Skip to contentSkip to conversation

KASAN: null-ptr-deref Write in binder_update_page_range

DISCUSSION Dae R. Jeong 4 participants 7 messages Aug 22, 2018 – Sep 12, 2018 Android drivers

lore
Share & tools
lore
#0 lore raw

Reporting the crash: KASAN: null-ptr-deref Write in binder_update_page_range

This crash has been found in v4.18-rc3 using RaceFuzzer (a modified version of Syzkaller), which we describe more at the end of this report.

Our analysis shows that the race occurs when invoking two syscalls concurrently, mmap$binder() and ioctl$BINDER_WRITE_READ. More specifically, since two code lines alloc->vma = vma; and alloc->vma_vm_mm = vma->vm_mm; in binder_alloc_mmap_handler() is not an atomic operation during mmap$binder() syscall, there is a time window that alloc->vma is assigned but alloc->vma_vm_mm isn't assigned. It causes the null pointer dereference in binder_alloc_new_buf_locked() since it checks whether alloc->vma is NULL, but it doesn't check that alloc->vma_vm_mm is NULL. More details on the thread interleaving and the crash log are follows.

Thread interleaving:
CPU0 (binder_alloc_mmap_handler) CPU1 (binder_alloc_new_buf_locked)
===== =====
// drivers/android/binder_alloc.c
// #L718 (v4.18-rc3)

alloc->vma = vma;
                                              // drivers/android/binder_alloc.c
                                              // #L346 (v4.18-rc3)
                                              if (alloc->vma == NULL) {
                                                  ...
                                                  // alloc->vma is not NULL at this point
                                                  return ERR_PTR(-ESRCH);
                                              }
                                              ...
                                              // #L438
                                              binder_update_page_range(alloc, 0,
                                                      (void *)PAGE_ALIGN((uintptr_t)buffer->data),
                                                      end_page_addr);

                                              // In binder_update_page_range() #L218
                                              // But still alloc->vma_vm_mm is NULL here
                                              if (need_mm && mmget_not_zero(alloc->vma_vm_mm))

alloc->vma_vm_mm = vma->vm_mm;

Crash Log:
==================================================================

BUG: KASAN: null-ptr-deref in __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
BUG: KASAN: null-ptr-deref in atomic_add_unless include/linux/atomic.h:533 [inline]
BUG: KASAN: null-ptr-deref in mmget_not_zero include/linux/sched/mm.h:75 [inline]
BUG: KASAN: null-ptr-deref in binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
Write of size 4 at addr 0000000000000058 by task syz-executor0/11184

CPU: 1 PID: 11184 Comm: syz-executor0 Not tainted 4.18.0-rc3 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x16e/0x22c lib/dump_stack.c:113
 kasan_report_error mm/kasan/report.c:352 [inline]
 kasan_report+0x163/0x380 mm/kasan/report.c:412
 check_memory_region_inline mm/kasan/kasan.c:260 [inline]
 check_memory_region+0x140/0x1a0 mm/kasan/kasan.c:267
 kasan_check_write+0x14/0x20 mm/kasan/kasan.c:278
 __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
 atomic_add_unless include/linux/atomic.h:533 [inline]
 mmget_not_zero include/linux/sched/mm.h:75 [inline]
 binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
 binder_alloc_new_buf_locked drivers/android/binder_alloc.c:443 [inline]
 binder_alloc_new_buf+0x467/0xc30 drivers/android/binder_alloc.c:513
 binder_transaction+0x125b/0x4fb0 drivers/android/binder.c:2957
 binder_thread_write+0xc08/0x2770 drivers/android/binder.c:3528
 binder_ioctl_write_read.isra.39+0x24f/0x8e0 drivers/android/binder.c:4456
 binder_ioctl+0xa86/0xf34 drivers/android/binder.c:4596
 vfs_ioctl fs/ioctl.c:46 [inline]
 do_vfs_ioctl+0x154/0xd40 fs/ioctl.c:686
 ksys_ioctl+0x94/0xb0 fs/ioctl.c:701
 __do_sys_ioctl fs/ioctl.c:708 [inline]
 __se_sys_ioctl fs/ioctl.c:706 [inline]
 __x64_sys_ioctl+0x43/0x50 fs/ioctl.c:706
 do_syscall_64+0x167/0x4b0 arch/x86/entry/common.c:290
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x456469
Code: 1d ba fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 eb b9 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007f575f268b28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 000000000072bfa0 RCX: 0000000000456469
RDX: 00000000200003c0 RSI: 00000000c0306201 RDI: 0000000000000016
RBP: 00000000000001a2 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f575f2696d4
R13: 00000000ffffffff R14: 00000000006f77d0 R15: 0000000000000000
==================================================================

= About RaceFuzzer

RaceFuzzer is a customized version of Syzkaller, specifically tailored to find race condition bugs in the Linux kernel. While we leverage many different technique, the notable feature of RaceFuzzer is in leveraging a custom hypervisor (QEMU/KVM) to interleave the scheduling. In particular, we modified the hypervisor to intentionally stall a per-core execution, which is similar to supporting per-core breakpoint functionality. This allows RaceFuzzer to force the kernel to deterministically trigger racy condition (which may rarely happen in practice due to randomness in scheduling).

RaceFuzzer's C repro always pinpoints two racy syscalls. Since C repro's scheduling synchronization should be performed at the user space, its reproducibility is limited (reproduction may take from 1 second to 10 minutes (or even more), depending on a bug). This is because, while RaceFuzzer precisely interleaves the scheduling at the kernel's instruction level when finding this bug, C repro cannot fully utilize such a feature. Please disregard all code related to "should_hypercall" in the C repro, as this is only for our debugging purposes using our own hypervisor.

#1 lore raw

Hi,

Quoted 101 lines · Dae R. Jeong

Reporting the crash: KASAN: null-ptr-deref Write in binder_update_page_range

This crash has been found in v4.18-rc3 using RaceFuzzer (a modified version of Syzkaller), which we describe more at the end of this report.

Our analysis shows that the race occurs when invoking two syscalls concurrently, mmap$binder() and ioctl$BINDER_WRITE_READ. More specifically, since two code lines alloc->vma = vma; and alloc->vma_vm_mm = vma->vm_mm; in binder_alloc_mmap_handler() is not an atomic operation during mmap$binder() syscall, there is a time window that alloc->vma is assigned but alloc->vma_vm_mm isn't assigned. It causes the null pointer dereference in binder_alloc_new_buf_locked() since it checks whether alloc->vma is NULL, but it doesn't check that alloc->vma_vm_mm is NULL. More details on the thread interleaving and the crash log are follows.

Thread interleaving:
CPU0 (binder_alloc_mmap_handler) CPU1 (binder_alloc_new_buf_locked)
===== =====
// drivers/android/binder_alloc.c
// #L718 (v4.18-rc3)

alloc->vma = vma;
                                              // drivers/android/binder_alloc.c
                                              // #L346 (v4.18-rc3)
                                              if (alloc->vma == NULL) {
                                                  ...
                                                  // alloc->vma is not NULL at this point
                                                  return ERR_PTR(-ESRCH);
                                              }
                                              ...
                                              // #L438
                                              binder_update_page_range(alloc, 0,
                                                      (void *)PAGE_ALIGN((uintptr_t)buffer->data),
                                                      end_page_addr);

                                              // In binder_update_page_range() #L218
                                              // But still alloc->vma_vm_mm is NULL here
                                              if (need_mm && mmget_not_zero(alloc->vma_vm_mm))

alloc->vma_vm_mm = vma->vm_mm;

Crash Log:
==================================================================

BUG: KASAN: null-ptr-deref in __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
BUG: KASAN: null-ptr-deref in atomic_add_unless include/linux/atomic.h:533 [inline]
BUG: KASAN: null-ptr-deref in mmget_not_zero include/linux/sched/mm.h:75 [inline]
BUG: KASAN: null-ptr-deref in binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
Write of size 4 at addr 0000000000000058 by task syz-executor0/11184

CPU: 1 PID: 11184 Comm: syz-executor0 Not tainted 4.18.0-rc3 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x16e/0x22c lib/dump_stack.c:113
 kasan_report_error mm/kasan/report.c:352 [inline]
 kasan_report+0x163/0x380 mm/kasan/report.c:412
 check_memory_region_inline mm/kasan/kasan.c:260 [inline]
 check_memory_region+0x140/0x1a0 mm/kasan/kasan.c:267
 kasan_check_write+0x14/0x20 mm/kasan/kasan.c:278
 __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
 atomic_add_unless include/linux/atomic.h:533 [inline]
 mmget_not_zero include/linux/sched/mm.h:75 [inline]
 binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
 binder_alloc_new_buf_locked drivers/android/binder_alloc.c:443 [inline]
 binder_alloc_new_buf+0x467/0xc30 drivers/android/binder_alloc.c:513
 binder_transaction+0x125b/0x4fb0 drivers/android/binder.c:2957
 binder_thread_write+0xc08/0x2770 drivers/android/binder.c:3528
 binder_ioctl_write_read.isra.39+0x24f/0x8e0 drivers/android/binder.c:4456
 binder_ioctl+0xa86/0xf34 drivers/android/binder.c:4596
 vfs_ioctl fs/ioctl.c:46 [inline]
 do_vfs_ioctl+0x154/0xd40 fs/ioctl.c:686
 ksys_ioctl+0x94/0xb0 fs/ioctl.c:701
 __do_sys_ioctl fs/ioctl.c:708 [inline]
 __se_sys_ioctl fs/ioctl.c:706 [inline]
 __x64_sys_ioctl+0x43/0x50 fs/ioctl.c:706
 do_syscall_64+0x167/0x4b0 arch/x86/entry/common.c:290
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x456469
Code: 1d ba fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 eb b9 fb ff c3 66 2e 0f 1f 84 00 00 00 00
RSP: 002b:00007f575f268b28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 000000000072bfa0 RCX: 0000000000456469
RDX: 00000000200003c0 RSI: 00000000c0306201 RDI: 0000000000000016
RBP: 00000000000001a2 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007f575f2696d4
R13: 00000000ffffffff R14: 00000000006f77d0 R15: 0000000000000000
==================================================================

= About RaceFuzzer

RaceFuzzer is a customized version of Syzkaller, specifically tailored to find race condition bugs in the Linux kernel. While we leverage many different technique, the notable feature of RaceFuzzer is in leveraging a custom hypervisor (QEMU/KVM) to interleave the scheduling. In particular, we modified the hypervisor to intentionally stall a per-core execution, which is similar to supporting per-core breakpoint functionality. This allows RaceFuzzer to force the kernel to deterministically trigger racy condition (which may rarely happen in practice due to randomness in scheduling).

Quoted 9 lines

RaceFuzzer's C repro always pinpoints two racy syscalls. Since C repro's scheduling synchronization should be performed at the user space, its reproducibility is limited (reproduction may take from 1 second to 10 minutes (or even more), depending on a bug). This is because, while RaceFuzzer precisely interleaves the scheduling at the kernel's instruction level when finding this bug, C repro cannot fully utilize such a feature. Please disregard all code related to "should_hypercall" in the C repro, as this is only for our debugging purposes using our own hypervisor.

What a amazing tool!

Could you test this patch? I found that bug a month ago but didn't submit yet.

From 03fb1de4784a0ae34ddd60ab0706ec43b7dfaf8d Mon Sep 17 00:00:00 2001
From: Minchan Kim <[email protected]>
Date: Thu, 23 Aug 2018 14:14:13 +0900
Subject: [PATCH] android: binder: fix the race mmap and alloc_new_buf_locked

There is RaceFuzzer report like below because we have no lock to close below the race between binder_mmap and binder_alloc_new_buf_locked. To close the race, let's use memory barrier so that if someone see alloc->vma is not NULL, alloc->vma_vm_mm should be never NULL.

(I didn't add stable mark intentionallybecause standard android userspace libraries that interact with binder (libbinder & libhwbinder) prevent the mmap/ioctl race. - from Todd)

"
Thread interleaving:
CPU0 (binder_alloc_mmap_handler) CPU1 (binder_alloc_new_buf_locked)
===== =====
// drivers/android/binder_alloc.c
// #L718 (v4.18-rc3)

alloc->vma = vma;
                                              // drivers/android/binder_alloc.c
                                              // #L346 (v4.18-rc3)
                                              if (alloc->vma == NULL) {
                                                  ...
                                                  // alloc->vma is not NULL at this point
                                                  return ERR_PTR(-ESRCH);
                                              }
                                              ...
                                              // #L438
                                              binder_update_page_range(alloc, 0,
                                                      (void *)PAGE_ALIGN((uintptr_t)buffer->data),
                                                      end_page_addr);

                                              // In binder_update_page_range() #L218
                                              // But still alloc->vma_vm_mm is NULL here
                                              if (need_mm && mmget_not_zero(alloc->vma_vm_mm))

alloc->vma_vm_mm = vma->vm_mm;

Crash Log:
==================================================================

BUG: KASAN: null-ptr-deref in __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
BUG: KASAN: null-ptr-deref in atomic_add_unless include/linux/atomic.h:533 [inline]
BUG: KASAN: null-ptr-deref in mmget_not_zero include/linux/sched/mm.h:75 [inline]
BUG: KASAN: null-ptr-deref in binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
Write of size 4 at addr 0000000000000058 by task syz-executor0/11184

CPU: 1 PID: 11184 Comm: syz-executor0 Not tainted 4.18.0-rc3 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x16e/0x22c lib/dump_stack.c:113
 kasan_report_error mm/kasan/report.c:352 [inline]
 kasan_report+0x163/0x380 mm/kasan/report.c:412
 check_memory_region_inline mm/kasan/kasan.c:260 [inline]
 check_memory_region+0x140/0x1a0 mm/kasan/kasan.c:267
 kasan_check_write+0x14/0x20 mm/kasan/kasan.c:278
 __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
 atomic_add_unless include/linux/atomic.h:533 [inline]
 mmget_not_zero include/linux/sched/mm.h:75 [inline]
 binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
 binder_alloc_new_buf_locked drivers/android/binder_alloc.c:443 [inline]
 binder_alloc_new_buf+0x467/0xc30 drivers/android/binder_alloc.c:513
 binder_transaction+0x125b/0x4fb0 drivers/android/binder.c:2957
 binder_thread_write+0xc08/0x2770 drivers/android/binder.c:3528
 binder_ioctl_write_read.isra.39+0x24f/0x8e0 drivers/android/binder.c:4456
 binder_ioctl+0xa86/0xf34 drivers/android/binder.c:4596
 vfs_ioctl fs/ioctl.c:46 [inline]
 do_vfs_ioctl+0x154/0xd40 fs/ioctl.c:686
 ksys_ioctl+0x94/0xb0 fs/ioctl.c:701
 __do_sys_ioctl fs/ioctl.c:708 [inline]
 __se_sys_ioctl fs/ioctl.c:706 [inline]
 __x64_sys_ioctl+0x43/0x50 fs/ioctl.c:706
 do_syscall_64+0x167/0x4b0 arch/x86/entry/common.c:290
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

"

---
 drivers/android/binder_alloc.c | 43 +++++++++++++++++++++++++++-------
 1 file changed, 35 insertions(+), 8 deletions(-)
drivers/android/binder_alloc.c +35 -8
index 3f3b7b253445..64fd96eada31 100644
@@ -332,6 +332,35 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate,
 	return vma ? -ENOMEM : -ESRCH;
 }
 
+
+static inline void binder_alloc_set_vma(struct binder_alloc *alloc,
+		struct vm_area_struct *vma)
+{
+	if (vma)
+		alloc->vma_vm_mm = vma->vm_mm;
+	/*
+	 * If we see alloc->vma is not NULL, buffer data structures set up
+	 * completely. Look at smp_rmb side binder_alloc_get_vma.
+	 * We also want to guarantee new alloc->vma_vm_mm is always visible
+	 * if alloc->vma is set.
+	 */
+	smp_wmb();
+	alloc->vma = vma;
+}
+
+static inline struct vm_area_struct *binder_alloc_get_vma(
+		struct binder_alloc *alloc)
+{
+	struct vm_area_struct *vma = NULL;
+
+	if (alloc->vma) {
+		/* Look at description in binder_alloc_set_vma */
+		smp_rmb();
+		vma = alloc->vma;
+	}
+	return vma;
+}
+
 static struct binder_buffer *binder_alloc_new_buf_locked(
 				struct binder_alloc *alloc,
 				size_t data_size,
@@ -348,7 +377,7 @@ static struct binder_buffer *binder_alloc_new_buf_locked(
 	size_t size, data_offsets_size;
 	int ret;
 
-	if (alloc->vma == NULL) {
+	if (!binder_alloc_get_vma(alloc)) {
 		binder_alloc_debug(BINDER_DEBUG_USER_ERROR,
 				   "%d: binder_alloc_buf, no vma\n",
 				   alloc->pid);
@@ -723,9 +752,7 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc,
 	buffer->free = 1;
 	binder_insert_free_buffer(alloc, buffer);
 	alloc->free_async_space = alloc->buffer_size / 2;
-	barrier();
-	alloc->vma = vma;
-	alloc->vma_vm_mm = vma->vm_mm;
+	binder_alloc_set_vma(alloc, vma);
 	mmgrab(alloc->vma_vm_mm);
 
 	return 0;
@@ -754,10 +781,10 @@ void binder_alloc_deferred_release(struct binder_alloc *alloc)
 	int buffers, page_count;
 	struct binder_buffer *buffer;
 
-	BUG_ON(alloc->vma);
-
 	buffers = 0;
 	mutex_lock(&alloc->mutex);
+	BUG_ON(alloc->vma);
+
 	while ((n = rb_first(&alloc->allocated_buffers))) {
 		buffer = rb_entry(n, struct binder_buffer, rb_node);
 
@@ -900,7 +927,7 @@ int binder_alloc_get_allocated_count(struct binder_alloc *alloc)
  */
 void binder_alloc_vma_close(struct binder_alloc *alloc)
 {
-	WRITE_ONCE(alloc->vma, NULL);
+	binder_alloc_set_vma(alloc, NULL);
 }
 
 /**
@@ -935,7 +962,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
 
 	index = page - alloc->pages;
 	page_addr = (uintptr_t)alloc->buffer + index * PAGE_SIZE;
-	vma = alloc->vma;
+	vma = binder_alloc_get_vma(alloc);
 	if (vma) {
 		if (!mmget_not_zero(alloc->vma_vm_mm))
 			goto err_mmget;
Signature · 1 line
2.18.0.1017.ga543ac7ca45-goog
#2 lore raw
Quoted 2 lines

Could you test this patch? I found that bug a month ago but didn't submit yet.

I don't have a reproducer now. I manually analzed a root cause of the crash using a fuzzer's log. The log reported a race on 'alloc->vma'. Because I don't have a reproducer, I can't test the patch. I'm sorry.

#3 lore raw
Quoted 6 lines · Dae R. Jeong
Quoted 2 lines

Could you test this patch? I found that bug a month ago but didn't submit yet.

I don't have a reproducer now. I manually analzed a root cause of the crash using a fuzzer's log. The log reported a race on 'alloc->vma'. Because I don't have a reproducer, I can't test the patch. I'm sorry.

Ah, Okay. Anyway, one of author for the patch is already binder maintainer. If other maintainers don't object it, let's fix the race in this chance.

Thanks.

Reviewed-by
#4 lore raw
In reply to Minchan Kim

Thanks Minchan!

Quoted 2 lines · Minchan Kim
Quoted 101 lines
---
 drivers/android/binder_alloc.c | 43 +++++++++++++++++++++++++++-------
 1 file changed, 35 insertions(+), 8 deletions(-)
drivers/android/binder_alloc.c +35 -9
index 3f3b7b253445..64fd96eada31 100644
@@ -332,6 +332,35 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate,
        return vma ? -ENOMEM : -ESRCH;
 }
 
+
+static inline void binder_alloc_set_vma(struct binder_alloc *alloc,
+               struct vm_area_struct *vma)
+{
+       if (vma)
+               alloc->vma_vm_mm = vma->vm_mm;
+       /*
+        * If we see alloc->vma is not NULL, buffer data structures set up
+        * completely. Look at smp_rmb side binder_alloc_get_vma.
+        * We also want to guarantee new alloc->vma_vm_mm is always visible
+        * if alloc->vma is set.
+        */
+       smp_wmb();
+       alloc->vma = vma;
+}
+
+static inline struct vm_area_struct *binder_alloc_get_vma(
+               struct binder_alloc *alloc)
+{
+       struct vm_area_struct *vma = NULL;
+
+       if (alloc->vma) {
+               /* Look at description in binder_alloc_set_vma */
+               smp_rmb();
+               vma = alloc->vma;
+       }
+       return vma;
+}
+
 static struct binder_buffer *binder_alloc_new_buf_locked(
                                struct binder_alloc *alloc,
                                size_t data_size,
@@ -348,7 +377,7 @@ static struct binder_buffer *binder_alloc_new_buf_locked(
        size_t size, data_offsets_size;
        int ret;
 
-       if (alloc->vma == NULL) {
+       if (!binder_alloc_get_vma(alloc)) {
                binder_alloc_debug(BINDER_DEBUG_USER_ERROR,
                                   "%d: binder_alloc_buf, no vma\n",
                                   alloc->pid);
@@ -723,9 +752,7 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc,
        buffer->free = 1;
        binder_insert_free_buffer(alloc, buffer);
        alloc->free_async_space = alloc->buffer_size / 2;
-       barrier();
-       alloc->vma = vma;
-       alloc->vma_vm_mm = vma->vm_mm;
+       binder_alloc_set_vma(alloc, vma);
        mmgrab(alloc->vma_vm_mm);
 
        return 0;
@@ -754,10 +781,10 @@ void binder_alloc_deferred_release(struct binder_alloc *alloc)
        int buffers, page_count;
        struct binder_buffer *buffer;
 
-       BUG_ON(alloc->vma);
-
        buffers = 0;
        mutex_lock(&alloc->mutex);
+       BUG_ON(alloc->vma);
+
        while ((n = rb_first(&alloc->allocated_buffers))) {
                buffer = rb_entry(n, struct binder_buffer, rb_node);
 
@@ -900,7 +927,7 @@ int binder_alloc_get_allocated_count(struct binder_alloc *alloc)
  */
 void binder_alloc_vma_close(struct binder_alloc *alloc)
 {
-       WRITE_ONCE(alloc->vma, NULL);
+       binder_alloc_set_vma(alloc, NULL);
 }
 
 /**
@@ -935,7 +962,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
 
        index = page - alloc->pages;
        page_addr = (uintptr_t)alloc->buffer + index * PAGE_SIZE;
-       vma = alloc->vma;
+       vma = binder_alloc_get_vma(alloc);
        if (vma) {
                if (!mmget_not_zero(alloc->vma_vm_mm))
                        goto err_mmget;
--

2.18.0.1017.ga543ac7ca45-goog

#5 lore raw

Thanks, Martijn,

Greg, could you have a look to pick up?

Quoted 108 lines · Martijn Coenen

Thanks Minchan!

Quoted 2 lines · Minchan Kim
Quoted 101 lines
---
 drivers/android/binder_alloc.c | 43 +++++++++++++++++++++++++++-------
 1 file changed, 35 insertions(+), 8 deletions(-)
drivers/android/binder_alloc.c +35 -9
index 3f3b7b253445..64fd96eada31 100644
@@ -332,6 +332,35 @@ static int binder_update_page_range(struct binder_alloc *alloc, int allocate,
        return vma ? -ENOMEM : -ESRCH;
 }
 
+
+static inline void binder_alloc_set_vma(struct binder_alloc *alloc,
+               struct vm_area_struct *vma)
+{
+       if (vma)
+               alloc->vma_vm_mm = vma->vm_mm;
+       /*
+        * If we see alloc->vma is not NULL, buffer data structures set up
+        * completely. Look at smp_rmb side binder_alloc_get_vma.
+        * We also want to guarantee new alloc->vma_vm_mm is always visible
+        * if alloc->vma is set.
+        */
+       smp_wmb();
+       alloc->vma = vma;
+}
+
+static inline struct vm_area_struct *binder_alloc_get_vma(
+               struct binder_alloc *alloc)
+{
+       struct vm_area_struct *vma = NULL;
+
+       if (alloc->vma) {
+               /* Look at description in binder_alloc_set_vma */
+               smp_rmb();
+               vma = alloc->vma;
+       }
+       return vma;
+}
+
 static struct binder_buffer *binder_alloc_new_buf_locked(
                                struct binder_alloc *alloc,
                                size_t data_size,
@@ -348,7 +377,7 @@ static struct binder_buffer *binder_alloc_new_buf_locked(
        size_t size, data_offsets_size;
        int ret;
 
-       if (alloc->vma == NULL) {
+       if (!binder_alloc_get_vma(alloc)) {
                binder_alloc_debug(BINDER_DEBUG_USER_ERROR,
                                   "%d: binder_alloc_buf, no vma\n",
                                   alloc->pid);
@@ -723,9 +752,7 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc,
        buffer->free = 1;
        binder_insert_free_buffer(alloc, buffer);
        alloc->free_async_space = alloc->buffer_size / 2;
-       barrier();
-       alloc->vma = vma;
-       alloc->vma_vm_mm = vma->vm_mm;
+       binder_alloc_set_vma(alloc, vma);
        mmgrab(alloc->vma_vm_mm);
 
        return 0;
@@ -754,10 +781,10 @@ void binder_alloc_deferred_release(struct binder_alloc *alloc)
        int buffers, page_count;
        struct binder_buffer *buffer;
 
-       BUG_ON(alloc->vma);
-
        buffers = 0;
        mutex_lock(&alloc->mutex);
+       BUG_ON(alloc->vma);
+
        while ((n = rb_first(&alloc->allocated_buffers))) {
                buffer = rb_entry(n, struct binder_buffer, rb_node);
 
@@ -900,7 +927,7 @@ int binder_alloc_get_allocated_count(struct binder_alloc *alloc)
  */
 void binder_alloc_vma_close(struct binder_alloc *alloc)
 {
-       WRITE_ONCE(alloc->vma, NULL);
+       binder_alloc_set_vma(alloc, NULL);
 }
 
 /**
@@ -935,7 +962,7 @@ enum lru_status binder_alloc_free_page(struct list_head *item,
 
        index = page - alloc->pages;
        page_addr = (uintptr_t)alloc->buffer + index * PAGE_SIZE;
-       vma = alloc->vma;
+       vma = binder_alloc_get_vma(alloc);
        if (vma) {
                if (!mmget_not_zero(alloc->vma_vm_mm))
                        goto err_mmget;
--

2.18.0.1017.ga543ac7ca45-goog

maintainer
#6 lore raw
Quoted 3 lines · Minchan Kim

Thanks, Martijn,

Greg, could you have a look to pick up?

Now queued up, thanks.

greg k-h