2022-06-15 10:33:27

by Pankaj Raghav

[permalink] [raw]
Subject: [PATCH v7 10/13] dm-table: use bdev_is_zone_start helper in device_area_is_invalid()

Use bdev_is_zone_start() helper that uses generic calculation to check
for zone alignment instead of using po2 based alignment check.

Signed-off-by: Pankaj Raghav <[email protected]>
Reviewed-by: Luis Chamberlain <[email protected]>
---
drivers/md/dm-table.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index bd539afbf..b553cdb6d 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -251,7 +251,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
if (bdev_is_zoned(bdev)) {
unsigned int zone_sectors = bdev_zone_sectors(bdev);

- if (start & (zone_sectors - 1)) {
+ if (blk_queue_is_zone_start(bdev_get_queue(bdev), start)) {
DMWARN("%s: start=%llu not aligned to h/w zone size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)start,
@@ -268,7 +268,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
* devices do not end up with a smaller zone in the middle of
* the sector range.
*/
- if (len & (zone_sectors - 1)) {
+ if (blk_queue_is_zone_start(bdev_get_queue(bdev), len)) {
DMWARN("%s: len=%llu not aligned to h/w zone size %u of %pg",
dm_device_name(ti->table->md),
(unsigned long long)len,
--
2.25.1


2022-06-15 12:14:33

by Damien Le Moal

[permalink] [raw]
Subject: Re: [PATCH v7 10/13] dm-table: use bdev_is_zone_start helper in device_area_is_invalid()

On 6/15/22 19:19, Pankaj Raghav wrote:
> Use bdev_is_zone_start() helper that uses generic calculation to check
> for zone alignment instead of using po2 based alignment check.
>
> Signed-off-by: Pankaj Raghav <[email protected]>
> Reviewed-by: Luis Chamberlain <[email protected]>
> ---
> drivers/md/dm-table.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
> index bd539afbf..b553cdb6d 100644
> --- a/drivers/md/dm-table.c
> +++ b/drivers/md/dm-table.c
> @@ -251,7 +251,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
> if (bdev_is_zoned(bdev)) {
> unsigned int zone_sectors = bdev_zone_sectors(bdev);
>
> - if (start & (zone_sectors - 1)) {
> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), start)) {

This is wrong. And you are changing this to the correct test in the next
patch.

> DMWARN("%s: start=%llu not aligned to h/w zone size %u of %pg",
> dm_device_name(ti->table->md),
> (unsigned long long)start,
> @@ -268,7 +268,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
> * devices do not end up with a smaller zone in the middle of
> * the sector range.
> */
> - if (len & (zone_sectors - 1)) {
> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), len)) {
> DMWARN("%s: len=%llu not aligned to h/w zone size %u of %pg",
> dm_device_name(ti->table->md),
> (unsigned long long)len,


--
Damien Le Moal
Western Digital Research

2022-06-16 10:25:53

by Pankaj Raghav

[permalink] [raw]
Subject: Re: [PATCH v7 10/13] dm-table: use bdev_is_zone_start helper in device_area_is_invalid()

drivers/md/dm-table.c
>> +++ b/drivers/md/dm-table.c
>> @@ -251,7 +251,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
>> if (bdev_is_zoned(bdev)) {
>> unsigned int zone_sectors = bdev_zone_sectors(bdev);
>>
>> - if (start & (zone_sectors - 1)) {
>> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), start)) {
>
> This is wrong. And you are changing this to the correct test in the next
> patch.
>
Yeah, I think I made a mistake while committing it. The next patch
should only have the removing po2 condition check and these changes
should have been only in this patch. I will fix it up!
>> DMWARN("%s: start=%llu not aligned to h/w zone size %u of %pg",
>> dm_device_name(ti->table->md),
>> (unsigned long long)start,
>> @@ -268,7 +268,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
>> * devices do not end up with a smaller zone in the middle of
>> * the sector range.
>> */
>> - if (len & (zone_sectors - 1)) {
>> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), len)) {
>> DMWARN("%s: len=%llu not aligned to h/w zone size %u of %pg",
>> dm_device_name(ti->table->md),
>> (unsigned long long)len,
>
>

2022-06-16 23:40:08

by Damien Le Moal

[permalink] [raw]
Subject: Re: [PATCH v7 10/13] dm-table: use bdev_is_zone_start helper in device_area_is_invalid()

On 6/16/22 18:55, Pankaj Raghav wrote:
> drivers/md/dm-table.c
>>> +++ b/drivers/md/dm-table.c
>>> @@ -251,7 +251,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
>>> if (bdev_is_zoned(bdev)) {
>>> unsigned int zone_sectors = bdev_zone_sectors(bdev);
>>>
>>> - if (start & (zone_sectors - 1)) {
>>> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), start)) {
>>
>> This is wrong. And you are changing this to the correct test in the next
>> patch.
>>
> Yeah, I think I made a mistake while committing it. The next patch
> should only have the removing po2 condition check and these changes
> should have been only in this patch. I will fix it up!

This one and the next patch should be squashed together. They both deal
with non power of 2 zone size.


>>> DMWARN("%s: start=%llu not aligned to h/w zone size %u of %pg",
>>> dm_device_name(ti->table->md),
>>> (unsigned long long)start,
>>> @@ -268,7 +268,7 @@ static int device_area_is_invalid(struct dm_target *ti, struct dm_dev *dev,
>>> * devices do not end up with a smaller zone in the middle of
>>> * the sector range.
>>> */
>>> - if (len & (zone_sectors - 1)) {
>>> + if (blk_queue_is_zone_start(bdev_get_queue(bdev), len)) {
>>> DMWARN("%s: len=%llu not aligned to h/w zone size %u of %pg",
>>> dm_device_name(ti->table->md),
>>> (unsigned long long)len,
>>
>>


--
Damien Le Moal
Western Digital Research