[patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[Thomas Gleixner]

This is V2 of the series. V1 can be found here:

     https://lore.kernel.org/all/20250620130144.351492917@linutronix.de

When looking into supporting auxiliary clocks in the PTP ioctl, the
inpenetrable ptp_ioctl() letter soup bothered me enough to clean it up.

The code (~400 lines!) is really hard to follow due to a gazillion of
local variables, which are only used in certain case scopes, and a
mixture of gotos, breaks and direct error return paths.

Clean it up by splitting out the IOCTL functionality into seperate
functions, which contain only the required local variables and are trivial
to follow. Complete the cleanup by converting the code to lock guards and
get rid of all gotos.

That reduces the code size by 48 lines and also the binary text size is
80 bytes smaller than the current maze.

The series is split up into one patch per IOCTL command group for easy
review.

Changes vs. V1:

  - Picked up Reviewed tags as appropriate

  - Dropped the pointless memset()s in GETFUNC/SETFUNC - Paolo

  - Dropped the __free() in ptp_open/read() - Jakub

Applies against v6.16-rc1 and also cleanly against next. It's also
available from git:

     git://git.kernel.org/pub/scm/linux/kernel/git/tglx/devel.git timers/ptp/driver

Thanks,

	tglx
---
 ptp_chardev.c |  734 ++++++++++++++++++++++++++--------------------------------
 1 file changed, 341 insertions(+), 393 deletions(-)

[patch V2 01/13] ptp: Split out PTP_CLOCK_GETCAPS ioctl code

[Thomas Gleixner]

ptp_ioctl() is an inpenetrable letter soup with a gazillion of case (scope)
specific variables defined at the top of the function and pointless breaks
and gotos.

Start cleaning it up by splitting out the PTP_CLOCK_GETCAPS ioctl code into
a helper function. Use a argument pointer with a single sparse compliant
type cast instead of proliferating the type cast all over the place.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |   41 +++++++++++++++++++++++------------------
 1 file changed, 23 insertions(+), 18 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -157,6 +157,26 @@ int ptp_release(struct posix_clock_conte
 	return 0;
 }
 
+static long ptp_clock_getcaps(struct ptp_clock *ptp, void __user *arg)
+{
+	struct ptp_clock_caps caps = {
+		.max_adj		= ptp->info->max_adj,
+		.n_alarm		= ptp->info->n_alarm,
+		.n_ext_ts		= ptp->info->n_ext_ts,
+		.n_per_out		= ptp->info->n_per_out,
+		.pps			= ptp->info->pps,
+		.n_pins			= ptp->info->n_pins,
+		.cross_timestamping	= ptp->info->getcrosststamp != NULL,
+		.adjust_phase		= ptp->info->adjphase != NULL &&
+					  ptp->info->getmaxphase != NULL,
+	};
+
+	if (caps.adjust_phase)
+		caps.max_phase_adj = ptp->info->getmaxphase(ptp->info);
+
+	return copy_to_user(arg, &caps, sizeof(caps)) ? -EFAULT : 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
@@ -171,37 +191,22 @@ long ptp_ioctl(struct posix_clock_contex
 	struct timestamp_event_queue *tsevq;
 	struct ptp_system_timestamp sts;
 	struct ptp_clock_request req;
-	struct ptp_clock_caps caps;
 	struct ptp_clock_time *pct;
 	struct ptp_pin_desc pd;
 	struct timespec64 ts;
 	int enable, err = 0;
+	void __user *argptr;
 
 	if (in_compat_syscall() && cmd != PTP_ENABLE_PPS && cmd != PTP_ENABLE_PPS2)
 		arg = (unsigned long)compat_ptr(arg);
+	argptr = (void __force __user *)arg;
 
 	tsevq = pccontext->private_clkdata;
 
 	switch (cmd) {
-
 	case PTP_CLOCK_GETCAPS:
 	case PTP_CLOCK_GETCAPS2:
-		memset(&caps, 0, sizeof(caps));
-
-		caps.max_adj = ptp->info->max_adj;
-		caps.n_alarm = ptp->info->n_alarm;
-		caps.n_ext_ts = ptp->info->n_ext_ts;
-		caps.n_per_out = ptp->info->n_per_out;
-		caps.pps = ptp->info->pps;
-		caps.n_pins = ptp->info->n_pins;
-		caps.cross_timestamping = ptp->info->getcrosststamp != NULL;
-		caps.adjust_phase = ptp->info->adjphase != NULL &&
-				    ptp->info->getmaxphase != NULL;
-		if (caps.adjust_phase)
-			caps.max_phase_adj = ptp->info->getmaxphase(ptp->info);
-		if (copy_to_user((void __user *)arg, &caps, sizeof(caps)))
-			err = -EFAULT;
-		break;
+		return ptp_clock_getcaps(ptp, argptr);
 
 	case PTP_EXTTS_REQUEST:
 	case PTP_EXTTS_REQUEST2:

[patch V2 02/13] ptp: Split out PTP_EXTTS_REQUEST ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_EXTTS_REQUEST
ioctl code into a helper function. Convert to a lock guard while at it.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |  105 +++++++++++++++++++++-------------------------
 1 file changed, 50 insertions(+), 55 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -177,12 +177,57 @@ static long ptp_clock_getcaps(struct ptp
 	return copy_to_user(arg, &caps, sizeof(caps)) ? -EFAULT : 0;
 }
 
