Skip to contentSkip to conversation

incoming

PATCH New Andrew Morton 1 participant 13 messages Jan 29, 2022

Subsystems (12)
lore
Share & tools
lore

Andrew Morton

author maintainer
#0 lore raw

12 patches, based on f8c7e4ede46fe63ff10000669652648aab09d112.

Subsystems affected by this patch series:

sysctl
binfmt
ia64
mm/memory-failure
mm/folios
selftests
mm/kasan
mm/psi
ocfs2

Subsystem: sysctl

Andrew Morton <[email protected]>:
include/linux/sysctl.h: fix register_sysctl_mount_point() return type

Subsystem: binfmt

Tong Zhang <[email protected]>:
binfmt_misc: fix crash when load/unload module

Subsystem: ia64

Randy Dunlap <[email protected]>:
ia64: make IA64_MCA_RECOVERY bool instead of tristate

Subsystem: mm/memory-failure

Joao Martins <[email protected]>:
memory-failure: fetch compound_head after pgmap_pfn_valid()

Subsystem: mm/folios

Wei Yang <[email protected]>:
mm: page->mapping folio->mapping should have the same offset

Subsystem: selftests

Maor Gottlieb <[email protected]>:
tools/testing/scatterlist: add missing defines

Subsystem: mm/kasan

Marco Elver <[email protected]>:
kasan: test: fix compatibility with FORTIFY_SOURCE

Peter Collingbourne <[email protected]>:
mm, kasan: use compare-exchange operation to set KASAN page tag

Subsystem: mm/psi

Suren Baghdasaryan <[email protected]>:
psi: fix "no previous prototype" warnings when CONFIG_CGROUPS=n
psi: fix "defined but not used" warnings when CONFIG_PROC_FS=n

Subsystem: ocfs2

Joseph Qi <[email protected]>:
Patch series "ocfs2: fix a deadlock case":
jbd2: export jbd2_journal_[grab|put]_journal_head
ocfs2: fix a deadlock when commit trans

 arch/ia64/Kconfig                    |    2 
 fs/binfmt_misc.c                     |    8 +--
 fs/jbd2/journal.c                    |    2 
 fs/ocfs2/suballoc.c                  |   25 ++++-------
 include/linux/mm.h                   |   17 +++++--
 include/linux/mm_types.h             |    1 
 include/linux/psi.h                  |   11 ++--
 include/linux/sysctl.h               |    2 
 kernel/sched/psi.c                   |   79 ++++++++++++++++++-----------------
 lib/test_kasan.c                     |    5 ++
 mm/memory-failure.c                  |    6 ++
 tools/testing/scatterlist/linux/mm.h |    3 -
 12 files changed, 91 insertions(+), 70 deletions(-)
author maintainer
#1 lore raw

PATCH 1/12 include/linux/sysctl.h: fix register_sysctl_mount_point() return type

From: Andrew Morton <[email protected]>
Subject: include/linux/sysctl.h: fix register_sysctl_mount_point() return type

The CONFIG_SYSCTL=n stub returns the wrong type.

Fixes ee9efac48a082 ("sysctl: add helper to register a sysctl mount point")
Cc Tong Zhang <[email protected]>

---

 include/linux/sysctl.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/sysctl.h renamed from include/linux/sysctl.h~include-linux-sysctlh-fix-register_sysctl_mount_point-return-type +1 -1
@@ -265,7 +265,7 @@ static inline struct ctl_table_header *r
 	return NULL;
 }
 
-static inline struct sysctl_header *register_sysctl_mount_point(const char *path)
+static inline struct ctl_table_header *register_sysctl_mount_point(const char *path)
 {
 	return NULL;
 }

_

author maintainer
#2 lore raw

PATCH 2/12 binfmt_misc: fix crash when load/unload module

In reply to Andrew Morton

From: Tong Zhang <[email protected]>
Subject: binfmt_misc: fix crash when load/unload module

We should unregister the table upon module unload otherwise something horrible will happen when we load binfmt_misc module again. Also note that we should keep value returned by register_sysctl_mount_point() and release it later, otherwise it will leak.

Also, per Christian's comment, to fully restore the old behavior that won't break userspace the check(binfmt_misc_header) should be eliminated.

reproduce:
modprobe binfmt_misc
modprobe -r binfmt_misc
modprobe binfmt_misc
modprobe -r binfmt_misc
modprobe binfmt_misc

[   18.032038] Call Trace:
[   18.032108]  <TASK>
[   18.032169]  dump_stack_lvl+0x34/0x44
[   18.032273]  __register_sysctl_table+0x6f4/0x720
[   18.032397]  ? preempt_count_sub+0xf/0xb0
[   18.032508]  ? 0xffffffffc0040000
[   18.032600]  init_misc_binfmt+0x2d/0x1000 [binfmt_misc]
[   18.042520] binfmt_misc: Failed to create fs/binfmt_misc sysctl mount point
modprobe: can't load module binfmt_misc (kernel/fs/binfmt_misc.ko): Cannot allocate memory
[   18.063549] binfmt_misc: Failed to create fs/binfmt_misc sysctl mount point
[   18.204779] BUG: unable to handle page fault for address: fffffbfff8004802
Fixes 3ba442d5331f ("fs: move binfmt_misc sysctl to its own file")
Cc Eric Biederman <[email protected]>
Cc Kees Cook <[email protected]>
Cc Iurii Zaikin <[email protected]>

---

 fs/binfmt_misc.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
fs/binfmt_misc.c renamed from fs/binfmt_misc.c~binfmt_misc-fix-crash-when-load-unload-module +4 -4
@@ -817,20 +817,20 @@ static struct file_system_type bm_fs_typ
 };
 MODULE_ALIAS_FS("binfmt_misc");
 
+static struct ctl_table_header *binfmt_misc_header;
+
 static int __init init_misc_binfmt(void)
 {
 	int err = register_filesystem(&bm_fs_type);
 	if (!err)
 		insert_binfmt(&misc_format);
-	if (!register_sysctl_mount_point("fs/binfmt_misc")) {
-		pr_warn("Failed to create fs/binfmt_misc sysctl mount point");
-		return -ENOMEM;
-	}
+	binfmt_misc_header = register_sysctl_mount_point("fs/binfmt_misc");
 	return 0;
 }
 
 static void __exit exit_misc_binfmt(void)
 {
+	unregister_sysctl_table(binfmt_misc_header);
 	unregister_binfmt(&misc_format);
 	unregister_filesystem(&bm_fs_type);
 }

_

author maintainer
#3 lore raw

PATCH 3/12 ia64: make IA64_MCA_RECOVERY bool instead of tristate

In reply to Andrew Morton

From: Randy Dunlap <[email protected]>
Subject: ia64: make IA64_MCA_RECOVERY bool instead of tristate

In linux-next, IA64_MCA_RECOVERY uses the (new) function make_task_dead(), which is not exported for use by modules. Instead of exporting it for one user, convert IA64_MCA_RECOVERY to be a bool Kconfig symbol.

In a config file from "kernel test robot <[email protected]>" for a different problem, this linker error was exposed when CONFIG_IA64_MCA_RECOVERY=m.

Fixes this build error:

ERROR: modpost: "make_task_dead" [arch/ia64/kernel/mca_recovery.ko] undefined!

---

 arch/ia64/Kconfig |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/ia64/Kconfig renamed from arch/ia64/Kconfig~ia64-make-ia64_mca_recovery-bool-instead-of-tristate +1 -1
@@ -318,7 +318,7 @@ config ARCH_PROC_KCORE_TEXT
 	depends on PROC_KCORE
 
 config IA64_MCA_RECOVERY
-	tristate "MCA recovery from errors other than TLB."
+	bool "MCA recovery from errors other than TLB."
 
 config IA64_PALINFO
 	tristate "/proc/pal support"

_

author maintainer
#4 lore raw

PATCH 4/12 memory-failure: fetch compound_head after pgmap_pfn_valid()

In reply to Andrew Morton

From: Joao Martins <[email protected]>
Subject: memory-failure: fetch compound_head after pgmap_pfn_valid()

memory_failure_dev_pagemap() at the moment assumes base pages (e.g. dax_lock_page()). For devmap with compound pages fetch the compound_head in case a tail page memory failure is being handled.

Currently this is a nop, but in the advent of compound pages in dev_pagemap it allows memory_failure_dev_pagemap() to keep working.

Without this fix memory-failure handling (i.e. MCEs on pmem) with device-dax configured namespaces will regress (and crash).

---

 mm/memory-failure.c |    6 ++++++
 1 file changed, 6 insertions(+)
mm/memory-failure.c renamed from mm/memory-failure.c~memory-failure-fetch-compound_head-after-pgmap_pfn_valid +6 -0
@@ -1596,6 +1596,12 @@ static int memory_failure_dev_pagemap(un
 	}
 
 	/*
+	 * Pages instantiated by device-dax (not filesystem-dax)
+	 * may be compound pages.
+	 */
+	page = compound_head(page);
+
+	/*
 	 * Prevent the inode from being freed while we are interrogating
 	 * the address_space, typically this would be handled by
 	 * lock_page(), but dax pages do not use the page lock. This

_

author maintainer
#5 lore raw

PATCH 5/12 mm: page->mapping folio->mapping should have the same offset

In reply to Andrew Morton

From: Wei Yang <[email protected]>
Subject: mm: page->mapping folio->mapping should have the same offset

As with the other members of folio, the offset of page->mapping and folio->mapping must be the same. The compile-time check was inadvertently removed during development. Add it back.

---

 include/linux/mm_types.h |    1 +
 1 file changed, 1 insertion(+)
include/linux/mm_types.h renamed from include/linux/mm_types.h~mm-page-mapping-folio-mapping-should-have-the-same-offset +1 -0
@@ -261,6 +261,7 @@ static_assert(sizeof(struct page) == siz
 	static_assert(offsetof(struct page, pg) == offsetof(struct folio, fl))
 FOLIO_MATCH(flags, flags);
 FOLIO_MATCH(lru, lru);
+FOLIO_MATCH(mapping, mapping);
 FOLIO_MATCH(compound_head, lru);
 FOLIO_MATCH(index, index);
 FOLIO_MATCH(private, private);

_

author maintainer
#6 lore raw

PATCH 6/12 tools/testing/scatterlist: add missing defines

In reply to Andrew Morton

From: Maor Gottlieb <[email protected]>
Subject: tools/testing/scatterlist: add missing defines

The cited commits replaced preemptible with pagefault_disabled and flush_kernel_dcache_page with flush_dcache_page respectively, hence need to update the corresponding defines in the test.

scatterlist.c: In function ‘sg_miter_stop’:
scatterlist.c:919:4: warning: implicit declaration of function ‘flush_dcache_page’ [-Wimplicit-function-declaration]

    flush_dcache_page(miter->page);
    ^~~~~~~~~~~~~~~~~
In file included from ./linux/scatterlist.h:8:0,
                 from scatterlist.c:9:
scatterlist.c:922:18: warning: implicit declaration of function ‘pagefault_disabled’ [-Wimplicit-function-declaration]
    WARN_ON_ONCE(!pagefault_disabled());
                  ^
./linux/mm.h:23:25: note: in definition of macro ‘WARN_ON_ONCE’
  int __ret_warn_on = !!(condition);                      \
                         ^~~~~~~~~
Fixes 723aca208516 ("mm/scatterlist: replace the !preemptible warning in sg_miter_stop()")
Fixes 0e84f5dbf8d6 ("scatterlist: replace flush_kernel_dcache_page with flush_dcache_page")
Cc Thomas Gleixner <[email protected]>
Cc Christoph Hellwig <[email protected]>

---

 tools/testing/scatterlist/linux/mm.h |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
tools/testing/scatterlist/linux/mm.h renamed from tools/testing/scatterlist/linux/mm.h~tools-testing-scatterlist-add-missing-defines +2 -1
@@ -74,7 +74,7 @@ static inline unsigned long page_to_phys
 	      __UNIQUE_ID(min1_), __UNIQUE_ID(min2_),   \
 	      x, y)
 
-#define preemptible() (1)
+#define pagefault_disabled() (0)
 
 static inline void *kmap(struct page *page)
 {
@@ -127,6 +127,7 @@ kmalloc_array(unsigned int n, unsigned i
 #define kmemleak_free(a)
 
 #define PageSlab(p) (0)
+#define flush_dcache_page(p)
 
 #define MAX_ERRNO	4095
 

_

author maintainer
#7 lore raw

PATCH 7/12 kasan: test: fix compatibility with FORTIFY_SOURCE

In reply to Andrew Morton

From: Marco Elver <[email protected]>
Subject: kasan: test: fix compatibility with FORTIFY_SOURCE

With CONFIG_FORTIFY_SOURCE enabled, string functions will also perform dynamic checks using __builtin_object_size(ptr), which when failed will panic the kernel.

Because the KASAN test deliberately performs out-of-bounds operations, the kernel panics with FORTIFY_SOURCE, for example:

 | kernel BUG at lib/string_helpers.c:910!
 | invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI
 | CPU: 1 PID: 137 Comm: kunit_try_catch Tainted: G    B             5.16.0-rc3+ #3
 | Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-2 04/01/2014
 | RIP: 0010:fortify_panic+0x19/0x1b
 | ...
 | Call Trace:
 |  <TASK>
 |  kmalloc_oob_in_memset.cold+0x16/0x16
 |  ...

Fix it by also hiding ptr from the optimizer, which will ensure that __builtin_object_size() does not return a valid size, preventing fortified string functions from panicking.

---

 lib/test_kasan.c |    5 +++++
 1 file changed, 5 insertions(+)
lib/test_kasan.c renamed from lib/test_kasan.c~kasan-test-fix-compatibility-with-fortify_source +5 -0
@@ -492,6 +492,7 @@ static void kmalloc_oob_in_memset(struct
 	ptr = kmalloc(size, GFP_KERNEL);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
+	OPTIMIZER_HIDE_VAR(ptr);
 	OPTIMIZER_HIDE_VAR(size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 				memset(ptr, 0, size + KASAN_GRANULE_SIZE));
@@ -515,6 +516,7 @@ static void kmalloc_memmove_negative_siz
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
 	memset((char *)ptr, 0, 64);
+	OPTIMIZER_HIDE_VAR(ptr);
 	OPTIMIZER_HIDE_VAR(invalid_size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		memmove((char *)ptr, (char *)ptr + 4, invalid_size));
@@ -531,6 +533,7 @@ static void kmalloc_memmove_invalid_size
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
 	memset((char *)ptr, 0, 64);
+	OPTIMIZER_HIDE_VAR(ptr);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		memmove((char *)ptr, (char *)ptr + 4, invalid_size));
 	kfree(ptr);
@@ -893,6 +896,7 @@ static void kasan_memchr(struct kunit *t
 	ptr = kmalloc(size, GFP_KERNEL | __GFP_ZERO);
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 
+	OPTIMIZER_HIDE_VAR(ptr);
 	OPTIMIZER_HIDE_VAR(size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		kasan_ptr_result = memchr(ptr, '1', size + 1));
@@ -919,6 +923,7 @@ static void kasan_memcmp(struct kunit *t
 	KUNIT_ASSERT_NOT_ERR_OR_NULL(test, ptr);
 	memset(arr, 0, sizeof(arr));
 
+	OPTIMIZER_HIDE_VAR(ptr);
 	OPTIMIZER_HIDE_VAR(size);
 	KUNIT_EXPECT_KASAN_FAIL(test,
 		kasan_int_result = memcmp(ptr, arr, size+1));

_

author maintainer
#8 lore raw

PATCH 8/12 mm, kasan: use compare-exchange operation to set KASAN page tag

In reply to Andrew Morton

From: Peter Collingbourne <[email protected]>
Subject: mm, kasan: use compare-exchange operation to set KASAN page tag

It has been reported that the tag setting operation on newly-allocated pages can cause the page flags to be corrupted when performed concurrently with other flag updates as a result of the use of non-atomic operations. Fix the problem by using a compare-exchange loop to update the tag.

