[Devel] [PATCH hci-8.1 v2 2/6] ms/vdso/timens: Move functions to new file

[Eva Kurchatova <eva.kurchatova@virtuozzo.com>]

As a preparation of the untangling of time namespaces and the vDSO, move
the glue functions between those subsystems into a new file.

While at it, switch the mutex lock and mmap_read_lock() in the vDSO
namespace code to guard().

Signed-off-by: Thomas Weißschuh <thomas.weissschuh@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@kernel.org>
Link: https://patch.msgid.link/20260326-vdso-timens-decoupling-v2-1-c82693a7775f@linutronix.de
(cherry picked from commit 5dc9cf835aba73c882348aa4f99be83b6e45ad9b)

https://virtuozzo.atlassian.net/browse/VSTOR-148112
Feature: fix ms/timens

Signed-off-by: Eva Kurchatova <eva.kurchatova@virtuozzo.com>
---
 MAINTAINERS                      |   2 +
 include/linux/time_namespace.h   |   8 --
 kernel/time/Makefile             |   2 +-
 kernel/time/namespace.c          | 119 +-------------------------
 kernel/time/namespace_internal.h |  13 +++
 kernel/time/namespace_vdso.c     | 141 +++++++++++++++++++++++++++++++
 lib/vdso/datastore.c             |  25 ------
 7 files changed, 161 insertions(+), 149 deletions(-)
 create mode 100644 kernel/time/namespace_internal.h
 create mode 100644 kernel/time/namespace_vdso.c

diff --git a/MAINTAINERS b/MAINTAINERS
index 9523da3c767c..ac649ee0c43c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -9683,6 +9683,7 @@ S:	Maintained
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git timers/vdso
 F:	include/asm-generic/vdso/vsyscall.h
 F:	include/vdso/
+F:	kernel/time/namespace_vdso.c
 F:	kernel/time/vsyscall.c
 F:	lib/vdso/
 F:	tools/testing/selftests/vDSO/
@@ -18576,6 +18577,7 @@ F:	include/trace/events/timer*
 F:	kernel/time/itimer.c
 F:	kernel/time/posix-*
 F:	kernel/time/namespace.c
+F:	kernel/time/namespace_vdso.c
 
 POWER MANAGEMENT CORE
 M:	"Rafael J. Wysocki" <rafael@kernel.org>
diff --git a/include/linux/time_namespace.h b/include/linux/time_namespace.h
index 3d3b529c5a98..23384eafebfb 100644
--- a/include/linux/time_namespace.h
+++ b/include/linux/time_namespace.h
@@ -35,8 +35,6 @@ extern struct time_namespace init_time_ns;
 extern struct mutex offset_lock;
 
 #ifdef CONFIG_TIME_NS
-extern int vdso_join_timens(struct task_struct *task,
-			    struct time_namespace *ns);
 extern void timens_commit(struct task_struct *tsk, struct time_namespace *ns);
 
 static inline struct time_namespace *get_time_ns(struct time_namespace *ns)
@@ -110,12 +108,6 @@ static inline ktime_t timens_ktime_to_host(clockid_t clockid, ktime_t tim)
 }
 
 #else
-static inline int vdso_join_timens(struct task_struct *task,
-				   struct time_namespace *ns)
-{
-	return 0;
-}
-
 static inline void timens_commit(struct task_struct *tsk,
 				 struct time_namespace *ns)
 {
diff --git a/kernel/time/Makefile b/kernel/time/Makefile
index fe0ae82124fe..494bd4e41781 100644
--- a/kernel/time/Makefile
+++ b/kernel/time/Makefile
@@ -23,6 +23,6 @@ endif
 obj-$(CONFIG_HAVE_GENERIC_VDSO)			+= vsyscall.o
 obj-$(CONFIG_DEBUG_FS)				+= timekeeping_debug.o
 obj-$(CONFIG_TEST_UDELAY)			+= test_udelay.o
-obj-$(CONFIG_TIME_NS)				+= namespace.o
+obj-$(CONFIG_TIME_NS)				+= namespace.o namespace_vdso.o
 obj-$(CONFIG_TEST_CLOCKSOURCE_WATCHDOG)		+= clocksource-wdtest.o
 obj-$(CONFIG_TIME_KUNIT_TEST)			+= time_test.o
diff --git a/kernel/time/namespace.c b/kernel/time/namespace.c
index e85f947d4063..c82dc9a5d1de 100644
--- a/kernel/time/namespace.c
+++ b/kernel/time/namespace.c
@@ -18,7 +18,7 @@
 #include <linux/err.h>
 #include <linux/mm.h>
 
-#include <vdso/datapage.h>
+#include "namespace_internal.h"
 
 ktime_t do_timens_ktime_to_host(clockid_t clockid, ktime_t tim,
 				struct timens_offsets *ns_offsets)
@@ -139,112 +139,7 @@ struct time_namespace *copy_time_ns(unsigned long flags,
 	return clone_time_ns(user_ns, old_ns);
 }
 
-static struct timens_offset offset_from_ts(struct timespec64 off)
-{
-	struct timens_offset ret;
-
-	ret.sec = off.tv_sec;
-	ret.nsec = off.tv_nsec;
-
-	return ret;
-}
-
-/*
- * A time namespace VVAR page has the same layout as the VVAR page which
- * contains the system wide VDSO data.
- *
- * For a normal task the VVAR pages are installed in the normal ordering:
- *     VVAR
- *     PVCLOCK
- *     HVCLOCK
- *     TIMENS   <- Not really required
- *
- * Now for a timens task the pages are installed in the following order:
- *     TIMENS
- *     PVCLOCK
- *     HVCLOCK
- *     VVAR
- *
- * The check for vdso_clock->clock_mode is in the unlikely path of
- * the seq begin magic. So for the non-timens case most of the time
- * 'seq' is even, so the branch is not taken.
- *
- * If 'seq' is odd, i.e. a concurrent update is in progress, the extra check
- * for vdso_clock->clock_mode is a non-issue. The task is spin waiting for the
- * update to finish and for 'seq' to become even anyway.
- *
- * Timens page has vdso_clock->clock_mode set to VDSO_CLOCKMODE_TIMENS which
- * enforces the time namespace handling path.
- */
-static void timens_setup_vdso_clock_data(struct vdso_clock *vc,
-					 struct time_namespace *ns)
-{
-	struct timens_offset *offset = vc->offset;
-	struct timens_offset monotonic = offset_from_ts(ns->offsets.monotonic);
-	struct timens_offset boottime = offset_from_ts(ns->offsets.boottime);
-
-	vc->seq				= 1;
-	vc->clock_mode			= VDSO_CLOCKMODE_TIMENS;
-	offset[CLOCK_MONOTONIC]		= monotonic;
-	offset[CLOCK_MONOTONIC_RAW]	= monotonic;
-	offset[CLOCK_MONOTONIC_COARSE]	= monotonic;
-	offset[CLOCK_BOOTTIME]		= boottime;
-	offset[CLOCK_BOOTTIME_ALARM]	= boottime;
-}
-
-struct page *find_timens_vvar_page(struct vm_area_struct *vma)
-{
-	if (likely(vma->vm_mm == current->mm))
-		return current->nsproxy->time_ns->vvar_page;
-
-	/*
-	 * VM_PFNMAP | VM_IO protect .fault() handler from being called
-	 * through interfaces like /proc/$pid/mem or
-	 * process_vm_{readv,writev}() as long as there's no .access()
-	 * in special_mapping_vmops().
-	 * For more details check_vma_flags() and __access_remote_vm()
-	 */
-
-	WARN(1, "vvar_page accessed remotely");
-
-	return NULL;
-}
-
-/*
- * Protects possibly multiple offsets writers racing each other
- * and tasks entering the namespace.
- */
-DEFINE_MUTEX(offset_lock);
-
-static void timens_set_vvar_page(struct task_struct *task,
-				struct time_namespace *ns)
-{
-	struct vdso_time_data *vdata;
-	struct vdso_clock *vc;
-	unsigned int i;
-
-	if (ns == &init_time_ns)
-		return;
-
-	/* Fast-path, taken by every task in namespace except the first. */
-	if (likely(ns->frozen_offsets))
-		return;
-
-	mutex_lock(&offset_lock);
-	/* Nothing to-do: vvar_page has been already initialized. */
-	if (ns->frozen_offsets)
-		goto out;
-
-	ns->frozen_offsets = true;
-	vdata = page_address(ns->vvar_page);
-	vc = vdata->clock_data;
-
-	for (i = 0; i < CS_BASES; i++)
-		timens_setup_vdso_clock_data(&vc[i], ns);
-
-out:
-	mutex_unlock(&offset_lock);
-}
+DEFINE_MUTEX(timens_offset_lock);
 
 void free_time_ns(struct time_namespace *ns)
 {
@@ -297,12 +192,6 @@ static void timens_put(struct ns_common *ns)
 	put_time_ns(to_time_ns(ns));
 }
 
-void timens_commit(struct task_struct *tsk, struct time_namespace *ns)
-{
-	timens_set_vvar_page(tsk, ns);
-	vdso_join_timens(tsk, ns);
-}
-
 static int timens_install(struct nsset *nsset, struct ns_common *new)
 {
 	struct nsproxy *nsproxy = nsset->nsproxy;
@@ -427,7 +316,7 @@ int proc_timens_set_offset(struct file *file, struct task_struct *p,
 			goto out;
 	}
 
-	mutex_lock(&offset_lock);
+	mutex_lock(&timens_offset_lock);
 	if (time_ns->frozen_offsets) {
 		err = -EACCES;
 		goto out_unlock;
@@ -452,7 +341,7 @@ int proc_timens_set_offset(struct file *file, struct task_struct *p,
 	}
 
 out_unlock:
-	mutex_unlock(&offset_lock);
+	mutex_unlock(&timens_offset_lock);
 out:
 	put_time_ns(time_ns);
 
diff --git a/kernel/time/namespace_internal.h b/kernel/time/namespace_internal.h
new file mode 100644
index 000000000000..e85da11abb4d
--- /dev/null
+++ b/kernel/time/namespace_internal.h
@@ -0,0 +1,13 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _TIME_NAMESPACE_INTERNAL_H
+#define _TIME_NAMESPACE_INTERNAL_H
+
+#include <linux/mutex.h>
+
+/*
+ * Protects possibly multiple offsets writers racing each other
+ * and tasks entering the namespace.
+ */
+extern struct mutex timens_offset_lock;
+
+#endif /* _TIME_NAMESPACE_INTERNAL_H */
diff --git a/kernel/time/namespace_vdso.c b/kernel/time/namespace_vdso.c
new file mode 100644
index 000000000000..558bf5208588
--- /dev/null
+++ b/kernel/time/namespace_vdso.c
@@ -0,0 +1,141 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Author: Andrei Vagin <avagin@openvz.org>
+ * Author: Dmitry Safonov <dima@arista.com>
+ */
+
+#include <linux/cleanup.h>
+#include <linux/mm.h>
+#include <linux/time_namespace.h>
+#include <linux/time.h>
+#include <linux/vdso_datastore.h>
+
+#include <vdso/clocksource.h>
+#include <vdso/datapage.h>
+
+#include "namespace_internal.h"
+
+static struct timens_offset offset_from_ts(struct timespec64 off)
+{
+	struct timens_offset ret;
+
+	ret.sec = off.tv_sec;
+	ret.nsec = off.tv_nsec;
+
+	return ret;
+}
+
+/*
+ * A time namespace VVAR page has the same layout as the VVAR page which
+ * contains the system wide VDSO data.
+ *
+ * For a normal task the VVAR pages are installed in the normal ordering:
+ *     VVAR
+ *     PVCLOCK
+ *     HVCLOCK
+ *     TIMENS   <- Not really required
+ *
+ * Now for a timens task the pages are installed in the following order:
+ *     TIMENS
+ *     PVCLOCK
+ *     HVCLOCK
+ *     VVAR
+ *
+ * The check for vdso_clock->clock_mode is in the unlikely path of
+ * the seq begin magic. So for the non-timens case most of the time
+ * 'seq' is even, so the branch is not taken.
+ *
+ * If 'seq' is odd, i.e. a concurrent update is in progress, the extra check
+ * for vdso_clock->clock_mode is a non-issue. The task is spin waiting for the
+ * update to finish and for 'seq' to become even anyway.
+ *
+ * Timens page has vdso_clock->clock_mode set to VDSO_CLOCKMODE_TIMENS which
+ * enforces the time namespace handling path.
+ */
+static void timens_setup_vdso_clock_data(struct vdso_clock *vc,
+					 struct time_namespace *ns)
+{
+	struct timens_offset *offset = vc->offset;
+	struct timens_offset monotonic = offset_from_ts(ns->offsets.monotonic);
+	struct timens_offset boottime = offset_from_ts(ns->offsets.boottime);
+
+	vc->seq				= 1;
+	vc->clock_mode			= VDSO_CLOCKMODE_TIMENS;
+	offset[CLOCK_MONOTONIC]		= monotonic;
