All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sundar Iyer <sundar.iyer@stericsson.com>
To: <linux-arm-kernel@lists.infradead.org>,
	<dmitry.torokhov@gmail.com>, <sameo@linux.intel.com>,
	<ben-linux@fluff.org>
Cc: <linux-input@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
	Sundar Iyer <sundar.iyer@stericsson.com>
Subject: [PATCH 18/20] mach-ux500: explicit enable MTU TCR in the kernel
Date: Fri, 3 Dec 2010 20:35:51 +0530	[thread overview]
Message-ID: <1291388753-14662-19-git-send-email-sundar.iyer@stericsson.com> (raw)
In-Reply-To: <1291388753-14662-1-git-send-email-sundar.iyer@stericsson.com>

PRCM_TCR enables the various timers in the system. This must be achieved
before any of the MTUs are usable for kernel usage. Explicit enabling of
this in the kernel makes it independent of bootloader actions.

Signed-off-by: Sundar Iyer <sundar.iyer@stericsson.com>
---
 arch/arm/mach-ux500/clock.c |   49 ++++++++-----------------------------------
 1 files changed, 9 insertions(+), 40 deletions(-)

diff --git a/arch/arm/mach-ux500/clock.c b/arch/arm/mach-ux500/clock.c
index 531de5c..6657b05 100644
--- a/arch/arm/mach-ux500/clock.c
+++ b/arch/arm/mach-ux500/clock.c
@@ -132,7 +132,7 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 {
 	void __iomem *addr = __io_address(UX500_PRCMU_BASE)
 		+ PRCM_TCR;
-	u32 tcr = readl(addr);
+	u32 tcr;
 	int mtu = (int) clk->data;
 	/*
 	 * One of these is selected eventually
@@ -143,6 +143,14 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 	unsigned long mturate;
 	unsigned long retclk;
 
+	/*
+	 * On a startup, always conifgure the TCR to the doze mode;
+	 * bootloaders do it for us. Do this in the kernel too.
+	 */
+	writel(PRCM_TCR_DOZE_MODE, addr);
+
+	tcr = readl(addr);
+
 	/* Get the rate from the parent as a default */
 	if (clk->parent_periph)
 		mturate = clk_get_rate(clk->parent_periph);
@@ -152,45 +160,6 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 		/* We need to be connected SOMEWHERE */
 		BUG();
 
-	/*
-	 * Are we in doze mode?
-	 * In this mode the parent peripheral or the fixed 32768 Hz
-	 * clock is fed into the block.
-	 */
-	if (!(tcr & PRCM_TCR_DOZE_MODE)) {
-		/*
-		 * Here we're using the clock input from the APE ULP
-		 * clock domain. But first: are the timers stopped?
-		 */
-		if (tcr & PRCM_TCR_STOPPED) {
-			clk32k = 0;
-			mturate = 0;
-		} else {
-			/* Else default mode: 0 and 2.4 MHz */
-			clk32k = 0;
-			if (cpu_is_u5500())
-				/* DB5500 divides by 8 */
-				mturate /= 8;
-			else if (cpu_is_u8500ed()) {
-				/*
-				 * This clocking setting must not be used
-				 * in the ED chip, it is simply not
-				 * connected anywhere!
-				 */
-				mturate = 0;
-				BUG();
-			} else
-				/*
-				 * In this mode the ulp38m4 clock is divided
-				 * by a factor 16, on the DB8500 typically
-				 * 38400000 / 16 ~ 2.4 MHz.
-				 * TODO: Replace the constant with a reference
-				 * to the ULP source once this is modeled.
-				 */
-				mturate = 38400000 / 16;
-		}
-	}
-
 	/* Return the clock selected for this MTU */
 	if (tcr & (1 << mtu))
 		retclk = clk32k;
-- 
1.7.2.dirty


WARNING: multiple messages have this Message-ID (diff)
From: Sundar Iyer <sundar.iyer@stericsson.com>
To: linux-arm-kernel@lists.infradead.org, dmitry.torokhov@gmail.com,
	sameo@linux.intel.com, ben-linux@fluff.org
Cc: linux-input@vger.kernel.org, linux-kernel@vger.kernel.org,
	Sundar Iyer <sundar.iyer@stericsson.com>
Subject: [PATCH 18/20] mach-ux500: explicit enable MTU TCR in the kernel
Date: Fri, 3 Dec 2010 20:35:51 +0530	[thread overview]
Message-ID: <1291388753-14662-19-git-send-email-sundar.iyer@stericsson.com> (raw)
In-Reply-To: <1291388753-14662-1-git-send-email-sundar.iyer@stericsson.com>

PRCM_TCR enables the various timers in the system. This must be achieved
before any of the MTUs are usable for kernel usage. Explicit enabling of
this in the kernel makes it independent of bootloader actions.

Signed-off-by: Sundar Iyer <sundar.iyer@stericsson.com>
---
 arch/arm/mach-ux500/clock.c |   49 ++++++++-----------------------------------
 1 files changed, 9 insertions(+), 40 deletions(-)

diff --git a/arch/arm/mach-ux500/clock.c b/arch/arm/mach-ux500/clock.c
index 531de5c..6657b05 100644
--- a/arch/arm/mach-ux500/clock.c
+++ b/arch/arm/mach-ux500/clock.c
@@ -132,7 +132,7 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 {
 	void __iomem *addr = __io_address(UX500_PRCMU_BASE)
 		+ PRCM_TCR;
-	u32 tcr = readl(addr);
+	u32 tcr;
 	int mtu = (int) clk->data;
 	/*
 	 * One of these is selected eventually
@@ -143,6 +143,14 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 	unsigned long mturate;
 	unsigned long retclk;
 
+	/*
+	 * On a startup, always conifgure the TCR to the doze mode;
+	 * bootloaders do it for us. Do this in the kernel too.
+	 */
+	writel(PRCM_TCR_DOZE_MODE, addr);
+
+	tcr = readl(addr);
+
 	/* Get the rate from the parent as a default */
 	if (clk->parent_periph)
 		mturate = clk_get_rate(clk->parent_periph);
@@ -152,45 +160,6 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 		/* We need to be connected SOMEWHERE */
 		BUG();
 
-	/*
-	 * Are we in doze mode?
-	 * In this mode the parent peripheral or the fixed 32768 Hz
-	 * clock is fed into the block.
-	 */
-	if (!(tcr & PRCM_TCR_DOZE_MODE)) {
-		/*
-		 * Here we're using the clock input from the APE ULP
-		 * clock domain. But first: are the timers stopped?
-		 */
-		if (tcr & PRCM_TCR_STOPPED) {
-			clk32k = 0;
-			mturate = 0;
-		} else {
-			/* Else default mode: 0 and 2.4 MHz */
-			clk32k = 0;
-			if (cpu_is_u5500())
-				/* DB5500 divides by 8 */
-				mturate /= 8;
-			else if (cpu_is_u8500ed()) {
-				/*
-				 * This clocking setting must not be used
-				 * in the ED chip, it is simply not
-				 * connected anywhere!
-				 */
-				mturate = 0;
-				BUG();
-			} else
-				/*
-				 * In this mode the ulp38m4 clock is divided
-				 * by a factor 16, on the DB8500 typically
-				 * 38400000 / 16 ~ 2.4 MHz.
-				 * TODO: Replace the constant with a reference
-				 * to the ULP source once this is modeled.
-				 */
-				mturate = 38400000 / 16;
-		}
-	}
-
 	/* Return the clock selected for this MTU */
 	if (tcr & (1 << mtu))
 		retclk = clk32k;
-- 
1.7.2.dirty

WARNING: multiple messages have this Message-ID (diff)
From: sundar.iyer@stericsson.com (Sundar Iyer)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 18/20] mach-ux500: explicit enable MTU TCR in the kernel
Date: Fri, 3 Dec 2010 20:35:51 +0530	[thread overview]
Message-ID: <1291388753-14662-19-git-send-email-sundar.iyer@stericsson.com> (raw)
In-Reply-To: <1291388753-14662-1-git-send-email-sundar.iyer@stericsson.com>

PRCM_TCR enables the various timers in the system. This must be achieved
before any of the MTUs are usable for kernel usage. Explicit enabling of
this in the kernel makes it independent of bootloader actions.

Signed-off-by: Sundar Iyer <sundar.iyer@stericsson.com>
---
 arch/arm/mach-ux500/clock.c |   49 ++++++++-----------------------------------
 1 files changed, 9 insertions(+), 40 deletions(-)

diff --git a/arch/arm/mach-ux500/clock.c b/arch/arm/mach-ux500/clock.c
index 531de5c..6657b05 100644
--- a/arch/arm/mach-ux500/clock.c
+++ b/arch/arm/mach-ux500/clock.c
@@ -132,7 +132,7 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 {
 	void __iomem *addr = __io_address(UX500_PRCMU_BASE)
 		+ PRCM_TCR;
-	u32 tcr = readl(addr);
+	u32 tcr;
 	int mtu = (int) clk->data;
 	/*
 	 * One of these is selected eventually
@@ -143,6 +143,14 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 	unsigned long mturate;
 	unsigned long retclk;
 
+	/*
+	 * On a startup, always conifgure the TCR to the doze mode;
+	 * bootloaders do it for us. Do this in the kernel too.
+	 */
+	writel(PRCM_TCR_DOZE_MODE, addr);
+
+	tcr = readl(addr);
+
 	/* Get the rate from the parent as a default */
 	if (clk->parent_periph)
 		mturate = clk_get_rate(clk->parent_periph);
@@ -152,45 +160,6 @@ static unsigned long clk_mtu_get_rate(struct clk *clk)
 		/* We need to be connected SOMEWHERE */
 		BUG();
 
-	/*
-	 * Are we in doze mode?
-	 * In this mode the parent peripheral or the fixed 32768 Hz
-	 * clock is fed into the block.
-	 */
-	if (!(tcr & PRCM_TCR_DOZE_MODE)) {
-		/*
-		 * Here we're using the clock input from the APE ULP
-		 * clock domain. But first: are the timers stopped?
-		 */
-		if (tcr & PRCM_TCR_STOPPED) {
-			clk32k = 0;
-			mturate = 0;
-		} else {
-			/* Else default mode: 0 and 2.4 MHz */
-			clk32k = 0;
-			if (cpu_is_u5500())
-				/* DB5500 divides by 8 */
-				mturate /= 8;
-			else if (cpu_is_u8500ed()) {
-				/*
-				 * This clocking setting must not be used
-				 * in the ED chip, it is simply not
-				 * connected anywhere!
-				 */
-				mturate = 0;
-				BUG();
-			} else
-				/*
-				 * In this mode the ulp38m4 clock is divided
-				 * by a factor 16, on the DB8500 typically
-				 * 38400000 / 16 ~ 2.4 MHz.
-				 * TODO: Replace the constant with a reference
-				 * to the ULP source once this is modeled.
-				 */
-				mturate = 38400000 / 16;
-		}
-	}
-
 	/* Return the clock selected for this MTU */
 	if (tcr & (1 << mtu))
 		retclk = clk32k;
-- 
1.7.2.dirty

  parent reply	other threads:[~2010-12-03 15:07 UTC|newest]

Thread overview: 100+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-03 15:05 [PATCH 00/20] ux500: platform data, TC3589x keypad driver Sundar Iyer
2010-12-03 15:05 ` Sundar Iyer
2010-12-03 15:05 ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 01/20] mfd/ab8500: remove spi support Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 14:29   ` Samuel Ortiz
2010-12-09 14:29     ` Samuel Ortiz
2010-12-09 14:35     ` Sundar R IYER
2010-12-09 14:35       ` Sundar R IYER
2010-12-09 15:36       ` Samuel Ortiz
2010-12-09 15:36         ` Samuel Ortiz
2010-12-19 20:48         ` Linus Walleij
2010-12-19 20:48           ` Linus Walleij
2010-12-24 10:53           ` Samuel Ortiz
2010-12-24 10:53             ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 02/20] mach-ux500: deprecate spi support for ab8500 Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 03/20] mach-ux500: move keymaps to new file Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-19 20:53   ` Linus Walleij
2010-12-19 20:53     ` Linus Walleij
2010-12-03 15:05 ` [PATCH 04/20] nomadik-gpio: allow sleep mode dir/pull to differ from normal mode Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 05/20] mach-ux500: add STMPE1601 platform data Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-19 20:54   ` Linus Walleij
2010-12-19 20:54     ` Linus Walleij
2010-12-03 15:05 ` [PATCH 06/20] mfd/tc35892: rename tc35892 header to tc3589x Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:41   ` Samuel Ortiz
2010-12-09 15:41     ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 07/20] mfd/tc35892: rename tc35892 core driver " Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:42   ` Samuel Ortiz
2010-12-09 15:42     ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 08/20] mfd/tc3589x: rename tc35892 structs/registers to tc359x Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:45   ` Samuel Ortiz
2010-12-09 15:45     ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 09/20] mfd/tc3589x: add block identifier for multiple child devices Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:48   ` Samuel Ortiz
2010-12-09 15:48     ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 10/20] input/tc3589x: add tc3589x keypad support Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-05 18:38   ` Trilok Soni
2010-12-05 18:38     ` Trilok Soni
2010-12-03 15:05 ` [PATCH 11/20] mfd/tc3589x: fix random interrupt misses Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:49   ` Samuel Ortiz
2010-12-09 15:49     ` Samuel Ortiz
2010-12-10  4:31     ` Sundar R IYER
2010-12-03 15:05 ` [PATCH 12/20] mfd/tc3589x: undo gpio module reset during chip init Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 15:50   ` Samuel Ortiz
2010-12-09 15:50     ` Samuel Ortiz
2010-12-03 15:05 ` [PATCH 13/20] mfd/tc3589x: add suspend/resume support Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-09 17:03   ` Samuel Ortiz
2010-12-09 17:03     ` Samuel Ortiz
2010-12-10  4:32     ` Sundar R IYER
2010-12-10  4:32       ` Sundar R IYER
2010-12-10  4:32       ` Sundar R IYER
2010-12-03 15:05 ` [PATCH 14/20] plat-nomadik/gpio: add expander gpio pins enumeration Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 15/20] mach-ux500: add touchscreen interfaces platform data Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 16/20] i2c/nomadik: add adapter name for updated sanity checkings Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 17/20] mach-ux500: add TC35893 keypad platform data Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` Sundar Iyer [this message]
2010-12-03 15:05   ` [PATCH 18/20] mach-ux500: explicit enable MTU TCR in the kernel Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 19/20] mach-ux500: clean up checkpatch spits Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05 ` [PATCH 20/20] i2c/nomadik: some checkpatch warnings Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-03 15:05   ` Sundar Iyer
2010-12-19 21:03 ` [PATCH 00/20] ux500: platform data, TC3589x keypad driver Linus Walleij
2010-12-19 21:03   ` Linus Walleij
2010-12-19 21:03   ` Linus Walleij

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1291388753-14662-19-git-send-email-sundar.iyer@stericsson.com \
    --to=sundar.iyer@stericsson.com \
    --cc=ben-linux@fluff.org \
    --cc=dmitry.torokhov@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-input@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sameo@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.