All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tanya Brokhman <tlinder@codeaurora.org>
To: Richard Weinberger <richard@nod.at>, dedekind1@gmail.com
Cc: linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 2/6] UBI: Fastmap: Ensure that only one fastmap work is scheduled
Date: Thu, 04 Dec 2014 18:14:08 +0200	[thread overview]
Message-ID: <54808850.5090200@codeaurora.org> (raw)
In-Reply-To: <1416835236-25185-3-git-send-email-richard@nod.at>

On 11/24/2014 3:20 PM, Richard Weinberger wrote:
> If the WL pool runs out of PEBs we schedule a fastmap write
> to refill it as soon as possible.
> Ensure that only one at a time is scheduled otherwise we might end in
> a fastmap write storm because writing the fastmap can schedule another
> write if bitflips are detected.
>

Reviewed-by: Tanya Brokhman <tlinder@codeaurora.org>

> Signed-off-by: Richard Weinberger <richard@nod.at>
> ---
>   drivers/mtd/ubi/ubi.h | 4 +++-
>   drivers/mtd/ubi/wl.c  | 8 +++++++-
>   2 files changed, 10 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
> index f80ffab..04c4c05 100644
> --- a/drivers/mtd/ubi/ubi.h
> +++ b/drivers/mtd/ubi/ubi.h
> @@ -427,6 +427,7 @@ struct ubi_debug_info {
>    * @fm_size: fastmap size in bytes
>    * @fm_sem: allows ubi_update_fastmap() to block EBA table changes
>    * @fm_work: fastmap work queue
> + * @fm_work_scheduled: non-zero if fastmap work was scheduled
>    *
>    * @used: RB-tree of used physical eraseblocks
>    * @erroneous: RB-tree of erroneous used physical eraseblocks
> @@ -438,7 +439,7 @@ struct ubi_debug_info {
>    * @pq_head: protection queue head
>    * @wl_lock: protects the @used, @free, @pq, @pq_head, @lookuptbl, @move_from,
>    *	     @move_to, @move_to_put @erase_pending, @wl_scheduled, @works,
> - *	     @erroneous, and @erroneous_peb_count fields
> + *	     @erroneous, @erroneous_peb_count, and @fm_work_scheduled fields
>    * @move_mutex: serializes eraseblock moves
>    * @work_sem: used to wait for all the scheduled works to finish and prevent
>    * new works from being submitted
> @@ -533,6 +534,7 @@ struct ubi_device {
>   	void *fm_buf;
>   	size_t fm_size;
>   	struct work_struct fm_work;
> +	int fm_work_scheduled;
>
>   	/* Wear-leveling sub-system's stuff */
>   	struct rb_root used;
> diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
> index 834f6fe..7f135df 100644
> --- a/drivers/mtd/ubi/wl.c
> +++ b/drivers/mtd/ubi/wl.c
> @@ -149,6 +149,9 @@ static void update_fastmap_work_fn(struct work_struct *wrk)
>   {
>   	struct ubi_device *ubi = container_of(wrk, struct ubi_device, fm_work);
>   	ubi_update_fastmap(ubi);
> +	spin_lock(&ubi->wl_lock);
> +	ubi->fm_work_scheduled = 0;
> +	spin_unlock(&ubi->wl_lock);
>   }
>
>   /**
> @@ -660,7 +663,10 @@ static struct ubi_wl_entry *get_peb_for_wl(struct ubi_device *ubi)
>   		/* We cannot update the fastmap here because this
>   		 * function is called in atomic context.
>   		 * Let's fail here and refill/update it as soon as possible. */
> -		schedule_work(&ubi->fm_work);
> +		if (!ubi->fm_work_scheduled) {
> +			ubi->fm_work_scheduled = 1;
> +			schedule_work(&ubi->fm_work);
> +		}
>   		return NULL;
>   	} else {
>   		pnum = pool->pebs[pool->used++];
>


Thanks,
Tanya Brokhman
-- 
Qualcomm Israel, on behalf of Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project

  parent reply	other threads:[~2014-12-04 16:14 UTC|newest]

