All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paul Walmsley <paul@pwsan.com>
To: linux-omap@vger.kernel.org, linux-arm-kernel@lists.infradead.org
Cc: Kevin Hilman <khilman@ti.com>
Subject: [PATCH 03/10] ARM: OMAP2xxx: PM: clean up some crufty powerstate programming code
Date: Sat, 08 Dec 2012 18:23:38 -0700	[thread overview]
Message-ID: <20121209012337.19716.8908.stgit@dusk.lan> (raw)
In-Reply-To: <20121209011755.19716.25244.stgit@dusk.lan>

Don't attempt to put clockdomains to sleep; this should be handled by the
clock framework.  It should be enough to program the next-power-state,
and then let the code in omap_pm_clkdms_setup() deal with the rest.

Start out by programming the MPU and CORE powerdomains to stay ON.
Then control the MPU and CORE powerdomain states directly in
omap2_enter_full_retention() and omap2_enter_mpu_retention().  Not the
most optimal way to do it, but certainly is the most conservative until
OMAP2xxx PM is working again.

Get rid of the open-coded PM_PWSTCTRL_MPU writes in
omap2_enter_mpu_retention(); use the powerdomain code instead.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Kevin Hilman <khilman@ti.com>
---
 arch/arm/mach-omap2/pm24xx.c |   30 ++++++++++--------------------
 1 file changed, 10 insertions(+), 20 deletions(-)

diff --git a/arch/arm/mach-omap2/pm24xx.c b/arch/arm/mach-omap2/pm24xx.c
index 83815dd..4c06d8a 100644
--- a/arch/arm/mach-omap2/pm24xx.c
+++ b/arch/arm/mach-omap2/pm24xx.c
@@ -89,11 +89,7 @@ static int omap2_enter_full_retention(void)
 	omap2_prm_write_mod_reg(0xffffffff, CORE_MOD, OMAP24XX_PM_WKST2);
 	omap2_prm_write_mod_reg(0xffffffff, WKUP_MOD, PM_WKST);
 
-	/*
-	 * Set MPU powerdomain's next power state to RETENTION;
-	 * preserve logic state during retention
-	 */
-	pwrdm_set_logic_retst(mpu_pwrdm, PWRDM_POWER_RET);
+	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_RET);
 	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
 
 	/* Workaround to kill USB */
@@ -136,6 +132,9 @@ no_sleep:
 	/* Mask future PRCM-to-MPU interrupts */
 	omap2_prm_write_mod_reg(0x0, OCP_MOD, OMAP2_PRCM_IRQSTATUS_MPU_OFFSET);
 
+	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
+	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_ON);
+
 	return 0;
 }
 
@@ -185,17 +184,16 @@ static void omap2_enter_mpu_retention(void)
 		omap2_prm_write_mod_reg(0xffffffff, WKUP_MOD, PM_WKST);
 
 		/* Try to enter MPU retention */
-		omap2_prm_write_mod_reg((0x01 << OMAP_POWERSTATE_SHIFT) |
-				  OMAP_LOGICRETSTATE_MASK,
-				  MPU_MOD, OMAP2_PM_PWSTCTRL);
+		pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
+
 	} else {
 		/* Block MPU retention */
-
-		omap2_prm_write_mod_reg(OMAP_LOGICRETSTATE_MASK, MPU_MOD,
-						 OMAP2_PM_PWSTCTRL);
+		pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
 	}
 
 	omap2_sram_idle();
+
+	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
 }
 
 static int omap2_can_sleep(void)
@@ -250,25 +248,17 @@ static void __init prcm_setup_regs(void)
 	for (i = 0; i < num_mem_banks; i++)
 		pwrdm_set_mem_retst(core_pwrdm, i, PWRDM_POWER_RET);
 
-	/* Set CORE powerdomain's next power state to RETENTION */
-	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_RET);
+	pwrdm_set_logic_retst(core_pwrdm, PWRDM_POWER_RET);
 
-	/*
-	 * Set MPU powerdomain's next power state to RETENTION;
-	 * preserve logic state during retention
-	 */
 	pwrdm_set_logic_retst(mpu_pwrdm, PWRDM_POWER_RET);
-	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
 
 	/* Force-power down DSP, GFX powerdomains */
 
 	pwrdm = clkdm_get_pwrdm(dsp_clkdm);
 	pwrdm_set_next_pwrst(pwrdm, PWRDM_POWER_OFF);
-	clkdm_sleep(dsp_clkdm);
 
 	pwrdm = clkdm_get_pwrdm(gfx_clkdm);
 	pwrdm_set_next_pwrst(pwrdm, PWRDM_POWER_OFF);
-	clkdm_sleep(gfx_clkdm);
 
 	/* Enable hardware-supervised idle for all clkdms */
 	clkdm_for_each(omap_pm_clkdms_setup, NULL);



WARNING: multiple messages have this Message-ID (diff)
From: paul@pwsan.com (Paul Walmsley)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 03/10] ARM: OMAP2xxx: PM: clean up some crufty powerstate programming code
Date: Sat, 08 Dec 2012 18:23:38 -0700	[thread overview]
Message-ID: <20121209012337.19716.8908.stgit@dusk.lan> (raw)
In-Reply-To: <20121209011755.19716.25244.stgit@dusk.lan>

Don't attempt to put clockdomains to sleep; this should be handled by the
clock framework.  It should be enough to program the next-power-state,
and then let the code in omap_pm_clkdms_setup() deal with the rest.

Start out by programming the MPU and CORE powerdomains to stay ON.
Then control the MPU and CORE powerdomain states directly in
omap2_enter_full_retention() and omap2_enter_mpu_retention().  Not the
most optimal way to do it, but certainly is the most conservative until
OMAP2xxx PM is working again.

Get rid of the open-coded PM_PWSTCTRL_MPU writes in
omap2_enter_mpu_retention(); use the powerdomain code instead.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Kevin Hilman <khilman@ti.com>
---
 arch/arm/mach-omap2/pm24xx.c |   30 ++++++++++--------------------
 1 file changed, 10 insertions(+), 20 deletions(-)

diff --git a/arch/arm/mach-omap2/pm24xx.c b/arch/arm/mach-omap2/pm24xx.c
index 83815dd..4c06d8a 100644
--- a/arch/arm/mach-omap2/pm24xx.c
+++ b/arch/arm/mach-omap2/pm24xx.c
@@ -89,11 +89,7 @@ static int omap2_enter_full_retention(void)
 	omap2_prm_write_mod_reg(0xffffffff, CORE_MOD, OMAP24XX_PM_WKST2);
 	omap2_prm_write_mod_reg(0xffffffff, WKUP_MOD, PM_WKST);
 
-	/*
-	 * Set MPU powerdomain's next power state to RETENTION;
-	 * preserve logic state during retention
-	 */
-	pwrdm_set_logic_retst(mpu_pwrdm, PWRDM_POWER_RET);
+	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_RET);
 	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
 
 	/* Workaround to kill USB */
@@ -136,6 +132,9 @@ no_sleep:
 	/* Mask future PRCM-to-MPU interrupts */
 	omap2_prm_write_mod_reg(0x0, OCP_MOD, OMAP2_PRCM_IRQSTATUS_MPU_OFFSET);
 
+	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
+	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_ON);
+
 	return 0;
 }
 
@@ -185,17 +184,16 @@ static void omap2_enter_mpu_retention(void)
 		omap2_prm_write_mod_reg(0xffffffff, WKUP_MOD, PM_WKST);
 
 		/* Try to enter MPU retention */
-		omap2_prm_write_mod_reg((0x01 << OMAP_POWERSTATE_SHIFT) |
-				  OMAP_LOGICRETSTATE_MASK,
-				  MPU_MOD, OMAP2_PM_PWSTCTRL);
+		pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
+
 	} else {
 		/* Block MPU retention */
-
-		omap2_prm_write_mod_reg(OMAP_LOGICRETSTATE_MASK, MPU_MOD,
-						 OMAP2_PM_PWSTCTRL);
+		pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
 	}
 
 	omap2_sram_idle();
+
+	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_ON);
 }
 
 static int omap2_can_sleep(void)
@@ -250,25 +248,17 @@ static void __init prcm_setup_regs(void)
 	for (i = 0; i < num_mem_banks; i++)
 		pwrdm_set_mem_retst(core_pwrdm, i, PWRDM_POWER_RET);
 
-	/* Set CORE powerdomain's next power state to RETENTION */
-	pwrdm_set_next_pwrst(core_pwrdm, PWRDM_POWER_RET);
+	pwrdm_set_logic_retst(core_pwrdm, PWRDM_POWER_RET);
 
-	/*
-	 * Set MPU powerdomain's next power state to RETENTION;
-	 * preserve logic state during retention
-	 */
 	pwrdm_set_logic_retst(mpu_pwrdm, PWRDM_POWER_RET);
-	pwrdm_set_next_pwrst(mpu_pwrdm, PWRDM_POWER_RET);
 
 	/* Force-power down DSP, GFX powerdomains */
 
 	pwrdm = clkdm_get_pwrdm(dsp_clkdm);
 	pwrdm_set_next_pwrst(pwrdm, PWRDM_POWER_OFF);
-	clkdm_sleep(dsp_clkdm);
 
 	pwrdm = clkdm_get_pwrdm(gfx_clkdm);
 	pwrdm_set_next_pwrst(pwrdm, PWRDM_POWER_OFF);
-	clkdm_sleep(gfx_clkdm);
 
 	/* Enable hardware-supervised idle for all clkdms */
 	clkdm_for_each(omap_pm_clkdms_setup, NULL);

  parent reply	other threads:[~2012-12-09  1:24 UTC|newest]

Thread overview: 56+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-09  1:23 [PATCH 00/10] ARM: OMAP2+: second set of PM fixes and cleanup for 3.9 Paul Walmsley
2012-12-09  1:23 ` Paul Walmsley
2012-12-09  1:23 ` [PATCH 02/10] ARM: OMAP2+: clockdomain: add pwrdm_state_switch() call to clkdm_sleep() Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-09  1:23 ` [PATCH 01/10] ARM: OMAP3/4: cpuidle: fix sparse and checkpatch warnings Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-12  8:31   ` Vaibhav Hiremath
2012-12-12  8:31     ` Vaibhav Hiremath
2012-12-13  5:41     ` Paul Walmsley
2012-12-13  5:41       ` Paul Walmsley
2012-12-13  5:55       ` Hiremath, Vaibhav
2012-12-13  5:55         ` Hiremath, Vaibhav
2012-12-13  7:29         ` Paul Walmsley
2012-12-13  7:29           ` Paul Walmsley
2012-12-13  9:18           ` Hiremath, Vaibhav
2012-12-13  9:18             ` Hiremath, Vaibhav
2012-12-12  9:28   ` Santosh Shilimkar
2012-12-12  9:28     ` Santosh Shilimkar
2012-12-09  1:23 ` Paul Walmsley [this message]
2012-12-09  1:23   ` [PATCH 03/10] ARM: OMAP2xxx: PM: clean up some crufty powerstate programming code Paul Walmsley
2012-12-09  1:23 ` [PATCH 04/10] ARM: OMAP2: PM/powerdomain: drop unnecessary pwrdm_wait_transition() Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-09  1:23 ` [PATCH 05/10] ARM: OMAP2+: PM/powerdomain: move omap_set_pwrdm_state() to powerdomain code Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-12  9:31   ` Jean Pihet
2012-12-12  9:31     ` Jean Pihet
2013-01-29 20:59     ` Paul Walmsley
2013-01-29 20:59       ` Paul Walmsley
2012-12-12 10:21   ` Vaibhav Hiremath
2012-12-12 10:21     ` Vaibhav Hiremath
2013-01-09 17:43     ` Russell King - ARM Linux
2013-01-09 17:43       ` Russell King - ARM Linux
2012-12-12 10:31   ` Jean Pihet
2012-12-12 10:31     ` Jean Pihet
2012-12-09  1:23 ` [PATCH 06/10] ARM: OMAP2+: powerdomain/clockdomain: add a per-powerdomain spinlock Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-12  9:41   ` Jean Pihet
2012-12-12  9:41     ` Jean Pihet
2013-01-29 21:13     ` Paul Walmsley
2013-01-29 21:13       ` Paul Walmsley
2012-12-12 10:28   ` Jean Pihet
2012-12-12 10:28     ` Jean Pihet
2012-12-09  1:23 ` [PATCH 07/10] ARM: OMAP2xxx: CM: remove autodep handling Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-09  1:23 ` [PATCH 08/10] ARM: OMAP2+: clockdomain: work on wkdep/sleepdep functions Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-09  1:23 ` [PATCH 09/10] ARM: OMAP2+: clockdomain: convert existing atomic usecounts into spinlock-protected shorts/ints Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2012-12-12 10:21   ` Vaibhav Hiremath
2012-12-12 10:21     ` Vaibhav Hiremath
2012-12-26  6:31   ` Bedia, Vaibhav
2012-12-26  6:31     ` Bedia, Vaibhav
2012-12-09  1:23 ` [PATCH 10/10] ARM: OMAP2+: powerdomain: fix whitespace, improve flag comments Paul Walmsley
2012-12-09  1:23   ` Paul Walmsley
2013-01-04 13:07 ` [PATCH 00/10] ARM: OMAP2+: second set of PM fixes and cleanup for 3.9 Tero Kristo
2013-01-04 13:07   ` Tero Kristo

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=20121209012337.19716.8908.stgit@dusk.lan \
    --to=paul@pwsan.com \
    --cc=khilman@ti.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-omap@vger.kernel.org \
    /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.