Merge branch 'for-3.17-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

Pull cgroup fixes from Tejun Heo: "This is quite late but these need to be backported anyway. This is the fix for a long-standing cpuset bug which existed from 2009. cpuset makes use of PF_SPREAD_{PAGE|SLAB} flags to modify the task's memory allocation behavior according to the settings of the cpuset it belongs to; unfortunately, when those flags have to be changed, cpuset did so directly even whlie the target task is running, which is obviously racy as task->flags may be modified by the task itself at any time. This obscure bug manifested as corrupt PF_USED_MATH flag leading to a weird crash. The bug is fixed by moving the flag to task->atomic_flags. The first two are prepatory ones to help defining atomic_flags accessors and the third one is the actual fix" * 'for-3.17-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: cpuset: PF_SPREAD_PAGE and PF_SPREAD_SLAB should be atomic flags sched: add macros to define bitops for task atomic flags sched: fix confusing PFA_NO_NEW_PRIVS constant

Merge branch 'for-3.17-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup
Pull cgroup fixes from Tejun Heo: "This is quite late but these need to be backported anyway. This is the fix for a long-standing cpuset bug which existed from 2009. cpuset makes use of PF_SPREAD_{PAGE|SLAB} flags to modify the task's memory allocation behavior according to the settings of the cpuset it belongs to; unfortunately, when those flags have to be changed, cpuset did so directly even whlie the target task is running, which is obviously racy as task->flags may be modified by the task itself at any time. This obscure bug manifested as corrupt PF_USED_MATH flag leading to a weird crash. The bug is fixed by moving the flag to task->atomic_flags. The first two are prepatory ones to help defining atomic_flags accessors and the third one is the actual fix" * 'for-3.17-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: cpuset: PF_SPREAD_PAGE and PF_SPREAD_SLAB should be atomic flags sched: add macros to define bitops for task atomic flags sched: fix confusing PFA_NO_NEW_PRIVS constant
Linus Torvalds
2 parents 8369289864 2ad654bc5e
Showing 6 changed files Side-by-side Diff
Documentation/cgroups/cpusets.txt
include/linux/cpuset.h
include/linux/sched.h
kernel/cpuset.c
mm/slab.c
scripts/tags.sh
@@ -345,14 +345,14 @@
 The implementation is simple.
  
 Setting the flag 'cpuset.memory_spread_page' turns on a per-process flag
-PF_SPREAD_PAGE for each task that is in that cpuset or subsequently
+PFA_SPREAD_PAGE for each task that is in that cpuset or subsequently
 joins that cpuset.  The page allocation calls for the page cache
-is modified to perform an inline check for this PF_SPREAD_PAGE task
+is modified to perform an inline check for this PFA_SPREAD_PAGE task
 flag, and if set, a call to a new routine cpuset_mem_spread_node()
 returns the node to prefer for the allocation.
  
 Similarly, setting 'cpuset.memory_spread_slab' turns on the flag
-PF_SPREAD_SLAB, and appropriately marked slab caches will allocate
+PFA_SPREAD_SLAB, and appropriately marked slab caches will allocate
 pages from the node returned by cpuset_mem_spread_node().
  
 The cpuset_mem_spread_node() routine is also simple.  It uses the
@@ -93,12 +93,12 @@
  
 static inline int cpuset_do_page_mem_spread(void)
 {
-	return current->flags & PF_SPREAD_PAGE;
+	return task_spread_page(current);
 }
  
 static inline int cpuset_do_slab_mem_spread(void)
 {
-	return current->flags & PF_SPREAD_SLAB;
+	return task_spread_slab(current);
 }
  
 extern int current_cpuset_is_being_rebound(void);
@@ -1903,8 +1903,6 @@
 #define PF_KTHREAD	0x00200000	/* I am a kernel thread */
 #define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
 #define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
-#define PF_SPREAD_PAGE	0x01000000	/* Spread page cache over cpuset */
-#define PF_SPREAD_SLAB	0x02000000	/* Spread some slab caches over cpuset */
 #define PF_NO_SETAFFINITY 0x04000000	/* Userland is not allowed to meddle with cpus_allowed */
 #define PF_MCE_EARLY    0x08000000      /* Early kill for mce process policy */
 #define PF_MUTEX_TESTER	0x20000000	/* Thread belongs to the rt mutex tester */
  
  
@@ -1957,17 +1955,31 @@
 }
  
 /* Per-process atomic flags. */
-#define PFA_NO_NEW_PRIVS 0x00000001	/* May not gain new privileges. */
+#define PFA_NO_NEW_PRIVS 0	/* May not gain new privileges. */
+#define PFA_SPREAD_PAGE  1      /* Spread page cache over cpuset */
+#define PFA_SPREAD_SLAB  2      /* Spread some slab caches over cpuset */
  
-static inline bool task_no_new_privs(struct task_struct *p)
-{
-	return test_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags);
-}
  
-static inline void task_set_no_new_privs(struct task_struct *p)
-{
-	set_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags);
-}
+#define TASK_PFA_TEST(name, func)					\
+	static inline bool task_##func(struct task_struct *p)		\
+	{ return test_bit(PFA_##name, &p->atomic_flags); }
+#define TASK_PFA_SET(name, func)					\
+	static inline void task_set_##func(struct task_struct *p)	\
+	{ set_bit(PFA_##name, &p->atomic_flags); }
+#define TASK_PFA_CLEAR(name, func)					\
+	static inline void task_clear_##func(struct task_struct *p)	\
+	{ clear_bit(PFA_##name, &p->atomic_flags); }
+
+TASK_PFA_TEST(NO_NEW_PRIVS, no_new_privs)
+TASK_PFA_SET(NO_NEW_PRIVS, no_new_privs)
+
+TASK_PFA_TEST(SPREAD_PAGE, spread_page)
+TASK_PFA_SET(SPREAD_PAGE, spread_page)
+TASK_PFA_CLEAR(SPREAD_PAGE, spread_page)
+
+TASK_PFA_TEST(SPREAD_SLAB, spread_slab)
+TASK_PFA_SET(SPREAD_SLAB, spread_slab)
+TASK_PFA_CLEAR(SPREAD_SLAB, spread_slab)
  
 /*
  * task->jobctl flags
@@ -365,13 +365,14 @@
 					struct task_struct *tsk)
 {
 	if (is_spread_page(cs))
-		tsk->flags |= PF_SPREAD_PAGE;
+		task_set_spread_page(tsk);
 	else
-		tsk->flags &= ~PF_SPREAD_PAGE;
+		task_clear_spread_page(tsk);
+
 	if (is_spread_slab(cs))
-		tsk->flags |= PF_SPREAD_SLAB;
+		task_set_spread_slab(tsk);
 	else
-		tsk->flags &= ~PF_SPREAD_SLAB;
+		task_clear_spread_slab(tsk);
 }
  
 /*
@@ -2987,7 +2987,7 @@
  
 #ifdef CONFIG_NUMA
 /*
- * Try allocating on another node if PF_SPREAD_SLAB is a mempolicy is set.
+ * Try allocating on another node if PFA_SPREAD_SLAB is a mempolicy is set.
  *
  * If we are in_interrupt, then process context, including cpusets and
  * mempolicy, may not apply and should not be used for allocation policy.
@@ -3219,7 +3219,7 @@
 {
 	void *objp;
  
