public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Ed Blake <ed.blake@sondrel.com>
To: tglx@linutronix.de, jason@lakedaemon.net, marc.zyngier@arm.com
Cc: linux-kernel@vger.kernel.org, Ed Blake <ed.blake@sondrel.com>
Subject: [PATCH 2/4] irqchip: imgpdc: Avoid immediate wake event during set_wake
Date: Mon,  2 Oct 2017 10:55:57 +0100	[thread overview]
Message-ID: <1506938159-466-3-git-send-email-ed.blake@sondrel.com> (raw)
In-Reply-To: <1506938159-466-1-git-send-email-ed.blake@sondrel.com>

When enabling sys wakes, an immediate power transition may occur if the
wake input has been active in the past. Work around this by doing the
following:

    1) Set SYS_WAKE_CONFIG to the current value of SOC_POWER so that an
       immediate power transition has no effect
    2) Enable and disable the sys wake to flush the stale wake event
    3) Set SYS_WAKE_CONFIG back to the original value

Signed-off-by: Ed Blake <ed.blake@sondrel.com>
---
 drivers/irqchip/irq-imgpdc.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/irqchip/irq-imgpdc.c b/drivers/irqchip/irq-imgpdc.c
index 2ac3210..22d8f8a 100644
--- a/drivers/irqchip/irq-imgpdc.c
+++ b/drivers/irqchip/irq-imgpdc.c
@@ -8,6 +8,7 @@
  */
 
 #include <linux/bitops.h>
+#include <linux/delay.h>
 #include <linux/interrupt.h>
 #include <linux/irqdomain.h>
 #include <linux/io.h>
@@ -18,6 +19,7 @@
 
 /* PDC interrupt register numbers */
 
+#define PDC_SOC_POWER			0x300
 #define PDC_IRQ_STATUS			0x310
 #define PDC_IRQ_ENABLE			0x314
 #define PDC_IRQ_CLEAR			0x318
@@ -204,6 +206,8 @@ static int pdc_irq_set_wake(struct irq_data *data, unsigned int on)
 	unsigned int mask = (1 << 16) << hw;
 	unsigned int dst_irq;
 	unsigned int *dst_irq_wake_depth;
+	unsigned int syswake, soc_sys_wake_cfg_regoff;
+	u32 soc_sys_wake_cfg, soc_power;
 	int ret;
 
 	/* control the destination IRQ wakeup too for standby mode */
@@ -216,6 +220,27 @@ static int pdc_irq_set_wake(struct irq_data *data, unsigned int on)
 	}
 
 	raw_spin_lock(&priv->lock);
+	/*
+	 * When enabling sys wakes, an immediate power transition may occur if
+	 * the wake input has been active in the past. Work around this:
+	 * 1) Set SYS_WAKE_CONFIG to the current value of SOC_POWER so that an
+	 *    immediate power transition has no effect
+	 * 2) Enable and disable the sys wake to flush the stale wake event
+	 * 3) Set SYS_WAKE_CONFIG back to the original value
+	 */
+	if (hwirq_is_syswake(hw) && on) {
+		syswake = hwirq_to_syswake(hw);
+		soc_sys_wake_cfg_regoff = PDC_SYS_WAKE_CONFIG_BASE +
+					  syswake * PDC_SYS_WAKE_CONFIG_STRIDE;
+		soc_sys_wake_cfg = pdc_read(priv, soc_sys_wake_cfg_regoff);
+		soc_power = pdc_read(priv, PDC_SOC_POWER);
+		pdc_write(priv, soc_sys_wake_cfg_regoff, soc_power);
+		pdc_write(priv, PDC_IRQ_ROUTE, priv->irq_route | mask);
+		udelay(31);
+		pdc_write(priv, PDC_IRQ_ROUTE, priv->irq_route);
+		pdc_write(priv, soc_sys_wake_cfg_regoff, soc_sys_wake_cfg);
+	}
+
 	if (on)
 		priv->irq_route |= mask;
 	else
-- 
1.9.1

  parent reply	other threads:[~2017-10-02  9:57 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-02  9:55 [PATCH 0/4] irqchip: imgpdc: Fix various issues Ed Blake
2017-10-02  9:55 ` [PATCH 1/4] irqchip: imgpdc: Avoid unbalanced irq wake disable Ed Blake
2017-10-02  9:55 ` Ed Blake [this message]
2017-10-02  9:55 ` [PATCH 3/4] irqchip: imgpdc: Set sys wake polarities to active high Ed Blake
2017-10-04 13:14   ` James Hogan
2017-10-05 14:37     ` Ed Blake
2017-10-02  9:55 ` [PATCH 4/4] irqchip: imgpdc: Pass on peripheral mask/unmasks to the parent Ed Blake
2017-10-04 14:03   ` James Hogan
2017-10-05 14:48     ` Ed Blake
2017-10-05 15:26       ` James Hogan
2017-10-05 15:43         ` Ed Blake
2017-10-05 16:42           ` Ed Blake
2017-10-03 20:32 ` [PATCH 0/4] irqchip: imgpdc: Fix various issues Thomas Gleixner

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=1506938159-466-3-git-send-email-ed.blake@sondrel.com \
    --to=ed.blake@sondrel.com \
    --cc=jason@lakedaemon.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marc.zyngier@arm.com \
    --cc=tglx@linutronix.de \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox