diff mbox series

[v2] mtd: mtd_blkdevs: Initialize rq.limits.discard_granularity

Message ID 20210615093905.3473709-1-chengzhihao1@huawei.com
State Accepted
Headers show
Series [v2] mtd: mtd_blkdevs: Initialize rq.limits.discard_granularity | expand

Commit Message

Zhihao Cheng June 15, 2021, 9:39 a.m. UTC
Since commit b35fd7422c2f8("block: check queue's limits.discard_granularity
in __blkdev_issue_discard()") checks rq.limits.discard_granularity in
__blkdev_issue_discard(), we may get following warnings on formatted ftl:

  WARNING: CPU: 2 PID: 7313 at block/blk-lib.c:51
  __blkdev_issue_discard+0x2a7/0x390

Reproducer:
  1. ftl_format /dev/mtd0
  2. modprobe ftl
  3. mkfs.vfat /dev/ftla
  4. mount -odiscard /dev/ftla temp
  5. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
  6. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct

Fix it by initializing rq.limits.discard_granularity if device supports
discard operation.

Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
---
 drivers/mtd/mtd_blkdevs.c | 1 +
 1 file changed, 1 insertion(+)

Comments

Miquel Raynal June 18, 2021, 7:52 a.m. UTC | #1
Hi Zhihao,

Zhihao Cheng <chengzhihao1@huawei.com> wrote on Tue, 15 Jun 2021
17:39:05 +0800:

> Since commit b35fd7422c2f8("block: check queue's limits.discard_granularity
> in __blkdev_issue_discard()") checks rq.limits.discard_granularity in
> __blkdev_issue_discard(), we may get following warnings on formatted ftl:
> 
>   WARNING: CPU: 2 PID: 7313 at block/blk-lib.c:51
>   __blkdev_issue_discard+0x2a7/0x390
> 
> Reproducer:
>   1. ftl_format /dev/mtd0
>   2. modprobe ftl
>   3. mkfs.vfat /dev/ftla
>   4. mount -odiscard /dev/ftla temp
>   5. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>   6. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
> 
> Fix it by initializing rq.limits.discard_granularity if device supports
> discard operation.

Can you please verify that this is still needed after Christoph's
additional fix:
https://patchwork.kernel.org/project/xen-devel/patch/20210602065345.355274-10-hch@lst.de/

> 
> Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
> ---
>  drivers/mtd/mtd_blkdevs.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
> index 01bf234e7f3a..6eac1c9c164d 100644
> --- a/drivers/mtd/mtd_blkdevs.c
> +++ b/drivers/mtd/mtd_blkdevs.c
> @@ -413,6 +413,7 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
>  	if (tr->discard) {
>  		blk_queue_flag_set(QUEUE_FLAG_DISCARD, new->rq);
>  		blk_queue_max_discard_sectors(new->rq, UINT_MAX);
> +		new->rq->limits.discard_granularity = tr->blksize;
>  	}
>  
>  	gd->queue = new->rq;




Thanks,
Miquèl
Zhihao Cheng June 18, 2021, 8:55 a.m. UTC | #2
在 2021/6/18 15:52, Miquel Raynal 写道:
> Hi Zhihao,
> 
> Zhihao Cheng <chengzhihao1@huawei.com> wrote on Tue, 15 Jun 2021
> 17:39:05 +0800:
> 
>> Since commit b35fd7422c2f8("block: check queue's limits.discard_granularity
>> in __blkdev_issue_discard()") checks rq.limits.discard_granularity in
>> __blkdev_issue_discard(), we may get following warnings on formatted ftl:
>>
>>    WARNING: CPU: 2 PID: 7313 at block/blk-lib.c:51
>>    __blkdev_issue_discard+0x2a7/0x390
>>
>> Reproducer:
>>    1. ftl_format /dev/mtd0
>>    2. modprobe ftl
>>    3. mkfs.vfat /dev/ftla
>>    4. mount -odiscard /dev/ftla temp
>>    5. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>>    6. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>>
>> Fix it by initializing rq.limits.discard_granularity if device supports
>> discard operation.
> 
> Can you please verify that this is still needed after Christoph's
> additional fix:
> https://patchwork.kernel.org/project/xen-devel/patch/20210602065345.355274-10-hch@lst.de/
> 
Hi, Miquel,
I can still reproduce the problem after [1] applied.
I look through this series patches and I didn't find out 
'rq.limits.discard_granularity' has been initialized somewhere. The 
patchset is aming at reconstructing to avoid boilerplate code in drivers?
So 'rq.limits.discard_granularity' still need to be initialized in 
mtdblock driver.

