Hi,
The following are a set of patches which removes semaphores from
scsi/smartpqi. These are part of a bigger effort to eliminate all
semaphores from the linux kernel.
Thanks,
Binoy
Binoy Jayan (2):
scsi: smartpqi: Replace semaphore sync_request_sem with mutex
scsi: smartpqi: Replace semaphore lun_reset_sem with mutex
drivers/scsi/smartpqi/smartpqi.h | 6 ++++--
drivers/scsi/smartpqi/smartpqi_init.c | 37 +++++++++--------------------------
2 files changed, 13 insertions(+), 30 deletions(-)
--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
Semaphores are going away in the future, so replace the semaphore
sync_request_sem with the a mutex lock. timeout_msecs is not used
for the lock sync_request_sem, so remove the timed locking too.
Signed-off-by: Binoy Jayan <[email protected]>
---
drivers/scsi/smartpqi/smartpqi.h | 4 +++-
drivers/scsi/smartpqi/smartpqi_init.c | 31 ++++++-------------------------
2 files changed, 9 insertions(+), 26 deletions(-)
diff --git a/drivers/scsi/smartpqi/smartpqi.h b/drivers/scsi/smartpqi/smartpqi.h
index 07b6444..b4559b1 100644
--- a/drivers/scsi/smartpqi/smartpqi.h
+++ b/drivers/scsi/smartpqi/smartpqi.h
@@ -19,6 +19,8 @@
#if !defined(_SMARTPQI_H)
#define _SMARTPQI_H
+#include <linux/mutex.h>
+
#pragma pack(1)
#define PQI_DEVICE_SIGNATURE "PQI DREG"
@@ -961,7 +963,7 @@ struct pqi_ctrl_info {
unsigned int num_heartbeats_requested;
struct timer_list heartbeat_timer;
- struct semaphore sync_request_sem;
+ struct mutex sync_request_mutex;
struct semaphore lun_reset_sem;
};
diff --git a/drivers/scsi/smartpqi/smartpqi_init.c b/drivers/scsi/smartpqi/smartpqi_init.c
index a535b26..4974f7e 100644
--- a/drivers/scsi/smartpqi/smartpqi_init.c
+++ b/drivers/scsi/smartpqi/smartpqi_init.c
@@ -3444,29 +3444,11 @@ static int pqi_submit_raid_request_synchronous(struct pqi_ctrl_info *ctrl_info,
unsigned long msecs_blocked;
size_t iu_length;
- /*
- * Note that specifying PQI_SYNC_FLAGS_INTERRUPTABLE and a timeout value
- * are mutually exclusive.
- */
-
- if (flags & PQI_SYNC_FLAGS_INTERRUPTABLE) {
- if (down_interruptible(&ctrl_info->sync_request_sem))
+ if (flags & PQI_SYNC_FLAGS_INTERRUPTABLE)
+ if (mutex_lock_interruptible(&ctrl_info->sync_request_mutex))
return -ERESTARTSYS;
- } else {
- if (timeout_msecs == NO_TIMEOUT) {
- down(&ctrl_info->sync_request_sem);
- } else {
- start_jiffies = jiffies;
- if (down_timeout(&ctrl_info->sync_request_sem,
- msecs_to_jiffies(timeout_msecs)))
- return -ETIMEDOUT;
- msecs_blocked =
- jiffies_to_msecs(jiffies - start_jiffies);
- if (msecs_blocked >= timeout_msecs)
- return -ETIMEDOUT;
- timeout_msecs -= msecs_blocked;
- }
- }
+ else
+ mutex_lock(&ctrl_info->sync_request_mutex);
io_request = pqi_alloc_io_request(ctrl_info);
@@ -3508,7 +3490,7 @@ static int pqi_submit_raid_request_synchronous(struct pqi_ctrl_info *ctrl_info,
pqi_free_io_request(io_request);
- up(&ctrl_info->sync_request_sem);
+ mutex_unlock(&ctrl_info->sync_request_mutex);
return rc;
}
@@ -5540,8 +5522,7 @@ static struct pqi_ctrl_info *pqi_alloc_ctrl_info(int numa_node)
INIT_DELAYED_WORK(&ctrl_info->rescan_work, pqi_rescan_worker);
INIT_DELAYED_WORK(&ctrl_info->update_time_work, pqi_update_time_worker);
- sema_init(&ctrl_info->sync_request_sem,
- PQI_RESERVED_IO_SLOTS_SYNCHRONOUS_REQUESTS);
+ mutex_init(&ctrl_info->sync_request_mutex);
sema_init(&ctrl_info->lun_reset_sem, PQI_RESERVED_IO_SLOTS_LUN_RESET);
ctrl_info->ctrl_id = atomic_inc_return(&pqi_controller_count) - 1;
--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
Semaphores are going away in the future, so replace the semaphore
lun_reset_sem with the a mutex lock.
Signed-off-by: Binoy Jayan <[email protected]>
---
drivers/scsi/smartpqi/smartpqi.h | 2 +-
drivers/scsi/smartpqi/smartpqi_init.c | 6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/scsi/smartpqi/smartpqi.h b/drivers/scsi/smartpqi/smartpqi.h
index b4559b1..3ecec27 100644
--- a/drivers/scsi/smartpqi/smartpqi.h
+++ b/drivers/scsi/smartpqi/smartpqi.h
@@ -964,7 +964,7 @@ struct pqi_ctrl_info {
struct timer_list heartbeat_timer;
struct mutex sync_request_mutex;
- struct semaphore lun_reset_sem;
+ struct mutex lun_reset_mutex;
};
enum pqi_ctrl_mode {
diff --git a/drivers/scsi/smartpqi/smartpqi_init.c b/drivers/scsi/smartpqi/smartpqi_init.c
index 4974f7e..1e5df85 100644
--- a/drivers/scsi/smartpqi/smartpqi_init.c
+++ b/drivers/scsi/smartpqi/smartpqi_init.c
@@ -4601,7 +4601,7 @@ static int pqi_lun_reset(struct pqi_ctrl_info *ctrl_info,
DECLARE_COMPLETION_ONSTACK(wait);
struct pqi_task_management_request *request;
- down(&ctrl_info->lun_reset_sem);
+ mutex_lock(&ctrl_info->lun_reset_mutex);
io_request = pqi_alloc_io_request(ctrl_info);
io_request->io_complete_callback = pqi_lun_reset_complete;
@@ -4627,7 +4627,7 @@ static int pqi_lun_reset(struct pqi_ctrl_info *ctrl_info,
rc = io_request->status;
pqi_free_io_request(io_request);
- up(&ctrl_info->lun_reset_sem);
+ mutex_unlock(&ctrl_info->lun_reset_mutex);
return rc;
}
@@ -5523,7 +5523,7 @@ static struct pqi_ctrl_info *pqi_alloc_ctrl_info(int numa_node)
INIT_DELAYED_WORK(&ctrl_info->update_time_work, pqi_update_time_worker);
mutex_init(&ctrl_info->sync_request_mutex);
- sema_init(&ctrl_info->lun_reset_sem, PQI_RESERVED_IO_SLOTS_LUN_RESET);
+ mutex_init(&ctrl_info->lun_reset_mutex);
ctrl_info->ctrl_id = atomic_inc_return(&pqi_controller_count) - 1;
ctrl_info->max_msix_vectors = PQI_MAX_MSIX_VECTORS;
--
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project
On Thursday, October 20, 2016 2:24:01 PM CEST Binoy Jayan wrote:
> Semaphores are going away in the future, so replace the semaphore
> sync_request_sem with the a mutex lock. timeout_msecs is not used
> for the lock sync_request_sem, so remove the timed locking too.
