[PATCH 0/2] Add cli and ioctl to ignore a scanned device

[PATCH 0/2] Add cli and ioctl to ignore a scanned device

[Anand Jain]

Patch 1/2 is preparatory patch to get a function to delete a device
  from the list.
Patch 2/2 adds the ioctl and feature to delete a device from the
device list.

Anand Jain (2):
  btrfs: refactor btrfs_free_stale_device() to get device list delete
  btrfs: introduce feature to ignore a btrfs device

 fs/btrfs/super.c           |  4 ++
 fs/btrfs/volumes.c         | 97 +++++++++++++++++++++++++++++++++++++++++-----
 fs/btrfs/volumes.h         |  2 +
 include/uapi/linux/btrfs.h |  2 +
 4 files changed, 95 insertions(+), 10 deletions(-)

-- 
2.13.1

[PATCH 1/2] btrfs: refactor btrfs_free_stale_device() to get device list delete

[Anand Jain]

We need to delete a device from the dev_list, so refactor
btrfs_free_stale_device() for delete_device_from_list().

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 fs/btrfs/volumes.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 8bac1cf17048..8ead85dba6f5 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -546,6 +546,22 @@ static void pending_bios_fn(struct btrfs_work *work)
 	run_scheduled_bios(device);
 }
 
+void delete_device_from_list(struct btrfs_device *dev)
+{
+	struct btrfs_fs_devices *fs_devs;
+	fs_devs = dev->fs_devices;
+
+	if (fs_devs->num_devices == 1) {
+		btrfs_sysfs_remove_fsid(fs_devs);
+		list_del(&fs_devs->list);
+		free_fs_devices(fs_devs);
+	} else {
+		fs_devs->num_devices--;
+		list_del(&dev->dev_list);
+		rcu_string_free(dev->name);
+		kfree(dev);
+	}
+}
 
 void btrfs_free_stale_device(struct btrfs_device *cur_dev)
 {
@@ -586,16 +602,7 @@ void btrfs_free_stale_device(struct btrfs_device *cur_dev)
 
 		if (!del) {
 			/* delete the stale device */
-			if (fs_devs->num_devices == 1) {
-				btrfs_sysfs_remove_fsid(fs_devs);
-				list_del(&fs_devs->list);
-				free_fs_devices(fs_devs);
-			} else {
-				fs_devs->num_devices--;
-				list_del(&dev->dev_list);
-				rcu_string_free(dev->name);
-				kfree(dev);
-			}
+			delete_device_from_list(dev);
 			break;
 		}
 	}
-- 
2.13.1

Re: [PATCH 1/2] btrfs: refactor btrfs_free_stale_device() to get device list delete

[Nikolay Borisov]

On 13.11.2017 07:44, Anand Jain wrote:
> We need to delete a device from the dev_list, so refactor
> btrfs_free_stale_device() for delete_device_from_list().
> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  fs/btrfs/volumes.c | 27 +++++++++++++++++----------
>  1 file changed, 17 insertions(+), 10 deletions(-)
> 
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index 8bac1cf17048..8ead85dba6f5 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -546,6 +546,22 @@ static void pending_bios_fn(struct btrfs_work *work)
>  	run_scheduled_bios(device);
>  }
>  
> +void delete_device_from_list(struct btrfs_device *dev)

In this patch and the other one this function is used only in volume.c
so make static. Also if it relies on some particular lock being held why
not add lockdep_assert_held(xxx) as a means of documenting that ?

> +{
> +	struct btrfs_fs_devices *fs_devs;
> +	fs_devs = dev->fs_devices;
> +
> +	if (fs_devs->num_devices == 1) {
> +		btrfs_sysfs_remove_fsid(fs_devs);
> +		list_del(&fs_devs->list);
> +		free_fs_devices(fs_devs);
> +	} else {
> +		fs_devs->num_devices--;
> +		list_del(&dev->dev_list);
> +		rcu_string_free(dev->name);
> +		kfree(dev);
> +	}
> +}
>  
>  void btrfs_free_stale_device(struct btrfs_device *cur_dev)
>  {
> @@ -586,16 +602,7 @@ void btrfs_free_stale_device(struct btrfs_device *cur_dev)
>  
>  		if (!del) {
>  			/* delete the stale device */
> -			if (fs_devs->num_devices == 1) {
> -				btrfs_sysfs_remove_fsid(fs_devs);
> -				list_del(&fs_devs->list);
> -				free_fs_devices(fs_devs);
> -			} else {
> -				fs_devs->num_devices--;
> -				list_del(&dev->dev_list);
> -				rcu_string_free(dev->name);
> -				kfree(dev);
> -			}
> +			delete_device_from_list(dev);
>  			break;
>  		}
>  	}
> 

Re: [PATCH 1/2] btrfs: refactor btrfs_free_stale_device() to get device list delete

[Anand Jain]

On 11/16/2017 09:59 PM, Nikolay Borisov wrote:
> 
> 
> On 13.11.2017 07:44, Anand Jain wrote:
>> We need to delete a device from the dev_list, so refactor
>> btrfs_free_stale_device() for delete_device_from_list().
>>
>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>> ---
>>   fs/btrfs/volumes.c | 27 +++++++++++++++++----------
>>   1 file changed, 17 insertions(+), 10 deletions(-)
>>
>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>> index 8bac1cf17048..8ead85dba6f5 100644
>> --- a/fs/btrfs/volumes.c
>> +++ b/fs/btrfs/volumes.c
>> @@ -546,6 +546,22 @@ static void pending_bios_fn(struct btrfs_work *work)
>>   	run_scheduled_bios(device);
>>   }
>>   
>> +void delete_device_from_list(struct btrfs_device *dev)
> 
> In this patch and the other one this function is used only in volume.c
> so make static.

  Right. Will do.

> Also if it relies on some particular lock being held why
> not add lockdep_assert_held(xxx) as a means of documenting that ?

  Actually, btrfs_free_stale_device() should be holding device_list_mutex
  but it didn't so I would like to fix it separately. So here I shall
  create a new function instead of peal from btrfs_free_stale_device(),
  which I think btrfs_free_stale_device() can use it later.

Thanks, Anand

>> +{
>> +	struct btrfs_fs_devices *fs_devs;
>> +	fs_devs = dev->fs_devices;
>> +
>> +	if (fs_devs->num_devices == 1) {
>> +		btrfs_sysfs_remove_fsid(fs_devs);
>> +		list_del(&fs_devs->list);
>> +		free_fs_devices(fs_devs);
>> +	} else {
>> +		fs_devs->num_devices--;
>> +		list_del(&dev->dev_list);
>> +		rcu_string_free(dev->name);
>> +		kfree(dev);
>> +	}
>> +}
>>   
>>   void btrfs_free_stale_device(struct btrfs_device *cur_dev)
>>   {
>> @@ -586,16 +602,7 @@ void btrfs_free_stale_device(struct btrfs_device *cur_dev)
>>   
>>   		if (!del) {
>>   			/* delete the stale device */
>> -			if (fs_devs->num_devices == 1) {
>> -				btrfs_sysfs_remove_fsid(fs_devs);
>> -				list_del(&fs_devs->list);
>> -				free_fs_devices(fs_devs);
>> -			} else {
>> -				fs_devs->num_devices--;
>> -				list_del(&dev->dev_list);
>> -				rcu_string_free(dev->name);
>> -				kfree(dev);
>> -			}
>> +			delete_device_from_list(dev);
>>   			break;
>>   		}
>>   	}
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

[PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

Support for a new command is being added here:
 btrfs dev ignore <dev>
Which shall undo the effects of the command
 btrfs dev scan <dev>

This cli/ioctl is needed as there is no way to continue to mount in
degraded mode if the device is already scanned, which is required to
recover from the split brain raid conditions.

