All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tony Lindgren <tony@atomide.com>
To: Tarun Kanti DebBarma <tarun.kanti@ti.com>
Cc: linux-omap@vger.kernel.org, khilman@ti.com,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v15 11/12] OMAP: dmtimer: extend spinlock to exported APIs
Date: Tue, 13 Sep 2011 16:15:40 -0700	[thread overview]
Message-ID: <20110913231540.GI24252@atomide.com> (raw)
In-Reply-To: <1315516098-29761-12-git-send-email-tarun.kanti@ti.com>

* Tarun Kanti DebBarma <tarun.kanti@ti.com> [110908 13:36]:
> Since the exported APIs can be called from interrupt context
> extend spinlock protection to some more relevant APIs to avoid
> race condition.

We should have locking for requesting and releasing a timer etc,
but I don't see need for that for the timer specific functions.

> @@ -317,9 +317,11 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_trigger);
>  void omap_dm_timer_start(struct omap_dm_timer *timer)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (timer->loses_context) {
>  		u32 ctx_loss_cnt_after =
>  			timer->get_context_loss_count(&timer->pdev->dev);

Here the caller already owns the timer being started. So there
should never be multiple users for a single timer. If there are,
then the caller should take care of locking.

>  void omap_dm_timer_stop(struct omap_dm_timer *timer)
>  {
> -	unsigned long rate = 0;
> +	unsigned long rate = 0, flags;
>  	struct dmtimer_platform_data *pdata = timer->pdev->dev.platform_data;
>  	bool is_omap2 = true;
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (pdata->needs_manual_reset)
>  		is_omap2 = false;
>  	else

Here too.

> @@ -389,8 +394,10 @@ void omap_dm_timer_set_load(struct omap_dm_timer *timer, int autoreload,
>  			    unsigned int load)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
>  	if (autoreload)
>  		l |= OMAP_TIMER_CTRL_AR;

And here.

> @@ -412,9 +420,11 @@ void omap_dm_timer_set_load_start(struct omap_dm_timer *timer, int autoreload,
>                              unsigned int load)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (timer->loses_context) {
>  		u32 ctx_loss_cnt_after =
>  			timer->get_context_loss_count(&timer->pdev->dev);

Not needed here either. And that's the case for all the other functions
too.

Regards,

Tony

WARNING: multiple messages have this Message-ID (diff)
From: tony@atomide.com (Tony Lindgren)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v15 11/12] OMAP: dmtimer: extend spinlock to exported APIs
Date: Tue, 13 Sep 2011 16:15:40 -0700	[thread overview]
Message-ID: <20110913231540.GI24252@atomide.com> (raw)
In-Reply-To: <1315516098-29761-12-git-send-email-tarun.kanti@ti.com>

* Tarun Kanti DebBarma <tarun.kanti@ti.com> [110908 13:36]:
> Since the exported APIs can be called from interrupt context
> extend spinlock protection to some more relevant APIs to avoid
> race condition.

We should have locking for requesting and releasing a timer etc,
but I don't see need for that for the timer specific functions.

> @@ -317,9 +317,11 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_trigger);
>  void omap_dm_timer_start(struct omap_dm_timer *timer)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (timer->loses_context) {
>  		u32 ctx_loss_cnt_after =
>  			timer->get_context_loss_count(&timer->pdev->dev);

Here the caller already owns the timer being started. So there
should never be multiple users for a single timer. If there are,
then the caller should take care of locking.

>  void omap_dm_timer_stop(struct omap_dm_timer *timer)
>  {
> -	unsigned long rate = 0;
> +	unsigned long rate = 0, flags;
>  	struct dmtimer_platform_data *pdata = timer->pdev->dev.platform_data;
>  	bool is_omap2 = true;
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (pdata->needs_manual_reset)
>  		is_omap2 = false;
>  	else

Here too.

> @@ -389,8 +394,10 @@ void omap_dm_timer_set_load(struct omap_dm_timer *timer, int autoreload,
>  			    unsigned int load)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
>  	if (autoreload)
>  		l |= OMAP_TIMER_CTRL_AR;

And here.

> @@ -412,9 +420,11 @@ void omap_dm_timer_set_load_start(struct omap_dm_timer *timer, int autoreload,
>                              unsigned int load)
>  {
>  	u32 l;
> +	unsigned long flags;
>  
>  	omap_dm_timer_enable(timer);
>  
> +	spin_lock_irqsave(&dm_timer_lock, flags);
>  	if (timer->loses_context) {
>  		u32 ctx_loss_cnt_after =
>  			timer->get_context_loss_count(&timer->pdev->dev);

Not needed here either. And that's the case for all the other functions
too.

Regards,

Tony

  reply	other threads:[~2011-09-13 23:15 UTC|newest]

Thread overview: 62+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-09-08 21:08 [PATCH v15 00/12] OMAP: dmtimer: adaptation to platform_driver Tarun Kanti DebBarma
2011-09-08 21:08 ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 01/12] OMAP2+: dmtimer: add device names to flck nodes Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 02/12] OMAP4: hwmod data: add dmtimer version information Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 03/12] OMAP1: dmtimer: conversion to platform devices Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 04/12] OMAP2+: dmtimer: convert " Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 05/12] OMAP: dmtimer: platform driver Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 06/12] OMAP: dmtimer: switch-over to platform device driver Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-14 21:45   ` Tony Lindgren
2011-09-14 21:45     ` Tony Lindgren
2011-09-14 22:24     ` Tony Lindgren
2011-09-14 22:24       ` Tony Lindgren
2011-09-15  5:42     ` DebBarma, Tarun Kanti
2011-09-15  5:42       ` DebBarma, Tarun Kanti
2011-09-15  8:46       ` Mohammed, Afzal
2011-09-15  8:46         ` Mohammed, Afzal
2011-09-15 16:58         ` Tony Lindgren
2011-09-15 16:58           ` Tony Lindgren
2011-09-15 17:05           ` Tony Lindgren
2011-09-15 17:05             ` Tony Lindgren
2011-09-16  3:54             ` DebBarma, Tarun Kanti
2011-09-16  3:54               ` DebBarma, Tarun Kanti
2011-09-16  9:30               ` DebBarma, Tarun Kanti
2011-09-16  9:30                 ` DebBarma, Tarun Kanti
2011-09-17  1:35                 ` [PATCH] ARM: OMAP: Add support for dmtimer v2 ip (Re: [PATCH v15 06/12] OMAP: dmtimer: switch-over to platform device driver) Tony Lindgren
2011-09-17  1:35                   ` Tony Lindgren
2011-09-18 12:27                   ` DebBarma, Tarun Kanti
2011-09-18 12:27                     ` DebBarma, Tarun Kanti
2011-09-19 17:43                     ` Tony Lindgren
2011-09-19 17:43                       ` Tony Lindgren
2011-09-19  4:06                   ` Pedanekar, Hemant
2011-09-19  4:06                     ` Pedanekar, Hemant
2011-09-19 17:40                     ` Tony Lindgren
2011-09-19 17:40                       ` Tony Lindgren
2011-09-19 23:41                       ` Pedanekar, Hemant
2011-09-19 23:41                         ` Pedanekar, Hemant
2011-09-19  5:22                   ` Mohammed, Afzal
2011-09-19  5:22                     ` Mohammed, Afzal
2011-09-19 17:43                     ` Tony Lindgren
2011-09-19 17:43                       ` Tony Lindgren
2011-09-08 21:08 ` [PATCH v15 07/12] OMAP: dmtimer: pm_runtime support Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 08/12] OMAP: dmtimer: add timeout to low-level routines Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 09/12] OMAP: dmtimer: mark clocksource and clockevent timers reserved Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 10/12] OMAP: dmtimer: low-power mode support Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-08 21:08 ` [PATCH v15 11/12] OMAP: dmtimer: extend spinlock to exported APIs Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma
2011-09-13 23:15   ` Tony Lindgren [this message]
2011-09-13 23:15     ` Tony Lindgren
2011-09-14  4:58     ` DebBarma, Tarun Kanti
2011-09-14  4:58       ` DebBarma, Tarun Kanti
2011-09-08 21:08 ` [PATCH v15 12/12] OMAP: dmtimer: add error handling to export APIs Tarun Kanti DebBarma
2011-09-08 21:08   ` Tarun Kanti DebBarma

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=20110913231540.GI24252@atomide.com \
    --to=tony@atomide.com \
    --cc=khilman@ti.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=tarun.kanti@ti.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.