>
> Signed-off-by: Binoy Jayan <[email protected]>
The patch looks correct to me, but I think if you remove the support
for handling timeouts, you should update the prototype of
pqi_submit_raid_request_synchronous to no longer pass the timeout
argument in the first place.
Arnd
On Thursday, October 20, 2016 2:24:02 PM CEST Binoy Jayan wrote:
> Semaphores are going away in the future, so replace the semaphore
> lun_reset_sem with the a mutex lock.
>
> Signed-off-by: Binoy Jayan <[email protected]>
>
Reviewed-by: Arnd Bergmann <[email protected]>
On Thursday, October 20, 2016 2:24:01 PM CEST Binoy Jayan wrote:
> - sema_init(&ctrl_info->sync_request_sem,
> - PQI_RESERVED_IO_SLOTS_SYNCHRONOUS_REQUESTS);
> + mutex_init(&ctrl_info->sync_request_mutex);
>
Looking at this again, I see that PQI_RESERVED_IO_SLOTS_SYNCHRONOUS_REQUESTS
is '3', so this is in fact a counting semaphore rather than a mutex,
and the conversion is changing the behavior.
The patch can't go in unless you either show that it should be
a normal mutex rather than a counting semaphore, or you find a way
to keep the behavior the same.
Arnd
Hi Arnd
On 20 October 2016 at 14:36, Arnd Bergmann <[email protected]> wrote:
> On Thursday, October 20, 2016 2:24:01 PM CEST Binoy Jayan wrote:
>> Semaphores are going away in the future, so replace the semaphore
>> sync_request_sem with the a mutex lock. timeout_msecs is not used
>> for the lock sync_request_sem, so remove the timed locking too.
>>
>> Signed-off-by: Binoy Jayan <[email protected]>
>
> The patch looks correct to me, but I think if you remove the support
> for handling timeouts, you should update the prototype of
> pqi_submit_raid_request_synchronous to no longer pass the timeout
> argument in the first place.
But we still need "timeout_msecs" in a call to
pqi_submit_raid_request_synchronous_with_io_request()
drivers/scsi/smartpqi/smartpqi_init.c +3484
-Binoy
On 20 October 2016 at 14:40, Arnd Bergmann <[email protected]> wrote:
> On Thursday, October 20, 2016 2:24:01 PM CEST Binoy Jayan wrote:
>> - sema_init(&ctrl_info->sync_request_sem,
>> - PQI_RESERVED_IO_SLOTS_SYNCHRONOUS_REQUESTS);
>> + mutex_init(&ctrl_info->sync_request_mutex);
>>
>
> Looking at this again, I see that PQI_RESERVED_IO_SLOTS_SYNCHRONOUS_REQUESTS
> is '3', so this is in fact a counting semaphore rather than a mutex,
> and the conversion is changing the behavior.
>
> The patch can't go in unless you either show that it should be
> a normal mutex rather than a counting semaphore, or you find a way
> to keep the behavior the same.
This still holds true, will try changing this accordingly.
-Binoy
On Monday, October 24, 2016 3:34:27 PM CEST Binoy Jayan wrote:
> Hi Arnd
>
> On 20 October 2016 at 14:36, Arnd Bergmann <[email protected]> wrote:
> > On Thursday, October 20, 2016 2:24:01 PM CEST Binoy Jayan wrote:
> >> Semaphores are going away in the future, so replace the semaphore
> >> sync_request_sem with the a mutex lock. timeout_msecs is not used
> >> for the lock sync_request_sem, so remove the timed locking too.
> >>
> >> Signed-off-by: Binoy Jayan <[email protected]>
> >
> > The patch looks correct to me, but I think if you remove the support
> > for handling timeouts, you should update the prototype of
> > pqi_submit_raid_request_synchronous to no longer pass the timeout
> > argument in the first place.
>
> But we still need "timeout_msecs" in a call to
> pqi_submit_raid_request_synchronous_with_io_request()
>
> drivers/scsi/smartpqi/smartpqi_init.c +3484
Why? If it's always zero, we can remove that too.
Arnd