This patch proposes to use ioctl #5 as it was empty.
	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
If #5 is reserved for some other purpose, I think I should change this.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 fs/btrfs/super.c           |  4 +++
 fs/btrfs/volumes.c         | 70 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/volumes.h         |  2 ++
 include/uapi/linux/btrfs.h |  2 ++
 4 files changed, 78 insertions(+)

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index fa7bad8b22da..64f4beb1526d 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
 		break;
+	case BTRFS_IOC_IGNORE_DEV:
+		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
+					    &btrfs_fs_type, &fs_devices);
+		break;
 	case BTRFS_IOC_DEVICES_READY:
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 8ead85dba6f5..1a06a17e111e 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
 	return 0;
 }
 
+static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
+{
+	int ret = 0;
+	struct btrfs_fs_devices *fs_devices;
+	struct btrfs_device *device;
+
+	fs_devices = find_fsid(disk_super->fsid);
+	if (!fs_devices)
+		return -ENOENT;
+
+	if (fs_devices->opened)
+		return -EPERM;
+
+	mutex_lock(&uuid_mutex);
+	mutex_lock(&fs_devices->device_list_mutex);
+	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
+	if (!device) {
+		ret = -ENOENT;
+		goto error;
+	}
+
+	delete_device_from_list(device);
+
+error:
+	mutex_unlock(&fs_devices->device_list_mutex);
+	mutex_unlock(&uuid_mutex);
+
+	return ret;
+}
+
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret)
+{
+	struct btrfs_super_block *disk_super;
+	struct block_device *bdev;
+	struct page *page;
+	int ret = -EINVAL;
+	u64 devid;
+	u64 bytenr;
+
+	bytenr = btrfs_sb_offset(0);
+	flags |= FMODE_EXCL;
+
+	bdev = blkdev_get_by_path(path, flags, holder);
+	if (IS_ERR(bdev)) {
+		ret = PTR_ERR(bdev);
+		goto error;
+	}
+
+	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
+		goto error_bdev_put;
+
+	devid = btrfs_stack_device_id(&disk_super->dev_item);
+
+	ret = device_list_remove(disk_super, devid);
+	if (ret)
+	pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to unscan: %d",
+		disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
+	else
+	pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
+		disk_super->fsid, path, disk_super->dev_item.uuid, devid);
+
+	btrfs_release_disk_super(page);
+
+error_bdev_put:
+	blkdev_put(bdev, flags);
+error:
+	return ret;
+}
+
 /*
  * Look for a btrfs signature on a device. This may be called out of the mount path
  * and we are not allowed to call set_blocksize during the scan. The superblock
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index 6108fdfec67f..340e40acdab5 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
 int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 			  struct btrfs_fs_devices **fs_devices_ret);
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret);
 int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
 void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
index 378230c163d5..928485c31aa4 100644
--- a/include/uapi/linux/btrfs.h
+++ b/include/uapi/linux/btrfs.h
@@ -739,6 +739,8 @@ enum btrfs_err_code {
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
 				   struct btrfs_ioctl_vol_args)
+#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
+				   struct btrfs_ioctl_vol_args)
 /* trans start and trans end are dangerous, and only for
  * use by applications that know how to avoid the
  * resulting deadlocks
-- 
2.13.1

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Nikolay Borisov]

On 13.11.2017 07:44, Anand Jain wrote:
> Support for a new command is being added here:
>  btrfs dev ignore <dev>
> Which shall undo the effects of the command
>  btrfs dev scan <dev>
> 
> This cli/ioctl is needed as there is no way to continue to mount in
> degraded mode if the device is already scanned, which is required to
> recover from the split brain raid conditions.
> 
> This patch proposes to use ioctl #5 as it was empty.
> 	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
> If #5 is reserved for some other purpose, I think I should change this.
> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  fs/btrfs/super.c           |  4 +++
>  fs/btrfs/volumes.c         | 70 ++++++++++++++++++++++++++++++++++++++++++++++
>  fs/btrfs/volumes.h         |  2 ++
>  include/uapi/linux/btrfs.h |  2 ++
>  4 files changed, 78 insertions(+)
> 
> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
> index fa7bad8b22da..64f4beb1526d 100644
> --- a/fs/btrfs/super.c
> +++ b/fs/btrfs/super.c
> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
>  		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>  					    &btrfs_fs_type, &fs_devices);
>  		break;
> +	case BTRFS_IOC_IGNORE_DEV:
> +		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
> +					    &btrfs_fs_type, &fs_devices);
> +		break;
>  	case BTRFS_IOC_DEVICES_READY:
>  		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>  					    &btrfs_fs_type, &fs_devices);
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index 8ead85dba6f5..1a06a17e111e 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
>  	return 0;
>  }
>  
> +static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
> +{
> +	int ret = 0;
> +	struct btrfs_fs_devices *fs_devices;
> +	struct btrfs_device *device;
> +
> +	fs_devices = find_fsid(disk_super->fsid);
> +	if (!fs_devices)
> +		return -ENOENT;
> +
> +	if (fs_devices->opened)
> +		return -EPERM;

Isn't -EBUSY more appropriate?

> +
> +	mutex_lock(&uuid_mutex);
> +	mutex_lock(&fs_devices->device_list_mutex);
> +	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
> +	if (!device) {
> +		ret = -ENOENT;
> +		goto error;
> +	}
> +
> +	delete_device_from_list(device);
> +
> +error:
> +	mutex_unlock(&fs_devices->device_list_mutex);
> +	mutex_unlock(&uuid_mutex);
> +
> +	return ret;
> +}
> +
> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
> +			  struct btrfs_fs_devices **fs_devices_ret)
> +{
> +	struct btrfs_super_block *disk_super;
> +	struct block_device *bdev;
> +	struct page *page;
> +	int ret = -EINVAL;
> +	u64 devid;
> +	u64 bytenr;
> +
> +	bytenr = btrfs_sb_offset(0);
> +	flags |= FMODE_EXCL;
> +
> +	bdev = blkdev_get_by_path(path, flags, holder);
> +	if (IS_ERR(bdev)) {
> +		ret = PTR_ERR(bdev);
> +		goto error;
> +	}
> +
> +	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
> +		goto error_bdev_put;
> +
> +	devid = btrfs_stack_device_id(&disk_super->dev_item);
> +
> +	ret = device_list_remove(disk_super, devid);
> +	if (ret)
> +	pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to unscan: %d",
> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
> +	else
> +	pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid);
> +
> +	btrfs_release_disk_super(page);
> +
> +error_bdev_put:
> +	blkdev_put(bdev, flags);
> +error:
> +	return ret;
> +}
> +
>  /*
>   * Look for a btrfs signature on a device. This may be called out of the mount path
>   * and we are not allowed to call set_blocksize during the scan. The superblock
> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
> index 6108fdfec67f..340e40acdab5 100644
> --- a/fs/btrfs/volumes.h
> +++ b/fs/btrfs/volumes.h
> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
>  		       fmode_t flags, void *holder);
>  int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
>  			  struct btrfs_fs_devices **fs_devices_ret);
> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
> +			  struct btrfs_fs_devices **fs_devices_ret);
>  int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>  void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
>  void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
> index 378230c163d5..928485c31aa4 100644
> --- a/include/uapi/linux/btrfs.h
> +++ b/include/uapi/linux/btrfs.h
> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>  				   struct btrfs_ioctl_vol_args)
>  #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>  				   struct btrfs_ioctl_vol_args)
> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
> +				   struct btrfs_ioctl_vol_args)
>  /* trans start and trans end are dangerous, and only for
>   * use by applications that know how to avoid the
>   * resulting deadlocks
> 

