[PATCH V1] md:For aborted a resync by request,using ->curr_resync as the last completed request rather than ->curr_resync_completed.

[PATCH V1] md:For aborted a resync by request,using ->curr_resync as the last completed request rather than ->curr_resync_completed.

[majianpeng]

V0-->V1:
As suggested by neil, it added a flag MD_RECOVERY_ERROR to differentiate between an
abort-due-to-error and an abort-due-to-request.We only deal with the later case.

Thanks!
Jianpeng

Aborting a resync maybe by io-error,or request(stop array or reboot
system).
So we add a flag MD_RECOVERY_ERROR to differentiate between an
abort-due-to-error and an abort-due-to-request.
For the later case,when resync aborted,it exec
wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active)).
So the ->curr_resync is the last completed request.

Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
---
 drivers/md/md.c |   10 ++++++++--
 drivers/md/md.h |    2 ++
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 9ab768a..5199657 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -7172,6 +7172,7 @@ void md_done_sync(struct mddev *mddev, int blocks, int ok)
 	wake_up(&mddev->recovery_wait);
 	if (!ok) {
 		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
+		set_bit(MD_RECOVERY_ERROR, &mddev->recovery);
 		md_wakeup_thread(mddev->thread);
 		// stop recovery, signal do_sync ....
 	}
@@ -7558,8 +7559,13 @@ void md_do_sync(struct md_thread *thread)
 					printk(KERN_INFO
 					       "md: checkpointing %s of %s.\n",
 					       desc, mdname(mddev));
-					mddev->recovery_cp =
-						mddev->curr_resync_completed;
+					if (test_bit(MD_RECOVERY_ERROR,
+						&mddev->recovery_cp))
+						mddev->recovery_cp =
+							mddev->curr_resync_completed;
+					else
+						mddev->recovery_cp =
+							mddev->curr_resync;
 				}
 			} else
 				mddev->recovery_cp = MaxSector;
diff --git a/drivers/md/md.h b/drivers/md/md.h
index af443ab..c29e62e 100644
--- a/drivers/md/md.h
+++ b/drivers/md/md.h
@@ -307,6 +307,7 @@ struct mddev {
 	 * REQUEST:  user-space has requested a sync (used with SYNC)
 	 * CHECK:    user-space request for check-only, no repair
 	 * RESHAPE:  A reshape is happening
+	 * ERROR:    sync-action interrupted because io-error
 	 *
 	 * If neither SYNC or RESHAPE are set, then it is a recovery.
 	 */
@@ -320,6 +321,7 @@ struct mddev {
 #define	MD_RECOVERY_CHECK	7
 #define MD_RECOVERY_RESHAPE	8
 #define	MD_RECOVERY_FROZEN	9
+#define	MD_RECOVERY_ERROR	10
 
 	unsigned long			recovery;
 	/* If a RAID personality determines that recovery (of a particular
-- 
1.7.9.5

Re: [PATCH V1] md:For aborted a resync by request,using ->curr_resync as the last completed request rather than ->curr_resync_completed.

[NeilBrown]

[-- Attachment #1: Type: text/plain, Size: 2790 bytes --]

On Mon, 19 Nov 2012 19:57:34 +0800 majianpeng <majianpeng@gmail.com> wrote:

> V0-->V1:
> As suggested by neil, it added a flag MD_RECOVERY_ERROR to differentiate between an
> abort-due-to-error and an abort-due-to-request.We only deal with the later case.
> 
> Thanks!
> Jianpeng
> 
> Aborting a resync maybe by io-error,or request(stop array or reboot
> system).
> So we add a flag MD_RECOVERY_ERROR to differentiate between an
> abort-due-to-error and an abort-due-to-request.
> For the later case,when resync aborted,it exec
> wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active)).
> So the ->curr_resync is the last completed request.
> 
> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
> ---
>  drivers/md/md.c |   10 ++++++++--
>  drivers/md/md.h |    2 ++
>  2 files changed, 10 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/md/md.c b/drivers/md/md.c
> index 9ab768a..5199657 100644
> --- a/drivers/md/md.c
> +++ b/drivers/md/md.c
> @@ -7172,6 +7172,7 @@ void md_done_sync(struct mddev *mddev, int blocks, int ok)
>  	wake_up(&mddev->recovery_wait);
>  	if (!ok) {
>  		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
> +		set_bit(MD_RECOVERY_ERROR, &mddev->recovery);
>  		md_wakeup_thread(mddev->thread);
>  		// stop recovery, signal do_sync ....
>  	}
> @@ -7558,8 +7559,13 @@ void md_do_sync(struct md_thread *thread)
>  					printk(KERN_INFO
>  					       "md: checkpointing %s of %s.\n",
>  					       desc, mdname(mddev));
> -					mddev->recovery_cp =
> -						mddev->curr_resync_completed;
> +					if (test_bit(MD_RECOVERY_ERROR,
> +						&mddev->recovery_cp))
> +						mddev->recovery_cp =
> +							mddev->curr_resync_completed;
> +					else
> +						mddev->recovery_cp =
> +							mddev->curr_resync;
>  				}
>  			} else
>  				mddev->recovery_cp = MaxSector;
> diff --git a/drivers/md/md.h b/drivers/md/md.h
> index af443ab..c29e62e 100644
> --- a/drivers/md/md.h
> +++ b/drivers/md/md.h
> @@ -307,6 +307,7 @@ struct mddev {
>  	 * REQUEST:  user-space has requested a sync (used with SYNC)
>  	 * CHECK:    user-space request for check-only, no repair
>  	 * RESHAPE:  A reshape is happening
> +	 * ERROR:    sync-action interrupted because io-error
>  	 *
>  	 * If neither SYNC or RESHAPE are set, then it is a recovery.
>  	 */
> @@ -320,6 +321,7 @@ struct mddev {
>  #define	MD_RECOVERY_CHECK	7
>  #define MD_RECOVERY_RESHAPE	8
>  #define	MD_RECOVERY_FROZEN	9
> +#define	MD_RECOVERY_ERROR	10
>  
>  	unsigned long			recovery;
>  	/* If a RAID personality determines that recovery (of a particular



The attachment still have MS-windows line endings ("\r\n" instead of just
"\n") but it wasn't too hard to strip those out.

Applied, thanks,
NeilBrown


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]

Re: Re: [PATCH V1] md:For aborted a resync by request,using ->curr_resync as the last completed request rather than ->curr_resync_completed.

[majianpeng]

>On Mon, 19 Nov 2012 19:57:34 +0800 majianpeng <majianpeng@gmail.com> wrote:
>
>> V0-->V1:
>> As suggested by neil, it added a flag MD_RECOVERY_ERROR to differentiate between an
>> abort-due-to-error and an abort-due-to-request.We only deal with the later case.
>> 
>> Thanks!
>> Jianpeng
>> 
>> Aborting a resync maybe by io-error,or request(stop array or reboot
>> system).
>> So we add a flag MD_RECOVERY_ERROR to differentiate between an
>> abort-due-to-error and an abort-due-to-request.
>> For the later case,when resync aborted,it exec
>> wait_event(mddev->recovery_wait, !atomic_read(&mddev->recovery_active)).
>> So the ->curr_resync is the last completed request.
>> 
>> Signed-off-by: Jianpeng Ma <majianpeng@gmail.com>
>> ---
>>  drivers/md/md.c |   10 ++++++++--
>>  drivers/md/md.h |    2 ++
>>  2 files changed, 10 insertions(+), 2 deletions(-)
>> 
>> diff --git a/drivers/md/md.c b/drivers/md/md.c
>> index 9ab768a..5199657 100644
>> --- a/drivers/md/md.c
>> +++ b/drivers/md/md.c
>> @@ -7172,6 +7172,7 @@ void md_done_sync(struct mddev *mddev, int blocks, int ok)
>>  	wake_up(&mddev->recovery_wait);
>>  	if (!ok) {
>>  		set_bit(MD_RECOVERY_INTR, &mddev->recovery);
>> +		set_bit(MD_RECOVERY_ERROR, &mddev->recovery);
>>  		md_wakeup_thread(mddev->thread);
>>  		// stop recovery, signal do_sync ....
>>  	}
>> @@ -7558,8 +7559,13 @@ void md_do_sync(struct md_thread *thread)
>>  					printk(KERN_INFO
>>  					       "md: checkpointing %s of %s.\n",
>>  					       desc, mdname(mddev));
>> -					mddev->recovery_cp =
>> -						mddev->curr_resync_completed;
>> +					if (test_bit(MD_RECOVERY_ERROR,
>> +						&mddev->recovery_cp))
>> +						mddev->recovery_cp =
>> +							mddev->curr_resync_completed;
>> +					else
>> +						mddev->recovery_cp =
>> +							mddev->curr_resync;
>>  				}
>>  			} else
>>  				mddev->recovery_cp = MaxSector;
>> diff --git a/drivers/md/md.h b/drivers/md/md.h
>> index af443ab..c29e62e 100644
>> --- a/drivers/md/md.h
>> +++ b/drivers/md/md.h
>> @@ -307,6 +307,7 @@ struct mddev {
>>  	 * REQUEST:  user-space has requested a sync (used with SYNC)
>>  	 * CHECK:    user-space request for check-only, no repair
>>  	 * RESHAPE:  A reshape is happening
>> +	 * ERROR:    sync-action interrupted because io-error
>>  	 *
>>  	 * If neither SYNC or RESHAPE are set, then it is a recovery.
>>  	 */
>> @@ -320,6 +321,7 @@ struct mddev {
>>  #define	MD_RECOVERY_CHECK	7
>>  #define MD_RECOVERY_RESHAPE	8
>>  #define	MD_RECOVERY_FROZEN	9
>> +#define	MD_RECOVERY_ERROR	10
>>  
>>  	unsigned long			recovery;
>>  	/* If a RAID personality determines that recovery (of a particular
>
>
>
>The attachment still have MS-windows line endings ("\r\n" instead of just
>"\n") but it wasn't too hard to strip those out.
>
>Applied, thanks,
>NeilBrown
>
>
Sorry for that, i will check it and fix.
Thanks!
Jianpeng