From: Alexandre Belloni <alexandre.belloni@free-electrons.com> To: Oleksij Rempel <linux@rempel-privat.de> Cc: Alessandro Zummo <a.zummo@towertech.it>, rtc-linux@googlegroups.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH 2/2] rtc: asm9260: rework locking Date: Wed, 31 Aug 2016 18:09:40 +0200 [thread overview] Message-ID: <20160831160940.eb3ip4t3jfrga3lv@piout.net> (raw) In-Reply-To: <5e6bfec7-d5b6-6311-9cab-76d9aa932408@rempel-privat.de> On 31/08/2016 at 17:14:27 +0200, Oleksij Rempel wrote : > Hi > > Am 31.08.2016 um 02:18 schrieb Alexandre Belloni: > > The rtc-asm9260 driver uses a discrete spinlock (wrongly uninitialized). > > Use the rtc mutex to lock mmio accesses instead. > > > > Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com> > > --- > > drivers/rtc/rtc-asm9260.c | 16 ++-------------- > > 1 file changed, 2 insertions(+), 14 deletions(-) > > > > diff --git a/drivers/rtc/rtc-asm9260.c b/drivers/rtc/rtc-asm9260.c > > index 5219916ce11d..812f131da694 100644 > > --- a/drivers/rtc/rtc-asm9260.c > > +++ b/drivers/rtc/rtc-asm9260.c > > @@ -112,8 +112,6 @@ struct asm9260_rtc_priv { > > void __iomem *iobase; > > struct rtc_device *rtc; > > struct clk *clk; > > - /* io lock */ > > - spinlock_t lock; > > }; > > > > static irqreturn_t asm9260_rtc_irq(int irq, void *dev_id) > > @@ -122,11 +120,13 @@ static irqreturn_t asm9260_rtc_irq(int irq, void *dev_id) > > u32 isr; > > unsigned long events = 0; > > > > + mutex_lock(&priv->rtc->ops_lock); > > isr = ioread32(priv->iobase + HW_CIIR); > > if (!isr) > > return IRQ_NONE; > > > > iowrite32(0, priv->iobase + HW_CIIR); > > + mutex_unlock(&priv->rtc->ops_lock); > > This means if we have shared or spurious interrupt we will get a > deadlock here? > True, I'll fix that. I actually got it right at first and then I toyed with sparse/smatch to see whether it was properly checking the locking. Seems like my setup was not correct. Thanks for the review ! > > > events |= RTC_AF | RTC_IRQF; > > > > @@ -139,9 +139,7 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > u32 ctime0, ctime1, ctime2; > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > ctime0 = ioread32(priv->iobase + HW_CTIME0); > > ctime1 = ioread32(priv->iobase + HW_CTIME1); > > ctime2 = ioread32(priv->iobase + HW_CTIME2); > > @@ -155,7 +153,6 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > ctime1 = ioread32(priv->iobase + HW_CTIME1); > > ctime2 = ioread32(priv->iobase + HW_CTIME2); > > } > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > tm->tm_sec = (ctime0 >> BM_CTIME0_SEC_S) & BM_CTIME0_SEC_M; > > tm->tm_min = (ctime0 >> BM_CTIME0_MIN_S) & BM_CTIME0_MIN_M; > > @@ -174,9 +171,7 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > /* > > * make sure SEC counter will not flip other counter on write time, > > * real value will be written at the enf of sequence. > > @@ -191,7 +186,6 @@ static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > iowrite32(tm->tm_hour, priv->iobase + HW_HOUR); > > iowrite32(tm->tm_min, priv->iobase + HW_MIN); > > iowrite32(tm->tm_sec, priv->iobase + HW_SEC); > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return 0; > > } > > @@ -199,9 +193,7 @@ static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > alrm->time.tm_year = ioread32(priv->iobase + HW_ALYEAR); > > alrm->time.tm_mon = ioread32(priv->iobase + HW_ALMON); > > alrm->time.tm_mday = ioread32(priv->iobase + HW_ALDOM); > > @@ -213,7 +205,6 @@ static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > > > alrm->enabled = ioread32(priv->iobase + HW_AMR) ? 1 : 0; > > alrm->pending = ioread32(priv->iobase + HW_CIIR) ? 