Skip to contentSkip to conversation

clk: imx8mp: Set the correct parent for audio_root_clk

PATCH In review Abel Vesa 2 participants 5 messages Apr 27, 2020 – May 22, 2020

Subsystems (3)
lore
Share & tools
lore

Abel Vesa

author
#0 lore raw

PATCH 1/1 clk: imx8mp: Set the correct parent for audio_root_clk

Instead of ipg_root, the parent needs to be ipg_audio_root.

---
 drivers/clk/imx/clk-imx8mp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/imx/clk-imx8mp.c +1 -1
index 41469e2..dcdfc9d 100644
@@ -727,7 +727,7 @@ static int imx8mp_clocks_probe(struct platform_device *pdev)
 	hws[IMX8MP_CLK_HDMI_ROOT] = imx_clk_hw_gate4("hdmi_root_clk", "hdmi_axi", ccm_base + 0x45f0, 0);
 	hws[IMX8MP_CLK_TSENSOR_ROOT] = imx_clk_hw_gate4("tsensor_root_clk", "ipg_root", ccm_base + 0x4620, 0);
 	hws[IMX8MP_CLK_VPU_ROOT] = imx_clk_hw_gate4("vpu_root_clk", "vpu_bus", ccm_base + 0x4630, 0);
-	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk", "ipg_root", ccm_base + 0x4650, 0);
+	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk", "ipg_audio_root", ccm_base + 0x4650, 0);
 
 	hws[IMX8MP_CLK_ARM] = imx_clk_hw_cpu("arm", "arm_a53_core",
 					     hws[IMX8MP_CLK_A53_CORE]->clk,
Signature · 1 line
2.7.4
#1 lore raw
Quoted 6 lines

From: Abel Vesa <[email protected]>
Sent: Monday, April 27, 2020 11:11 PM

Instead of ipg_root, the parent needs to be ipg_audio_root.

I have a few doubts about this patch:
1. From latest RM, it seems CCGR101 (0x4650) is a shared gate for many audio instances.
2. If this patch is about AUDIO_AHB_CLK_ROOT, then it's parent is AHB[POST_PODF] from the clock tree in RM. Not quite understand why this patch changes to IPG[POST_PODF]. Is this RM incorrect issue?

BTW, if this patch is taken from someone else, we usually better keep the original author if not fundamental changes.

Regards
Aisheng

Quoted 25 lines
---
 drivers/clk/imx/clk-imx8mp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/imx/clk-imx8mp.c index renamed from drivers/clk/imx/clk-imx8mp.c +0 -0

41469e2..dcdfc9d 100644

drivers/clk/imx/clk-imx8mp.c +0 -1
@@ -727,7 +727,7 @@ static int imx8mp_clocks_probe(struct platform_device
 *pdev)
 	hws[IMX8MP_CLK_HDMI_ROOT] = imx_clk_hw_gate4("hdmi_root_clk",
 "hdmi_axi", ccm_base + 0x45f0, 0);
 	hws[IMX8MP_CLK_TSENSOR_ROOT] =
 imx_clk_hw_gate4("tsensor_root_clk", "ipg_root", ccm_base + 0x4620, 0);
 	hws[IMX8MP_CLK_VPU_ROOT] = imx_clk_hw_gate4("vpu_root_clk",
 "vpu_bus", ccm_base + 0x4630, 0);
-	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",

"ipg_root", ccm_base + 0x4650, 0);

drivers/clk/imx/clk-imx8mp.c +2 -1
+	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",
+"ipg_audio_root", ccm_base + 0x4650, 0);
 
 	hws[IMX8MP_CLK_ARM] = imx_clk_hw_cpu("arm", "arm_a53_core",
 					     hws[IMX8MP_CLK_A53_CORE]->clk,
--

2.7.4

author
#2 lore raw
Quoted 5 lines · Aisheng Dong
Quoted 5 lines

From: Abel Vesa <[email protected]>
Sent: Monday, April 27, 2020 11:11 PM

Instead of ipg_root, the parent needs to be ipg_audio_root.

