[KJ] [PATCH] Add poll_event* interfaces (was Re: [patch] fix common

* [KJ] [PATCH] Add poll_event* interfaces (was Re: [patch] fix common
@ 2006-08-12  6:28 Nishanth Aravamudan
  2006-08-13 22:42 ` [KJ] [PATCH] Add poll_event* interfaces (was Re: [patch] fix Pavel Machek
                   ` (7 more replies)
  0 siblings, 8 replies; 9+ messages in thread
From: Nishanth Aravamudan @ 2006-08-12  6:28 UTC (permalink / raw)
  To: kernel-janitors

On 12.08.2006 [14:26:14 +1000], Darren Jenkins wrote:
> G'day
> 
> On 8/10/06, Alexey Dobriyan <adobriyan@gmail.com> wrote:
> > On Thu, Aug 10, 2006 at 08:48:19AM +1000, Darren Jenkins wrote:
> > > On 8/8/06, Pavel Machek <pavel@ucw.cz> wrote:
> > > > > Sorry I did not realise that was your problem with the code.
> > > > > Would it help if we just explicitly added a
> > > > >
> > > > > if (ready())
> > > > >        return 0;
> > > > >
> > > > > after the loop, in the example code? so people wont miss adding
> > > > > something like that in?
> > > >
> > > > Yes, that would do the trick.
> > > >                                                                 Pavel
> > >
> > >
> > > I was going to make a patch for the TODO, but it seems this subject
> > > has since been removed.
> > > Any chance it is going to be reinstated ? or is this subject
> > > considered not so much janitorial work?
> >
> > I need to read through those threads so correct version will be present
> > in TODO. So far, testing for OK condition was forgotten. Anything else?
> 
> No the problem that Pavel Machek showed us was the only one I saw.
> 
> >         unsigned long timeout = jiffies + HZ/2;
> >        for (;;) {
> >                if (ready())
> >                        return 0;
> > [IMAGINE HALF A SECOND DELAY HERE]
> >                if (time_after(timeout, jiffies))
> >                        break;
> >                msleep(10);
> >        }
> >
> > Oops
> 
> 
> 
> There were three proposed solutions though.

<snip>

FWIW, here is my poll_event() patch from some time ago updated to
2.6.18-rc4. I think it's closest in spirit to 3). Comments, thoughts?

Description: Add the poll_event(), poll_event_timeout(),
poll_event_interruptible() and poll_event_interruptible_timeout()
interfaces. These macros encapsulate typical sleeping code usage in a
sane way.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>

diff -urpN 2.6.18-rc4/include/linux/delay.h 2.6.18-rc4-dev/include/linux/delay.h

--- 2.6.18-rc4/include/linux/delay.h	2006-08-11 22:40:57.000000000 -0700
+++ 2.6.18-rc4-dev/include/linux/delay.h	2006-08-11 23:26:23.000000000 -0700
@@ -44,4 +44,99 @@ static inline void ssleep(unsigned int s
 	msleep(seconds * 1000);
 }
 
+/*
+ * poll_event* check if a @condition is true every @interval milliseconds,
+ * up to a @timeout maximum, if specified, else forever.
+ * The *_interruptible versions will sleep in TASK_INTERRUPTIBLE
+ * (instead of TASK_UNINTERRUPTIBLE) and will return early on signals
+ * These interfaces are intentionally very similar to wait_event*
+ */
+
+/**
+ * poll_event - check a condition at regular intervals
+ * @condition: Condition to check
+ * @interval: Number of milliseconds between checks
+ */
+#define poll_event(condition, interval)				\
+do {								\
+	while (!(condition))					\
+		msleep(interval);				\
+} while(0)
+
+/**
+ * poll_event_timeout - check a condition at regular intervals with a timeout
+ * @condition: Condition to check
+ * @interval: Number of milliseconds between checks
+ * @timeout: Time in jiffies to stop checking at
+ * returns: 0, if condition caused return
+ * 	    -ETIME, if timeout
+ */
+#define poll_event_timeout(condition, interval, timeout)		\
+({									\
+	int __ret = 0;							\
+ 									\
+	for (;;) {							\
+		if (condition)						\
+			break;						\
+ 		msleep(interval);					\
+		if (time_after(jiffies, timeout)) {			\
+			__ret = -ETIME;					\
+ 			break;						\
+		}							\
+	}								\
+	__ret;								\
+})
+
+/**
+ * poll_event_interruptible - check a condition at regular intervals, returning early on signals
+ * @condition: Condition to check
+ * @interval: Number of milliseconds between checks
+ * returns: 0, if condition caused return
+ * 	    -EINTR, if a signal
+ */
+#define poll_event_interruptible(condition, interval)		\
+({								\
+	int __ret = 0;						\
+								\
+	for (;;) {						\
+		if (condition)					\
+			break;					\
+ 		msleep_interruptible(interval);			\
+		if (signal_pending(current)) {			\
+			__ret = -EINTR;				\
+ 			break;					\
+		}						\
+	}							\
+	__ret;
+})
+
+/**
+ * poll_event_interruptible_timeout - check a condition at regular intervals with a timeout, returning early on signals
+ * @condition: Condition to check
+ * @interval: Number of milliseconds between checks
+ * @timeout: Time in jiffies to stop checking at
+ * returns 0, if condition caused return
+ * 	   -EINTR, if a signal
+ * 	   -ETIME, if timeout
+ */
+#define poll_event_interruptible_timeout(conditition, interval, timeout)	\
+({										\
+	int __ret = 0;								\
+										\
+	for (;;) {								\
+		if (condition)							\
+			break;							\
+ 		msleep_interruptible(interval);					\
+		if (signal_pending(current)) {					\
+			__ret = -EINTR;						\
+			break;							\
+		}								\
+		if (time_after(jiffies, timeout)) {				\
+			__ret = -ETIME;						\
+			break;							\
+		}								\
+	}									\
+	__ret;									\
+})
+
 #endif /* defined(_LINUX_DELAY_H) */


-- 
Nishanth Aravamudan <nacc@us.ibm.com>
IBM Linux Technology Center
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
https://lists.osdl.org/mailman/listinfo/kernel-janitors

^ permalink raw reply	[flat|nested] 9+ messages in thread