[PATCH v2] relay: Use per CPU constructs for the relay channel buffer pointers

[akash . goel]

From: Akash Goel <akash.goel@xxxxxxxxx>

relay essentially needs to maintain the per CPU array of channel buffer
pointers but it manually creates that array.
Instead its better to avail the per CPU constructs, provided by the
kernel, to allocate & access the array of pointer to channel buffers.

v2: Include <linux/percpu.h> in relay.h so that it pulls in the percpu
    api explicitly. (Chris)

Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Cc: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
Signed-off-by: Akash Goel <akash.goel@xxxxxxxxx>
Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
---
 include/linux/relay.h | 17 +++++++-----
 kernel/relay.c        | 74 +++++++++++++++++++++++++++++----------------------
 2 files changed, 52 insertions(+), 39 deletions(-)

diff --git a/include/linux/relay.h b/include/linux/relay.h
index d7c8359..eb295e3 100644
--- a/include/linux/relay.h
+++ b/include/linux/relay.h
@@ -19,6 +19,7 @@
 #include <linux/fs.h>
 #include <linux/poll.h>
 #include <linux/kref.h>
+#include <linux/percpu.h>
 
 /*
  * Tracks changes to rchan/rchan_buf structs
@@ -63,7 +64,7 @@ struct rchan
 	struct kref kref;		/* channel refcount */
 	void *private_data;		/* for user-defined data */
 	size_t last_toobig;		/* tried to log event > subbuf size */
-	struct rchan_buf *buf[NR_CPUS]; /* per-cpu channel buffers */
+	struct rchan_buf ** __percpu buf; /* per-cpu channel buffers */
 	int is_global;			/* One global buffer ? */
 	struct list_head list;		/* for channel list */
 	struct dentry *parent;		/* parent dentry passed to open */
@@ -204,7 +205,7 @@ static inline void relay_write(struct rchan *chan,
 	struct rchan_buf *buf;
 
 	local_irq_save(flags);
-	buf = chan->buf[smp_processor_id()];
+	buf = *this_cpu_ptr(chan->buf);
 	if (unlikely(buf->offset + length > chan->subbuf_size))
 		length = relay_switch_subbuf(buf, length);
 	memcpy(buf->data + buf->offset, data, length);
@@ -230,12 +231,12 @@ static inline void __relay_write(struct rchan *chan,
 {
 	struct rchan_buf *buf;
 
-	buf = chan->buf[get_cpu()];
+	buf = *get_cpu_ptr(chan->buf);
 	if (unlikely(buf->offset + length > buf->chan->subbuf_size))
 		length = relay_switch_subbuf(buf, length);
 	memcpy(buf->data + buf->offset, data, length);
 	buf->offset += length;
-	put_cpu();
+	put_cpu_ptr(chan->buf);
 }
 
 /**
@@ -251,17 +252,19 @@ static inline void __relay_write(struct rchan *chan,
  */
 static inline void *relay_reserve(struct rchan *chan, size_t length)
 {
-	void *reserved;
-	struct rchan_buf *buf = chan->buf[smp_processor_id()];
+	void *reserved = NULL;
+	struct rchan_buf *buf = *get_cpu_ptr(chan->buf);
 
 	if (unlikely(buf->offset + length > buf->chan->subbuf_size)) {
 		length = relay_switch_subbuf(buf, length);
 		if (!length)
-			return NULL;
+			goto end;
 	}
 	reserved = buf->data + buf->offset;
 	buf->offset += length;
 
+end:
+	put_cpu_ptr(chan->buf);
 	return reserved;
 }
 
diff --git a/kernel/relay.c b/kernel/relay.c
index d797502..f55ab82 100644
--- a/kernel/relay.c
+++ b/kernel/relay.c
@@ -214,7 +214,7 @@ static void relay_destroy_buf(struct rchan_buf *buf)
 			__free_page(buf->page_array[i]);
 		relay_free_page_array(buf->page_array);
 	}
-	chan->buf[buf->cpu] = NULL;
+	*per_cpu_ptr(chan->buf, buf->cpu) = NULL;
 	kfree(buf->padding);
 	kfree(buf);
 	kref_put(&chan->kref, relay_destroy_channel);
@@ -382,20 +382,21 @@ static void __relay_reset(struct rchan_buf *buf, unsigned int init)
  */
 void relay_reset(struct rchan *chan)
 {
+	struct rchan_buf *buf;
 	unsigned int i;
 
 	if (!chan)
 		return;
 