Quoted 9 lines
Quoted 1 line

I have a few doubts about this patch:
1. From latest RM, it seems CCGR101 (0x4650) is a shared gate for many audio instances.
2. If this patch is about AUDIO_AHB_CLK_ROOT, then it's parent is AHB[POST_PODF] from the clock tree in RM. Not quite understand why this patch changes to IPG[POST_PODF]. Is this RM incorrect issue?

BTW, if this patch is taken from someone else, we usually better keep the original author if not fundamental changes.

I made this change at the suggestion from S.j. Wang. I'm the original author in linux-nxp (internal tree).

Quoted 29 lines

Regards
Aisheng

Quoted 25 lines
---
 drivers/clk/imx/clk-imx8mp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/imx/clk-imx8mp.c index renamed from drivers/clk/imx/clk-imx8mp.c +0 -0

41469e2..dcdfc9d 100644

drivers/clk/imx/clk-imx8mp.c +0 -1
@@ -727,7 +727,7 @@ static int imx8mp_clocks_probe(struct platform_device
 *pdev)
 	hws[IMX8MP_CLK_HDMI_ROOT] = imx_clk_hw_gate4("hdmi_root_clk",
 "hdmi_axi", ccm_base + 0x45f0, 0);
 	hws[IMX8MP_CLK_TSENSOR_ROOT] =
 imx_clk_hw_gate4("tsensor_root_clk", "ipg_root", ccm_base + 0x4620, 0);
 	hws[IMX8MP_CLK_VPU_ROOT] = imx_clk_hw_gate4("vpu_root_clk",
 "vpu_bus", ccm_base + 0x4630, 0);
-	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",

"ipg_root", ccm_base + 0x4650, 0);

drivers/clk/imx/clk-imx8mp.c +2 -1
+	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",
+"ipg_audio_root", ccm_base + 0x4650, 0);
 
 	hws[IMX8MP_CLK_ARM] = imx_clk_hw_cpu("arm", "arm_a53_core",
 					     hws[IMX8MP_CLK_A53_CORE]->clk,
--

2.7.4

#3 lore raw
Quoted 17 lines

From: Abel Vesa <[email protected]>
Sent: Thursday, April 30, 2020 6:11 PM

Quoted 9 lines · Aisheng Dong
Quoted 6 lines

From: Abel Vesa <[email protected]>
Sent: Monday, April 27, 2020 11:11 PM

Instead of ipg_root, the parent needs to be ipg_audio_root.

I have a few doubts about this patch:
1. From latest RM, it seems CCGR101 (0x4650) is a shared gate for many audio

instances.

Quoted 1 line

2. If this patch is about AUDIO_AHB_CLK_ROOT, then it's parent is

AHB[POST_PODF] from the clock tree in RM.

Quoted 1 line

Not quite understand why this patch changes to IPG[POST_PODF]. Is this RM

Quoted 9 lines

incorrect issue?

Quoted 2 lines

BTW, if this patch is taken from someone else, we usually better keep the

original author if not fundamental changes.

Quoted 1 line

I made this change at the suggestion from S.j. Wang. I'm the original author in linux-nxp (internal tree).

That's fine. Then please ignore my "stupid" reminder. 😊 BTW, how about other questions?

Regards
Aisheng

Quoted 31 lines
Quoted 7 lines

Regards
Aisheng

Quoted 4 lines
---
 drivers/clk/imx/clk-imx8mp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 8 lines
Quoted 8 lines
(quoted diff) +0 -0

b/drivers/clk/imx/clk-imx8mp.c index 41469e2..dcdfc9d 100644

drivers/clk/imx/clk-imx8mp.c +0 -0
@@ -727,7 +727,7 @@ static int imx8mp_clocks_probe(struct

platform_device
*pdev)
hws[IMX8MP_CLK_HDMI_ROOT] =

imx_clk_hw_gate4("hdmi_root_clk",

Quoted 15 lines
Quoted 14 lines

"hdmi_axi", ccm_base + 0x45f0, 0);
hws[IMX8MP_CLK_TSENSOR_ROOT] =
imx_clk_hw_gate4("tsensor_root_clk", "ipg_root", ccm_base + 0x4620, 0);
hws[IMX8MP_CLK_VPU_ROOT] = imx_clk_hw_gate4("vpu_root_clk",
"vpu_bus", ccm_base + 0x4630, 0);

drivers/clk/imx/clk-imx8mp.c +0 -1
-	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",

"ipg_root", ccm_base + 0x4650, 0);

drivers/clk/imx/clk-imx8mp.c +2 -1
+	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",
+"ipg_audio_root", ccm_base + 0x4650, 0);
 
 	hws[IMX8MP_CLK_ARM] = imx_clk_hw_cpu("arm", "arm_a53_core",
 					     hws[IMX8MP_CLK_A53_CORE]->clk,
--

2.7.4

author
#4 lore raw
Quoted 3 lines · Aisheng Dong
Quoted 3 lines

From: Abel Vesa <[email protected]>
Sent: Thursday, April 30, 2020 6:11 PM

Sorry I didn't answer the other ones earlier. See below.

Quoted 11 lines
Quoted 11 lines · Aisheng Dong
Quoted 9 lines
Quoted 6 lines

From: Abel Vesa <[email protected]>
Sent: Monday, April 27, 2020 11:11 PM

Instead of ipg_root, the parent needs to be ipg_audio_root.

I have a few doubts about this patch:
1. From latest RM, it seems CCGR101 (0x4650) is a shared gate for many audio

instances.

Yes. This is true. I'm working on adding the shared gate now.

Quoted 4 lines
Quoted 4 lines
Quoted 1 line

2. If this patch is about AUDIO_AHB_CLK_ROOT, then it's parent is

AHB[POST_PODF] from the clock tree in RM.

Quoted 1 line

Not quite understand why this patch changes to IPG[POST_PODF]. Is this RM

incorrect issue?

Yes. Right again. It needs to be the IMX8MP_CLK_AUDIO_AHB. Will change that now.

Quoted 46 lines
Quoted 8 lines
Quoted 2 lines

BTW, if this patch is taken from someone else, we usually better keep the

original author if not fundamental changes.

Quoted 1 line

I made this change at the suggestion from S.j. Wang. I'm the original author in linux-nxp (internal tree).

That's fine. Then please ignore my "stupid" reminder. 😊 BTW, how about other questions?

Regards
Aisheng

Quoted 31 lines
Quoted 7 lines

Regards
Aisheng

Quoted 4 lines
---
 drivers/clk/imx/clk-imx8mp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Quoted 8 lines
Quoted 8 lines
(quoted diff) +0 -0

b/drivers/clk/imx/clk-imx8mp.c index 41469e2..dcdfc9d 100644

drivers/clk/imx/clk-imx8mp.c +0 -0
@@ -727,7 +727,7 @@ static int imx8mp_clocks_probe(struct

platform_device
*pdev)
hws[IMX8MP_CLK_HDMI_ROOT] =

imx_clk_hw_gate4("hdmi_root_clk",

Quoted 15 lines
Quoted 14 lines

"hdmi_axi", ccm_base + 0x45f0, 0);
hws[IMX8MP_CLK_TSENSOR_ROOT] =
imx_clk_hw_gate4("tsensor_root_clk", "ipg_root", ccm_base + 0x4620, 0);
hws[IMX8MP_CLK_VPU_ROOT] = imx_clk_hw_gate4("vpu_root_clk",
"vpu_bus", ccm_base + 0x4630, 0);

drivers/clk/imx/clk-imx8mp.c +0 -1
-	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",

"ipg_root", ccm_base + 0x4650, 0);

drivers/clk/imx/clk-imx8mp.c +2 -1
+	hws[IMX8MP_CLK_AUDIO_ROOT] = imx_clk_hw_gate4("audio_root_clk",
+"ipg_audio_root", ccm_base + 0x4650, 0);
 
 	hws[IMX8MP_CLK_ARM] = imx_clk_hw_cpu("arm", "arm_a53_core",
 					     hws[IMX8MP_CLK_A53_CORE]->clk,
--

2.7.4