[PATCH v2 1/1] sched/numa: Fix memory leak due to the overwritten vma->numab_state

[PATCH v2 1/1] sched/numa: Fix memory leak due to the overwritten vma->numab_state

[Adrian Huang]

From: Adrian Huang <ahuang12@lenovo.com>

[Problem Description]
When running the hackbench program of LTP, the following memory leak is
reported by kmemleak.

  # /opt/ltp/testcases/bin/hackbench 20 thread 1000
  Running with 20*40 (== 800) tasks.

  # dmesg | grep kmemleak
  ...
  kmemleak: 480 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
  kmemleak: 665 new suspected memory leaks (see /sys/kernel/debug/kmemleak)

  # cat /sys/kernel/debug/kmemleak
  unreferenced object 0xffff888cd8ca2c40 (size 64):
    comm "hackbench", pid 17142, jiffies 4299780315
    hex dump (first 32 bytes):
      ac 74 49 00 01 00 00 00 4c 84 49 00 01 00 00 00  .tI.....L.I.....
      00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    backtrace (crc bff18fd4):
      [<ffffffff81419a89>] __kmalloc_cache_noprof+0x2f9/0x3f0
      [<ffffffff8113f715>] task_numa_work+0x725/0xa00
      [<ffffffff8110f878>] task_work_run+0x58/0x90
      [<ffffffff81ddd9f8>] syscall_exit_to_user_mode+0x1c8/0x1e0
      [<ffffffff81dd78d5>] do_syscall_64+0x85/0x150
      [<ffffffff81e0012b>] entry_SYSCALL_64_after_hwframe+0x76/0x7e
  ...

This issue can be consistently reproduced on three different servers:
  * a 448-core server
  * a 256-core server
  * a 192-core server

[Root Cause]
Since multiple threads are created by the hackbench program (along with
the command argument 'thread'), a shared vma might be accessed by two or
more cores simultaneously. When two or more cores observe that
vma->numab_state is NULL at the same time, vma->numab_state will be
overwritten.

Although current code ensures that only one thread scans the VMAs in a
single 'numa_scan_period', there might be a chance for another thread
to enter in the next 'numa_scan_period' while we have not gotten till
numab_state allocation [1].

Note that the command `/opt/ltp/testcases/bin/hackbench 50 process 1000`
cannot the reproduce the issue. It is verified with 200+ test runs.

[Solution]
Use the cmpxchg atomic operation to ensure that only one thread executes
the vma->numab_state assignment.

[1] https://lore.kernel.org/lkml/1794be3c-358c-4cdc-a43d-a1f841d91ef7@amd.com/

Fixes: ef6a22b70f6d ("sched/numa: apply the scan delay to every new vma")
Reported-by: Jiwei Sun <sunjw10@lenovo.com>
Signed-off-by: Adrian Huang <ahuang12@lenovo.com>
Reviewed-by: Raghavendra K T <raghavendra.kt@amd.com>
---
 kernel/sched/fair.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 3356315d7e64..7f99df294583 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3399,10 +3399,16 @@ static void task_numa_work(struct callback_head *work)
 
 		/* Initialise new per-VMA NUMAB state. */
 		if (!vma->numab_state) {
-			vma->numab_state = kzalloc(sizeof(struct vma_numab_state),
-				GFP_KERNEL);
-			if (!vma->numab_state)
+			struct vma_numab_state *ptr;
+
+			ptr = kzalloc(sizeof(*ptr), GFP_KERNEL);
+			if (!ptr)
+				continue;
+
+			if (cmpxchg(&vma->numab_state, NULL, ptr)) {
+				kfree(ptr);
 				continue;
+			}
 
 			vma->numab_state->start_scan_seq = mm->numa_scan_seq;
 
-- 
2.34.1

Re: [PATCH v2 1/1] sched/numa: Fix memory leak due to the overwritten vma->numab_state

[Raghavendra K T]

+ Vlastimil

Looks like he was unintentionally missed in CC. He has added Reviewed-by 
to V1

On 11/13/2024 3:51 PM, Adrian Huang wrote:
> From: Adrian Huang <ahuang12@lenovo.com>
> 
> [Problem Description]
> When running the hackbench program of LTP, the following memory leak is
> reported by kmemleak.
> 
>    # /opt/ltp/testcases/bin/hackbench 20 thread 1000
>    Running with 20*40 (== 800) tasks.
> 
>    # dmesg | grep kmemleak
>    ...
>    kmemleak: 480 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
>    kmemleak: 665 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
> 
>    # cat /sys/kernel/debug/kmemleak
>    unreferenced object 0xffff888cd8ca2c40 (size 64):
>      comm "hackbench", pid 17142, jiffies 4299780315
>      hex dump (first 32 bytes):
>        ac 74 49 00 01 00 00 00 4c 84 49 00 01 00 00 00  .tI.....L.I.....
>        00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>      backtrace (crc bff18fd4):
>        [<ffffffff81419a89>] __kmalloc_cache_noprof+0x2f9/0x3f0
>        [<ffffffff8113f715>] task_numa_work+0x725/0xa00
>        [<ffffffff8110f878>] task_work_run+0x58/0x90
>        [<ffffffff81ddd9f8>] syscall_exit_to_user_mode+0x1c8/0x1e0
>        [<ffffffff81dd78d5>] do_syscall_64+0x85/0x150
>        [<ffffffff81e0012b>] entry_SYSCALL_64_after_hwframe+0x76/0x7e
>    ...
> 
> This issue can be consistently reproduced on three different servers:
>    * a 448-core server
>    * a 256-core server
>    * a 192-core server
> 
> [Root Cause]
> Since multiple threads are created by the hackbench program (along with
> the command argument 'thread'), a shared vma might be accessed by two or
> more cores simultaneously. When two or more cores observe that
> vma->numab_state is NULL at the same time, vma->numab_state will be
> overwritten.
> 
> Although current code ensures that only one thread scans the VMAs in a
> single 'numa_scan_period', there might be a chance for another thread
> to enter in the next 'numa_scan_period' while we have not gotten till
> numab_state allocation [1].
> 
> Note that the command `/opt/ltp/testcases/bin/hackbench 50 process 1000`
> cannot the reproduce the issue. It is verified with 200+ test runs.
> 
> [Solution]
> Use the cmpxchg atomic operation to ensure that only one thread executes
> the vma->numab_state assignment.
> 
> [1] https://lore.kernel.org/lkml/1794be3c-358c-4cdc-a43d-a1f841d91ef7@amd.com/
> 
> Fixes: ef6a22b70f6d ("sched/numa: apply the scan delay to every new vma")
> Reported-by: Jiwei Sun <sunjw10@lenovo.com>
> Signed-off-by: Adrian Huang <ahuang12@lenovo.com>
> Reviewed-by: Raghavendra K T <raghavendra.kt@amd.com>
> ---
>   kernel/sched/fair.c | 12 +++++++++---
>   1 file changed, 9 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 3356315d7e64..7f99df294583 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3399,10 +3399,16 @@ static void task_numa_work(struct callback_head *work)
>   
>   		/* Initialise new per-VMA NUMAB state. */
>   		if (!vma->numab_state) {
> -			vma->numab_state = kzalloc(sizeof(struct vma_numab_state),
> -				GFP_KERNEL);
> -			if (!vma->numab_state)
> +			struct vma_numab_state *ptr;
> +
> +			ptr = kzalloc(sizeof(*ptr), GFP_KERNEL);
> +			if (!ptr)
> +				continue;
> +
> +			if (cmpxchg(&vma->numab_state, NULL, ptr)) {
> +				kfree(ptr);
>   				continue;
> +			}
>   
>   			vma->numab_state->start_scan_seq = mm->numa_scan_seq;
>   

Re: [PATCH v2 1/1] sched/numa: Fix memory leak due to the overwritten vma->numab_state

[Vlastimil Babka]

On 11/15/24 11:45, Raghavendra K T wrote:
> + Vlastimil
> 
> Looks like he was unintentionally missed in CC. He has added Reviewed-by 
> to V1

Thanks, seems I did it for v1 when v2 was already sent, and without also cc
linux-mm I didn't notice it was.

> On 11/13/2024 3:51 PM, Adrian Huang wrote:
>> From: Adrian Huang <ahuang12@lenovo.com>
>> 
>> [Problem Description]
>> When running the hackbench program of LTP, the following memory leak is
>> reported by kmemleak.
>> 
>>    # /opt/ltp/testcases/bin/hackbench 20 thread 1000
>>    Running with 20*40 (== 800) tasks.
>> 
>>    # dmesg | grep kmemleak
>>    ...
>>    kmemleak: 480 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
>>    kmemleak: 665 new suspected memory leaks (see /sys/kernel/debug/kmemleak)
>> 
>>    # cat /sys/kernel/debug/kmemleak
>>    unreferenced object 0xffff888cd8ca2c40 (size 64):
>>      comm "hackbench", pid 17142, jiffies 4299780315
>>      hex dump (first 32 bytes):
>>        ac 74 49 00 01 00 00 00 4c 84 49 00 01 00 00 00  .tI.....L.I.....
>>        00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>>      backtrace (crc bff18fd4):
>>        [<ffffffff81419a89>] __kmalloc_cache_noprof+0x2f9/0x3f0
>>        [<ffffffff8113f715>] task_numa_work+0x725/0xa00
>>        [<ffffffff8110f878>] task_work_run+0x58/0x90
>>        [<ffffffff81ddd9f8>] syscall_exit_to_user_mode+0x1c8/0x1e0
>>        [<ffffffff81dd78d5>] do_syscall_64+0x85/0x150
>>        [<ffffffff81e0012b>] entry_SYSCALL_64_after_hwframe+0x76/0x7e
>>    ...
>> 
>> This issue can be consistently reproduced on three different servers:
>>    * a 448-core server
>>    * a 256-core server
>>    * a 192-core server
>> 
>> [Root Cause]
>> Since multiple threads are created by the hackbench program (along with
>> the command argument 'thread'), a shared vma might be accessed by two or
>> more cores simultaneously. When two or more cores observe that
>> vma->numab_state is NULL at the same time, vma->numab_state will be
>> overwritten.
>> 
>> Although current code ensures that only one thread scans the VMAs in a
>> single 'numa_scan_period', there might be a chance for another thread
>> to enter in the next 'numa_scan_period' while we have not gotten till
>> numab_state allocation [1].
>> 
>> Note that the command `/opt/ltp/testcases/bin/hackbench 50 process 1000`
>> cannot the reproduce the issue. It is verified with 200+ test runs.
>> 
>> [Solution]
>> Use the cmpxchg atomic operation to ensure that only one thread executes
>> the vma->numab_state assignment.
>> 
>> [1] https://lore.kernel.org/lkml/1794be3c-358c-4cdc-a43d-a1f841d91ef7@amd.com/
>> 
>> Fixes: ef6a22b70f6d ("sched/numa: apply the scan delay to every new vma")
>> Reported-by: Jiwei Sun <sunjw10@lenovo.com>
>> Signed-off-by: Adrian Huang <ahuang12@lenovo.com>
>> Reviewed-by: Raghavendra K T <raghavendra.kt@amd.com>

Reviewed-by: Vlastimil Babka <vbabka@suse.cz>

>> ---
>>   kernel/sched/fair.c | 12 +++++++++---
>>   1 file changed, 9 insertions(+), 3 deletions(-)
>> 
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index 3356315d7e64..7f99df294583 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -3399,10 +3399,16 @@ static void task_numa_work(struct callback_head *work)
>>   
>>   		/* Initialise new per-VMA NUMAB state. */
>>   		if (!vma->numab_state) {
>> -			vma->numab_state = kzalloc(sizeof(struct vma_numab_state),
>> -				GFP_KERNEL);
>> -			if (!vma->numab_state)
>> +			struct vma_numab_state *ptr;
>> +
>> +			ptr = kzalloc(sizeof(*ptr), GFP_KERNEL);
>> +			if (!ptr)
>> +				continue;
>> +
>> +			if (cmpxchg(&vma->numab_state, NULL, ptr)) {
>> +				kfree(ptr);
>>   				continue;
>> +			}
>>   
>>   			vma->numab_state->start_scan_seq = mm->numa_scan_seq;
>>   

Re: [PATCH v2 1/1] sched/numa: Fix memory leak due to the overwritten vma->numab_state

[Huang Adrian]

On Fri, Nov 15, 2024 at 7:02 PM Vlastimil Babka <vbabka@suse.cz> wrote:
>
> On 11/15/24 11:45, Raghavendra K T wrote:
> > + Vlastimil
> >
> > Looks like he was unintentionally missed in CC. He has added Reviewed-by
> > to V1
>
> Thanks, seems I did it for v1 when v2 was already sent, and without also cc
> linux-mm I didn't notice it was.
>

Sorry about that, I fat-fingered it.

And, thanks for Raghu's and Vlastimil's reviewed-by.

-- Adrian