+static long ptp_extts_request(struct ptp_clock *ptp, unsigned int cmd, void __user *arg)
+{
+	struct ptp_clock_request req = { .type = PTP_CLK_REQ_EXTTS };
+	struct ptp_clock_info *ops = ptp->info;
+	unsigned int supported_extts_flags;
+
+	if (copy_from_user(&req.extts, arg, sizeof(req.extts)))
+		return -EFAULT;
+
+	if (cmd == PTP_EXTTS_REQUEST2) {
+		/* Tell the drivers to check the flags carefully. */
+		req.extts.flags |= PTP_STRICT_FLAGS;
+		/* Make sure no reserved bit is set. */
+		if ((req.extts.flags & ~PTP_EXTTS_VALID_FLAGS) ||
+		    req.extts.rsv[0] || req.extts.rsv[1])
+			return -EINVAL;
+
+		/* Ensure one of the rising/falling edge bits is set. */
+		if ((req.extts.flags & PTP_ENABLE_FEATURE) &&
+		    (req.extts.flags & PTP_EXTTS_EDGES) == 0)
+			return -EINVAL;
+	} else {
+		req.extts.flags &= PTP_EXTTS_V1_VALID_FLAGS;
+		memset(req.extts.rsv, 0, sizeof(req.extts.rsv));
+	}
+
+	if (req.extts.index >= ops->n_ext_ts)
+		return -EINVAL;
+
+	supported_extts_flags = ptp->info->supported_extts_flags;
+	/* The PTP_ENABLE_FEATURE flag is always supported. */
+	supported_extts_flags |= PTP_ENABLE_FEATURE;
+	/* If the driver does not support strictly checking flags, the
+	 * PTP_RISING_EDGE and PTP_FALLING_EDGE flags are merely hints
+	 * which are not enforced.
+	 */
+	if (!(supported_extts_flags & PTP_STRICT_FLAGS))
+		supported_extts_flags |= PTP_EXTTS_EDGES;
+	/* Reject unsupported flags */
+	if (req.extts.flags & ~supported_extts_flags)
+		return -EOPNOTSUPP;
+
+	scoped_cond_guard(mutex_intr, return -ERESTARTSYS, &ptp->pincfg_mux)
+		return ops->enable(ops, &req, req.extts.flags & PTP_ENABLE_FEATURE ? 1 : 0);
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
-	unsigned int i, pin_index, supported_extts_flags;
 	struct ptp_sys_offset_extended *extoff = NULL;
 	struct ptp_sys_offset_precise precise_offset;
 	struct system_device_crosststamp xtstamp;
@@ -192,6 +237,7 @@ long ptp_ioctl(struct posix_clock_contex
 	struct ptp_system_timestamp sts;
 	struct ptp_clock_request req;
 	struct ptp_clock_time *pct;
+	unsigned int i, pin_index;
 	struct ptp_pin_desc pd;
 	struct timespec64 ts;
 	int enable, err = 0;
@@ -210,60 +256,9 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_EXTTS_REQUEST:
 	case PTP_EXTTS_REQUEST2:
-		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0) {
-			err = -EACCES;
-			break;
-		}
-		memset(&req, 0, sizeof(req));
-
-		if (copy_from_user(&req.extts, (void __user *)arg,
-				   sizeof(req.extts))) {
-			err = -EFAULT;
-			break;
-		}
-		if (cmd == PTP_EXTTS_REQUEST2) {
-			/* Tell the drivers to check the flags carefully. */
-			req.extts.flags |= PTP_STRICT_FLAGS;
-			/* Make sure no reserved bit is set. */
-			if ((req.extts.flags & ~PTP_EXTTS_VALID_FLAGS) ||
-			    req.extts.rsv[0] || req.extts.rsv[1]) {
-				err = -EINVAL;
-				break;
-			}
-			/* Ensure one of the rising/falling edge bits is set. */
-			if ((req.extts.flags & PTP_ENABLE_FEATURE) &&
-			    (req.extts.flags & PTP_EXTTS_EDGES) == 0) {
-				err = -EINVAL;
-				break;
-			}
-		} else if (cmd == PTP_EXTTS_REQUEST) {
-			req.extts.flags &= PTP_EXTTS_V1_VALID_FLAGS;
-			req.extts.rsv[0] = 0;
-			req.extts.rsv[1] = 0;
-		}
-		if (req.extts.index >= ops->n_ext_ts) {
-			err = -EINVAL;
-			break;
-		}
-		supported_extts_flags = ptp->info->supported_extts_flags;
-		/* The PTP_ENABLE_FEATURE flag is always supported. */
-		supported_extts_flags |= PTP_ENABLE_FEATURE;
-		/* If the driver does not support strictly checking flags, the
-		 * PTP_RISING_EDGE and PTP_FALLING_EDGE flags are merely
-		 * hints which are not enforced.
-		 */
-		if (!(supported_extts_flags & PTP_STRICT_FLAGS))
-			supported_extts_flags |= PTP_EXTTS_EDGES;
-		/* Reject unsupported flags */
-		if (req.extts.flags & ~supported_extts_flags)
-			return -EOPNOTSUPP;
-		req.type = PTP_CLK_REQ_EXTTS;
-		enable = req.extts.flags & PTP_ENABLE_FEATURE ? 1 : 0;
-		if (mutex_lock_interruptible(&ptp->pincfg_mux))
-			return -ERESTARTSYS;
-		err = ops->enable(ops, &req, enable);
-		mutex_unlock(&ptp->pincfg_mux);
-		break;
+		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0)
+			return -EACCES;
+		return ptp_extts_request(ptp, cmd, argptr);
 
 	case PTP_PEROUT_REQUEST:
 	case PTP_PEROUT_REQUEST2:

[patch V2 03/13] ptp: Split out PTP_PEROUT_REQUEST ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_PEROUT_REQUEST
ioctl code into a helper function. Convert to a lock guard while at it.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |  129 ++++++++++++++++++++--------------------------
 1 file changed, 58 insertions(+), 71 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -223,6 +223,61 @@ static long ptp_extts_request(struct ptp
 		return ops->enable(ops, &req, req.extts.flags & PTP_ENABLE_FEATURE ? 1 : 0);
 }
 
+static long ptp_perout_request(struct ptp_clock *ptp, unsigned int cmd, void __user *arg)
+{
+	struct ptp_clock_request req = { .type = PTP_CLK_REQ_PEROUT };
+	struct ptp_perout_request *perout = &req.perout;
+	struct ptp_clock_info *ops = ptp->info;
+
+	if (copy_from_user(perout, arg, sizeof(*perout)))
+		return -EFAULT;
+
+	if (cmd == PTP_PEROUT_REQUEST2) {
+		if (perout->flags & ~PTP_PEROUT_VALID_FLAGS)
+			return -EINVAL;
+
+		/*
+		 * The "on" field has undefined meaning if
+		 * PTP_PEROUT_DUTY_CYCLE isn't set, we must still treat it
+		 * as reserved, which must be set to zero.
+		 */
+		if (!(perout->flags & PTP_PEROUT_DUTY_CYCLE) &&
+		    !mem_is_zero(perout->rsv, sizeof(perout->rsv)))
+			return -EINVAL;
+
+		if (perout->flags & PTP_PEROUT_DUTY_CYCLE) {
+			/* The duty cycle must be subunitary. */
+			if (perout->on.sec > perout->period.sec ||
+			    (perout->on.sec == perout->period.sec &&
+			     perout->on.nsec > perout->period.nsec))
+				return -ERANGE;
+		}
+
+		if (perout->flags & PTP_PEROUT_PHASE) {
+			/*
+			 * The phase should be specified modulo the period,
+			 * therefore anything equal or larger than 1 period
+			 * is invalid.
+			 */
+			if (perout->phase.sec > perout->period.sec ||
+			    (perout->phase.sec == perout->period.sec &&
+			     perout->phase.nsec >= perout->period.nsec))
+				return -ERANGE;
+		}
+	} else {
+		perout->flags &= PTP_PEROUT_V1_VALID_FLAGS;
+		memset(perout->rsv, 0, sizeof(perout->rsv));
+	}
+
+	if (perout->index >= ops->n_per_out)
+		return -EINVAL;
+	if (perout->flags & ~ops->supported_perout_flags)
+		return -EOPNOTSUPP;
+
+	scoped_cond_guard(mutex_intr, return -ERESTARTSYS, &ptp->pincfg_mux)
+		return ops->enable(ops, &req, perout->period.sec || perout->period.nsec);
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
@@ -262,77 +317,9 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_PEROUT_REQUEST:
 	case PTP_PEROUT_REQUEST2:
-		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0) {
-			err = -EACCES;
-			break;
-		}
-		memset(&req, 0, sizeof(req));
-
-		if (copy_from_user(&req.perout, (void __user *)arg,
-				   sizeof(req.perout))) {
-			err = -EFAULT;
-			break;
-		}
-		if (cmd == PTP_PEROUT_REQUEST2) {
-			struct ptp_perout_request *perout = &req.perout;
-
-			if (perout->flags & ~PTP_PEROUT_VALID_FLAGS) {
-				err = -EINVAL;
-				break;
-			}
-			/*
-			 * The "on" field has undefined meaning if
-			 * PTP_PEROUT_DUTY_CYCLE isn't set, we must still treat
-			 * it as reserved, which must be set to zero.
-			 */
-			if (!(perout->flags & PTP_PEROUT_DUTY_CYCLE) &&
-			    (perout->rsv[0] || perout->rsv[1] ||
-			     perout->rsv[2] || perout->rsv[3])) {
-				err = -EINVAL;
-				break;
-			}
-			if (perout->flags & PTP_PEROUT_DUTY_CYCLE) {
-				/* The duty cycle must be subunitary. */
-				if (perout->on.sec > perout->period.sec ||
-				    (perout->on.sec == perout->period.sec &&
-				     perout->on.nsec > perout->period.nsec)) {
-					err = -ERANGE;
-					break;
-				}
-			}
-			if (perout->flags & PTP_PEROUT_PHASE) {
-				/*
-				 * The phase should be specified modulo the
-				 * period, therefore anything equal or larger
-				 * than 1 period is invalid.
-				 */
-				if (perout->phase.sec > perout->period.sec ||
-				    (perout->phase.sec == perout->period.sec &&
-				     perout->phase.nsec >= perout->period.nsec)) {
-					err = -ERANGE;
-					break;
-				}
-			}
-		} else if (cmd == PTP_PEROUT_REQUEST) {
-			req.perout.flags &= PTP_PEROUT_V1_VALID_FLAGS;
-			req.perout.rsv[0] = 0;
-			req.perout.rsv[1] = 0;
-			req.perout.rsv[2] = 0;
-			req.perout.rsv[3] = 0;
-		}
-		if (req.perout.index >= ops->n_per_out) {
-			err = -EINVAL;
-			break;
-		}
-		if (req.perout.flags & ~ptp->info->supported_perout_flags)
-			return -EOPNOTSUPP;
-		req.type = PTP_CLK_REQ_PEROUT;
-		enable = req.perout.period.sec || req.perout.period.nsec;
-		if (mutex_lock_interruptible(&ptp->pincfg_mux))
-			return -ERESTARTSYS;
-		err = ops->enable(ops, &req, enable);
-		mutex_unlock(&ptp->pincfg_mux);
-		break;
+		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0)
+			return -EACCES;
+		return ptp_perout_request(ptp, cmd, argptr);
 
 	case PTP_ENABLE_PPS:
 	case PTP_ENABLE_PPS2:

[patch V2 04/13] ptp: Split out PTP_ENABLE_PPS ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_ENABLE_PPS
ioctl code into a helper function. Convert to a lock guard while at it.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |   33 ++++++++++++++++-----------------
 1 file changed, 16 insertions(+), 17 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -278,6 +278,18 @@ static long ptp_perout_request(struct pt
 		return ops->enable(ops, &req, perout->period.sec || perout->period.nsec);
 }
 
+static long ptp_enable_pps(struct ptp_clock *ptp, bool enable)
+{
+	struct ptp_clock_request req = { .type = PTP_CLK_REQ_PPS };
+	struct ptp_clock_info *ops = ptp->info;
+
+	if (!capable(CAP_SYS_TIME))
+		return -EPERM;
+
+	scoped_cond_guard(mutex_intr, return -ERESTARTSYS, &ptp->pincfg_mux)
+		return ops->enable(ops, &req, enable);
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
@@ -290,13 +302,12 @@ long ptp_ioctl(struct posix_clock_contex
 	struct ptp_sys_offset *sysoff = NULL;
 	struct timestamp_event_queue *tsevq;
 	struct ptp_system_timestamp sts;
-	struct ptp_clock_request req;
 	struct ptp_clock_time *pct;
 	unsigned int i, pin_index;
 	struct ptp_pin_desc pd;
 	struct timespec64 ts;
-	int enable, err = 0;
 	void __user *argptr;
+	int err = 0;
 
 	if (in_compat_syscall() && cmd != PTP_ENABLE_PPS && cmd != PTP_ENABLE_PPS2)
 		arg = (unsigned long)compat_ptr(arg);
@@ -323,21 +334,9 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_ENABLE_PPS:
 	case PTP_ENABLE_PPS2:
-		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0) {
-			err = -EACCES;
-			break;
-		}
-		memset(&req, 0, sizeof(req));
-
-		if (!capable(CAP_SYS_TIME))
-			return -EPERM;
-		req.type = PTP_CLK_REQ_PPS;
-		enable = arg ? 1 : 0;
-		if (mutex_lock_interruptible(&ptp->pincfg_mux))
-			return -ERESTARTSYS;
-		err = ops->enable(ops, &req, enable);
-		mutex_unlock(&ptp->pincfg_mux);
-		break;
+		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0)
+			return -EACCES;
+		return ptp_enable_pps(ptp, !!arg);
 
 	case PTP_SYS_OFFSET_PRECISE:
 	case PTP_SYS_OFFSET_PRECISE2:

[patch V2 05/13] ptp: Split out PTP_SYS_OFFSET_PRECISE ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_SYS_OFFSET_PRECISE
ioctl code into a helper function.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |   53 +++++++++++++++++++++++++---------------------
 1 file changed, 29 insertions(+), 24 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -291,14 +291,40 @@ static long ptp_enable_pps(struct ptp_cl
 		return ops->enable(ops, &req, enable);
 }
 
+static long ptp_sys_offset_precise(struct ptp_clock *ptp, void __user *arg)
+{
+	struct ptp_sys_offset_precise precise_offset;
+	struct system_device_crosststamp xtstamp;
+	struct timespec64 ts;
+	int err;
+
+	if (!ptp->info->getcrosststamp)
+		return -EOPNOTSUPP;
+
+	err = ptp->info->getcrosststamp(ptp->info, &xtstamp);
+	if (err)
+		return err;
+
+	memset(&precise_offset, 0, sizeof(precise_offset));
+	ts = ktime_to_timespec64(xtstamp.device);
+	precise_offset.device.sec = ts.tv_sec;
+	precise_offset.device.nsec = ts.tv_nsec;
+	ts = ktime_to_timespec64(xtstamp.sys_realtime);
+	precise_offset.sys_realtime.sec = ts.tv_sec;
+	precise_offset.sys_realtime.nsec = ts.tv_nsec;
+	ts = ktime_to_timespec64(xtstamp.sys_monoraw);
+	precise_offset.sys_monoraw.sec = ts.tv_sec;
+	precise_offset.sys_monoraw.nsec = ts.tv_nsec;
+
+	return copy_to_user(arg, &precise_offset, sizeof(precise_offset)) ? -EFAULT : 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
 	struct ptp_sys_offset_extended *extoff = NULL;
-	struct ptp_sys_offset_precise precise_offset;
-	struct system_device_crosststamp xtstamp;
 	struct ptp_clock_info *ops = ptp->info;
 	struct ptp_sys_offset *sysoff = NULL;
 	struct timestamp_event_queue *tsevq;
@@ -341,28 +367,7 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_SYS_OFFSET_PRECISE:
 	case PTP_SYS_OFFSET_PRECISE2:
-		if (!ptp->info->getcrosststamp) {
-			err = -EOPNOTSUPP;
-			break;
-		}
-		err = ptp->info->getcrosststamp(ptp->info, &xtstamp);
-		if (err)
-			break;
-
-		memset(&precise_offset, 0, sizeof(precise_offset));
-		ts = ktime_to_timespec64(xtstamp.device);
-		precise_offset.device.sec = ts.tv_sec;
-		precise_offset.device.nsec = ts.tv_nsec;
-		ts = ktime_to_timespec64(xtstamp.sys_realtime);
-		precise_offset.sys_realtime.sec = ts.tv_sec;
-		precise_offset.sys_realtime.nsec = ts.tv_nsec;
-		ts = ktime_to_timespec64(xtstamp.sys_monoraw);
-		precise_offset.sys_monoraw.sec = ts.tv_sec;
-		precise_offset.sys_monoraw.nsec = ts.tv_nsec;
-		if (copy_to_user((void __user *)arg, &precise_offset,
-				 sizeof(precise_offset)))
-			err = -EFAULT;
-		break;
+		return ptp_sys_offset_precise(ptp, argptr);
 
 	case PTP_SYS_OFFSET_EXTENDED:
 	case PTP_SYS_OFFSET_EXTENDED2:

[patch V2 06/13] ptp: Split out PTP_SYS_OFFSET_EXTENDED ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the
PTP_SYS_OFFSET_EXTENDED ioctl code into a helper function.

Convert it to __free() to avoid gotos.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |   75 +++++++++++++++++++++++-----------------------
 1 file changed, 39 insertions(+), 36 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -319,16 +319,52 @@ static long ptp_sys_offset_precise(struc
 	return copy_to_user(arg, &precise_offset, sizeof(precise_offset)) ? -EFAULT : 0;
 }
 
+static long ptp_sys_offset_extended(struct ptp_clock *ptp, void __user *arg)
+{
+	struct ptp_sys_offset_extended *extoff __free(kfree) = NULL;
+	struct ptp_system_timestamp sts;
+
+	if (!ptp->info->gettimex64)
+		return -EOPNOTSUPP;
+
+	extoff = memdup_user(arg, sizeof(*extoff));
+	if (IS_ERR(extoff))
+		return PTR_ERR(extoff);
+
+	if (extoff->n_samples > PTP_MAX_SAMPLES ||
+	    extoff->rsv[0] || extoff->rsv[1] ||
+	    (extoff->clockid != CLOCK_REALTIME &&
+	     extoff->clockid != CLOCK_MONOTONIC &&
+	     extoff->clockid != CLOCK_MONOTONIC_RAW))
+		return -EINVAL;
+
+	sts.clockid = extoff->clockid;
+	for (unsigned int i = 0; i < extoff->n_samples; i++) {
+		struct timespec64 ts;
+		int err;
+
+		err = ptp->info->gettimex64(ptp->info, &ts, &sts);
+		if (err)
+			return err;
+		extoff->ts[i][0].sec = sts.pre_ts.tv_sec;
+		extoff->ts[i][0].nsec = sts.pre_ts.tv_nsec;
+		extoff->ts[i][1].sec = ts.tv_sec;
+		extoff->ts[i][1].nsec = ts.tv_nsec;
+		extoff->ts[i][2].sec = sts.post_ts.tv_sec;
+		extoff->ts[i][2].nsec = sts.post_ts.tv_nsec;
+	}
+
+	return copy_to_user(arg, extoff, sizeof(*extoff)) ? -EFAULT : 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
-	struct ptp_sys_offset_extended *extoff = NULL;
 	struct ptp_clock_info *ops = ptp->info;
 	struct ptp_sys_offset *sysoff = NULL;
 	struct timestamp_event_queue *tsevq;
-	struct ptp_system_timestamp sts;
 	struct ptp_clock_time *pct;
 	unsigned int i, pin_index;
 	struct ptp_pin_desc pd;
@@ -371,39 +407,7 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_SYS_OFFSET_EXTENDED:
 	case PTP_SYS_OFFSET_EXTENDED2:
-		if (!ptp->info->gettimex64) {
-			err = -EOPNOTSUPP;
-			break;
-		}
-		extoff = memdup_user((void __user *)arg, sizeof(*extoff));
-		if (IS_ERR(extoff)) {
-			err = PTR_ERR(extoff);
-			extoff = NULL;
-			break;
-		}
-		if (extoff->n_samples > PTP_MAX_SAMPLES ||
-		    extoff->rsv[0] || extoff->rsv[1] ||
-		    (extoff->clockid != CLOCK_REALTIME &&
-		     extoff->clockid != CLOCK_MONOTONIC &&
-		     extoff->clockid != CLOCK_MONOTONIC_RAW)) {
-			err = -EINVAL;
-			break;
-		}
-		sts.clockid = extoff->clockid;
-		for (i = 0; i < extoff->n_samples; i++) {
-			err = ptp->info->gettimex64(ptp->info, &ts, &sts);
-			if (err)
-				goto out;
-			extoff->ts[i][0].sec = sts.pre_ts.tv_sec;
-			extoff->ts[i][0].nsec = sts.pre_ts.tv_nsec;
-			extoff->ts[i][1].sec = ts.tv_sec;
-			extoff->ts[i][1].nsec = ts.tv_nsec;
-			extoff->ts[i][2].sec = sts.post_ts.tv_sec;
-			extoff->ts[i][2].nsec = sts.post_ts.tv_nsec;
-		}
-		if (copy_to_user((void __user *)arg, extoff, sizeof(*extoff)))
-			err = -EFAULT;
-		break;
+		return ptp_sys_offset_extended(ptp, argptr);
 
 	case PTP_SYS_OFFSET:
 	case PTP_SYS_OFFSET2:
@@ -528,7 +532,6 @@ long ptp_ioctl(struct posix_clock_contex
 	}
 
 out:
-	kfree(extoff);
 	kfree(sysoff);
 	return err;
 }

[patch V2 07/13] ptp: Split out PTP_SYS_OFFSET ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_SYS_OFFSET ioctl
code into a helper function.

Convert it to __free() to avoid gotos.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 drivers/ptp/ptp_chardev.c |   78 +++++++++++++++++++++++-----------------------
 1 file changed, 40 insertions(+), 38 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -357,18 +357,54 @@ static long ptp_sys_offset_extended(stru
 	return copy_to_user(arg, extoff, sizeof(*extoff)) ? -EFAULT : 0;
 }
 
+static long ptp_sys_offset(struct ptp_clock *ptp, void __user *arg)
+{
+	struct ptp_sys_offset *sysoff __free(kfree) = NULL;
+	struct ptp_clock_time *pct;
+	struct timespec64 ts;
+
+	sysoff = memdup_user(arg, sizeof(*sysoff));
+	if (IS_ERR(sysoff))
+		return PTR_ERR(sysoff);
+
+	if (sysoff->n_samples > PTP_MAX_SAMPLES)
+		return -EINVAL;
+
+	pct = &sysoff->ts[0];
+	for (unsigned int i = 0; i < sysoff->n_samples; i++) {
+		struct ptp_clock_info *ops = ptp->info;
+		int err;
+
+		ktime_get_real_ts64(&ts);
+		pct->sec = ts.tv_sec;
+		pct->nsec = ts.tv_nsec;
+		pct++;
+		if (ops->gettimex64)
+			err = ops->gettimex64(ops, &ts, NULL);
+		else
+			err = ops->gettime64(ops, &ts);
+		if (err)
+			return err;
+		pct->sec = ts.tv_sec;
+		pct->nsec = ts.tv_nsec;
+		pct++;
+	}
+	ktime_get_real_ts64(&ts);
+	pct->sec = ts.tv_sec;
+	pct->nsec = ts.tv_nsec;
+
+	return copy_to_user(arg, sysoff, sizeof(*sysoff)) ? -EFAULT : 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
 	struct ptp_clock_info *ops = ptp->info;
-	struct ptp_sys_offset *sysoff = NULL;
 	struct timestamp_event_queue *tsevq;
-	struct ptp_clock_time *pct;
 	unsigned int i, pin_index;
 	struct ptp_pin_desc pd;
-	struct timespec64 ts;
 	void __user *argptr;
 	int err = 0;
 
@@ -411,38 +447,7 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_SYS_OFFSET:
 	case PTP_SYS_OFFSET2:
-		sysoff = memdup_user((void __user *)arg, sizeof(*sysoff));
-		if (IS_ERR(sysoff)) {
-			err = PTR_ERR(sysoff);
-			sysoff = NULL;
-			break;
-		}
-		if (sysoff->n_samples > PTP_MAX_SAMPLES) {
-			err = -EINVAL;
-			break;
-		}
-		pct = &sysoff->ts[0];
-		for (i = 0; i < sysoff->n_samples; i++) {
-			ktime_get_real_ts64(&ts);
-			pct->sec = ts.tv_sec;
-			pct->nsec = ts.tv_nsec;
-			pct++;
-			if (ops->gettimex64)
-				err = ops->gettimex64(ops, &ts, NULL);
-			else
-				err = ops->gettime64(ops, &ts);
-			if (err)
-				goto out;
-			pct->sec = ts.tv_sec;
-			pct->nsec = ts.tv_nsec;
-			pct++;
-		}
-		ktime_get_real_ts64(&ts);
-		pct->sec = ts.tv_sec;
-		pct->nsec = ts.tv_nsec;
-		if (copy_to_user((void __user *)arg, sysoff, sizeof(*sysoff)))
-			err = -EFAULT;
-		break;
+		return ptp_sys_offset(ptp, argptr);
 
 	case PTP_PIN_GETFUNC:
 	case PTP_PIN_GETFUNC2:
@@ -530,9 +535,6 @@ long ptp_ioctl(struct posix_clock_contex
 		err = -ENOTTY;
 		break;
 	}
-
-out:
-	kfree(sysoff);
 	return err;
 }
 

[patch V2 08/13] ptp: Split out PTP_PIN_GETFUNC ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_PIN_GETFUNC ioctl
code into a helper function. Convert to lock guard while at it and remove
the pointless memset of the pd::rsv because nothing uses it.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
---
V2: Remove the pointless memset() - Paolo
---
 drivers/ptp/ptp_chardev.c |   50 +++++++++++++++++++---------------------------
 1 file changed, 21 insertions(+), 29 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -395,6 +395,26 @@ static long ptp_sys_offset(struct ptp_cl
 	return copy_to_user(arg, sysoff, sizeof(*sysoff)) ? -EFAULT : 0;
 }
 
+static long ptp_pin_getfunc(struct ptp_clock *ptp, unsigned int cmd, void __user *arg)
+{
+	struct ptp_clock_info *ops = ptp->info;
+	struct ptp_pin_desc pd;
+
+	if (copy_from_user(&pd, arg, sizeof(pd)))
+		return -EFAULT;
+
+	if (cmd == PTP_PIN_GETFUNC2 && !mem_is_zero(pd.rsv, sizeof(pd.rsv)))
+		return -EINVAL;
+
+	if (pd.index >= ops->n_pins)
+		return -EINVAL;
+
+	scoped_cond_guard(mutex_intr, return -ERESTARTSYS, &ptp->pincfg_mux)
+		pd = ops->pin_config[array_index_nospec(pd.index, ops->n_pins)];
+
+	return copy_to_user(arg, &pd, sizeof(pd)) ? -EFAULT : 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
@@ -450,35 +470,7 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_PIN_GETFUNC:
 	case PTP_PIN_GETFUNC2:
-		if (copy_from_user(&pd, (void __user *)arg, sizeof(pd))) {
-			err = -EFAULT;
-			break;
-		}
-		if ((pd.rsv[0] || pd.rsv[1] || pd.rsv[2]
-				|| pd.rsv[3] || pd.rsv[4])
-			&& cmd == PTP_PIN_GETFUNC2) {
-			err = -EINVAL;
-			break;
-		} else if (cmd == PTP_PIN_GETFUNC) {
-			pd.rsv[0] = 0;
-			pd.rsv[1] = 0;
-			pd.rsv[2] = 0;
-			pd.rsv[3] = 0;
-			pd.rsv[4] = 0;
-		}
-		pin_index = pd.index;
-		if (pin_index >= ops->n_pins) {
-			err = -EINVAL;
-			break;
-		}
-		pin_index = array_index_nospec(pin_index, ops->n_pins);
-		if (mutex_lock_interruptible(&ptp->pincfg_mux))
-			return -ERESTARTSYS;
-		pd = ops->pin_config[pin_index];
-		mutex_unlock(&ptp->pincfg_mux);
-		if (!err && copy_to_user((void __user *)arg, &pd, sizeof(pd)))
-			err = -EFAULT;
-		break;
+		return ptp_pin_getfunc(ptp, cmd, argptr);
 
 	case PTP_PIN_SETFUNC:
 	case PTP_PIN_SETFUNC2:

[patch V2 09/13] ptp: Split out PTP_PIN_SETFUNC ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_PIN_SETFUNC ioctl
code into a helper function. Convert to lock guard while at it and remove
the pointless memset of the pd::rsv because nothing uses it.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
---
V2: Remove the pointless memset() - Paolo
---
 drivers/ptp/ptp_chardev.c |   58 +++++++++++++++++++---------------------------
 1 file changed, 24 insertions(+), 34 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -415,16 +415,34 @@ static long ptp_pin_getfunc(struct ptp_c
 	return copy_to_user(arg, &pd, sizeof(pd)) ? -EFAULT : 0;
 }
 
+static long ptp_pin_setfunc(struct ptp_clock *ptp, unsigned int cmd, void __user *arg)
+{
+	struct ptp_clock_info *ops = ptp->info;
+	struct ptp_pin_desc pd;
+	unsigned int pin_index;
+
+	if (copy_from_user(&pd, arg, sizeof(pd)))
+		return -EFAULT;
+
+	if (cmd == PTP_PIN_SETFUNC2 && !mem_is_zero(pd.rsv, sizeof(pd.rsv)))
+		return -EINVAL;
+
+	if (pd.index >= ops->n_pins)
+		return -EINVAL;
+
+	pin_index = array_index_nospec(pd.index, ops->n_pins);
+	scoped_cond_guard(mutex_intr, return -ERESTARTSYS, &ptp->pincfg_mux)
+		return ptp_set_pinfunc(ptp, pin_index, pd.func, pd.chan);
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
-	struct ptp_clock_info *ops = ptp->info;
 	struct timestamp_event_queue *tsevq;
-	unsigned int i, pin_index;
-	struct ptp_pin_desc pd;
 	void __user *argptr;
+	unsigned int i;
 	int err = 0;
 
 	if (in_compat_syscall() && cmd != PTP_ENABLE_PPS && cmd != PTP_ENABLE_PPS2)
@@ -474,37 +492,9 @@ long ptp_ioctl(struct posix_clock_contex
 
 	case PTP_PIN_SETFUNC:
 	case PTP_PIN_SETFUNC2:
-		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0) {
-			err = -EACCES;
-			break;
-		}
-		if (copy_from_user(&pd, (void __user *)arg, sizeof(pd))) {
-			err = -EFAULT;
-			break;
-		}
-		if ((pd.rsv[0] || pd.rsv[1] || pd.rsv[2]
-				|| pd.rsv[3] || pd.rsv[4])
-			&& cmd == PTP_PIN_SETFUNC2) {
-			err = -EINVAL;
-			break;
-		} else if (cmd == PTP_PIN_SETFUNC) {
-			pd.rsv[0] = 0;
-			pd.rsv[1] = 0;
-			pd.rsv[2] = 0;
-			pd.rsv[3] = 0;
-			pd.rsv[4] = 0;
-		}
-		pin_index = pd.index;
-		if (pin_index >= ops->n_pins) {
-			err = -EINVAL;
-			break;
-		}
-		pin_index = array_index_nospec(pin_index, ops->n_pins);
-		if (mutex_lock_interruptible(&ptp->pincfg_mux))
-			return -ERESTARTSYS;
-		err = ptp_set_pinfunc(ptp, pin_index, pd.func, pd.chan);
-		mutex_unlock(&ptp->pincfg_mux);
-		break;
+		if ((pccontext->fp->f_mode & FMODE_WRITE) == 0)
+			return -EACCES;
+		return ptp_pin_setfunc(ptp, cmd, argptr);
 
 	case PTP_MASK_CLEAR_ALL:
 		bitmap_clear(tsevq->mask, 0, PTP_MAX_CHANNELS);

[patch V2 10/13] ptp: Split out PTP_MASK_CLEAR_ALL ioctl code

[Thomas Gleixner]

Continue the ptp_ioctl() cleanup by splitting out the PTP_MASK_CLEAR_ALL ioctl
code into a helper function.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |    9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -442,6 +442,12 @@ static long ptp_pin_setfunc(struct ptp_c
 		return ptp_set_pinfunc(ptp, pin_index, pd.func, pd.chan);
 }
 
+static long ptp_mask_clear_all(struct timestamp_event_queue *tsevq)
+{
+	bitmap_clear(tsevq->mask, 0, PTP_MAX_CHANNELS);
+	return 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
@@ -504,8 +510,7 @@ long ptp_ioctl(struct posix_clock_contex
 		return ptp_pin_setfunc(ptp, cmd, argptr);
 
 	case PTP_MASK_CLEAR_ALL:
-		bitmap_clear(tsevq->mask, 0, PTP_MAX_CHANNELS);
-		break;
+		return ptp_mask_clear_all(pccontext->private_clkdata);
 
 	case PTP_MASK_EN_SINGLE:
 		if (copy_from_user(&i, (void __user *)arg, sizeof(i))) {

[patch V2 11/13] ptp: Split out PTP_MASK_EN_SINGLE ioctl code

[Thomas Gleixner]

Finish the ptp_ioctl() cleanup by splitting out the PTP_MASK_EN_SINGLE
ioctl code and removing the remaining local variables and return
statements.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>

---
 drivers/ptp/ptp_chardev.c |   35 +++++++++++++++--------------------
 1 file changed, 15 insertions(+), 20 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -448,22 +448,28 @@ static long ptp_mask_clear_all(struct ti
 	return 0;
 }
 
+static long ptp_mask_en_single(struct timestamp_event_queue *tsevq, void __user *arg)
+{
+	unsigned int channel;
+
+	if (copy_from_user(&channel, arg, sizeof(channel)))
+		return -EFAULT;
+	if (channel >= PTP_MAX_CHANNELS)
+		return -EFAULT;
+	set_bit(channel, tsevq->mask);
+	return 0;
+}
+
 long ptp_ioctl(struct posix_clock_context *pccontext, unsigned int cmd,
 	       unsigned long arg)
 {
-	struct ptp_clock *ptp =
-		container_of(pccontext->clk, struct ptp_clock, clock);
-	struct timestamp_event_queue *tsevq;
+	struct ptp_clock *ptp = container_of(pccontext->clk, struct ptp_clock, clock);
 	void __user *argptr;
-	unsigned int i;
-	int err = 0;
 
 	if (in_compat_syscall() && cmd != PTP_ENABLE_PPS && cmd != PTP_ENABLE_PPS2)
 		arg = (unsigned long)compat_ptr(arg);
 	argptr = (void __force __user *)arg;
 
-	tsevq = pccontext->private_clkdata;
-
 	switch (cmd) {
 	case PTP_CLOCK_GETCAPS:
 	case PTP_CLOCK_GETCAPS2:
@@ -513,22 +519,11 @@ long ptp_ioctl(struct posix_clock_contex
 		return ptp_mask_clear_all(pccontext->private_clkdata);
 
 	case PTP_MASK_EN_SINGLE:
-		if (copy_from_user(&i, (void __user *)arg, sizeof(i))) {
-			err = -EFAULT;
-			break;
-		}
-		if (i >= PTP_MAX_CHANNELS) {
-			err = -EFAULT;
-			break;
-		}
-		set_bit(i, tsevq->mask);
-		break;
+		return ptp_mask_en_single(pccontext->private_clkdata, argptr);
 
 	default:
-		err = -ENOTTY;
-		break;
+		return -ENOTTY;
 	}
-	return err;
 }
 
 __poll_t ptp_poll(struct posix_clock_context *pccontext, struct file *fp,

[patch V2 12/13] ptp: Convert chardev code to lock guards

[Thomas Gleixner]

Convert the various spin_lock_irqsave() protected critical regions to
scoped guards. Use spinlock_irq instead of spinlock_irqsave as all the
functions are invoked in thread context with interrupts enabled.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>

---
 drivers/ptp/ptp_chardev.c |   34 +++++++++++++---------------------
 1 file changed, 13 insertions(+), 21 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -110,7 +110,6 @@ int ptp_open(struct posix_clock_context
 		container_of(pccontext->clk, struct ptp_clock, clock);
 	struct timestamp_event_queue *queue;
 	char debugfsname[32];
-	unsigned long flags;
 
 	queue = kzalloc(sizeof(*queue), GFP_KERNEL);
 	if (!queue)
@@ -122,9 +121,8 @@ int ptp_open(struct posix_clock_context
 	}
 	bitmap_set(queue->mask, 0, PTP_MAX_CHANNELS);
 	spin_lock_init(&queue->lock);
-	spin_lock_irqsave(&ptp->tsevqs_lock, flags);
-	list_add_tail(&queue->qlist, &ptp->tsevqs);
-	spin_unlock_irqrestore(&ptp->tsevqs_lock, flags);
+	scoped_guard(spinlock_irq, &ptp->tsevqs_lock)
+		list_add_tail(&queue->qlist, &ptp->tsevqs);
 	pccontext->private_clkdata = queue;
 
 	/* Debugfs contents */
@@ -143,15 +141,13 @@ int ptp_open(struct posix_clock_context
 int ptp_release(struct posix_clock_context *pccontext)
 {
 	struct timestamp_event_queue *queue = pccontext->private_clkdata;
-	unsigned long flags;
 	struct ptp_clock *ptp =
 		container_of(pccontext->clk, struct ptp_clock, clock);
 
 	debugfs_remove(queue->debugfs_instance);
 	pccontext->private_clkdata = NULL;
-	spin_lock_irqsave(&ptp->tsevqs_lock, flags);
-	list_del(&queue->qlist);
-	spin_unlock_irqrestore(&ptp->tsevqs_lock, flags);
+	scoped_guard(spinlock_irq, &ptp->tsevqs_lock)
+		list_del(&queue->qlist);
 	bitmap_free(queue->mask);
 	kfree(queue);
 	return 0;
@@ -548,8 +544,6 @@ ssize_t ptp_read(struct posix_clock_cont
 		container_of(pccontext->clk, struct ptp_clock, clock);
 	struct timestamp_event_queue *queue;
 	struct ptp_extts_event *event;
-	unsigned long flags;
-	size_t qcnt, i;
 	int result;
 
 	queue = pccontext->private_clkdata;
@@ -584,21 +578,19 @@ ssize_t ptp_read(struct posix_clock_cont
 		goto exit;
 	}
 
-	spin_lock_irqsave(&queue->lock, flags);
+	scoped_guard(spinlock_irq, &queue->lock) {
+		size_t qcnt = queue_cnt(queue);
 
-	qcnt = queue_cnt(queue);
+		if (cnt > qcnt)
+			cnt = qcnt;
 
-	if (cnt > qcnt)
-		cnt = qcnt;
-
-	for (i = 0; i < cnt; i++) {
-		event[i] = queue->buf[queue->head];
-		/* Paired with READ_ONCE() in queue_cnt() */
-		WRITE_ONCE(queue->head, (queue->head + 1) % PTP_MAX_TIMESTAMPS);
+		for (size_t i = 0; i < cnt; i++) {
+			event[i] = queue->buf[queue->head];
+			/* Paired with READ_ONCE() in queue_cnt() */
+			WRITE_ONCE(queue->head, (queue->head + 1) % PTP_MAX_TIMESTAMPS);
+		}
 	}
 
-	spin_unlock_irqrestore(&queue->lock, flags);
-
 	cnt = cnt * sizeof(struct ptp_extts_event);
 
 	result = cnt;

[patch V2 13/13] ptp: Simplify ptp_read()

[Thomas Gleixner]

The mixture of gotos and direct return codes is inconsistent and just makes
the code harder to read. Let it consistently return error codes directly and
tidy the code flow up accordingly.

No functional change intended.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
V2: Fix the return value - Paolo
    Drop the __free() part - Jakub
---
 drivers/ptp/ptp_chardev.c |   54 +++++++++++++---------------------------------
 1 file changed, 16 insertions(+), 38 deletions(-)

--- a/drivers/ptp/ptp_chardev.c
+++ b/drivers/ptp/ptp_chardev.c
@@ -106,8 +106,7 @@ int ptp_set_pinfunc(struct ptp_clock *pt
 
 int ptp_open(struct posix_clock_context *pccontext, fmode_t fmode)
 {
-	struct ptp_clock *ptp =
-		container_of(pccontext->clk, struct ptp_clock, clock);
+	struct ptp_clock *ptp = container_of(pccontext->clk, struct ptp_clock, clock);
 	struct timestamp_event_queue *queue;
 	char debugfsname[32];
 
@@ -536,67 +535,46 @@ long ptp_ioctl(struct posix_clock_contex
 ssize_t ptp_read(struct posix_clock_context *pccontext, uint rdflags,
 		 char __user *buf, size_t cnt)
 {
-	struct ptp_clock *ptp =
-		container_of(pccontext->clk, struct ptp_clock, clock);
+	struct ptp_clock *ptp =	container_of(pccontext->clk, struct ptp_clock, clock);
 	struct timestamp_event_queue *queue;
 	struct ptp_extts_event *event;
-	int result;
+	ssize_t result;
 
 	queue = pccontext->private_clkdata;
-	if (!queue) {
-		result = -EINVAL;
-		goto exit;
-	}
+	if (!queue)
+		return -EINVAL;
 
-	if (cnt % sizeof(struct ptp_extts_event) != 0) {
-		result = -EINVAL;
-		goto exit;
-	}
+	if (cnt % sizeof(*event) != 0)
+		return -EINVAL;
 
 	if (cnt > EXTTS_BUFSIZE)
 		cnt = EXTTS_BUFSIZE;
 
-	cnt = cnt / sizeof(struct ptp_extts_event);
-
-	if (wait_event_interruptible(ptp->tsev_wq,
-				     ptp->defunct || queue_cnt(queue))) {
+	if (wait_event_interruptible(ptp->tsev_wq, ptp->defunct || queue_cnt(queue)))
 		return -ERESTARTSYS;
-	}
 
-	if (ptp->defunct) {
-		result = -ENODEV;
-		goto exit;
-	}
+	if (ptp->defunct)
+		return -ENODEV;
 
 	event = kmalloc(EXTTS_BUFSIZE, GFP_KERNEL);
-	if (!event) {
-		result = -ENOMEM;
-		goto exit;
-	}
+	if (!event)
+		return -ENOMEM;
 
 	scoped_guard(spinlock_irq, &queue->lock) {
-		size_t qcnt = queue_cnt(queue);
-
-		if (cnt > qcnt)
-			cnt = qcnt;
+		size_t qcnt = min((size_t)queue_cnt(queue), cnt / sizeof(*event));
 
-		for (size_t i = 0; i < cnt; i++) {
+		for (size_t i = 0; i < qcnt; i++) {
 			event[i] = queue->buf[queue->head];
 			/* Paired with READ_ONCE() in queue_cnt() */
 			WRITE_ONCE(queue->head, (queue->head + 1) % PTP_MAX_TIMESTAMPS);
 		}
+		cnt = qcnt * sizeof(*event);
 	}
 
-	cnt = cnt * sizeof(struct ptp_extts_event);
-
 	result = cnt;
-	if (copy_to_user(buf, event, cnt)) {
+	if (copy_to_user(buf, event, cnt))
 		result = -EFAULT;
-		goto free_event;
-	}
 
-free_event:
 	kfree(event);
-exit:
 	return result;
 }

Re: [patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[Richard Cochran]

On Wed, Jun 25, 2025 at 01:52:23PM +0200, Thomas Gleixner wrote:
> The code (~400 lines!) is really hard to follow due to a gazillion of
> local variables, which are only used in certain case scopes, and a
> mixture of gotos, breaks and direct error return paths.

But it is 100% organically grown and therefore healthy for you!

Thanks,
Richard

Re: [patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[Thomas Gleixner]

On Wed, Jun 25 2025 at 07:40, Richard Cochran wrote:

> On Wed, Jun 25, 2025 at 01:52:23PM +0200, Thomas Gleixner wrote:
>> The code (~400 lines!) is really hard to follow due to a gazillion of
>> local variables, which are only used in certain case scopes, and a
>> mixture of gotos, breaks and direct error return paths.
>
> But it is 100% organically grown and therefore healthy for you!

You sure there were no forbidden plants involved?

Re: [patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[Richard Cochran]

On Wed, Jun 25, 2025 at 06:05:44PM +0200, Thomas Gleixner wrote:
> You sure there were no forbidden plants involved?

Not in Calyfornia :^)

Cheers,
Richard

Re: [patch V2 00/13] ptp: Belated spring cleaning of the chardev driver

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:

On Wed, 25 Jun 2025 13:52:23 +0200 (CEST) you wrote:
> This is V2 of the series. V1 can be found here:
> 
>      https://lore.kernel.org/all/20250620130144.351492917@linutronix.de
> 
> When looking into supporting auxiliary clocks in the PTP ioctl, the
> inpenetrable ptp_ioctl() letter soup bothered me enough to clean it up.
> 
> [...]

Here is the summary with links:
  - [V2,01/13] ptp: Split out PTP_CLOCK_GETCAPS ioctl code
    https://git.kernel.org/netdev/net-next/c/7ca2ac4953fd
  - [V2,02/13] ptp: Split out PTP_EXTTS_REQUEST ioctl code
    https://git.kernel.org/netdev/net-next/c/f6b3e1bc6ed3
  - [V2,03/13] ptp: Split out PTP_PEROUT_REQUEST ioctl code
    https://git.kernel.org/netdev/net-next/c/3afc2caceaf7
  - [V2,04/13] ptp: Split out PTP_ENABLE_PPS ioctl code
    https://git.kernel.org/netdev/net-next/c/47aaa73d25ea
  - [V2,05/13] ptp: Split out PTP_SYS_OFFSET_PRECISE ioctl code
    https://git.kernel.org/netdev/net-next/c/e4355e314c94
  - [V2,06/13] ptp: Split out PTP_SYS_OFFSET_EXTENDED ioctl code
    https://git.kernel.org/netdev/net-next/c/37e42f8dd07d
  - [V2,07/13] ptp: Split out PTP_SYS_OFFSET ioctl code
    https://git.kernel.org/netdev/net-next/c/4b676af26e9b
  - [V2,08/13] ptp: Split out PTP_PIN_GETFUNC ioctl code
    https://git.kernel.org/netdev/net-next/c/b246e09f5fe1
  - [V2,09/13] ptp: Split out PTP_PIN_SETFUNC ioctl code
    https://git.kernel.org/netdev/net-next/c/d713f1ff64d1
  - [V2,10/13] ptp: Split out PTP_MASK_CLEAR_ALL ioctl code
    https://git.kernel.org/netdev/net-next/c/6a0f480478a7
  - [V2,11/13] ptp: Split out PTP_MASK_EN_SINGLE ioctl code
    https://git.kernel.org/netdev/net-next/c/745e3c751c4d
  - [V2,12/13] ptp: Convert chardev code to lock guards
    https://git.kernel.org/netdev/net-next/c/4838bc9e279c
  - [V2,13/13] ptp: Simplify ptp_read()
    https://git.kernel.org/netdev/net-next/c/b66d28142dc4

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html