1 : 0; > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return rtc_valid_tm(&alrm->time); > > } > > @@ -221,9 +212,7 @@ static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > static int asm9260_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > iowrite32(alrm->time.tm_year, priv->iobase + HW_ALYEAR); > > iowrite32(alrm->time.tm_mon, priv->iobase + HW_ALMON); > > iowrite32(alrm->time.tm_mday, priv->iobase + HW_ALDOM); > > @@ -234,7 +223,6 @@ static int asm9260_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > iowrite32(alrm->time.tm_sec, priv->iobase + HW_ALSEC); > > > > iowrite32(alrm->enabled ? 0 : BM_AMR_OFF, priv->iobase + HW_AMR); > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return 0; > > } > > > > > -- > Regards, > Oleksij > -- Alexandre Belloni, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com
WARNING: multiple messages have this Message-ID (diff)
From: Alexandre Belloni <alexandre.belloni@free-electrons.com> To: Oleksij Rempel <linux@rempel-privat.de> Cc: Alessandro Zummo <a.zummo@towertech.it>, rtc-linux@googlegroups.com, linux-kernel@vger.kernel.org Subject: [rtc-linux] Re: [PATCH 2/2] rtc: asm9260: rework locking Date: Wed, 31 Aug 2016 18:09:40 +0200 [thread overview] Message-ID: <20160831160940.eb3ip4t3jfrga3lv@piout.net> (raw) In-Reply-To: <5e6bfec7-d5b6-6311-9cab-76d9aa932408@rempel-privat.de> On 31/08/2016 at 17:14:27 +0200, Oleksij Rempel wrote : > Hi > > Am 31.08.2016 um 02:18 schrieb Alexandre Belloni: > > The rtc-asm9260 driver uses a discrete spinlock (wrongly uninitialized). > > Use the rtc mutex to lock mmio accesses instead. > > > > Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com> > > --- > > drivers/rtc/rtc-asm9260.c | 16 ++-------------- > > 1 file changed, 2 insertions(+), 14 deletions(-) > > > > diff --git a/drivers/rtc/rtc-asm9260.c b/drivers/rtc/rtc-asm9260.c > > index 5219916ce11d..812f131da694 100644 > > --- a/drivers/rtc/rtc-asm9260.c > > +++ b/drivers/rtc/rtc-asm9260.c > > @@ -112,8 +112,6 @@ struct asm9260_rtc_priv { > > void __iomem *iobase; > > struct rtc_device *rtc; > > struct clk *clk; > > - /* io lock */ > > - spinlock_t lock; > > }; > > > > static irqreturn_t asm9260_rtc_irq(int irq, void *dev_id) > > @@ -122,11 +120,13 @@ static irqreturn_t asm9260_rtc_irq(int irq, void *dev_id) > > u32 isr; > > unsigned long events = 0; > > > > + mutex_lock(&priv->rtc->ops_lock); > > isr = ioread32(priv->iobase + HW_CIIR); > > if (!isr) > > return IRQ_NONE; > > > > iowrite32(0, priv->iobase + HW_CIIR); > > + mutex_unlock(&priv->rtc->ops_lock); > > This means if we have shared or spurious interrupt we will get a > deadlock here? > True, I'll fix that. I actually got it right at first and then I toyed with sparse/smatch to see whether it was properly checking the locking. Seems like my setup was not correct. Thanks for the review ! > > > events |= RTC_AF | RTC_IRQF; > > > > @@ -139,9 +139,7 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > u32 ctime0, ctime1, ctime2; > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > ctime0 = ioread32(priv->iobase + HW_CTIME0); > > ctime1 = ioread32(priv->iobase + HW_CTIME1); > > ctime2 = ioread32(priv->iobase + HW_CTIME2); > > @@ -155,7 +153,6 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > ctime1 = ioread32(priv->iobase + HW_CTIME1); > > ctime2 = ioread32(priv->iobase + HW_CTIME2); > > } > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > tm->tm_sec = (ctime0 >> BM_CTIME0_SEC_S) & BM_CTIME0_SEC_M; > > tm->tm_min = (ctime0 >> BM_CTIME0_MIN_S) & BM_CTIME0_MIN_M; > > @@ -174,9 +171,7 @@ static int asm9260_rtc_read_time(struct device *dev, struct rtc_time *tm) > > static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > /* > > * make sure SEC counter will not flip other counter on write time, > > * real value will be written at the enf of sequence. > > @@ -191,7 +186,6 @@ static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > iowrite32(tm->tm_hour, priv->iobase + HW_HOUR); > > iowrite32(tm->tm_min, priv->iobase + HW_MIN); > > iowrite32(tm->tm_sec, priv->iobase + HW_SEC); > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return 0; > > } > > @@ -199,9 +193,7 @@ static int asm9260_rtc_set_time(struct device *dev, struct rtc_time *tm) > > static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > alrm->time.tm_year = ioread32(priv->iobase + HW_ALYEAR); > > alrm->time.tm_mon = ioread32(priv->iobase + HW_ALMON); > > alrm->time.tm_mday = ioread32(priv->iobase + HW_ALDOM); > > @@ -213,7 +205,6 @@ static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > > > alrm->enabled = ioread32(priv->iobase + HW_AMR) ? 1 : 0; > > alrm->pending = ioread32(priv->iobase + HW_CIIR) ? 1 : 0; > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return rtc_valid_tm(&alrm->time); > > } > > @@ -221,9 +212,7 @@ static int asm9260_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > static int asm9260_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > { > > struct asm9260_rtc_priv *priv = dev_get_drvdata(dev); > > - unsigned long irq_flags; > > > > - spin_lock_irqsave(&priv->lock, irq_flags); > > iowrite32(alrm->time.tm_year, priv->iobase + HW_ALYEAR); > > iowrite32(alrm->time.tm_mon, priv->iobase + HW_ALMON); > > iowrite32(alrm->time.tm_mday, priv->iobase + HW_ALDOM); > > @@ -234,7 +223,6 @@ static int asm9260_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alrm) > > iowrite32(alrm->time.tm_sec, priv->iobase + HW_ALSEC); > > > > iowrite32(alrm->enabled ? 0 : BM_AMR_OFF, priv->iobase + HW_AMR); > > - spin_unlock_irqrestore(&priv->lock, irq_flags); > > > > return 0; > > } > > > > > -- > Regards, > Oleksij > -- Alexandre Belloni, Free Electrons Embedded Linux and Kernel engineering http://free-electrons.com -- You received this message because you are subscribed to "rtc-linux". Membership options at http://groups.google.com/group/rtc-linux . Please read http://groups.google.com/group/rtc-linux/web/checklist before submitting a driver. --- You received this message because you are subscribed to the Google Groups "rtc-linux" group. To unsubscribe from this group and stop receiving emails from it, send an email to rtc-linux+unsubscribe@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
next prev parent reply other threads:[~2016-08-31 16:09 UTC|newest] Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-08-31 0:18 [PATCH 1/2] rtc: asm9260: allow COMPILE_TEST Alexandre Belloni 2016-08-31 0:18 ` [rtc-linux] " Alexandre Belloni 2016-08-31 0:18 ` [PATCH 2/2] rtc: asm9260: rework locking Alexandre Belloni 2016-08-31 0:18 ` [rtc-linux] " Alexandre Belloni 2016-08-31 15:14 ` Oleksij Rempel 2016-08-31 15:14 ` [rtc-linux] " Oleksij Rempel 2016-08-31 16:09 ` Alexandre Belloni [this message] 2016-08-31 16:09 ` Alexandre Belloni
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=20160831160940.eb3ip4t3jfrga3lv@piout.net \ --to=alexandre.belloni@free-electrons.com \ --cc=a.zummo@towertech.it \ --cc=linux-kernel@vger.kernel.org \ --cc=linux@rempel-privat.de \ --cc=rtc-linux@googlegroups.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: linkBe 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.