-	if (chan->is_global && chan->buf[0]) {
-		__relay_reset(chan->buf[0], 0);
+	if (chan->is_global && (buf = *per_cpu_ptr(chan->buf, 0))) {
+		__relay_reset(buf, 0);
 		return;
 	}
 
 	mutex_lock(&relay_channels_mutex);
 	for_each_possible_cpu(i)
-		if (chan->buf[i])
-			__relay_reset(chan->buf[i], 0);
+		if ((buf = *per_cpu_ptr(chan->buf, i)))
+			__relay_reset(buf, 0);
 	mutex_unlock(&relay_channels_mutex);
 }
 EXPORT_SYMBOL_GPL(relay_reset);
@@ -440,7 +441,7 @@ static struct rchan_buf *relay_open_buf(struct rchan *chan, unsigned int cpu)
 	struct dentry *dentry;
 
  	if (chan->is_global)
-		return chan->buf[0];
+		return *per_cpu_ptr(chan->buf, 0);
 
 	buf = relay_create_buf(chan);
 	if (!buf)
@@ -464,7 +465,7 @@ static struct rchan_buf *relay_open_buf(struct rchan *chan, unsigned int cpu)
  	__relay_reset(buf, 1);
 
  	if(chan->is_global) {
- 		chan->buf[0] = buf;
+		*per_cpu_ptr(chan->buf, 0) = buf;
  		buf->cpu = 0;
   	}
 
@@ -526,22 +527,24 @@ static int relay_hotcpu_callback(struct notifier_block *nb,
 {
 	unsigned int hotcpu = (unsigned long)hcpu;
 	struct rchan *chan;
+	struct rchan_buf *buf;
 
 	switch(action) {
 	case CPU_UP_PREPARE:
 	case CPU_UP_PREPARE_FROZEN:
 		mutex_lock(&relay_channels_mutex);
 		list_for_each_entry(chan, &relay_channels, list) {
-			if (chan->buf[hotcpu])
+			if ((buf = *per_cpu_ptr(chan->buf, hotcpu)))
 				continue;
-			chan->buf[hotcpu] = relay_open_buf(chan, hotcpu);
-			if(!chan->buf[hotcpu]) {
+			buf = relay_open_buf(chan, hotcpu);
+			if(!buf) {
 				printk(KERN_ERR
 					"relay_hotcpu_callback: cpu %d buffer "
 					"creation failed\n", hotcpu);
 				mutex_unlock(&relay_channels_mutex);
 				return notifier_from_errno(-ENOMEM);
 			}
+			*per_cpu_ptr(chan->buf, hotcpu) = buf;
 		}
 		mutex_unlock(&relay_channels_mutex);
 		break;
@@ -583,6 +586,7 @@ struct rchan *relay_open(const char *base_filename,
 {
 	unsigned int i;
 	struct rchan *chan;
+	struct rchan_buf *buf;
 
 	if (!(subbuf_size && n_subbufs))
 		return NULL;
@@ -593,6 +597,7 @@ struct rchan *relay_open(const char *base_filename,
 	if (!chan)
 		return NULL;
 
+	chan->buf = alloc_percpu(struct rchan_buf *);
 	chan->version = RELAYFS_CHANNEL_VERSION;
 	chan->n_subbufs = n_subbufs;
 	chan->subbuf_size = subbuf_size;
@@ -608,9 +613,10 @@ struct rchan *relay_open(const char *base_filename,
 
 	mutex_lock(&relay_channels_mutex);
 	for_each_online_cpu(i) {
-		chan->buf[i] = relay_open_buf(chan, i);
-		if (!chan->buf[i])
+		buf = relay_open_buf(chan, i);
+		if (!buf)
 			goto free_bufs;
+		*per_cpu_ptr(chan->buf, i) = buf;
 	}
 	list_add(&chan->list, &relay_channels);
 	mutex_unlock(&relay_channels_mutex);
@@ -619,8 +625,8 @@ struct rchan *relay_open(const char *base_filename,
 
 free_bufs:
 	for_each_possible_cpu(i) {
-		if (chan->buf[i])
-			relay_close_buf(chan->buf[i]);
+		if ((buf = *per_cpu_ptr(chan->buf, i)))
+			relay_close_buf(buf);
 	}
 
 	kref_put(&chan->kref, relay_destroy_channel);
@@ -666,6 +672,7 @@ int relay_late_setup_files(struct rchan *chan,
 	unsigned int i, curr_cpu;
 	unsigned long flags;
 	struct dentry *dentry;
+	struct rchan_buf *buf;
 	struct rchan_percpu_buf_dispatcher disp;
 