Thread overview: 66+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-24 13:20 Fastmap update v2 (pile 1) Richard Weinberger
2014-11-24 13:20 ` [PATCH 1/6] UBI: Fastmap: Care about the protection queue Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-11-27 14:54   ` Artem Bityutskiy
2015-01-09 21:23   ` Ezequiel Garcia
2015-01-09 21:31     ` Richard Weinberger
2015-01-09 21:34       ` Ezequiel Garcia
2014-11-24 13:20 ` [PATCH 2/6] UBI: Fastmap: Ensure that only one fastmap work is scheduled Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-11-27 15:27   ` Artem Bityutskiy
2014-11-27 16:13     ` Richard Weinberger
2014-11-27 16:35       ` Artem Bityutskiy
2014-11-27 16:39         ` Richard Weinberger
2014-11-27 16:49           ` Artem Bityutskiy
2014-12-04 16:14   ` Tanya Brokhman [this message]
2014-12-17 13:51   ` Guido Martínez
2014-12-17 13:51     ` Guido Martínez
2014-11-24 13:20 ` [PATCH 3/6] UBI: Fastmap: Ensure that all fastmap work is done upon WL shutdown Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-11-27 15:38   ` Artem Bityutskiy
2014-11-27 16:08     ` Richard Weinberger
2014-11-27 16:29       ` Artem Bityutskiy
2014-11-27 16:35         ` Richard Weinberger
2014-11-27 16:47           ` Artem Bityutskiy
2014-11-28  9:53             ` Richard Weinberger
2014-12-04 16:44     ` Tanya Brokhman
2014-12-04 17:21       ` Richard Weinberger
2014-12-17 14:26   ` Guido Martínez
2014-12-17 14:26     ` Guido Martínez
2015-01-09 21:32   ` Ezequiel Garcia
2015-01-09 21:37     ` Richard Weinberger
2015-01-09 21:39       ` Ezequiel Garcia
2014-11-24 13:20 ` [PATCH 4/6] UBI: Fastmap: Fix races in ubi_wl_get_peb() Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-12-05 13:09   ` Tanya Brokhman
2014-12-05 13:20     ` Richard Weinberger
2014-12-05 16:54       ` Tanya Brokhman
2014-12-05 21:08         ` Richard Weinberger
2014-12-07  7:36           ` Tanya Brokhman
2014-12-07  9:45             ` Richard Weinberger
2014-11-24 13:20 ` [PATCH 5/6] UBI: Split __wl_get_peb() Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-12-05 17:41   ` Tanya Brokhman
2014-12-05 21:02     ` Richard Weinberger
2014-12-17 15:03   ` Guido Martínez
2014-12-17 15:03     ` Guido Martínez
2014-11-24 13:20 ` [PATCH 6/6] UBI: Fastmap: Make ubi_refill_pools() fair Richard Weinberger
2014-11-24 13:20   ` Richard Weinberger
2014-12-05 17:55   ` Tanya Brokhman
2014-12-05 20:56     ` Richard Weinberger
2014-12-07  7:55       ` Tanya Brokhman
2014-12-07  9:49         ` Richard Weinberger
2014-12-17 15:48   ` Guido Martínez
2014-12-17 15:48     ` Guido Martínez
2014-11-27 14:53 ` Fastmap update v2 (pile 1) Artem Bityutskiy
2014-11-27 14:59   ` Richard Weinberger
2014-12-10  8:21 ` Richard Weinberger
2014-12-10  8:21   ` Richard Weinberger
2015-01-05 10:37   ` Richard Weinberger
2015-01-05 10:37     ` Richard Weinberger
2015-01-09 21:38 ` Ezequiel Garcia
2015-01-09 21:55   ` Richard Weinberger
2015-01-09 22:09     ` Ezequiel Garcia
2015-01-09 22:20       ` Richard Weinberger
2015-03-29 10:46 ` Richard Weinberger
2015-03-29 10:46   ` Richard Weinberger

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=54808850.5090200@codeaurora.org \
    --to=tlinder@codeaurora.org \
    --cc=dedekind1@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=richard@nod.at \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.