-	if (current->mempolicy || unlikely(current->flags & PF_SPREAD_SLAB)) {
+	if (current->mempolicy || cpuset_do_slab_mem_spread()) {
 		objp = alternate_node_alloc(cache, flags);
 		if (objp)
 			goto out;
@@ -197,6 +197,9 @@
 	--regex-c++='/SETPCGFLAG\(([^,)]*).*/SetPageCgroup\1/'		\
 	--regex-c++='/CLEARPCGFLAG\(([^,)]*).*/ClearPageCgroup\1/'	\
 	--regex-c++='/TESTCLEARPCGFLAG\(([^,)]*).*/TestClearPageCgroup\1/' \
+	--regex-c++='/TASK_PFA_TEST\([^,]*,\s*([^)]*)\)/task_\1/'	\
+	--regex-c++='/TASK_PFA_SET\([^,]*,\s*([^)]*)\)/task_set_\1/'	\
+	--regex-c++='/TASK_PFA_CLEAR\([^,]*,\s*([^)]*)\)/task_clear_\1/'\
 	--regex-c='/PCI_OP_READ\((\w*).*[1-4]\)/pci_bus_read_config_\1/' \
 	--regex-c='/PCI_OP_WRITE\((\w*).*[1-4]\)/pci_bus_write_config_\1/' \
 	--regex-c='/DEFINE_(MUTEX|SEMAPHORE|SPINLOCK)\((\w*)/\2/v/'	\
@@ -260,6 +263,9 @@
 	--regex='/SETPCGFLAG\(([^,)]*).*/SetPageCgroup\1/'	\
 	--regex='/CLEARPCGFLAG\(([^,)]*).*/ClearPageCgroup\1/'	\
 	--regex='/TESTCLEARPCGFLAG\(([^,)]*).*/TestClearPageCgroup\1/' \
+	--regex='/TASK_PFA_TEST\([^,]*,\s*([^)]*)\)/task_\1/'		\
+	--regex='/TASK_PFA_SET\([^,]*,\s*([^)]*)\)/task_set_\1/'	\
+	--regex='/TASK_PFA_CLEAR\([^,]*,\s*([^)]*)\)/task_clear_\1/'	\
 	--regex='/_PE(\([^,)]*\).*/PEVENT_ERRNO__\1/'		\
 	--regex='/PCI_OP_READ(\([a-z]*[a-z]\).*[1-4])/pci_bus_read_config_\1/' \
 	--regex='/PCI_OP_WRITE(\([a-z]*[a-z]\).*[1-4])/pci_bus_write_config_\1/'\
...	...	@@ -345,14 +345,14 @@
345	345	The implementation is simple.
346	346
347	347	Setting the flag 'cpuset.memory_spread_page' turns on a per-process flag
348		-PF_SPREAD_PAGE for each task that is in that cpuset or subsequently
	348	+PFA_SPREAD_PAGE for each task that is in that cpuset or subsequently
349	349	joins that cpuset. The page allocation calls for the page cache
350		-is modified to perform an inline check for this PF_SPREAD_PAGE task
	350	+is modified to perform an inline check for this PFA_SPREAD_PAGE task
351	351	flag, and if set, a call to a new routine cpuset_mem_spread_node()
352	352	returns the node to prefer for the allocation.
353	353
354	354	Similarly, setting 'cpuset.memory_spread_slab' turns on the flag
355		-PF_SPREAD_SLAB, and appropriately marked slab caches will allocate
	355	+PFA_SPREAD_SLAB, and appropriately marked slab caches will allocate
356	356	pages from the node returned by cpuset_mem_spread_node().
357	357
358	358	The cpuset_mem_spread_node() routine is also simple. It uses the
...	...	@@ -93,12 +93,12 @@
93	93
94	94	static inline int cpuset_do_page_mem_spread(void)
95	95	{
96		- return current->flags & PF_SPREAD_PAGE;
	96	+ return task_spread_page(current);
97	97	}
98	98
99	99	static inline int cpuset_do_slab_mem_spread(void)
100	100	{
101		- return current->flags & PF_SPREAD_SLAB;
	101	+ return task_spread_slab(current);
102	102	}
103	103
104	104	extern int current_cpuset_is_being_rebound(void);
...	...	@@ -1903,8 +1903,6 @@
1903	1903	#define PF_KTHREAD 0x00200000 /* I am a kernel thread */
1904	1904	#define PF_RANDOMIZE 0x00400000 /* randomize virtual address space */
1905	1905	#define PF_SWAPWRITE 0x00800000 /* Allowed to write to swap */
1906		-#define PF_SPREAD_PAGE 0x01000000 /* Spread page cache over cpuset */
1907		-#define PF_SPREAD_SLAB 0x02000000 /* Spread some slab caches over cpuset */
1908	1906	#define PF_NO_SETAFFINITY 0x04000000 /* Userland is not allowed to meddle with cpus_allowed */
1909	1907	#define PF_MCE_EARLY 0x08000000 /* Early kill for mce process policy */
1910	1908	#define PF_MUTEX_TESTER 0x20000000 /* Thread belongs to the rt mutex tester */
1911	1909
1912	1910
...	...	@@ -1957,17 +1955,31 @@
1957	1955	}
1958	1956
1959	1957	/* Per-process atomic flags. */
1960		-#define PFA_NO_NEW_PRIVS 0x00000001 /* May not gain new privileges. */
	1958	+#define PFA_NO_NEW_PRIVS 0 /* May not gain new privileges. */
	1959	+#define PFA_SPREAD_PAGE 1 /* Spread page cache over cpuset */
	1960	+#define PFA_SPREAD_SLAB 2 /* Spread some slab caches over cpuset */
1961	1961
1962		-static inline bool task_no_new_privs(struct task_struct *p)
1963		-{
1964		- return test_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags);
1965		-}
1966	1962
1967		-static inline void task_set_no_new_privs(struct task_struct *p)
1968		-{
1969		- set_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags);
1970		-}
	1963	+#define TASK_PFA_TEST(name, func) \
	1964	+ static inline bool task_##func(struct task_struct *p) \
	1965	+ { return test_bit(PFA_##name, &p->atomic_flags); }
	1966	+#define TASK_PFA_SET(name, func) \
	1967	+ static inline void task_set_##func(struct task_struct *p) \
	1968	+ { set_bit(PFA_##name, &p->atomic_flags); }
	1969	+#define TASK_PFA_CLEAR(name, func) \
	1970	+ static inline void task_clear_##func(struct task_struct *p) \
	1971	+ { clear_bit(PFA_##name, &p->atomic_flags); }
	1972	+
	1973	+TASK_PFA_TEST(NO_NEW_PRIVS, no_new_privs)
	1974	+TASK_PFA_SET(NO_NEW_PRIVS, no_new_privs)
	1975	+
	1976	+TASK_PFA_TEST(SPREAD_PAGE, spread_page)
	1977	+TASK_PFA_SET(SPREAD_PAGE, spread_page)
	1978	+TASK_PFA_CLEAR(SPREAD_PAGE, spread_page)
	1979	+
	1980	+TASK_PFA_TEST(SPREAD_SLAB, spread_slab)
	1981	+TASK_PFA_SET(SPREAD_SLAB, spread_slab)
	1982	+TASK_PFA_CLEAR(SPREAD_SLAB, spread_slab)
1971	1983
1972	1984	/*
1973	1985	* task->jobctl flags
...	...	@@ -365,13 +365,14 @@
365	365	struct task_struct *tsk)
366	366	{
367	367	if (is_spread_page(cs))
368		- tsk->flags \|= PF_SPREAD_PAGE;
	368	+ task_set_spread_page(tsk);
369	369	else
370		- tsk->flags &= ~PF_SPREAD_PAGE;
	370	+ task_clear_spread_page(tsk);
	371	+
371	372	if (is_spread_slab(cs))
372		- tsk->flags \|= PF_SPREAD_SLAB;
	373	+ task_set_spread_slab(tsk);
373	374	else
374		- tsk->flags &= ~PF_SPREAD_SLAB;
	375	+ task_clear_spread_slab(tsk);
375	376	}
376	377
377	378	/*
...	...	@@ -2987,7 +2987,7 @@
2987	2987
2988	2988	#ifdef CONFIG_NUMA
2989	2989	/*
2990		- * Try allocating on another node if PF_SPREAD_SLAB is a mempolicy is set.
	2990	+ * Try allocating on another node if PFA_SPREAD_SLAB is a mempolicy is set.
2991	2991	*
2992	2992	* If we are in_interrupt, then process context, including cpusets and
2993	2993	* mempolicy, may not apply and should not be used for allocation policy.
...	...	@@ -3219,7 +3219,7 @@
3219	3219	{
3220	3220	void *objp;
3221	3221
3222		- if (current->mempolicy \|\| unlikely(current->flags & PF_SPREAD_SLAB)) {
	3222	+ if (current->mempolicy \|\| cpuset_do_slab_mem_spread()) {
3223	3223	objp = alternate_node_alloc(cache, flags);
3224	3224	if (objp)
3225	3225	goto out;
...	...	@@ -197,6 +197,9 @@
197	197	--regex-c++='/SETPCGFLAG\(([^,)])./SetPageCgroup\1/' \
198	198	--regex-c++='/CLEARPCGFLAG\(([^,)])./ClearPageCgroup\1/' \
199	199	--regex-c++='/TESTCLEARPCGFLAG\(([^,)])./TestClearPageCgroup\1/' \
	200	+ --regex-c++='/TASK_PFA_TEST\([^,],\s([^)]*)\)/task_\1/' \
	201	+ --regex-c++='/TASK_PFA_SET\([^,],\s([^)]*)\)/task_set_\1/' \
	202	+ --regex-c++='/TASK_PFA_CLEAR\([^,],\s([^)]*)\)/task_clear_\1/'\
200	203	--regex-c='/PCI_OP_READ\((\w).[1-4]\)/pci_bus_read_config_\1/' \
201	204	--regex-c='/PCI_OP_WRITE\((\w).[1-4]\)/pci_bus_write_config_\1/' \
202	205	--regex-c='/DEFINE_(MUTEX\|SEMAPHORE\|SPINLOCK)\((\w*)/\2/v/' \
...	...	@@ -260,6 +263,9 @@
260	263	--regex='/SETPCGFLAG\(([^,)])./SetPageCgroup\1/' \
261	264	--regex='/CLEARPCGFLAG\(([^,)])./ClearPageCgroup\1/' \
262	265	--regex='/TESTCLEARPCGFLAG\(([^,)])./TestClearPageCgroup\1/' \
	266	+ --regex='/TASK_PFA_TEST\([^,],\s([^)]*)\)/task_\1/' \
	267	+ --regex='/TASK_PFA_SET\([^,],\s([^)]*)\)/task_set_\1/' \
	268	+ --regex='/TASK_PFA_CLEAR\([^,],\s([^)]*)\)/task_clear_\1/' \
263	269	--regex='/_PE(\([^,)]\)./PEVENT_ERRNO__\1/' \
264	270	--regex='/PCI_OP_READ(\([a-z][a-z]\).[1-4])/pci_bus_read_config_\1/' \
265	271	--regex='/PCI_OP_WRITE(\([a-z][a-z]\).[1-4])/pci_bus_write_config_\1/'\