 6 insertions(+), 1 deletion(-)
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +0 -0
index ee1a460..6b24608 100644
@@ -37,7 +37,7 @@
                 };
 
Quoted 6 lines

arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts line 40

Quoted 3 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +1 -1
                 gfx_memory: framebuffer {
-                       size = <0x01000000>;
+                       size = <0x04000000>;

You're using the memory region we have set aside for the gfx device - the thing that you plug a monitor into the back of a machine with. Are you sure that is what we want to do?

Quoted 4 lines

It still has to be allocated in the kernel, right? I'm assuming that the kernel won't give memory areas we've grabbed with the gfx driver to the video driver, and same in reverse. I could add a separate memory region too, this just seemed convenient.

Lets go with a new node.

Quoted 20 lines
Quoted 18 lines
Quoted 18 lines
arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts +5 -1
                         alignment = <0x01000000>;
                         compatible = "shared-dma-pool";
                         reusable;
@@ -315,4 +315,9 @@
       memory-region = <&gfx_memory>;
  };
 
+&video {
+       status = "okay";
+       memory-region = <&gfx_memory>;
+};
+
  #include "ibm-power9-dual.dtsi"
--

1.8.3.1