+	offset[CLOCK_MONOTONIC_RAW]	= monotonic;
+	offset[CLOCK_MONOTONIC_COARSE]	= monotonic;
+	offset[CLOCK_BOOTTIME]		= boottime;
+	offset[CLOCK_BOOTTIME_ALARM]	= boottime;
+}
+
+struct page *find_timens_vvar_page(struct vm_area_struct *vma)
+{
+	if (likely(vma->vm_mm == current->mm))
+		return current->nsproxy->time_ns->vvar_page;
+
+	/*
+	 * VM_PFNMAP | VM_IO protect .fault() handler from being called
+	 * through interfaces like /proc/$pid/mem or
+	 * process_vm_{readv,writev}() as long as there's no .access()
+	 * in special_mapping_vmops().
+	 * For more details check_vma_flags() and __access_remote_vm()
+	 */
+
+	WARN(1, "vvar_page accessed remotely");
+
+	return NULL;
+}
+
+static void timens_set_vvar_page(struct task_struct *task,
+				struct time_namespace *ns)
+{
+	struct vdso_time_data *vdata;
+	struct vdso_clock *vc;
+	unsigned int i;
+
+	if (ns == &init_time_ns)
+		return;
+
+	/* Fast-path, taken by every task in namespace except the first. */
+	if (likely(ns->frozen_offsets))
+		return;
+
+	guard(mutex)(&timens_offset_lock);
+	/* Nothing to-do: vvar_page has been already initialized. */
+	if (ns->frozen_offsets)
+		return;
+
+	ns->frozen_offsets = true;
+	vdata = page_address(ns->vvar_page);
+	vc = vdata->clock_data;
+
+	for (i = 0; i < CS_BASES; i++)
+		timens_setup_vdso_clock_data(&vc[i], ns);
+}
+
+/*
+ * The vvar page layout depends on whether a task belongs to the root or
+ * non-root time namespace. Whenever a task changes its namespace, the VVAR
+ * page tables are cleared and then they will be re-faulted with a
+ * corresponding layout.
+ * See also the comment near timens_setup_vdso_clock_data() for details.
+ */
+static int vdso_join_timens(struct task_struct *task, struct time_namespace *ns)
+{
+	struct mm_struct *mm = task->mm;
+	struct vm_area_struct *vma;
+	VMA_ITERATOR(vmi, mm, 0);
+
+	guard(mmap_read_lock)(mm);
+	for_each_vma(vmi, vma) {
+		if (vma_is_special_mapping(vma, &vdso_vvar_mapping))
+			zap_vma_pages(vma);
+	}
+	return 0;
+}
+
+void timens_commit(struct task_struct *tsk, struct time_namespace *ns)
+{
+	timens_set_vvar_page(tsk, ns);
+	vdso_join_timens(tsk, ns);
+}
diff --git a/lib/vdso/datastore.c b/lib/vdso/datastore.c
index 46192c1d8542..7713fd392dd0 100644
--- a/lib/vdso/datastore.c
+++ b/lib/vdso/datastore.c
@@ -102,28 +102,3 @@ struct vm_area_struct *vdso_install_vvar_mapping(struct mm_struct *mm, unsigned
 					VM_READ | VM_MAYREAD | VM_IO | VM_DONTDUMP | VM_PFNMAP,
 					&vdso_vvar_mapping);
 }
-
-#ifdef CONFIG_TIME_NS
-/*
- * The vvar page layout depends on whether a task belongs to the root or
- * non-root time namespace. Whenever a task changes its namespace, the VVAR
- * page tables are cleared and then they will be re-faulted with a
- * corresponding layout.
- * See also the comment near timens_setup_vdso_clock_data() for details.
- */
-int vdso_join_timens(struct task_struct *task, struct time_namespace *ns)
-{
-	struct mm_struct *mm = task->mm;
-	struct vm_area_struct *vma;
-	VMA_ITERATOR(vmi, mm, 0);
-
-	mmap_read_lock(mm);
-	for_each_vma(vmi, vma) {
-		if (vma_is_special_mapping(vma, &vdso_vvar_mapping))
-			zap_vma_pages(vma);
-	}
-	mmap_read_unlock(mm);
-
-	return 0;
-}
-#endif
-- 
2.56.0

_______________________________________________
Devel mailing list
Devel@openvz.org
https://lists.openvz.org/mailman/listinfo/devel