 	if (!chan || !base_filename)
@@ -684,10 +691,11 @@ int relay_late_setup_files(struct rchan *chan,
 
 	if (chan->is_global) {
 		err = -EINVAL;
-		if (!WARN_ON_ONCE(!chan->buf[0])) {
-			dentry = relay_create_buf_file(chan, chan->buf[0], 0);
+		buf = *per_cpu_ptr(chan->buf, 0);
+		if (!WARN_ON_ONCE(!buf)) {
+			dentry = relay_create_buf_file(chan, buf, 0);
 			if (dentry && !WARN_ON_ONCE(!chan->is_global)) {
-				relay_set_buf_dentry(chan->buf[0], dentry);
+				relay_set_buf_dentry(buf, dentry);
 				err = 0;
 			}
 		}
@@ -702,13 +710,14 @@ int relay_late_setup_files(struct rchan *chan,
 	 * on all currently online CPUs.
 	 */
 	for_each_online_cpu(i) {
-		if (unlikely(!chan->buf[i])) {
+		buf = *per_cpu_ptr(chan->buf, i);
+		if (unlikely(!buf)) {
 			WARN_ONCE(1, KERN_ERR "CPU has no buffer!\n");
 			err = -EINVAL;
 			break;
 		}
 
-		dentry = relay_create_buf_file(chan, chan->buf[i], i);
+		dentry = relay_create_buf_file(chan, buf, i);
 		if (unlikely(!dentry)) {
 			err = -EINVAL;
 			break;
@@ -716,10 +725,10 @@ int relay_late_setup_files(struct rchan *chan,
 
 		if (curr_cpu == i) {
 			local_irq_save(flags);
-			relay_set_buf_dentry(chan->buf[i], dentry);
+			relay_set_buf_dentry(buf, dentry);
 			local_irq_restore(flags);
 		} else {
-			disp.buf = chan->buf[i];
+			disp.buf = buf;
 			disp.dentry = dentry;
 			smp_mb();
 			/* relay_channels_mutex must be held, so wait. */
@@ -822,11 +831,10 @@ void relay_subbufs_consumed(struct rchan *chan,
 	if (!chan)
 		return;
 
-	if (cpu >= NR_CPUS || !chan->buf[cpu] ||
-					subbufs_consumed > chan->n_subbufs)
+	buf = *per_cpu_ptr(chan->buf, cpu);
+	if (cpu >= NR_CPUS || !buf || subbufs_consumed > chan->n_subbufs)
 		return;
 
-	buf = chan->buf[cpu];
 	if (subbufs_consumed > buf->subbufs_produced - buf->subbufs_consumed)
 		buf->subbufs_consumed = buf->subbufs_produced;
 	else
@@ -842,18 +850,19 @@ EXPORT_SYMBOL_GPL(relay_subbufs_consumed);
  */
 void relay_close(struct rchan *chan)
 {
+	struct rchan_buf *buf;
 	unsigned int i;
 
 	if (!chan)
 		return;
 
 	mutex_lock(&relay_channels_mutex);
-	if (chan->is_global && chan->buf[0])
-		relay_close_buf(chan->buf[0]);
+	if (chan->is_global && (buf = *per_cpu_ptr(chan->buf, 0)))
+		relay_close_buf(buf);
 	else
 		for_each_possible_cpu(i)
-			if (chan->buf[i])
-				relay_close_buf(chan->buf[i]);
+			if ((buf = *per_cpu_ptr(chan->buf, i)))
+				relay_close_buf(buf);
 
 	if (chan->last_toobig)
 		printk(KERN_WARNING "relay: one or more items not logged "
@@ -874,20 +883,21 @@ EXPORT_SYMBOL_GPL(relay_close);
  */
 void relay_flush(struct rchan *chan)
 {
+	struct rchan_buf *buf;
 	unsigned int i;
 
 	if (!chan)
 		return;
 
-	if (chan->is_global && chan->buf[0]) {
-		relay_switch_subbuf(chan->buf[0], 0);
+	if (chan->is_global && (buf = *per_cpu_ptr(chan->buf, 0))) {
+		relay_switch_subbuf(buf, 0);
 		return;
 	}
 
 	mutex_lock(&relay_channels_mutex);
 	for_each_possible_cpu(i)
-		if (chan->buf[i])
-			relay_switch_subbuf(chan->buf[i], 0);
+		if ((buf = *per_cpu_ptr(chan->buf, i)))
+			relay_switch_subbuf(buf, 0);
 	mutex_unlock(&relay_channels_mutex);
 }
 EXPORT_SYMBOL_GPL(relay_flush);
-- 
1.9.2