[RFC-PATCH] pstore: defer inserting OOPS entries into pstore

[Luck, Tony]

Life is simple for all the kernel terminating types of kmsg_dump
call backs - pstore just saves the tail end of the console log. But
for "oops" the situation is more complex - the kernel may carry on
running (possibly for ever).  So we'd like to make the logged copy
of the oops appear in the pstore filesystem - so that the user has
a handle to clear the entry from the persistent backing store (if
we don't, the store may fill with "oops" entries (that are also
safely stashed in /var/log/messages) leaving no space for real
errors.

Current code calls pstore_mkfile() immediately. But this may
not be safe. The oops could have happened with arbitrary locks
held, or in interrupt or NMI context. So allocating memory and
calling into generic filesystem code seems unwise.

This patch attempts to defer making the entry appear. At the time
of the oops, we merely set a flag "pstore_new_entry" noting that
a new entry has been added. A periodic timer checks once a minute
to see if the flag is set - if so, it schedules a work queue to
rescan the backing store and make all new entries appear in the
pstore filesystem.

Signed-off-by: Tony Luck <tony.luck@xxxxxxxxx>

Q: Is the "flag -> timer -> workqueue" indirection overkill? I was
unsure whether it was safe to call schedule_work() while
processing the OOPS.

---

diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c
index 893b961..379a02d 100644
--- a/fs/pstore/inode.c
+++ b/fs/pstore/inode.c
@@ -24,6 +24,7 @@
 #include <linux/highmem.h>
 #include <linux/time.h>
 #include <linux/init.h>
+#include <linux/list.h>
 #include <linux/string.h>
 #include <linux/mount.h>
 #include <linux/ramfs.h>
@@ -32,13 +33,18 @@
 #include <linux/magic.h>
 #include <linux/pstore.h>
 #include <linux/slab.h>
+#include <linux/spinlock.h>
 #include <linux/uaccess.h>
 
 #include "internal.h"
 
 #define	PSTORE_NAMELEN	64
 
+static DEFINE_SPINLOCK(allpstore_lock);
+static LIST_HEAD(allpstore);
+
 struct pstore_private {
+	struct list_head list;
 	struct pstore_info *psi;
 	enum pstore_type_id type;
 	u64	id;
@@ -81,8 +87,16 @@ static int pstore_unlink(struct inode *dir, struct dentry *dentry)
 
 static void pstore_evict_inode(struct inode *inode)
 {
+	struct pstore_private	*p = inode->i_private;
+	unsigned long		flags;
+
 	end_writeback(inode);
-	kfree(inode->i_private);
+	if (p) {
+		spin_lock_irqsave(&allpstore_lock, flags);
+		list_del(&p->list);
+		spin_unlock_irqrestore(&allpstore_lock, flags);
+		kfree(p);
+	}
 }
 
 static const struct inode_operations pstore_dir_inode_operations = {
@@ -182,9 +196,23 @@ int pstore_mkfile(enum pstore_type_id type, char *psname, u64 id,
 	struct dentry		*root = pstore_sb->s_root;
 	struct dentry		*dentry;
 	struct inode		*inode;
-	int			rc;
+	int			rc = 0;
 	char			name[PSTORE_NAMELEN];
-	struct pstore_private	*private;
+	struct pstore_private	*private, *pos;
+	unsigned long		flags;
+
+	spin_lock_irqsave(&allpstore_lock, flags);
+	list_for_each_entry(pos, &allpstore, list) {
+		if (pos->type == type &&
+		    pos->id == id &&
+		    pos->psi == psi) {
+			rc = -EEXIST;
+			break;
+		}
+	}
+	spin_unlock_irqrestore(&allpstore_lock, flags);
+	if (rc)
+		return rc;
 
 	rc = -ENOMEM;
 	inode = pstore_get_inode(pstore_sb, root->d_inode, S_IFREG | 0444, 0);
@@ -229,6 +257,10 @@ int pstore_mkfile(enum pstore_type_id type, char *psname, u64 id,
 
 	d_add(dentry, inode);
 
+	spin_lock_irqsave(&allpstore_lock, flags);
+	list_add(&private->list, &allpstore);
+	spin_unlock_irqrestore(&allpstore_lock, flags);
+
 	mutex_unlock(&root->d_inode->i_mutex);
 
 	return 0;
@@ -277,7 +309,7 @@ int pstore_fill_super(struct super_block *sb, void *data, int silent)
 		goto fail;
 	}
 
-	pstore_get_records();
+	pstore_get_records(0);
 
 	return 0;
 fail:
diff --git a/fs/pstore/internal.h b/fs/pstore/internal.h
index 611c1b3..3bde461 100644
--- a/fs/pstore/internal.h
+++ b/fs/pstore/internal.h
@@ -1,5 +1,5 @@
 extern void	pstore_set_kmsg_bytes(int);
-extern void	pstore_get_records(void);
+extern void	pstore_get_records(int);
 extern int	pstore_mkfile(enum pstore_type_id, char *psname, u64 id,
 			      char *data, size_t size,
 			      struct timespec time, struct pstore_info *psi);
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index c5300ec..ca60ebc 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -25,12 +25,29 @@
 #include <linux/module.h>
 #include <linux/pstore.h>
 #include <linux/string.h>
+#include <linux/timer.h>
 #include <linux/slab.h>
 #include <linux/uaccess.h>
+#include <linux/workqueue.h>
 
 #include "internal.h"
 
 /*
+ * We defer making "oops" entries appear in pstore - see
+ * whether the system is actually still running well enough
+ * to let someone see the entry
+ */
+#define	PSTORE_INTERVAL	(60 * HZ)
+
+static int pstore_new_entry;
+
+static void pstore_timefunc(unsigned long);
+static DEFINE_TIMER(pstore_timer, pstore_timefunc, 0, 0);
+
+static void pstore_dowork(struct work_struct *);
+static DECLARE_WORK(pstore_work, pstore_dowork);
+
+/*
  * pstore_lock just protects "psinfo" during
  * calls to pstore_register()
  */
@@ -100,9 +117,7 @@ static void pstore_dump(struct kmsg_dumper *dumper,
 		id = psinfo->write(PSTORE_TYPE_DMESG, part,
 				   hsize + l1_cpy + l2_cpy, psinfo);
 		if (reason == KMSG_DUMP_OOPS && pstore_is_mounted())
-			pstore_mkfile(PSTORE_TYPE_DMESG, psinfo->name, id,
-				      psinfo->buf, hsize + l1_cpy + l2_cpy,
-				      CURRENT_TIME, psinfo);
+			pstore_new_entry = 1;
 		l1 -= l1_cpy;
 		l2 -= l2_cpy;
 		total += l1_cpy + l2_cpy;
@@ -148,19 +163,24 @@ int pstore_register(struct pstore_info *psi)
 	}
 
 	if (pstore_is_mounted())
-		pstore_get_records();
+		pstore_get_records(0);
 
 	kmsg_dump_register(&pstore_dumper);
 
+	pstore_timer.expires = jiffies + PSTORE_INTERVAL;
+	add_timer(&pstore_timer);
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(pstore_register);
 
 /*
- * Read all the records from the persistent store. Create and
- * file files in our filesystem.
+ * Read all the records from the persistent store. Create
+ * files in our filesystem.  Don't warn about -EEXIST errors
+ * when we are re-scanning the backing store looking to add new
+ * error records.
  */
-void pstore_get_records(void)
+void pstore_get_records(int quiet)
 {
 	struct pstore_info *psi = psinfo;
 	ssize_t			size;
@@ -178,8 +198,9 @@ void pstore_get_records(void)
 		goto out;
 
 	while ((size = psi->read(&id, &type, &time, psi)) > 0) {
-		if (pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size,
-				  time, psi))
+		rc = pstore_mkfile(type, psi->name, id, psi->buf, (size_t)size,
+				  time, psi);
+		if (rc && (rc != -EEXIST || !quiet))
 			failed++;
 	}
 	psi->close(psi);
@@ -191,6 +212,21 @@ out:
 		       failed, psi->name);
 }
 
+static void pstore_dowork(struct work_struct *work)
+{
+	pstore_get_records(1);
+}
+
+static void pstore_timefunc(unsigned long dummy)
+{
+	if (pstore_new_entry) {
+		pstore_new_entry = 0;
+		schedule_work(&pstore_work);
+	}
+
+	mod_timer(&pstore_timer, jiffies + PSTORE_INTERVAL);
+}
+
 /*
  * Call platform driver to write a record to the
  * persistent store.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/