From: John Garry <john.g.garry@oracle.com>
To: Xingui Yang <yangxingui@huawei.com>,
jejb@linux.ibm.com, martin.petersen@oracle.com,
damien.lemoal@opensource.wdc.com, dlemoal@kernel.org
Cc: linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
linuxarm@huawei.com, prime.zeng@hisilicon.com,
kangfenglong@huawei.com
Subject: Re: [PATCH] ata: libata-scsi: Fix get identity data failed
Date: Thu, 4 May 2023 15:24:20 +0100 [thread overview]
Message-ID: <4ffe09f9-df67-f238-75cf-d0d16476e618@oracle.com> (raw)
In-Reply-To: <20230504131545.3409-1-yangxingui@huawei.com>
On 04/05/2023 14:15, Xingui Yang wrote:
> The function ata_get_identity() uses the helper ata_scsi_find_dev() to get
> the ata_device structure of a scsi device. However, when the ata device
> is managed by libsas, ata_scsi_find_dev() returns NULL, turning
> ata_get_identity() into a nop and always returns -ENOMSG.
>
> Fix this by replacing the pointer to the scsi_device struct argument with a
> pointer to the ata_device struct in ata_sas_scsi_ioctl() and
> ata_get_identity(). This pointer is provided by ata_scsi_ioctl() using
> ata_scsi_find_dev() in the case of a libata managed device and by
> sas_ioctl() using sas_to_ata_dev() in the case of a libsas managed ata
> device.
We made a similar change in commit 141f3d6256e5 ("ata: libata-sata: Fix
device queue depth control"). Is there some better way to do this to
make it all more scalable such that we don't need to keep replacing
calls to ata_scsi_find_dev()?
This was discussed earlier, along with queue depth issue (which does not
seem fixed, AFAIK):
https://lore.kernel.org/linux-scsi/13e5e5e5-7dc2-8f14-3dd2-43366343842d@huawei.com/
Thanks,
John
>
> Signed-off-by: Xingui Yang <yangxingui@huawei.com>
> ---
> drivers/ata/libata-scsi.c | 22 +++++++++++-----------
> drivers/scsi/libsas/sas_scsi_host.c | 3 ++-
> include/linux/libata.h | 2 +-
> 3 files changed, 14 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
> index 7bb12deab70c..68f2404e61d0 100644
> --- a/drivers/ata/libata-scsi.c
> +++ b/drivers/ata/libata-scsi.c
> @@ -327,8 +327,7 @@ EXPORT_SYMBOL_GPL(ata_scsi_unlock_native_capacity);
>
> /**
> * ata_get_identity - Handler for HDIO_GET_IDENTITY ioctl
> - * @ap: target port
> - * @sdev: SCSI device to get identify data for
> + * @dev: ATA device to get identify data for
> * @arg: User buffer area for identify data
> *
> * LOCKING:
> @@ -337,10 +336,8 @@ EXPORT_SYMBOL_GPL(ata_scsi_unlock_native_capacity);
> * RETURNS:
> * Zero on success, negative errno on error.
> */
> -static int ata_get_identity(struct ata_port *ap, struct scsi_device *sdev,
> - void __user *arg)
> +static int ata_get_identity(struct ata_device *dev, void __user *arg)
> {
> - struct ata_device *dev = ata_scsi_find_dev(ap, sdev);
> u16 __user *dst = arg;
> char buf[40];
>
> @@ -573,7 +570,7 @@ static bool ata_ioc32(struct ata_port *ap)
> * This handles both native and compat commands, so anything added
> * here must have a compatible argument, or check in_compat_syscall()
> */
> -int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *scsidev,
> +int ata_sas_scsi_ioctl(struct ata_port *ap, struct ata_device *dev,
> unsigned int cmd, void __user *arg)
> {
> unsigned long val;
> @@ -608,17 +605,17 @@ int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *scsidev,
> return rc;
>
> case HDIO_GET_IDENTITY:
> - return ata_get_identity(ap, scsidev, arg);
> + return ata_get_identity(dev, arg);
>
> case HDIO_DRIVE_CMD:
> if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
> return -EACCES;
> - return ata_cmd_ioctl(scsidev, arg);
> + return ata_cmd_ioctl(dev->sdev, arg);
>
> case HDIO_DRIVE_TASK:
> if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
> return -EACCES;
> - return ata_task_ioctl(scsidev, arg);
> + return ata_task_ioctl(dev->sdev, arg);
>
> default:
> rc = -ENOTTY;
> @@ -632,8 +629,11 @@ EXPORT_SYMBOL_GPL(ata_sas_scsi_ioctl);
> int ata_scsi_ioctl(struct scsi_device *scsidev, unsigned int cmd,
> void __user *arg)
> {
> - return ata_sas_scsi_ioctl(ata_shost_to_port(scsidev->host),
> - scsidev, cmd, arg);
> + struct ata_port *ap = ata_shost_to_port(scsidev->host);
> +
> + return ata_sas_scsi_ioctl(ap,
> + ata_scsi_find_dev(ap, scsidev),
> + cmd, arg);
> }
> EXPORT_SYMBOL_GPL(ata_scsi_ioctl);
>
> diff --git a/drivers/scsi/libsas/sas_scsi_host.c b/drivers/scsi/libsas/sas_scsi_host.c
> index a36fa1c128a8..c7a44ce7b2e2 100644
> --- a/drivers/scsi/libsas/sas_scsi_host.c
> +++ b/drivers/scsi/libsas/sas_scsi_host.c
> @@ -789,7 +789,8 @@ int sas_ioctl(struct scsi_device *sdev, unsigned int cmd, void __user *arg)
> struct domain_device *dev = sdev_to_domain_dev(sdev);
>
> if (dev_is_sata(dev))
> - return ata_sas_scsi_ioctl(dev->sata_dev.ap, sdev, cmd, arg);
> + return ata_sas_scsi_ioctl(dev->sata_dev.ap,
> + sas_to_ata_dev(dev), cmd, arg);
>
> return -EINVAL;
> }
> diff --git a/include/linux/libata.h b/include/linux/libata.h
> index 311cd93377c7..d5dd60530a24 100644
> --- a/include/linux/libata.h
> +++ b/include/linux/libata.h
> @@ -1085,7 +1085,7 @@ bool ata_scsi_dma_need_drain(struct request *rq);
> #else
> #define ata_scsi_dma_need_drain NULL
> #endif
> -extern int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *dev,
> +extern int ata_sas_scsi_ioctl(struct ata_port *ap, struct ata_device *dev,
> unsigned int cmd, void __user *arg);
> extern bool ata_link_online(struct ata_link *link);
> extern bool ata_link_offline(struct ata_link *link);
next prev parent reply other threads:[~2023-05-04 14:25 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-04 13:15 Xingui Yang
2023-05-04 14:24 ` John Garry [this message]
2023-05-05 2:17 ` yangxingui
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=4ffe09f9-df67-f238-75cf-d0d16476e618@oracle.com \
--to=john.g.garry@oracle.com \
--cc=damien.lemoal@opensource.wdc.com \
--cc=dlemoal@kernel.org \
--cc=jejb@linux.ibm.com \
--cc=kangfenglong@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=martin.petersen@oracle.com \
--cc=prime.zeng@hisilicon.com \
--cc=yangxingui@huawei.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®