---

 include/linux/mm.h |   17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
include/linux/mm.h renamed from include/linux/mm.h~mm-use-compare-exchange-operation-to-set-kasan-page-tag +12 -5
@@ -1506,11 +1506,18 @@ static inline u8 page_kasan_tag(const st
 
 static inline void page_kasan_tag_set(struct page *page, u8 tag)
 {
-	if (kasan_enabled()) {
-		tag ^= 0xff;
-		page->flags &= ~(KASAN_TAG_MASK << KASAN_TAG_PGSHIFT);
-		page->flags |= (tag & KASAN_TAG_MASK) << KASAN_TAG_PGSHIFT;
-	}
+	unsigned long old_flags, flags;
+
+	if (!kasan_enabled())
+		return;
+
+	tag ^= 0xff;
+	old_flags = READ_ONCE(page->flags);
+	do {
+		flags = old_flags;
+		flags &= ~(KASAN_TAG_MASK << KASAN_TAG_PGSHIFT);
+		flags |= (tag & KASAN_TAG_MASK) << KASAN_TAG_PGSHIFT;
+	} while (unlikely(!try_cmpxchg(&page->flags, &old_flags, flags)));
 }
 
 static inline void page_kasan_tag_reset(struct page *page)

_

author maintainer
#9 lore raw

PATCH 9/12 psi: fix "no previous prototype" warnings when CONFIG_CGROUPS=n

In reply to Andrew Morton

From: Suren Baghdasaryan <[email protected]>
Subject: psi: fix "no previous prototype" warnings when CONFIG_CGROUPS=n

When CONFIG_CGROUPS is disabled psi code generates the following warnings:

kernel/sched/psi.c:1112:21: warning: no previous prototype for 'psi_trigger_create' [-Wmissing-prototypes]
1112 | struct psi_trigger *psi_trigger_create(struct psi_group *group,
| ^~~~~~~~~~~~~~~~~~
kernel/sched/psi.c:1182:6: warning: no previous prototype for 'psi_trigger_destroy' [-Wmissing-prototypes]
1182 | void psi_trigger_destroy(struct psi_trigger *t)
| ^~~~~~~~~~~~~~~~~~~
kernel/sched/psi.c:1249:10: warning: no previous prototype for 'psi_trigger_poll' [-Wmissing-prototypes]

    1249 | __poll_t psi_trigger_poll(void **trigger_ptr,
         |          ^~~~~~~~~~~~~~~~

Change declarations of these functions in the header to provide the prototypes even when they are unused.

---

 include/linux/psi.h |   11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
include/linux/psi.h renamed from include/linux/psi.h~psi-fix-no-previous-prototype-warnings-when-config_cgroups=n +5 -6
@@ -25,18 +25,17 @@ void psi_memstall_enter(unsigned long *f
 void psi_memstall_leave(unsigned long *flags);
 
 int psi_show(struct seq_file *s, struct psi_group *group, enum psi_res res);
-
-#ifdef CONFIG_CGROUPS
-int psi_cgroup_alloc(struct cgroup *cgrp);
-void psi_cgroup_free(struct cgroup *cgrp);
-void cgroup_move_task(struct task_struct *p, struct css_set *to);
-
 struct psi_trigger *psi_trigger_create(struct psi_group *group,
 			char *buf, size_t nbytes, enum psi_res res);
 void psi_trigger_destroy(struct psi_trigger *t);
 
 __poll_t psi_trigger_poll(void **trigger_ptr, struct file *file,
 			poll_table *wait);
+
+#ifdef CONFIG_CGROUPS
+int psi_cgroup_alloc(struct cgroup *cgrp);
+void psi_cgroup_free(struct cgroup *cgrp);
+void cgroup_move_task(struct task_struct *p, struct css_set *to);
 #endif
 
 #else /* CONFIG_PSI */

_

author maintainer
#10 lore raw

PATCH 10/12 psi: fix "defined but not used" warnings when CONFIG_PROC_FS=n

In reply to Andrew Morton

From: Suren Baghdasaryan <[email protected]>
Subject: psi: fix "defined but not used" warnings when CONFIG_PROC_FS=n

When CONFIG_PROC_FS is disabled psi code generates the following warnings:

kernel/sched/psi.c:1364:30: warning: 'psi_cpu_proc_ops' defined but not used [-Wunused-const-variable=]

    1364 | static const struct proc_ops psi_cpu_proc_ops = {
         |                              ^~~~~~~~~~~~~~~~
kernel/sched/psi.c:1355:30: warning: 'psi_memory_proc_ops' defined but not used [-Wunused-const-variable=]
    1355 | static const struct proc_ops psi_memory_proc_ops = {
         |                              ^~~~~~~~~~~~~~~~~~~
kernel/sched/psi.c:1346:30: warning: 'psi_io_proc_ops' defined but not used [-Wunused-const-variable=]
    1346 | static const struct proc_ops psi_io_proc_ops = {
         |                              ^~~~~~~~~~~~~~~

Make definitions of these structures and related functions conditional on CONFIG_PROC_FS config.

---

 kernel/sched/psi.c |   79 ++++++++++++++++++++++---------------------
 1 file changed, 41 insertions(+), 38 deletions(-)
kernel/sched/psi.c renamed from kernel/sched/psi.c~psi-fix-defined-but-not-used-warnings-when-config_proc_fs=n +41 -38
@@ -1082,44 +1082,6 @@ int psi_show(struct seq_file *m, struct
 	return 0;
 }
 
-static int psi_io_show(struct seq_file *m, void *v)
-{
-	return psi_show(m, &psi_system, PSI_IO);
-}
-
-static int psi_memory_show(struct seq_file *m, void *v)
-{
-	return psi_show(m, &psi_system, PSI_MEM);
-}
-
-static int psi_cpu_show(struct seq_file *m, void *v)
-{
-	return psi_show(m, &psi_system, PSI_CPU);
-}
-
-static int psi_open(struct file *file, int (*psi_show)(struct seq_file *, void *))
-{
-	if (file->f_mode & FMODE_WRITE && !capable(CAP_SYS_RESOURCE))
-		return -EPERM;
-
-	return single_open(file, psi_show, NULL);
-}
-
-static int psi_io_open(struct inode *inode, struct file *file)
-{
-	return psi_open(file, psi_io_show);
-}
-
-static int psi_memory_open(struct inode *inode, struct file *file)
-{
-	return psi_open(file, psi_memory_show);
-}
-
-static int psi_cpu_open(struct inode *inode, struct file *file)
-{
-	return psi_open(file, psi_cpu_show);
-}
-
 struct psi_trigger *psi_trigger_create(struct psi_group *group,
 			char *buf, size_t nbytes, enum psi_res res)
 {
@@ -1278,6 +1240,45 @@ __poll_t psi_trigger_poll(void **trigger
 	return ret;
 }
 
+#ifdef CONFIG_PROC_FS
+static int psi_io_show(struct seq_file *m, void *v)
+{
+	return psi_show(m, &psi_system, PSI_IO);
+}
+
+static int psi_memory_show(struct seq_file *m, void *v)
+{
+	return psi_show(m, &psi_system, PSI_MEM);
+}
+
+static int psi_cpu_show(struct seq_file *m, void *v)
+{
+	return psi_show(m, &psi_system, PSI_CPU);
+}
+
+static int psi_open(struct file *file, int (*psi_show)(struct seq_file *, void *))
+{
+	if (file->f_mode & FMODE_WRITE && !capable(CAP_SYS_RESOURCE))
+		return -EPERM;
+
+	return single_open(file, psi_show, NULL);
+}
+
+static int psi_io_open(struct inode *inode, struct file *file)
+{
+	return psi_open(file, psi_io_show);
+}
+
+static int psi_memory_open(struct inode *inode, struct file *file)
+{
+	return psi_open(file, psi_memory_show);
+}
+
+static int psi_cpu_open(struct inode *inode, struct file *file)
+{
+	return psi_open(file, psi_cpu_show);
+}
+
 static ssize_t psi_write(struct file *file, const char __user *user_buf,
 			 size_t nbytes, enum psi_res res)
 {
@@ -1392,3 +1393,5 @@ static int __init psi_proc_init(void)
 	return 0;
 }
 module_init(psi_proc_init);
+
+#endif /* CONFIG_PROC_FS */

_

author maintainer
#11 lore raw

PATCH 11/12 jbd2: export jbd2_journal_[grab|put]_journal_head

In reply to Andrew Morton

From: Joseph Qi <[email protected]>
Subject: jbd2: export jbd2_journal_[grab|put]_journal_head

Patch series "ocfs2: fix a deadlock case".

This fixes a deadlock case in ocfs2. We firstly export jbd2 symbols jbd2_journal_[grab|put]_journal_head as preparation and later use them in ocfs2 insread of jbd_[lock|unlock]_bh_journal_head to fix the deadlock.

This patch (of 2):

This exports symbols jbd2_journal_[grab|put]_journal_head, which will be used outside modules, e.g. ocfs2.

Cc Mark Fasheh <[email protected]>
Cc Joel Becker <[email protected]>
Cc Junxiao Bi <[email protected]>
Cc Changwei Ge <[email protected]>
Cc Gang He <[email protected]>
Cc Jun Piao <[email protected]>
Cc Andreas Dilger <[email protected]>
Cc Gautham Ananthakrishna <[email protected]>
Cc Saeed Mirzamohammadi <[email protected]>
Cc "Theodore Ts'o" <[email protected]>

---

 fs/jbd2/journal.c |    2 ++
 1 file changed, 2 insertions(+)
fs/jbd2/journal.c renamed from fs/jbd2/journal.c~jbd2-export-jbd2_journal__journal_head +2 -0
@@ -2972,6 +2972,7 @@ struct journal_head *jbd2_journal_grab_j
 	jbd_unlock_bh_journal_head(bh);
 	return jh;
 }
+EXPORT_SYMBOL(jbd2_journal_grab_journal_head);
 
 static void __journal_remove_journal_head(struct buffer_head *bh)
 {
@@ -3024,6 +3025,7 @@ void jbd2_journal_put_journal_head(struc
 		jbd_unlock_bh_journal_head(bh);
 	}
 }
+EXPORT_SYMBOL(jbd2_journal_put_journal_head);
 
 /*
  * Initialize jbd inode head

_

author maintainer
#12 lore raw

PATCH 12/12 ocfs2: fix a deadlock when commit trans

In reply to Andrew Morton

From: Joseph Qi <[email protected]>
Subject: ocfs2: fix a deadlock when commit trans

commit 6f1b228529ae introduces a regression which can deadlock as follows:

Task1:                              Task2:
jbd2_journal_commit_transaction     ocfs2_test_bg_bit_allocatable
spin_lock(&jh->b_state_lock)        jbd_lock_bh_journal_head
__jbd2_journal_remove_checkpoint    spin_lock(&jh->b_state_lock)

jbd2_journal_put_journal_head
jbd_lock_bh_journal_head

Task1 and Task2 lock bh->b_state and jh->b_state_lock in different order, which finally result in a deadlock.

So use jbd2_journal_[grab|put]_journal_head instead in ocfs2_test_bg_bit_allocatable() to fix it.

Fixes 6f1b228529ae ("ocfs2: fix race between searching chunks and release journal_head from buffer_head")
Cc "Theodore Ts'o" <[email protected]>
Cc Andreas Dilger <[email protected]>
Cc Changwei Ge <[email protected]>
Cc Gang He <[email protected]>
Cc Joel Becker <[email protected]>
Cc Jun Piao <[email protected]>
Cc Junxiao Bi <[email protected]>
Cc Mark Fasheh <[email protected]>

---

 fs/ocfs2/suballoc.c |   25 +++++++++++--------------
 1 file changed, 11 insertions(+), 14 deletions(-)
fs/ocfs2/suballoc.c renamed from fs/ocfs2/suballoc.c~ocfs2-fix-a-deadlock-when-commit-trans +11 -14
@@ -1251,26 +1251,23 @@ static int ocfs2_test_bg_bit_allocatable
 {
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
 	struct journal_head *jh;
-	int ret = 1;
+	int ret;
 
 	if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
 		return 0;
 
-	if (!buffer_jbd(bg_bh))
+	jh = jbd2_journal_grab_journal_head(bg_bh);
+	if (!jh)
 		return 1;
 
-	jbd_lock_bh_journal_head(bg_bh);
-	if (buffer_jbd(bg_bh)) {
-		jh = bh2jh(bg_bh);
-		spin_lock(&jh->b_state_lock);
-		bg = (struct ocfs2_group_desc *) jh->b_committed_data;
-		if (bg)
-			ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
-		else
-			ret = 1;
-		spin_unlock(&jh->b_state_lock);
-	}
-	jbd_unlock_bh_journal_head(bg_bh);
+	spin_lock(&jh->b_state_lock);
+	bg = (struct ocfs2_group_desc *) jh->b_committed_data;
+	if (bg)
+		ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+	else
+		ret = 1;
+	spin_unlock(&jh->b_state_lock);
+	jbd2_journal_put_journal_head(jh);
 
 	return ret;
 }

_