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

On 11/16/2017 10:03 PM, Nikolay Borisov wrote:
> 
> 
> On 13.11.2017 07:44, Anand Jain wrote:
>> Support for a new command is being added here:
>>   btrfs dev ignore <dev>
>> Which shall undo the effects of the command
>>   btrfs dev scan <dev>
>>
>> This cli/ioctl is needed as there is no way to continue to mount in
>> degraded mode if the device is already scanned, which is required to
>> recover from the split brain raid conditions.
>>
>> This patch proposes to use ioctl #5 as it was empty.
>> 	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
>> If #5 is reserved for some other purpose, I think I should change this.
>>
>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>> ---
>>   fs/btrfs/super.c           |  4 +++
>>   fs/btrfs/volumes.c         | 70 ++++++++++++++++++++++++++++++++++++++++++++++
>>   fs/btrfs/volumes.h         |  2 ++
>>   include/uapi/linux/btrfs.h |  2 ++
>>   4 files changed, 78 insertions(+)
>>
>> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
>> index fa7bad8b22da..64f4beb1526d 100644
>> --- a/fs/btrfs/super.c
>> +++ b/fs/btrfs/super.c
>> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
>>   		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>   					    &btrfs_fs_type, &fs_devices);
>>   		break;
>> +	case BTRFS_IOC_IGNORE_DEV:
>> +		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
>> +					    &btrfs_fs_type, &fs_devices);
>> +		break;
>>   	case BTRFS_IOC_DEVICES_READY:
>>   		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>   					    &btrfs_fs_type, &fs_devices);
>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>> index 8ead85dba6f5..1a06a17e111e 100644
>> --- a/fs/btrfs/volumes.c
>> +++ b/fs/btrfs/volumes.c
>> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
>>   	return 0;
>>   }
>>   
>> +static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
>> +{
>> +	int ret = 0;
>> +	struct btrfs_fs_devices *fs_devices;
>> +	struct btrfs_device *device;
>> +
>> +	fs_devices = find_fsid(disk_super->fsid);
>> +	if (!fs_devices)
>> +		return -ENOENT;
>> +
>> +	if (fs_devices->opened)
>> +		return -EPERM;
> 
> Isn't -EBUSY more appropriate?

  Right added.

Thanks, Anand

> 
>> +
>> +	mutex_lock(&uuid_mutex);
>> +	mutex_lock(&fs_devices->device_list_mutex);
>> +	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
>> +	if (!device) {
>> +		ret = -ENOENT;
>> +		goto error;
>> +	}
>> +
>> +	delete_device_from_list(device);
>> +
>> +error:
>> +	mutex_unlock(&fs_devices->device_list_mutex);
>> +	mutex_unlock(&uuid_mutex);
>> +
>> +	return ret;
>> +}
>> +
>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
>> +			  struct btrfs_fs_devices **fs_devices_ret)
>> +{
>> +	struct btrfs_super_block *disk_super;
>> +	struct block_device *bdev;
>> +	struct page *page;
>> +	int ret = -EINVAL;
>> +	u64 devid;
>> +	u64 bytenr;
>> +
>> +	bytenr = btrfs_sb_offset(0);
>> +	flags |= FMODE_EXCL;
>> +
>> +	bdev = blkdev_get_by_path(path, flags, holder);
>> +	if (IS_ERR(bdev)) {
>> +		ret = PTR_ERR(bdev);
>> +		goto error;
>> +	}
>> +
>> +	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
>> +		goto error_bdev_put;
>> +
>> +	devid = btrfs_stack_device_id(&disk_super->dev_item);
>> +
>> +	ret = device_list_remove(disk_super, devid);
>> +	if (ret)
>> +	pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to unscan: %d",
>> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
>> +	else
>> +	pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
>> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid);
>> +
>> +	btrfs_release_disk_super(page);
>> +
>> +error_bdev_put:
>> +	blkdev_put(bdev, flags);
>> +error:
>> +	return ret;
>> +}
>> +
>>   /*
>>    * Look for a btrfs signature on a device. This may be called out of the mount path
>>    * and we are not allowed to call set_blocksize during the scan. The superblock
>> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
>> index 6108fdfec67f..340e40acdab5 100644
>> --- a/fs/btrfs/volumes.h
>> +++ b/fs/btrfs/volumes.h
>> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
>>   		       fmode_t flags, void *holder);
>>   int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
>>   			  struct btrfs_fs_devices **fs_devices_ret);
>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
>> +			  struct btrfs_fs_devices **fs_devices_ret);
>>   int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>>   void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
>>   void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
>> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
>> index 378230c163d5..928485c31aa4 100644
>> --- a/include/uapi/linux/btrfs.h
>> +++ b/include/uapi/linux/btrfs.h
>> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>>   				   struct btrfs_ioctl_vol_args)
>>   #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>>   				   struct btrfs_ioctl_vol_args)
>> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
>> +				   struct btrfs_ioctl_vol_args)
>>   /* trans start and trans end are dangerous, and only for
>>    * use by applications that know how to avoid the
>>    * resulting deadlocks
>>

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Nikolay Borisov]

On 13.11.2017 07:44, Anand Jain wrote:
> Support for a new command is being added here:
>  btrfs dev ignore <dev>
> Which shall undo the effects of the command
>  btrfs dev scan <dev>
> 
> This cli/ioctl is needed as there is no way to continue to mount in
> degraded mode if the device is already scanned, which is required to
> recover from the split brain raid conditions.
> 
> This patch proposes to use ioctl #5 as it was empty.
> 	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
> If #5 is reserved for some other purpose, I think I should change this.
> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
>  fs/btrfs/super.c           |  4 +++
>  fs/btrfs/volumes.c         | 70 ++++++++++++++++++++++++++++++++++++++++++++++
>  fs/btrfs/volumes.h         |  2 ++
>  include/uapi/linux/btrfs.h |  2 ++
>  4 files changed, 78 insertions(+)
> 
> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
> index fa7bad8b22da..64f4beb1526d 100644
> --- a/fs/btrfs/super.c
> +++ b/fs/btrfs/super.c
> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
>  		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>  					    &btrfs_fs_type, &fs_devices);
>  		break;
> +	case BTRFS_IOC_IGNORE_DEV:
> +		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
> +					    &btrfs_fs_type, &fs_devices);
> +		break;
>  	case BTRFS_IOC_DEVICES_READY:
>  		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>  					    &btrfs_fs_type, &fs_devices);
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index 8ead85dba6f5..1a06a17e111e 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
>  	return 0;
>  }
>  
> +static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
> +{
> +	int ret = 0;
> +	struct btrfs_fs_devices *fs_devices;
> +	struct btrfs_device *device;
> +
> +	fs_devices = find_fsid(disk_super->fsid);
> +	if (!fs_devices)
> +		return -ENOENT;
> +
> +	if (fs_devices->opened)
> +		return -EPERM;
> +
> +	mutex_lock(&uuid_mutex);
> +	mutex_lock(&fs_devices->device_list_mutex);
> +	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
> +	if (!device) {
> +		ret = -ENOENT;
> +		goto error;
> +	}
> +
> +	delete_device_from_list(device);
> +
> +error:
> +	mutex_unlock(&fs_devices->device_list_mutex);
> +	mutex_unlock(&uuid_mutex);
> +
> +	return ret;
> +}
> +
> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
> +			  struct btrfs_fs_devices **fs_devices_ret)
> +{
> +	struct btrfs_super_block *disk_super;
> +	struct block_device *bdev;
> +	struct page *page;
> +	int ret = -EINVAL;
> +	u64 devid;
> +	u64 bytenr;
> +
> +	bytenr = btrfs_sb_offset(0);
> +	flags |= FMODE_EXCL;
> +
> +	bdev = blkdev_get_by_path(path, flags, holder);
> +	if (IS_ERR(bdev)) {
> +		ret = PTR_ERR(bdev);
> +		goto error;
> +	}
> +
> +	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
> +		goto error_bdev_put;
> +
> +	devid = btrfs_stack_device_id(&disk_super->dev_item);
> +
> +	ret = device_list_remove(disk_super, devid);
> +	if (ret)
> +	pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to unscan: %d",
> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
> +	else
> +	pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid);

There is already btrfs_err and btrfs_info, use those. This will allow to
uniquely identify the failing btrfs file system also the pr_* statements
need to be indented after the if/else.

