* [PATCH V4] brd: check and limit max_part par
@ 2020-01-21 4:04 Zhiqiang Liu
2020-02-03 1:53 ` Zhiqiang Liu
2020-02-03 12:26 ` Ming Lei
0 siblings, 2 replies; 5+ messages in thread
From: Zhiqiang Liu @ 2020-01-21 4:04 UTC (permalink / raw)
To: Ming Lei, Jens Axboe
Cc: linux-block, linux-kernel@vger.kernel.org, Mingfangsen, Guiyao,
wubo40, Louhongxiang
In brd_init func, rd_nr num of brd_device are firstly allocated
and add in brd_devices, then brd_devices are traversed to add each
brd_device by calling add_disk func. When allocating brd_device,
the disk->first_minor is set to i * max_part, if rd_nr * max_part
is larger than MINORMASK, two different brd_device may have the same
devt, then only one of them can be successfully added.
when rmmod brd.ko, it will cause oops when calling brd_exit.
Follow those steps:
# modprobe brd rd_nr=3 rd_size=102400 max_part=1048576
# rmmod brd
then, the oops will appear.
Oops log:
[ 726.613722] Call trace:
[ 726.614175] kernfs_find_ns+0x24/0x130
[ 726.614852] kernfs_find_and_get_ns+0x44/0x68
[ 726.615749] sysfs_remove_group+0x38/0xb0
[ 726.616520] blk_trace_remove_sysfs+0x1c/0x28
[ 726.617320] blk_unregister_queue+0x98/0x100
[ 726.618105] del_gendisk+0x144/0x2b8
[ 726.618759] brd_exit+0x68/0x560 [brd]
[ 726.619501] __arm64_sys_delete_module+0x19c/0x2a0
[ 726.620384] el0_svc_common+0x78/0x130
[ 726.621057] el0_svc_handler+0x38/0x78
[ 726.621738] el0_svc+0x8/0xc
[ 726.622259] Code: aa0203f6 aa0103f7 aa1e03e0 d503201f (7940e260)
Here, we add brd_check_and_reset_par func to check and limit max_part par.
--
V3->V4:(suggested by Ming Lei)
- remove useless change
- add one limit of max_part
V2->V3: (suggested by Ming Lei)
- clear .minors when running out of consecutive minor space in brd_alloc
- remove limit of rd_nr
V1->V2: add more checks in brd_check_par_valid as suggested by Ming Lei.
Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
---
drivers/block/brd.c | 27 +++++++++++++++++++++++----
1 file changed, 23 insertions(+), 4 deletions(-)
diff --git a/drivers/block/brd.c b/drivers/block/brd.c
index df8103dd40ac..4684f95e3369 100644
--- a/drivers/block/brd.c
+++ b/drivers/block/brd.c
@@ -389,11 +389,12 @@ static struct brd_device *brd_alloc(int i)
* is harmless)
*/
blk_queue_physical_block_size(brd->brd_queue, PAGE_SIZE);
- disk = brd->brd_disk = alloc_disk(max_part);
+ disk = brd->brd_disk = alloc_disk(((i * max_part) & ~MINORMASK) ?
+ 0 : max_part);
if (!disk)
goto out_free_queue;
disk->major = RAMDISK_MAJOR;
- disk->first_minor = i * max_part;
+ disk->first_minor = i * disk->minors;
disk->fops = &brd_fops;
disk->private_data = brd;
disk->queue = brd->brd_queue;
@@ -468,6 +469,25 @@ static struct kobject *brd_probe(dev_t dev, int *part, void *data)
return kobj;
}
+static inline void brd_check_and_reset_par(void)
+{
+ if (unlikely(!max_part))
+ max_part = 1;
+
+ if (max_part > DISK_MAX_PARTS) {
+ pr_info("brd: max_part can't be larger than %d, reset max_part = %d.\n",
+ DISK_MAX_PARTS, DISK_MAX_PARTS);
+ max_part = DISK_MAX_PARTS;
+ }
+
+ /*
+ * make sure 'max_part' can be divided exactly by (1U << MINORBITS),
+ * otherwise, it is possiable to get same dev_t when adding partitions.
+ */
+ if ((1U << MINORBITS) % max_part != 0)
+ max_part = 1UL << fls(max_part);
+}
+
static int __init brd_init(void)
{
struct brd_device *brd, *next;
@@ -491,8 +511,7 @@ static int __init brd_init(void)
if (register_blkdev(RAMDISK_MAJOR, "ramdisk"))
return -EIO;
- if (unlikely(!max_part))
- max_part = 1;
+ brd_check_and_reset_par();
for (i = 0; i < rd_nr; i++) {
brd = brd_alloc(i);
--
2.19.1
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH V4] brd: check and limit max_part par
2020-01-21 4:04 [PATCH V4] brd: check and limit max_part par Zhiqiang Liu
@ 2020-02-03 1:53 ` Zhiqiang Liu
2020-02-03 12:36 ` Bob Liu
2020-02-03 12:26 ` Ming Lei
1 sibling, 1 reply; 5+ messages in thread
From: Zhiqiang Liu @ 2020-02-03 1:53 UTC (permalink / raw)
To: Ming Lei, Jens Axboe
Cc: linux-block, linux-kernel@vger.kernel.org, Mingfangsen, Guiyao,
wubo40, Louhongxiang
Friendly ping...
On 2020/1/21 12:04, Zhiqiang Liu wrote:
>
> In brd_init func, rd_nr num of brd_device are firstly allocated
> and add in brd_devices, then brd_devices are traversed to add each
> brd_device by calling add_disk func. When allocating brd_device,
> the disk->first_minor is set to i * max_part, if rd_nr * max_part
> is larger than MINORMASK, two different brd_device may have the same
> devt, then only one of them can be successfully added.
> when rmmod brd.ko, it will cause oops when calling brd_exit.
>
> Follow those steps:
> # modprobe brd rd_nr=3 rd_size=102400 max_part=1048576
> # rmmod brd
> then, the oops will appear.
>
> Oops log:
> [ 726.613722] Call trace:
> [ 726.614175] kernfs_find_ns+0x24/0x130
> [ 726.614852] kernfs_find_and_get_ns+0x44/0x68
> [ 726.615749] sysfs_remove_group+0x38/0xb0
> [ 726.616520] blk_trace_remove_sysfs+0x1c/0x28
> [ 726.617320] blk_unregister_queue+0x98/0x100
> [ 726.618105] del_gendisk+0x144/0x2b8
> [ 726.618759] brd_exit+0x68/0x560 [brd]
> [ 726.619501] __arm64_sys_delete_module+0x19c/0x2a0
> [ 726.620384] el0_svc_common+0x78/0x130
> [ 726.621057] el0_svc_handler+0x38/0x78
> [ 726.621738] el0_svc+0x8/0xc
> [ 726.622259] Code: aa0203f6 aa0103f7 aa1e03e0 d503201f (7940e260)
>
> Here, we add brd_check_and_reset_par func to check and limit max_part par.
>
> --
> V3->V4:(suggested by Ming Lei)
> - remove useless change
> - add one limit of max_part
>
> V2->V3: (suggested by Ming Lei)
> - clear .minors when running out of consecutive minor space in brd_alloc
> - remove limit of rd_nr
>
> V1->V2: add more checks in brd_check_par_valid as suggested by Ming Lei.
>
> Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
> ---
> drivers/block/brd.c | 27 +++++++++++++++++++++++----
> 1 file changed, 23 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/block/brd.c b/drivers/block/brd.c
> index df8103dd40ac..4684f95e3369 100644
> --- a/drivers/block/brd.c
> +++ b/drivers/block/brd.c
> @@ -389,11 +389,12 @@ static struct brd_device *brd_alloc(int i)
> * is harmless)
> */
> blk_queue_physical_block_size(brd->brd_queue, PAGE_SIZE);
> - disk = brd->brd_disk = alloc_disk(max_part);
> + disk = brd->brd_disk = alloc_disk(((i * max_part) & ~MINORMASK) ?
> + 0 : max_part);
> if (!disk)
> goto out_free_queue;
> disk->major = RAMDISK_MAJOR;
> - disk->first_minor = i * max_part;
> + disk->first_minor = i * disk->minors;
> disk->fops = &brd_fops;
> disk->private_data = brd;
> disk->queue = brd->brd_queue;
> @@ -468,6 +469,25 @@ static struct kobject *brd_probe(dev_t dev, int *part, void *data)
> return kobj;
> }
>
> +static inline void brd_check_and_reset_par(void)
> +{
> + if (unlikely(!max_part))
> + max_part = 1;
> +
> + if (max_part > DISK_MAX_PARTS) {
> + pr_info("brd: max_part can't be larger than %d, reset max_part = %d.\n",
> + DISK_MAX_PARTS, DISK_MAX_PARTS);
> + max_part = DISK_MAX_PARTS;
> + }
> +
> + /*
> + * make sure 'max_part' can be divided exactly by (1U << MINORBITS),
> + * otherwise, it is possiable to get same dev_t when adding partitions.
> + */
> + if ((1U << MINORBITS) % max_part != 0)
> + max_part = 1UL << fls(max_part);
> +}
> +
> static int __init brd_init(void)
> {
> struct brd_device *brd, *next;
> @@ -491,8 +511,7 @@ static int __init brd_init(void)
> if (register_blkdev(RAMDISK_MAJOR, "ramdisk"))
> return -EIO;
>
> - if (unlikely(!max_part))
> - max_part = 1;
> + brd_check_and_reset_par();
>
> for (i = 0; i < rd_nr; i++) {
> brd = brd_alloc(i);
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH V4] brd: check and limit max_part par
2020-01-21 4:04 [PATCH V4] brd: check and limit max_part par Zhiqiang Liu
2020-02-03 1:53 ` Zhiqiang Liu
@ 2020-02-03 12:26 ` Ming Lei
2020-02-04 1:21 ` Zhiqiang Liu
1 sibling, 1 reply; 5+ messages in thread
From: Ming Lei @ 2020-02-03 12:26 UTC (permalink / raw)
To: Zhiqiang Liu
Cc: Jens Axboe, linux-block, linux-kernel@vger.kernel.org,
Mingfangsen, Guiyao, Louhongxiang
On Tue, Jan 21, 2020 at 12:04:41PM +0800, Zhiqiang Liu wrote:
>
> In brd_init func, rd_nr num of brd_device are firstly allocated
> and add in brd_devices, then brd_devices are traversed to add each
> brd_device by calling add_disk func. When allocating brd_device,
> the disk->first_minor is set to i * max_part, if rd_nr * max_part
> is larger than MINORMASK, two different brd_device may have the same
> devt, then only one of them can be successfully added.
> when rmmod brd.ko, it will cause oops when calling brd_exit.
>
> Follow those steps:
> # modprobe brd rd_nr=3 rd_size=102400 max_part=1048576
> # rmmod brd
> then, the oops will appear.
>
> Oops log:
> [ 726.613722] Call trace:
> [ 726.614175] kernfs_find_ns+0x24/0x130
> [ 726.614852] kernfs_find_and_get_ns+0x44/0x68
> [ 726.615749] sysfs_remove_group+0x38/0xb0
> [ 726.616520] blk_trace_remove_sysfs+0x1c/0x28
> [ 726.617320] blk_unregister_queue+0x98/0x100
> [ 726.618105] del_gendisk+0x144/0x2b8
> [ 726.618759] brd_exit+0x68/0x560 [brd]
> [ 726.619501] __arm64_sys_delete_module+0x19c/0x2a0
> [ 726.620384] el0_svc_common+0x78/0x130
> [ 726.621057] el0_svc_handler+0x38/0x78
> [ 726.621738] el0_svc+0x8/0xc
> [ 726.622259] Code: aa0203f6 aa0103f7 aa1e03e0 d503201f (7940e260)
>
> Here, we add brd_check_and_reset_par func to check and limit max_part par.
>
> --
> V3->V4:(suggested by Ming Lei)
> - remove useless change
> - add one limit of max_part
>
> V2->V3: (suggested by Ming Lei)
> - clear .minors when running out of consecutive minor space in brd_alloc
> - remove limit of rd_nr
>
> V1->V2: add more checks in brd_check_par_valid as suggested by Ming Lei.
>
> Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
> ---
> drivers/block/brd.c | 27 +++++++++++++++++++++++----
> 1 file changed, 23 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/block/brd.c b/drivers/block/brd.c
> index df8103dd40ac..4684f95e3369 100644
> --- a/drivers/block/brd.c
> +++ b/drivers/block/brd.c
> @@ -389,11 +389,12 @@ static struct brd_device *brd_alloc(int i)
> * is harmless)
> */
> blk_queue_physical_block_size(brd->brd_queue, PAGE_SIZE);
> - disk = brd->brd_disk = alloc_disk(max_part);
> + disk = brd->brd_disk = alloc_disk(((i * max_part) & ~MINORMASK) ?
> + 0 : max_part);
> if (!disk)
> goto out_free_queue;
> disk->major = RAMDISK_MAJOR;
> - disk->first_minor = i * max_part;
> + disk->first_minor = i * disk->minors;
The above change isn't necessary.
> disk->fops = &brd_fops;
> disk->private_data = brd;
> disk->queue = brd->brd_queue;
> @@ -468,6 +469,25 @@ static struct kobject *brd_probe(dev_t dev, int *part, void *data)
> return kobj;
> }
>
> +static inline void brd_check_and_reset_par(void)
> +{
> + if (unlikely(!max_part))
> + max_part = 1;
> +
> + if (max_part > DISK_MAX_PARTS) {
> + pr_info("brd: max_part can't be larger than %d, reset max_part = %d.\n",
> + DISK_MAX_PARTS, DISK_MAX_PARTS);
> + max_part = DISK_MAX_PARTS;
> + }
> +
> + /*
> + * make sure 'max_part' can be divided exactly by (1U << MINORBITS),
> + * otherwise, it is possiable to get same dev_t when adding partitions.
> + */
> + if ((1U << MINORBITS) % max_part != 0)
> + max_part = 1UL << fls(max_part);
> +}
You should move the above change before capping it to DISK_MAX_PARTS
since 1UL << fls() may increase 'max_part'.
Thanks,
Ming
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH V4] brd: check and limit max_part par
2020-02-03 1:53 ` Zhiqiang Liu
@ 2020-02-03 12:36 ` Bob Liu
0 siblings, 0 replies; 5+ messages in thread
From: Bob Liu @ 2020-02-03 12:36 UTC (permalink / raw)
To: Zhiqiang Liu, Ming Lei, Jens Axboe
Cc: linux-block, linux-kernel@vger.kernel.org, Mingfangsen, Guiyao,
wubo40, Louhongxiang
On 2/3/20 9:53 AM, Zhiqiang Liu wrote:
> Friendly ping...
>
> On 2020/1/21 12:04, Zhiqiang Liu wrote:
>>
>> In brd_init func, rd_nr num of brd_device are firstly allocated
>> and add in brd_devices, then brd_devices are traversed to add each
>> brd_device by calling add_disk func. When allocating brd_device,
>> the disk->first_minor is set to i * max_part, if rd_nr * max_part
>> is larger than MINORMASK, two different brd_device may have the same
>> devt, then only one of them can be successfully added.
>> when rmmod brd.ko, it will cause oops when calling brd_exit.
>>
>> Follow those steps:
>> # modprobe brd rd_nr=3 rd_size=102400 max_part=1048576
>> # rmmod brd
>> then, the oops will appear.
>>
>> Oops log:
>> [ 726.613722] Call trace:
>> [ 726.614175] kernfs_find_ns+0x24/0x130
>> [ 726.614852] kernfs_find_and_get_ns+0x44/0x68
>> [ 726.615749] sysfs_remove_group+0x38/0xb0
>> [ 726.616520] blk_trace_remove_sysfs+0x1c/0x28
>> [ 726.617320] blk_unregister_queue+0x98/0x100
>> [ 726.618105] del_gendisk+0x144/0x2b8
>> [ 726.618759] brd_exit+0x68/0x560 [brd]
>> [ 726.619501] __arm64_sys_delete_module+0x19c/0x2a0
>> [ 726.620384] el0_svc_common+0x78/0x130
>> [ 726.621057] el0_svc_handler+0x38/0x78
>> [ 726.621738] el0_svc+0x8/0xc
>> [ 726.622259] Code: aa0203f6 aa0103f7 aa1e03e0 d503201f (7940e260)
>>
>> Here, we add brd_check_and_reset_par func to check and limit max_part par.
>>
>> --
>> V3->V4:(suggested by Ming Lei)
>> - remove useless change
>> - add one limit of max_part
>>
>> V2->V3: (suggested by Ming Lei)
>> - clear .minors when running out of consecutive minor space in brd_alloc
>> - remove limit of rd_nr
>>
>> V1->V2: add more checks in brd_check_par_valid as suggested by Ming Lei.
>>
>> Signed-off-by: Zhiqiang Liu <liuzhiqiang26@huawei.com>
>> ---
>> drivers/block/brd.c | 27 +++++++++++++++++++++++----
>> 1 file changed, 23 insertions(+), 4 deletions(-)
>>
Looks good to me.
Reviewed-by: Bob Liu <bob.liu@oracle.com>
>> diff --git a/drivers/block/brd.c b/drivers/block/brd.c
>> index df8103dd40ac..4684f95e3369 100644
>> --- a/drivers/block/brd.c
>> +++ b/drivers/block/brd.c
>> @@ -389,11 +389,12 @@ static struct brd_device *brd_alloc(int i)
>> * is harmless)
>> */
>> blk_queue_physical_block_size(brd->brd_queue, PAGE_SIZE);
>> - disk = brd->brd_disk = alloc_disk(max_part);
>> + disk = brd->brd_disk = alloc_disk(((i * max_part) & ~MINORMASK) ?
>> + 0 : max_part);
>> if (!disk)
>> goto out_free_queue;
>> disk->major = RAMDISK_MAJOR;
>> - disk->first_minor = i * max_part;
>> + disk->first_minor = i * disk->minors;
>> disk->fops = &brd_fops;
>> disk->private_data = brd;
>> disk->queue = brd->brd_queue;
>> @@ -468,6 +469,25 @@ static struct kobject *brd_probe(dev_t dev, int *part, void *data)
>> return kobj;
>> }
>>
>> +static inline void brd_check_and_reset_par(void)
>> +{
>> + if (unlikely(!max_part))
>> + max_part = 1;
>> +
>> + if (max_part > DISK_MAX_PARTS) {
>> + pr_info("brd: max_part can't be larger than %d, reset max_part = %d.\n",
>> + DISK_MAX_PARTS, DISK_MAX_PARTS);
>> + max_part = DISK_MAX_PARTS;
>> + }
>> +
>> + /*
>> + * make sure 'max_part' can be divided exactly by (1U << MINORBITS),
>> + * otherwise, it is possiable to get same dev_t when adding partitions.
>> + */
>> + if ((1U << MINORBITS) % max_part != 0)
>> + max_part = 1UL << fls(max_part);
>> +}
>> +
>> static int __init brd_init(void)
>> {
>> struct brd_device *brd, *next;
>> @@ -491,8 +511,7 @@ static int __init brd_init(void)
>> if (register_blkdev(RAMDISK_MAJOR, "ramdisk"))
>> return -EIO;
>>
>> - if (unlikely(!max_part))
>> - max_part = 1;
>> + brd_check_and_reset_par();
>>
>> for (i = 0; i < rd_nr; i++) {
>> brd = brd_alloc(i);
>>
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH V4] brd: check and limit max_part par
2020-02-03 12:26 ` Ming Lei
@ 2020-02-04 1:21 ` Zhiqiang Liu
0 siblings, 0 replies; 5+ messages in thread
From: Zhiqiang Liu @ 2020-02-04 1:21 UTC (permalink / raw)
To: Ming Lei
Cc: Jens Axboe, linux-block, linux-kernel@vger.kernel.org,
Mingfangsen, Guiyao, Louhongxiang
On 2020/2/3 20:26, Ming Lei wrote:
> On Tue, Jan 21, 2020 at 12:04:41PM +0800, Zhiqiang Liu wrote:
>>
>> In brd_init func, rd_nr num of brd_device are firstly allocated
>> and add in brd_devices, then brd_devices are traversed to add each
>> brd_device by calling add_disk func. When allocating brd_device,
>> the disk->first_minor is set to i * max_part, if rd_nr * max_part
>> is larger than MINORMASK, two different brd_device may have the same
>> devt, then only one of them can be successfully added.
>> when rmmod brd.ko, it will cause oops when calling brd_exit.
>>
>> +static inline void brd_check_and_reset_par(void)
>> +{
>> + if (unlikely(!max_part))
>> + max_part = 1;
>> +
>> + if (max_part > DISK_MAX_PARTS) {
>> + pr_info("brd: max_part can't be larger than %d, reset max_part = %d.\n",
>> + DISK_MAX_PARTS, DISK_MAX_PARTS);
>> + max_part = DISK_MAX_PARTS;
>> + }
>> +
>> + /*
>> + * make sure 'max_part' can be divided exactly by (1U << MINORBITS),
>> + * otherwise, it is possiable to get same dev_t when adding partitions.
>> + */
>> + if ((1U << MINORBITS) % max_part != 0)
>> + max_part = 1UL << fls(max_part);
>> +}
>
> You should move the above change before capping it to DISK_MAX_PARTS
> since 1UL << fls() may increase 'max_part'.
>
Thanks for your suggestion. I will send the v5 patch.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2020-02-04 1:21 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2020-01-21 4:04 [PATCH V4] brd: check and limit max_part par Zhiqiang Liu
2020-02-03 1:53 ` Zhiqiang Liu
2020-02-03 12:36 ` Bob Liu
2020-02-03 12:26 ` Ming Lei
2020-02-04 1:21 ` Zhiqiang Liu
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox