Skip to contentSkip to conversation

clk: tegra: remove USB from clk init table

PATCH RESEND In review Lucas Stach 6 participants 14 messages Apr 15, 2013 – May 6, 2013 6 messages not in the archive

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH RESEND 1/2 clk: tegra: remove USB from clk init table

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

---
Trace produced by system with 13MHz clk_m:

tegra_init_from_table: Failed to set rate 12000000 of usbd

------------[ cut here ]------------
WARNING: at drivers/clk/tegra/clk.c:64
tegra_init_from_table+0xc0/0x158()
Modules linked in:
[<c0013a84>] (unwind_backtrace+0x0/0xf8) from
[<c0021a24>](warn_slowpath_common+0x4c/0x64)
[<c0021a24>] (warn_slowpath_common+0x4c/0x64) from [<c0021a58>]
(warn_slowpath_null+0x1c/0x24)
[<c0021a58>] (warn_slowpath_null+0x1c/0x24) from [<c069343c>]
(tegra_init_from_table+0xc0/0x158)
[<c069343c>] (tegra_init_from_table+0xc0/0x158) from [<c0694878>]
(tegra20_clock_init+0x1398/0x13d4)
[<c0694878>] (tegra20_clock_init+0x1398/0x13d4) from [<c0693298>]
(of_clk_init+0x30/0x58)
[<c0693298>] (of_clk_init+0x30/0x58) from [<c0681e5c>]
(tegra_dt_init_irq+0x8/0x1c)
[<c0681e5c>] (tegra_dt_init_irq+0x8/0x1c) from [<c067d334>]
(init_IRQ+0x14/0x1c)
[<c067d334>] (init_IRQ+0x14/0x1c) from [<c067b6b4>]
(start_kernel+0x1a0/0x2f8)
[<c067b6b4>] (start_kernel+0x1a0/0x2f8) from [<0000807c>] (0x807c)
---[ end trace 1b75b31a2719ed1c ]---
---
 drivers/clk/tegra/clk-tegra20.c | 3 ---
 1 file changed, 3 deletions(-)
drivers/clk/tegra/clk-tegra20.c +0 -3
index f873dce..a73278f 100644
@@ -1259,9 +1259,6 @@ static __initdata struct tegra_clk_init_table init_table[] = {
 	{uartc, pll_p, 0, 0},
 	{uartd, pll_p, 0, 0},
 	{uarte, pll_p, 0, 0},
-	{usbd, clk_max, 12000000, 0},
-	{usb2, clk_max, 12000000, 0},
-	{usb3, clk_max, 12000000, 0},
 	{pll_a, clk_max, 56448000, 1},
 	{pll_a_out0, clk_max, 11289600, 1},
 	{cdev1, clk_max, 0, 1},
Signature · 1 line
1.8.1.4

A message that is not in the archive (sent off-list or not yet received): 1366011105-2351-1-git-send-email-dev-8ppwABl0HbeELgA04lAiVw@public.gmane.org

#2 lore raw

PATCH RESEND 2/2 clk: tegra: add ac97 controller clock

AC97 controller clock is hardwired to pll_a_out0.

---
 drivers/clk/tegra/clk-tegra20.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/clk/tegra/clk-tegra20.c +8 -0
index a73278f..bbcca91 100644
@@ -897,6 +897,14 @@ static void __init tegra20_periph_clk_init(void)
 	struct clk *clk;
 	int i;
 
+	/* ac97 */
+	clk = tegra_clk_register_periph_gate("ac97", "pll_a_out0",
+				    TEGRA_PERIPH_ON_APB,
+				    clk_base, 0, 3, &periph_l_regs,
+				    periph_clk_enb_refcnt);
+	clk_register_clkdev(clk, NULL, "tegra20-ac97");
+	clks[ac97] = clk;
+
 	/* apbdma */
 	clk = tegra_clk_register_periph_gate("apbdma", "pclk", 0, clk_base,
 				    0, 34, &periph_h_regs,
Signature · 1 line
1.8.1.4

A message that is not in the archive (sent off-list or not yet received): 1366011105-2351-2-git-send-email-dev-8ppwABl0HbeELgA04lAiVw@public.gmane.org

#4 lore raw

Re: [PATCH resend 2/2] clk: tegra: add ac97 controller clock

Quoted 2 lines · Lucas Stach

AC97 controller clock is hardwired to pll_a_out0.

Prashant, did you just forget to add this clock?

Quoted 14 lines
---
 drivers/clk/tegra/clk-tegra20.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/clk/tegra/clk-tegra20.c +1 -0
index a73278f..bbcca91 100644
@@ -897,6 +897,14 @@ static void __init tegra20_periph_clk_init(void)
 	struct clk *clk;
 	int i;
 
+	/* ac97 */

drivers/clk/tegra/clk-tegra20.c static void __init tegra20_periph_clk_init(void) · line 910

Quoted 10 lines
drivers/clk/tegra/clk-tegra20.c +7 -0
+	clk = tegra_clk_register_periph_gate("ac97", "pll_a_out0",
+				    TEGRA_PERIPH_ON_APB,
+				    clk_base, 0, 3, &periph_l_regs,
+				    periph_clk_enb_refcnt);
+	clk_register_clkdev(clk, NULL, "tegra20-ac97");
+	clks[ac97] = clk;
+
 	/* apbdma */
 	clk = tegra_clk_register_periph_gate("apbdma", "pclk", 0, clk_base,
 				    0, 34, &periph_h_regs,

Otherwise:

Reviewed-by
#5 lore raw

Re: [PATCH resend 2/2] clk: tegra: add ac97 controller clock

In reply to a message not in the archive
Quoted 4 lines · Lucas Stach

AC97 controller clock is hardwired to pll_a_out0.

---

It was not there in previous implementation neither do we implement it in our downstream kernel.

Lucas, are you using this clock anywhere?

Quoted 22 lines
  drivers/clk/tegra/clk-tegra20.c | 8 ++++++++
  1 file changed, 8 insertions(+)
drivers/clk/tegra/clk-tegra20.c +8 -0
index a73278f..bbcca91 100644
@@ -897,6 +897,14 @@ static void __init tegra20_periph_clk_init(void)
  	struct clk *clk;
  	int i;
 
+	/* ac97 */
+	clk = tegra_clk_register_periph_gate("ac97", "pll_a_out0",
+				    TEGRA_PERIPH_ON_APB,
+				    clk_base, 0, 3, &periph_l_regs,
+				    periph_clk_enb_refcnt);
+	clk_register_clkdev(clk, NULL, "tegra20-ac97");
+	clks[ac97] = clk;
+
  	/* apbdma */
  	clk = tegra_clk_register_periph_gate("apbdma", "pclk", 0, clk_base,
  				    0, 34, &periph_h_regs,

A message that is not in the archive (sent off-list or not yet received): [email protected]

#7 lore raw

Re: [PATCH resend 2/2] clk: tegra: add ac97 controller clock

Quoted 10 lines · Prashant Gaikwad
Quoted 4 lines · Lucas Stach

AC97 controller clock is hardwired to pll_a_out0.

---

It was not there in previous implementation neither do we implement it in our downstream kernel.

Lucas, are you using this clock anywhere?

Yes, there's an AC'97 driver upstream now, which Lucas wrote. It's used on the Colibri T20 board.

#8 lore raw
In reply to a message not in the archive
Quoted 9 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

less entries in the clock init table is always good.

Cheers,

Peter.

#9 lore raw
In reply to a message not in the archive
Quoted 8 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Peter, Prashant, I'd like to confirm that the usb* clocks really do have clk_m as their parent; we're sure they aren't driven by the 12MHz PLL_U output?

Either way, I guess it's safe to take this patch since the clock would be fixed rate; I'd just like to make sure the clock driver is accurate.

A message that is not in the archive (sent off-list or not yet received): [email protected]

#11 lore raw
Quoted 9 lines · Stephen Warren
Quoted 8 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Quoted 6 lines

Peter, Prashant, I'd like to confirm that the usb* clocks really do have clk_m as their parent; we're sure they aren't driven by the 12MHz PLL_U output?

Either way, I guess it's safe to take this patch since the clock would be fixed rate; I'd just like to make sure the clock driver is accurate.

These are controller clocks and are not driven by PLL_U.

A message that is not in the archive (sent off-list or not yet received): [email protected]

#13 lore raw
Quoted 10 lines · Prashant Gaikwad
Quoted 9 lines · Stephen Warren
Quoted 8 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Quoted 8 lines
Quoted 6 lines

Peter, Prashant, I'd like to confirm that the usb* clocks really do have clk_m as their parent; we're sure they aren't driven by the 12MHz PLL_U output?

Either way, I guess it's safe to take this patch since the clock would be fixed rate; I'd just like to make sure the clock driver is accurate.

These are controller clocks and are not driven by PLL_U.

So just to confirm: does that mean that they truly /are/ direct children of clk_m, just like the driver says right now?

Reviewed-by
#14 lore raw
In reply to a message not in the archive
Quoted 5 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

Quoted 7 lines

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

---
Trace produced by system with 13MHz clk_m:

Quoted 41 lines

tegra_init_from_table: Failed to set rate 12000000 of usbd

------------[ cut here ]------------
WARNING: at drivers/clk/tegra/clk.c:64
tegra_init_from_table+0xc0/0x158()
Modules linked in:
[<c0013a84>] (unwind_backtrace+0x0/0xf8) from
[<c0021a24>](warn_slowpath_common+0x4c/0x64)
[<c0021a24>] (warn_slowpath_common+0x4c/0x64) from [<c0021a58>]
(warn_slowpath_null+0x1c/0x24)
[<c0021a58>] (warn_slowpath_null+0x1c/0x24) from [<c069343c>]
(tegra_init_from_table+0xc0/0x158)
[<c069343c>] (tegra_init_from_table+0xc0/0x158) from [<c0694878>]
(tegra20_clock_init+0x1398/0x13d4)
[<c0694878>] (tegra20_clock_init+0x1398/0x13d4) from [<c0693298>]
(of_clk_init+0x30/0x58)
[<c0693298>] (of_clk_init+0x30/0x58) from [<c0681e5c>]
(tegra_dt_init_irq+0x8/0x1c)
[<c0681e5c>] (tegra_dt_init_irq+0x8/0x1c) from [<c067d334>]
(init_IRQ+0x14/0x1c)
[<c067d334>] (init_IRQ+0x14/0x1c) from [<c067b6b4>]
(start_kernel+0x1a0/0x2f8)
[<c067b6b4>] (start_kernel+0x1a0/0x2f8) from [<0000807c>] (0x807c)
---[ end trace 1b75b31a2719ed1c ]---
---
  drivers/clk/tegra/clk-tegra20.c | 3 ---
  1 file changed, 3 deletions(-)
drivers/clk/tegra/clk-tegra20.c +0 -3
index f873dce..a73278f 100644
@@ -1259,9 +1259,6 @@ static __initdata struct tegra_clk_init_table init_table[] = {
  	{uartc, pll_p, 0, 0},
  	{uartd, pll_p, 0, 0},
  	{uarte, pll_p, 0, 0},
-	{usbd, clk_max, 12000000, 0},
-	{usb2, clk_max, 12000000, 0},
-	{usb3, clk_max, 12000000, 0},
  	{pll_a, clk_max, 56448000, 1},
  	{pll_a_out0, clk_max, 11289600, 1},
  	{cdev1, clk_max, 0, 1},
Acked-by Tested-by
#15 lore raw
In reply to a message not in the archive
Quoted 8 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Lucas, these two patches look good for me. I'm just waiting on Mike's
ack to forward them to arm-soc. If that doesn't happen today, feel free
to forward them to [email protected] when it does, with a note requesting they be applied since I'm away on vacation. It might be useful for them
to know that they're best applied wherever Tegra's for-3.10/clk branch
was merged.

(Tested for regressions; I didn't actually test using the AC'97 clock, since I don't have the HW)

A message that is not in the archive (sent off-list or not yet received): [email protected]

#17 lore raw

Quoting Stephen Warren (2013-04-17 10:29:47)

Quoted 17 lines · Lucas Stach
Quoted 8 lines

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Lucas, these two patches look good for me. I'm just waiting on Mike's
ack to forward them to arm-soc. If that doesn't happen today, feel free
to forward them to [email protected] when it does, with a note requesting they be applied since I'm away on vacation. It might be useful for them
to know that they're best applied wherever Tegra's for-3.10/clk branch
was merged.

Quoted 5 lines

(Tested for regressions; I didn't actually test using the AC'97 clock, since I don't have the HW)

#18 lore raw
In reply to a message not in the archive
Quoted 17 lines · Stephen Warren
Quoted 8 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Lucas, these two patches look good for me. I'm just waiting on Mike's
ack to forward them to arm-soc. If that doesn't happen today, feel free
to forward them to [email protected] when it does, with a note requesting they be applied since I'm away on vacation. It might be useful for them
to know that they're best applied wherever Tegra's for-3.10/clk branch
was merged.

Quoted 5 lines

(Tested for regressions; I didn't actually test using the AC'97 clock, since I don't have the HW)

Ok, I take back my earlier agreement to apply patches that I just get a fresh acked-by cc for. This doesn't seem to scale very well since I have to go from mutt in a terminal back to gmail, find the patch, make sure I have the right version of that patch, add the thread to my inbox, run offlineimap, wait for it to show up in mutt and _then_ "git am" it, adding any acked-by or other tags.

So, if you want me to apply patches from email, please do so by (re)sending me the patch directly instead.

-Olof

#19 lore raw
In reply to a message not in the archive
Quoted 6 lines · Stephen Warren
Quoted 5 lines · Lucas Stach

The USB clocks are just clock gates, so no need to set a specific clock. In fact trying to set a specific clock is just a NOP if the requested clockrate is the same as those of the parent (clk_m) or will trigger a WARN_ON() if rates don't match up.

Quoted 8 lines
Quoted 3 lines

As we are not setting a specific rate, nor activating the clocks at init, there is no point in keeping the the usb entries in the clock init table.

Lucas, these two patches look good for me. I'm just waiting on Mike's
ack to forward them to arm-soc. If that doesn't happen today, feel free
to forward them to [email protected] when it does, with a note requesting they be applied since I'm away on vacation. ...

It looks like this didn't happen, so I'll forward the patches today.