> +
> +	btrfs_release_disk_super(page);
> +
> +error_bdev_put:
> +	blkdev_put(bdev, flags);
> +error:
> +	return ret;
> +}
> +
>  /*
>   * Look for a btrfs signature on a device. This may be called out of the mount path
>   * and we are not allowed to call set_blocksize during the scan. The superblock
> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
> index 6108fdfec67f..340e40acdab5 100644
> --- a/fs/btrfs/volumes.h
> +++ b/fs/btrfs/volumes.h
> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
>  		       fmode_t flags, void *holder);
>  int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
>  			  struct btrfs_fs_devices **fs_devices_ret);
> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
> +			  struct btrfs_fs_devices **fs_devices_ret);
>  int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>  void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
>  void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
> index 378230c163d5..928485c31aa4 100644
> --- a/include/uapi/linux/btrfs.h
> +++ b/include/uapi/linux/btrfs.h
> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>  				   struct btrfs_ioctl_vol_args)
>  #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>  				   struct btrfs_ioctl_vol_args)
> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
> +				   struct btrfs_ioctl_vol_args)
>  /* trans start and trans end are dangerous, and only for
>   * use by applications that know how to avoid the
>   * resulting deadlocks
> 

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

On 11/16/2017 10:11 PM, Nikolay Borisov wrote:
> 
> 
> On 13.11.2017 07:44, Anand Jain wrote:
>> Support for a new command is being added here:
>>   btrfs dev ignore <dev>
>> Which shall undo the effects of the command
>>   btrfs dev scan <dev>
>>
>> This cli/ioctl is needed as there is no way to continue to mount in
>> degraded mode if the device is already scanned, which is required to
>> recover from the split brain raid conditions.
>>
>> This patch proposes to use ioctl #5 as it was empty.
>> 	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
>> If #5 is reserved for some other purpose, I think I should change this.
>>
>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>> ---
>>   fs/btrfs/super.c           |  4 +++
>>   fs/btrfs/volumes.c         | 70 ++++++++++++++++++++++++++++++++++++++++++++++
>>   fs/btrfs/volumes.h         |  2 ++
>>   include/uapi/linux/btrfs.h |  2 ++
>>   4 files changed, 78 insertions(+)
>>
>> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
>> index fa7bad8b22da..64f4beb1526d 100644
>> --- a/fs/btrfs/super.c
>> +++ b/fs/btrfs/super.c
>> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
>>   		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>   					    &btrfs_fs_type, &fs_devices);
>>   		break;
>> +	case BTRFS_IOC_IGNORE_DEV:
>> +		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
>> +					    &btrfs_fs_type, &fs_devices);
>> +		break;
>>   	case BTRFS_IOC_DEVICES_READY:
>>   		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>   					    &btrfs_fs_type, &fs_devices);
>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>> index 8ead85dba6f5..1a06a17e111e 100644
>> --- a/fs/btrfs/volumes.c
>> +++ b/fs/btrfs/volumes.c
>> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
>>   	return 0;
>>   }
>>   
>> +static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
>> +{
>> +	int ret = 0;
>> +	struct btrfs_fs_devices *fs_devices;
>> +	struct btrfs_device *device;
>> +
>> +	fs_devices = find_fsid(disk_super->fsid);
>> +	if (!fs_devices)
>> +		return -ENOENT;
>> +
>> +	if (fs_devices->opened)
>> +		return -EPERM;
>> +
>> +	mutex_lock(&uuid_mutex);
>> +	mutex_lock(&fs_devices->device_list_mutex);
>> +	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
>> +	if (!device) {
>> +		ret = -ENOENT;
>> +		goto error;
>> +	}
>> +
>> +	delete_device_from_list(device);
>> +
>> +error:
>> +	mutex_unlock(&fs_devices->device_list_mutex);
>> +	mutex_unlock(&uuid_mutex);
>> +
>> +	return ret;
>> +}
>> +
>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
>> +			  struct btrfs_fs_devices **fs_devices_ret)
>> +{
>> +	struct btrfs_super_block *disk_super;
>> +	struct block_device *bdev;
>> +	struct page *page;
>> +	int ret = -EINVAL;
>> +	u64 devid;
>> +	u64 bytenr;
>> +
>> +	bytenr = btrfs_sb_offset(0);
>> +	flags |= FMODE_EXCL;
>> +
>> +	bdev = blkdev_get_by_path(path, flags, holder);
>> +	if (IS_ERR(bdev)) {
>> +		ret = PTR_ERR(bdev);
>> +		goto error;
>> +	}
>> +
>> +	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
>> +		goto error_bdev_put;
>> +
>> +	devid = btrfs_stack_device_id(&disk_super->dev_item);
>> +
>> +	ret = device_list_remove(disk_super, devid);
>> +	if (ret)
>> +	pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to unscan: %d",
>> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
>> +	else
>> +	pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
>> +		disk_super->fsid, path, disk_super->dev_item.uuid, devid);
> 
> There is already btrfs_err and btrfs_info, use those. This will allow to
> uniquely identify the failing btrfs file system

  We don't have fs_info in this context, btrfs_err/info can't be used.

> also the pr_* statements
> need to be indented after the if/else.

  Will do.

Thanks, Anand

>> +
>> +	btrfs_release_disk_super(page);
>> +
>> +error_bdev_put:
>> +	blkdev_put(bdev, flags);
>> +error:
>> +	return ret;
>> +}
>> +
>>   /*
>>    * Look for a btrfs signature on a device. This may be called out of the mount path
>>    * and we are not allowed to call set_blocksize during the scan. The superblock
>> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
>> index 6108fdfec67f..340e40acdab5 100644
>> --- a/fs/btrfs/volumes.h
>> +++ b/fs/btrfs/volumes.h
>> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
>>   		       fmode_t flags, void *holder);
>>   int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
>>   			  struct btrfs_fs_devices **fs_devices_ret);
>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
>> +			  struct btrfs_fs_devices **fs_devices_ret);
>>   int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>>   void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
>>   void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
>> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
>> index 378230c163d5..928485c31aa4 100644
>> --- a/include/uapi/linux/btrfs.h
>> +++ b/include/uapi/linux/btrfs.h
>> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>>   				   struct btrfs_ioctl_vol_args)
>>   #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>>   				   struct btrfs_ioctl_vol_args)
>> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
>> +				   struct btrfs_ioctl_vol_args)
>>   /* trans start and trans end are dangerous, and only for
>>    * use by applications that know how to avoid the
>>    * resulting deadlocks
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Nikolay Borisov]

On 18.11.2017 15:50, Anand Jain wrote:
> 
> 
> On 11/16/2017 10:11 PM, Nikolay Borisov wrote:
>>
>>
>> On 13.11.2017 07:44, Anand Jain wrote:
>>> Support for a new command is being added here:
>>>   btrfs dev ignore <dev>
>>> Which shall undo the effects of the command
>>>   btrfs dev scan <dev>
>>>
>>> This cli/ioctl is needed as there is no way to continue to mount in
>>> degraded mode if the device is already scanned, which is required to
>>> recover from the split brain raid conditions.
>>>
>>> This patch proposes to use ioctl #5 as it was empty.
>>>     IOW(BTRFS_IOCTL_MAGIC, 5, ..)
>>> If #5 is reserved for some other purpose, I think I should change this.
>>>
>>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>>> ---
>>>   fs/btrfs/super.c           |  4 +++
>>>   fs/btrfs/volumes.c         | 70
>>> ++++++++++++++++++++++++++++++++++++++++++++++
>>>   fs/btrfs/volumes.h         |  2 ++
>>>   include/uapi/linux/btrfs.h |  2 ++
>>>   4 files changed, 78 insertions(+)
>>>
>>> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
>>> index fa7bad8b22da..64f4beb1526d 100644
>>> --- a/fs/btrfs/super.c
>>> +++ b/fs/btrfs/super.c
>>> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file
>>> *file, unsigned int cmd,
>>>           ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>>                           &btrfs_fs_type, &fs_devices);
>>>           break;
>>> +    case BTRFS_IOC_IGNORE_DEV:
>>> +        ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
>>> +                        &btrfs_fs_type, &fs_devices);
>>> +        break;
>>>       case BTRFS_IOC_DEVICES_READY:
>>>           ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>>                           &btrfs_fs_type, &fs_devices);
>>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>>> index 8ead85dba6f5..1a06a17e111e 100644
>>> --- a/fs/btrfs/volumes.c
>>> +++ b/fs/btrfs/volumes.c
>>> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device
>>> *bdev, u64 bytenr,
>>>       return 0;
>>>   }
>>>   +static int device_list_remove(struct btrfs_super_block
>>> *disk_super, u64 devid)
>>> +{
>>> +    int ret = 0;
>>> +    struct btrfs_fs_devices *fs_devices;
>>> +    struct btrfs_device *device;
>>> +
>>> +    fs_devices = find_fsid(disk_super->fsid);
>>> +    if (!fs_devices)
>>> +        return -ENOENT;
>>> +
>>> +    if (fs_devices->opened)
>>> +        return -EPERM;
>>> +
>>> +    mutex_lock(&uuid_mutex);
>>> +    mutex_lock(&fs_devices->device_list_mutex);
>>> +    device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
>>> +    if (!device) {
>>> +        ret = -ENOENT;
>>> +        goto error;
>>> +    }
>>> +
>>> +    delete_device_from_list(device);
>>> +
>>> +error:
>>> +    mutex_unlock(&fs_devices->device_list_mutex);
>>> +    mutex_unlock(&uuid_mutex);
>>> +
>>> +    return ret;
>>> +}
>>> +
>>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void
>>> *holder,
>>> +              struct btrfs_fs_devices **fs_devices_ret)
>>> +{
>>> +    struct btrfs_super_block *disk_super;
>>> +    struct block_device *bdev;
>>> +    struct page *page;
>>> +    int ret = -EINVAL;
>>> +    u64 devid;
>>> +    u64 bytenr;
>>> +
>>> +    bytenr = btrfs_sb_offset(0);
>>> +    flags |= FMODE_EXCL;
>>> +
>>> +    bdev = blkdev_get_by_path(path, flags, holder);
>>> +    if (IS_ERR(bdev)) {
>>> +        ret = PTR_ERR(bdev);
>>> +        goto error;
>>> +    }
>>> +
>>> +    if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
>>> +        goto error_bdev_put;
>>> +
>>> +    devid = btrfs_stack_device_id(&disk_super->dev_item);
>>> +
>>> +    ret = device_list_remove(disk_super, devid);
>>> +    if (ret)
>>> +    pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to
>>> unscan: %d",
>>> +        disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
>>> +    else
>>> +    pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
>>> +        disk_super->fsid, path, disk_super->dev_item.uuid, devid);
>>
>> There is already btrfs_err and btrfs_info, use those. This will allow to
>> uniquely identify the failing btrfs file system
> 
>  We don't have fs_info in this context, btrfs_err/info can't be used.

Yes, but we have struct file in btrfs_control_ioctl then can't we do :

struct btrfs_fs_info = btrfs_sb(file_inode(file)->i_sb)


> 
>> also the pr_* statements
>> need to be indented after the if/else.
> 
>  Will do.
> 
> Thanks, Anand
> 
>>> +
>>> +    btrfs_release_disk_super(page);
>>> +
>>> +error_bdev_put:
>>> +    blkdev_put(bdev, flags);
>>> +error:
>>> +    return ret;
>>> +}
>>> +
>>>   /*
>>>    * Look for a btrfs signature on a device. This may be called out
>>> of the mount path
>>>    * and we are not allowed to call set_blocksize during the scan.
>>> The superblock
>>> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
>>> index 6108fdfec67f..340e40acdab5 100644
>>> --- a/fs/btrfs/volumes.h
>>> +++ b/fs/btrfs/volumes.h
>>> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices
>>> *fs_devices,
>>>                  fmode_t flags, void *holder);
>>>   int btrfs_scan_one_device(const char *path, fmode_t flags, void
>>> *holder,
>>>                 struct btrfs_fs_devices **fs_devices_ret);
>>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void
>>> *holder,
>>> +              struct btrfs_fs_devices **fs_devices_ret);
>>>   int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>>>   void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices,
>>> int step);
>>>   void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
>>> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
>>> index 378230c163d5..928485c31aa4 100644
>>> --- a/include/uapi/linux/btrfs.h
>>> +++ b/include/uapi/linux/btrfs.h
>>> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>>>                      struct btrfs_ioctl_vol_args)
>>>   #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>>>                      struct btrfs_ioctl_vol_args)
>>> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
>>> +                   struct btrfs_ioctl_vol_args)
>>>   /* trans start and trans end are dangerous, and only for
>>>    * use by applications that know how to avoid the
>>>    * resulting deadlocks
>>>
>> -- 
>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
> 

Re: [PATCH 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

On 11/18/2017 10:28 PM, Nikolay Borisov wrote:
> 
> 
> On 18.11.2017 15:50, Anand Jain wrote:
>>
>>
>> On 11/16/2017 10:11 PM, Nikolay Borisov wrote:
>>>
>>>
>>> On 13.11.2017 07:44, Anand Jain wrote:
>>>> Support for a new command is being added here:
>>>>    btrfs dev ignore <dev>
>>>> Which shall undo the effects of the command
>>>>    btrfs dev scan <dev>
>>>>
>>>> This cli/ioctl is needed as there is no way to continue to mount in
>>>> degraded mode if the device is already scanned, which is required to
>>>> recover from the split brain raid conditions.
>>>>
>>>> This patch proposes to use ioctl #5 as it was empty.
>>>>      IOW(BTRFS_IOCTL_MAGIC, 5, ..)
>>>> If #5 is reserved for some other purpose, I think I should change this.
>>>>
>>>> Signed-off-by: Anand Jain <anand.jain@oracle.com>
>>>> ---
>>>>    fs/btrfs/super.c           |  4 +++
>>>>    fs/btrfs/volumes.c         | 70
>>>> ++++++++++++++++++++++++++++++++++++++++++++++
>>>>    fs/btrfs/volumes.h         |  2 ++
>>>>    include/uapi/linux/btrfs.h |  2 ++
>>>>    4 files changed, 78 insertions(+)
>>>>
>>>> diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
>>>> index fa7bad8b22da..64f4beb1526d 100644
>>>> --- a/fs/btrfs/super.c
>>>> +++ b/fs/btrfs/super.c
>>>> @@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file
>>>> *file, unsigned int cmd,
>>>>            ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>>>                            &btrfs_fs_type, &fs_devices);
>>>>            break;
>>>> +    case BTRFS_IOC_IGNORE_DEV:
>>>> +        ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
>>>> +                        &btrfs_fs_type, &fs_devices);
>>>> +        break;
>>>>        case BTRFS_IOC_DEVICES_READY:
>>>>            ret = btrfs_scan_one_device(vol->name, FMODE_READ,
>>>>                            &btrfs_fs_type, &fs_devices);
>>>> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
>>>> index 8ead85dba6f5..1a06a17e111e 100644
>>>> --- a/fs/btrfs/volumes.c
>>>> +++ b/fs/btrfs/volumes.c
>>>> @@ -1181,6 +1181,76 @@ int btrfs_read_disk_super(struct block_device
>>>> *bdev, u64 bytenr,
>>>>        return 0;
>>>>    }
>>>>    +static int device_list_remove(struct btrfs_super_block
>>>> *disk_super, u64 devid)
>>>> +{
>>>> +    int ret = 0;
>>>> +    struct btrfs_fs_devices *fs_devices;
>>>> +    struct btrfs_device *device;
>>>> +
>>>> +    fs_devices = find_fsid(disk_super->fsid);
>>>> +    if (!fs_devices)
>>>> +        return -ENOENT;
>>>> +
>>>> +    if (fs_devices->opened)
>>>> +        return -EPERM;
>>>> +
>>>> +    mutex_lock(&uuid_mutex);
>>>> +    mutex_lock(&fs_devices->device_list_mutex);
>>>> +    device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
>>>> +    if (!device) {
>>>> +        ret = -ENOENT;
>>>> +        goto error;
>>>> +    }
>>>> +
>>>> +    delete_device_from_list(device);
>>>> +
>>>> +error:
>>>> +    mutex_unlock(&fs_devices->device_list_mutex);
>>>> +    mutex_unlock(&uuid_mutex);
>>>> +
>>>> +    return ret;
>>>> +}
>>>> +
>>>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void
>>>> *holder,
>>>> +              struct btrfs_fs_devices **fs_devices_ret)
>>>> +{
>>>> +    struct btrfs_super_block *disk_super;
>>>> +    struct block_device *bdev;
>>>> +    struct page *page;
>>>> +    int ret = -EINVAL;
>>>> +    u64 devid;
>>>> +    u64 bytenr;
>>>> +
>>>> +    bytenr = btrfs_sb_offset(0);
>>>> +    flags |= FMODE_EXCL;
>>>> +
>>>> +    bdev = blkdev_get_by_path(path, flags, holder);
>>>> +    if (IS_ERR(bdev)) {
>>>> +        ret = PTR_ERR(bdev);
>>>> +        goto error;
>>>> +    }
>>>> +
>>>> +    if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
>>>> +        goto error_bdev_put;
>>>> +
>>>> +    devid = btrfs_stack_device_id(&disk_super->dev_item);
>>>> +
>>>> +    ret = device_list_remove(disk_super, devid);
>>>> +    if (ret)
>>>> +    pr_err("BTRFS: %pU: device %s uuid %pU devid %llu failed to
>>>> unscan: %d",
>>>> +        disk_super->fsid, path, disk_super->dev_item.uuid, devid, ret);
>>>> +    else
>>>> +    pr_info("BTRFS: %pU: device %s uuid %pU devid %llu ran undo scan",
>>>> +        disk_super->fsid, path, disk_super->dev_item.uuid, devid);
>>>
>>> There is already btrfs_err and btrfs_info, use those. This will allow to
>>> uniquely identify the failing btrfs file system
>>
>>   We don't have fs_info in this context, btrfs_err/info can't be used.
> 
> Yes, but we have struct file in btrfs_control_ioctl then can't we do :
> 
> struct btrfs_fs_info = btrfs_sb(file_inode(file)->i_sb)

Nikolay, We are in the device scan context where btrfs_fs_info does
not exist yet.

Thanks, Anand


> 
>>
>>> also the pr_* statements
>>> need to be indented after the if/else.
>>
>>   Will do.
>>
>> Thanks, Anand
>>
>>>> +
>>>> +    btrfs_release_disk_super(page);
>>>> +
>>>> +error_bdev_put:
>>>> +    blkdev_put(bdev, flags);
>>>> +error:
>>>> +    return ret;
>>>> +}
>>>> +
>>>>    /*
>>>>     * Look for a btrfs signature on a device. This may be called out
>>>> of the mount path
>>>>     * and we are not allowed to call set_blocksize during the scan.
>>>> The superblock
>>>> diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
>>>> index 6108fdfec67f..340e40acdab5 100644
>>>> --- a/fs/btrfs/volumes.h
>>>> +++ b/fs/btrfs/volumes.h
>>>> @@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices
>>>> *fs_devices,
>>>>                   fmode_t flags, void *holder);
>>>>    int btrfs_scan_one_device(const char *path, fmode_t flags, void
>>>> *holder,
>>>>                  struct btrfs_fs_devices **fs_devices_ret);
>>>> +int btrfs_ignore_one_device(const char *path, fmode_t flags, void
>>>> *holder,
>>>> +              struct btrfs_fs_devices **fs_devices_ret);
>>>>    int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
>>>>    void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices,
>>>> int step);
>>>>    void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
>>>> diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
>>>> index 378230c163d5..928485c31aa4 100644
>>>> --- a/include/uapi/linux/btrfs.h
>>>> +++ b/include/uapi/linux/btrfs.h
>>>> @@ -739,6 +739,8 @@ enum btrfs_err_code {
>>>>                       struct btrfs_ioctl_vol_args)
>>>>    #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
>>>>                       struct btrfs_ioctl_vol_args)
>>>> +#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
>>>> +                   struct btrfs_ioctl_vol_args)
>>>>    /* trans start and trans end are dangerous, and only for
>>>>     * use by applications that know how to avoid the
>>>>     * resulting deadlocks
>>>>
>>> -- 
>>> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
>>> the body of a message to majordomo@vger.kernel.org
>>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>
>>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

[PATCH] btrfs-progs: add 'btrfs device ignore' cli

[Anand Jain]

This patch adds
  btrfs device ignode <dev>
so that a device can be ignored/missed during mount if the device is
already been scanned. Basically, this command will undo the effect
of the command
  btrfs device scan <dev>

This change is compatible with older kernel without the ioctl
BTRFS_IOC_IGNORE_DEV which is added here. At which it shall
report 'Inappropriate ioctl for device'. And will report
'Operation not permitted' on mounted device.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
 cmds-device.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 ioctl.h       |  2 ++
 2 files changed, 56 insertions(+)

diff --git a/cmds-device.c b/cmds-device.c
index 4337eb272037..78280e099492 100644
--- a/cmds-device.c
+++ b/cmds-device.c
@@ -320,6 +320,59 @@ out:
 	return !!ret;
 }
 
+static const char * const cmd_device_ignore_usage[] = {
+	"btrfs device ignore [<device>]",
+	"Ignore a device in the btrfs kernel module.",
+	NULL
+};
+
+static int btrfs_ignore_one_device(char *path)
+{
+	struct btrfs_ioctl_vol_args args;
+	int fd;
+	int ret;
+
+	fd = open("/dev/btrfs-control", O_RDWR);
+	if (fd < 0)
+		return -errno;
+
+	memset(&args, 0, sizeof(args));
+	strncpy_null(args.name, path);
+	ret = ioctl(fd, BTRFS_IOC_IGNORE_DEV, &args);
+	if (ret)
+		ret = -errno;
+	close(fd);
+	return ret;
+}
+
+static int cmd_device_ignore(int argc, char **argv)
+{
+	int ret = 0;
+	char *path;
+
+	if (check_argc_max(argc - optind, 1))
+		usage(cmd_device_ignore_usage);
+
+	if (is_block_device(argv[1]) != 1) {
+		error("Not a block device: %s", argv[1]);
+		return -ENOENT;
+	}
+
+	path = canonicalize_path(argv[1]);
+	if (!path) {
+		error("Could not canonicalize path '%s': %s",
+					argv[1], strerror(errno));
+		return -ENOENT;
+	}
+
+	ret  = btrfs_ignore_one_device(path);
+	if (ret)
+		error("Can't ignore %s: %s", path, strerror(-ret));
+
+	free(path);
+	return ret;
+}
+
 static const char * const cmd_device_ready_usage[] = {
 	"btrfs device ready <device>",
 	"Check device to see if it has all of its devices in cache for mounting",
@@ -595,6 +648,7 @@ const struct cmd_group device_cmd_group = {
 			CMD_ALIAS },
 		{ "remove", cmd_device_remove, cmd_device_remove_usage, NULL, 0 },
 		{ "scan", cmd_device_scan, cmd_device_scan_usage, NULL, 0 },
+		{ "ignore", cmd_device_ignore, cmd_device_ignore_usage, NULL, 0 },
 		{ "ready", cmd_device_ready, cmd_device_ready_usage, NULL, 0 },
 		{ "stats", cmd_device_stats, cmd_device_stats_usage, NULL, 0 },
 		{ "usage", cmd_device_usage,
diff --git a/ioctl.h b/ioctl.h
index 709e996f401c..bef8508119aa 100644
--- a/ioctl.h
+++ b/ioctl.h
@@ -721,6 +721,8 @@ static inline char *btrfs_err_str(enum btrfs_err_code err_code)
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
 				   struct btrfs_ioctl_vol_args)
+#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
+				   struct btrfs_ioctl_vol_args)
 /* trans start and trans end are dangerous, and only for
  * use by applications that know how to avoid the
  * resulting deadlocks
-- 
2.13.1

[PATCH v2 1/2] btrfs: add function to device list delete

[Anand Jain]

We need device delete from the dev_list so create a new function,
instead of refactor of btrfs_free_stale_device() because,
btrfs_free_stale_device() doesn't hold device_list_mutex which
is actually needed here.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v1: title of this patch
  btrfs: refactor btrfs_free_stale_device() to get device list delete
v2:
 add new function instead of peal from
 btrfs_free_stale_device() to delete a device from the device list,
 since btrfs_free_stale_device() does not hold device_list_mutex,
 may be it can be fixed later.


 /btrfs/volumes.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index bddf75d5e6a0..68f1dd44385b 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -546,6 +546,27 @@ static void pending_bios_fn(struct btrfs_work *work)
 	run_scheduled_bios(device);
 }
 
+static void delete_device_from_list(struct btrfs_device *device)
+{
+	struct btrfs_fs_devices *fs_devices;
+	fs_devices = device->fs_devices;
+
+	lockdep_assert_held(&uuid_mutex);
+
+	mutex_lock(&fs_devices->device_list_mutex);
+	fs_devices->num_devices--;
+	list_del(&device->dev_list);
+	mutex_unlock(&fs_devices->device_list_mutex);
+
+	rcu_string_free(device->name);
+	kfree(device);
+
+	if (fs_devices->num_devices == 0) {
+		btrfs_sysfs_remove_fsid(fs_devices);
+		list_del(&fs_devices->list);
+		free_fs_devices(fs_devices);
+	}
+}
 
 void btrfs_free_stale_device(struct btrfs_device *cur_dev)
 {
-- 
2.15.0

[PATCH v2 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

Support for a new command is being added here:
 btrfs dev ignore <dev>
Which shall undo the effects of the command
 btrfs dev scan <dev>

This cli/ioctl is needed as there is no way to continue to mount in
degraded mode if the device is already scanned, which is required to
recover from the split brain raid conditions.

This patch proposes to use ioctl #5 as it was empty.
	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
If #5 is reserved for some other purpose, I think I should change this.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v2: Use -EBUSY instead of -ENOENT
    Since now delete_device_from_list() holds device_list_mutex more
    fine grined as required, so dont hole device_list_mutex here.
    Reword and indent pr_err/info.

 fs/btrfs/super.c           |  4 +++
 fs/btrfs/volumes.c         | 64 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/volumes.h         |  2 ++
 include/uapi/linux/btrfs.h |  2 ++
 4 files changed, 72 insertions(+)

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index fa7bad8b22da..64f4beb1526d 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
 		break;
+	case BTRFS_IOC_IGNORE_DEV:
+		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
+					    &btrfs_fs_type, &fs_devices);
+		break;
 	case BTRFS_IOC_DEVICES_READY:
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 68f1dd44385b..e82b61da90e8 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1214,6 +1214,70 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
 	return 0;
 }
 
+static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
+{
+	int ret = 0;
+	struct btrfs_fs_devices *fs_devices;
+	struct btrfs_device *device;
+
+	fs_devices = find_fsid(disk_super->fsid);
+	if (!fs_devices)
+		return -ENOENT;
+
+	if (fs_devices->opened)
+		return -EBUSY;
+
+	mutex_lock(&uuid_mutex);
+
+	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
+	if (device)
+		delete_device_from_list(device);
+
+	mutex_unlock(&uuid_mutex);
+
+	return ret;
+}
+
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret)
+{
+	struct btrfs_super_block *disk_super;
+	struct block_device *bdev;
+	struct page *page;
+	int ret = -EINVAL;
+	u64 devid;
+	u64 bytenr;
+
+	bytenr = btrfs_sb_offset(0);
+	flags |= FMODE_EXCL;
+
+	bdev = blkdev_get_by_path(path, flags, holder);
+	if (IS_ERR(bdev)) {
+		ret = PTR_ERR(bdev);
+		goto error;
+	}
+
+	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
+		goto error_bdev_put;
+
+	devid = btrfs_stack_device_id(&disk_super->dev_item);
+
+	ret = device_list_remove(disk_super, devid);
+	if (ret)
+		pr_err("BTRFS: %pU device %s devid %llu failed to ignore: %d\n",
+			disk_super->fsid, path, devid, ret);
+	else
+		pr_info("BTRFS: %pU device %s devid %llu ignored\n",
+			disk_super->fsid, path, devid);
+
+	btrfs_release_disk_super(page);
+
+error_bdev_put:
+	blkdev_put(bdev, flags);
+error:
+	return ret;
+}
+
 /*
  * Look for a btrfs signature on a device. This may be called out of the mount path
  * and we are not allowed to call set_blocksize during the scan. The superblock
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index ff15208344a7..c5df0e6c2231 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
 int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 			  struct btrfs_fs_devices **fs_devices_ret);
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret);
 int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
 void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
index 6cdfd12cd14c..673ba6e6b331 100644
--- a/include/uapi/linux/btrfs.h
+++ b/include/uapi/linux/btrfs.h
@@ -740,6 +740,8 @@ enum btrfs_err_code {
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
 				   struct btrfs_ioctl_vol_args)
+#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
+				   struct btrfs_ioctl_vol_args)
 /* trans start and trans end are dangerous, and only for
  * use by applications that know how to avoid the
  * resulting deadlocks
-- 
2.15.0

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo

Re: [PATCH v2 1/2] btrfs: add function to device list delete

[Anand Jain]

   Wrong ML. Pls ignore.

Thanks, Anand


On 11/18/2017 10:03 PM, Anand Jain wrote:
> We need device delete from the dev_list so create a new function,
> instead of refactor of btrfs_free_stale_device() because,
> btrfs_free_stale_device() doesn't hold device_list_mutex which
> is actually needed here.
> 
> Signed-off-by: Anand Jain <anand.jain@oracle.com>
> ---
> v1: title of this patch
>    btrfs: refactor btrfs_free_stale_device() to get device list delete
> v2:
>   add new function instead of peal from
>   btrfs_free_stale_device() to delete a device from the device list,
>   since btrfs_free_stale_device() does not hold device_list_mutex,
>   may be it can be fixed later.
> 
> 
>   /btrfs/volumes.c | 21 +++++++++++++++++++++
>   1 file changed, 21 insertions(+)
> 
> diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> index bddf75d5e6a0..68f1dd44385b 100644
> --- a/fs/btrfs/volumes.c
> +++ b/fs/btrfs/volumes.c
> @@ -546,6 +546,27 @@ static void pending_bios_fn(struct btrfs_work *work)
>   	run_scheduled_bios(device);
>   }
>   
> +static void delete_device_from_list(struct btrfs_device *device)
> +{
> +	struct btrfs_fs_devices *fs_devices;
> +	fs_devices = device->fs_devices;
> +
> +	lockdep_assert_held(&uuid_mutex);
> +
> +	mutex_lock(&fs_devices->device_list_mutex);
> +	fs_devices->num_devices--;
> +	list_del(&device->dev_list);
> +	mutex_unlock(&fs_devices->device_list_mutex);
> +
> +	rcu_string_free(device->name);
> +	kfree(device);
> +
> +	if (fs_devices->num_devices == 0) {
> +		btrfs_sysfs_remove_fsid(fs_devices);
> +		list_del(&fs_devices->list);
> +		free_fs_devices(fs_devices);
> +	}
> +}
>   
>   void btrfs_free_stale_device(struct btrfs_device *cur_dev)
>   {
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-

[PATCH v3 1/2] btrfs: add function to device list delete

[Anand Jain]

We need device delete from the dev_list so create a new function.
New instead of refactor of btrfs_free_stale_device() because,
btrfs_free_stale_device() doesn't hold device_list_mutex which
is actually needed here.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v1: title of this patch
  btrfs: refactor btrfs_free_stale_device() to get device list delete
v2:
 delete_device_from_list() is not pealed from btrfs_free_stale_device()
 as this needs device_list_mutex. And its static now.
v3: Send to correct ML

 fs/btrfs/volumes.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index bddf75d5e6a0..68f1dd44385b 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -546,6 +546,27 @@ static void pending_bios_fn(struct btrfs_work *work)
 	run_scheduled_bios(device);
 }
 
+static void delete_device_from_list(struct btrfs_device *device)
+{
+	struct btrfs_fs_devices *fs_devices;
+	fs_devices = device->fs_devices;
+
+	lockdep_assert_held(&uuid_mutex);
+
+	mutex_lock(&fs_devices->device_list_mutex);
+	fs_devices->num_devices--;
+	list_del(&device->dev_list);
+	mutex_unlock(&fs_devices->device_list_mutex);
+
+	rcu_string_free(device->name);
+	kfree(device);
+
+	if (fs_devices->num_devices == 0) {
+		btrfs_sysfs_remove_fsid(fs_devices);
+		list_del(&fs_devices->list);
+		free_fs_devices(fs_devices);
+	}
+}
 
 void btrfs_free_stale_device(struct btrfs_device *cur_dev)
 {
-- 
2.15.0

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-

[PATCH v3 2/2] btrfs: introduce feature to ignore a btrfs device

[Anand Jain]

Support for a new command is being added here:
 btrfs dev ignore <dev>
Which shall undo the effects of the command
 btrfs dev scan <dev>

This cli/ioctl is needed as there is no way to continue to mount in
degraded mode if the device is already scanned, which is required to
recover from the split brain raid conditions.

This patch proposes to use ioctl #5 as it was empty.
	IOW(BTRFS_IOCTL_MAGIC, 5, ..)
If #5 is reserved for some other purpose, I think I should change this.

Signed-off-by: Anand Jain <anand.jain@oracle.com>
---
v2: Use -EBUSY instead of -ENOENT
    Since now delete_device_from_list() holds device_list_mutex
    so dont hold device_list_mutex in its parent. Reword and indent
    pr_err/info.
v3: Send to correct ML

 fs/btrfs/super.c           |  4 +++
 fs/btrfs/volumes.c         | 64 ++++++++++++++++++++++++++++++++++++++++++++++
 fs/btrfs/volumes.h         |  2 ++
 include/uapi/linux/btrfs.h |  2 ++
 4 files changed, 72 insertions(+)

diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index fa7bad8b22da..64f4beb1526d 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -2183,6 +2183,10 @@ static long btrfs_control_ioctl(struct file *file, unsigned int cmd,
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
 		break;
+	case BTRFS_IOC_IGNORE_DEV:
+		ret = btrfs_ignore_one_device(vol->name, FMODE_READ,
+					    &btrfs_fs_type, &fs_devices);
+		break;
 	case BTRFS_IOC_DEVICES_READY:
 		ret = btrfs_scan_one_device(vol->name, FMODE_READ,
 					    &btrfs_fs_type, &fs_devices);
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 68f1dd44385b..e82b61da90e8 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -1214,6 +1214,70 @@ int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr,
 	return 0;
 }
 
+static int device_list_remove(struct btrfs_super_block *disk_super, u64 devid)
+{
+	int ret = 0;
+	struct btrfs_fs_devices *fs_devices;
+	struct btrfs_device *device;
+
+	fs_devices = find_fsid(disk_super->fsid);
+	if (!fs_devices)
+		return -ENOENT;
+
+	if (fs_devices->opened)
+		return -EBUSY;
+
+	mutex_lock(&uuid_mutex);
+
+	device = find_device(fs_devices, devid, disk_super->dev_item.uuid);
+	if (device)
+		delete_device_from_list(device);
+
+	mutex_unlock(&uuid_mutex);
+
+	return ret;
+}
+
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret)
+{
+	struct btrfs_super_block *disk_super;
+	struct block_device *bdev;
+	struct page *page;
+	int ret = -EINVAL;
+	u64 devid;
+	u64 bytenr;
+
+	bytenr = btrfs_sb_offset(0);
+	flags |= FMODE_EXCL;
+
+	bdev = blkdev_get_by_path(path, flags, holder);
+	if (IS_ERR(bdev)) {
+		ret = PTR_ERR(bdev);
+		goto error;
+	}
+
+	if (btrfs_read_disk_super(bdev, bytenr, &page, &disk_super))
+		goto error_bdev_put;
+
+	devid = btrfs_stack_device_id(&disk_super->dev_item);
+
+	ret = device_list_remove(disk_super, devid);
+	if (ret)
+		pr_err("BTRFS: %pU device %s devid %llu failed to ignore: %d\n",
+			disk_super->fsid, path, devid, ret);
+	else
+		pr_info("BTRFS: %pU device %s devid %llu ignored\n",
+			disk_super->fsid, path, devid);
+
+	btrfs_release_disk_super(page);
+
+error_bdev_put:
+	blkdev_put(bdev, flags);
+error:
+	return ret;
+}
+
 /*
  * Look for a btrfs signature on a device. This may be called out of the mount path
  * and we are not allowed to call set_blocksize during the scan. The superblock
diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h
index ff15208344a7..c5df0e6c2231 100644
--- a/fs/btrfs/volumes.h
+++ b/fs/btrfs/volumes.h
@@ -421,6 +421,8 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices,
 		       fmode_t flags, void *holder);
 int btrfs_scan_one_device(const char *path, fmode_t flags, void *holder,
 			  struct btrfs_fs_devices **fs_devices_ret);
+int btrfs_ignore_one_device(const char *path, fmode_t flags, void *holder,
+			  struct btrfs_fs_devices **fs_devices_ret);
 int btrfs_close_devices(struct btrfs_fs_devices *fs_devices);
 void btrfs_close_extra_devices(struct btrfs_fs_devices *fs_devices, int step);
 void btrfs_assign_next_active_device(struct btrfs_fs_info *fs_info,
diff --git a/include/uapi/linux/btrfs.h b/include/uapi/linux/btrfs.h
index 6cdfd12cd14c..673ba6e6b331 100644
--- a/include/uapi/linux/btrfs.h
+++ b/include/uapi/linux/btrfs.h
@@ -740,6 +740,8 @@ enum btrfs_err_code {
 				   struct btrfs_ioctl_vol_args)
 #define BTRFS_IOC_SCAN_DEV _IOW(BTRFS_IOCTL_MAGIC, 4, \
 				   struct btrfs_ioctl_vol_args)
+#define BTRFS_IOC_IGNORE_DEV _IOW(BTRFS_IOCTL_MAGIC, 5, \
+				   struct btrfs_ioctl_vol_args)
 /* trans start and trans end are dangerous, and only for
  * use by applications that know how to avoid the
  * resulting deadlocks
-- 
2.15.0

--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo