When trying to issue write_zeroes command against TARGET with a 4K block
size, it ends up hitting the following condition at __blkdev_issue_zeroout:
if ((sector | nr_sects) & bs_mask)
return -EINVAL;
Causing the command to always fail.
Considering we need to add 1 to get the correct block count, that addition
needs to be performed in the native format, so we moved the +1 to within
le16_to_cpu prior to converting to 512b.
Signed-off-by: Rodrigo R. Galvao <[email protected]>
---
drivers/nvme/target/io-cmd.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/nvme/target/io-cmd.c b/drivers/nvme/target/io-cmd.c
index 28bbdff..5292bc3 100644
--- a/drivers/nvme/target/io-cmd.c
+++ b/drivers/nvme/target/io-cmd.c
@@ -173,8 +173,8 @@ static void nvmet_execute_write_zeroes(struct nvmet_req *req)
sector = le64_to_cpu(write_zeroes->slba) <<
(req->ns->blksize_shift - 9);
- nr_sector = (((sector_t)le16_to_cpu(write_zeroes->length)) <<
- (req->ns->blksize_shift - 9)) + 1;
+ nr_sector = (((sector_t)le16_to_cpu(write_zeroes->length + 1)) <<
+ (req->ns->blksize_shift - 9));
if (__blkdev_issue_zeroout(req->ns->bdev, sector, nr_sector,
GFP_KERNEL, &bio, 0))
--
2.7.4
On Mon, Apr 02, 2018 at 11:49:41AM -0300, Rodrigo R. Galvao wrote:
> When trying to issue write_zeroes command against TARGET with a 4K block
> size, it ends up hitting the following condition at __blkdev_issue_zeroout:
>
> if ((sector | nr_sects) & bs_mask)
> return -EINVAL;
>
> Causing the command to always fail.
> Considering we need to add 1 to get the correct block count, that addition
> needs to be performed in the native format, so we moved the +1 to within
> le16_to_cpu prior to converting to 512b.
>
> Signed-off-by: Rodrigo R. Galvao <[email protected]>
> ---
> drivers/nvme/target/io-cmd.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/nvme/target/io-cmd.c b/drivers/nvme/target/io-cmd.c
> index 28bbdff..5292bc3 100644
> --- a/drivers/nvme/target/io-cmd.c
> +++ b/drivers/nvme/target/io-cmd.c
> @@ -173,8 +173,8 @@ static void nvmet_execute_write_zeroes(struct nvmet_req *req)
>
> sector = le64_to_cpu(write_zeroes->slba) <<
> (req->ns->blksize_shift - 9);
> - nr_sector = (((sector_t)le16_to_cpu(write_zeroes->length)) <<
> - (req->ns->blksize_shift - 9)) + 1;
> + nr_sector = (((sector_t)le16_to_cpu(write_zeroes->length + 1)) <<
> + (req->ns->blksize_shift - 9));
I'm terribly sorry, but the +1 actually needs to be outside the
le16_to_cpu. The above will work on little-endian machines, but not big.
> I'm terribly sorry, but the +1 actually needs to be outside the
> le16_to_cpu. The above will work on little-endian machines, but not big.
No problem at all. Thank you for helping on this.
I'm going to send a V3 with this change.
--
Rodrigo R. Galvão
Intern - Linux Technology Center - IBM