[1] 
https://patchwork.kernel.org/project/xen-devel/patch/20210602065345.355274-10-hch@lst.de/

>
Zhihao Cheng July 12, 2021, 1:39 p.m. UTC | #3
在 2021/6/18 16:55, Zhihao Cheng 写道:
> 在 2021/6/18 15:52, Miquel Raynal 写道:
>> Hi Zhihao,
>>
>> Zhihao Cheng <chengzhihao1@huawei.com> wrote on Tue, 15 Jun 2021
>> 17:39:05 +0800:
>>
>>> Since commit b35fd7422c2f8("block: check queue's 
>>> limits.discard_granularity
>>> in __blkdev_issue_discard()") checks rq.limits.discard_granularity in
>>> __blkdev_issue_discard(), we may get following warnings on formatted 
>>> ftl:
>>>
>>>    WARNING: CPU: 2 PID: 7313 at block/blk-lib.c:51
>>>    __blkdev_issue_discard+0x2a7/0x390
>>>
>>> Reproducer:
>>>    1. ftl_format /dev/mtd0
>>>    2. modprobe ftl
>>>    3. mkfs.vfat /dev/ftla
>>>    4. mount -odiscard /dev/ftla temp
>>>    5. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>>>    6. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>>>
>>> Fix it by initializing rq.limits.discard_granularity if device supports
>>> discard operation.
>>
>> Can you please verify that this is still needed after Christoph's
>> additional fix:
>> https://patchwork.kernel.org/project/xen-devel/patch/20210602065345.355274-10-hch@lst.de/ 
>>
>>
> Hi, Miquel,
> I can still reproduce the problem after [1] applied.
> I look through this series patches and I didn't find out 
> 'rq.limits.discard_granularity' has been initialized somewhere. The 
> patchset is aming at reconstructing to avoid boilerplate code in drivers?
> So 'rq.limits.discard_granularity' still need to be initialized in 
> mtdblock driver.
>
> [1] 
> https://patchwork.kernel.org/project/xen-devel/patch/20210602065345.355274-10-hch@lst.de/
>
friendly
Miquel Raynal July 15, 2021, 10:50 p.m. UTC | #4
On Tue, 2021-06-15 at 09:39:05 UTC, Zhihao Cheng wrote:
> Since commit b35fd7422c2f8("block: check queue's limits.discard_granularity
> in __blkdev_issue_discard()") checks rq.limits.discard_granularity in
> __blkdev_issue_discard(), we may get following warnings on formatted ftl:
> 
>   WARNING: CPU: 2 PID: 7313 at block/blk-lib.c:51
>   __blkdev_issue_discard+0x2a7/0x390
> 
> Reproducer:
>   1. ftl_format /dev/mtd0
>   2. modprobe ftl
>   3. mkfs.vfat /dev/ftla
>   4. mount -odiscard /dev/ftla temp
>   5. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
>   6. dd if=/dev/zero of=temp/tst bs=1M count=10 oflag=direct
> 
> Fix it by initializing rq.limits.discard_granularity if device supports
> discard operation.
> 
> Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>

Applied to https://git.kernel.org/pub/scm/linux/kernel/git/mtd/linux.git mtd/fixes, thanks.

Miquel
diff mbox series

Patch

diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index 01bf234e7f3a..6eac1c9c164d 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -413,6 +413,7 @@  int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
 	if (tr->discard) {
 		blk_queue_flag_set(QUEUE_FLAG_DISCARD, new->rq);
 		blk_queue_max_discard_sectors(new->rq, UINT_MAX);
+		new->rq->limits.discard_granularity = tr->blksize;
 	}
 
 	gd->queue = new->rq;