block: genhd: Add disk/partition specific uevent callbacks for partition info
PATCH 1/1 block: genhd: Add disk/partition specific uevent callbacks for partition info
From: San Mehat <[email protected]>
This patch has been carried in the Android tree for quite some time and is one of the few patches required to get a mainline kernel up and running with an exsiting Android userspace. So I wanted to submit it for review and consideration if it should be merged.
For disk devices, a new uevent parameter 'NPARTS' specifies the number of partitions detected by the kernel. Partition devices get 'PARTN' which specifies the partitions index in the table, and 'PARTNAME', which specifies PARTNAME specifices the partition name of a partition device
Android's userspace uses this for creating device node links from the partition name and number: ie:
/dev/block/platform/soc/by-name/system
or
/dev/block/platform/soc/by-num/p1
One can see its usage here:
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#355
and
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#494
Thoughts and feedback would be greatly appreciated!
--- block/genhd.c | 17 +++++++++++++++++ block/partition-generic.c | 11 +++++++++++ 2 files changed, 28 insertions(+)
block/genhd.c +17 -0
@@ -1115,6 +1115,22 @@ static void disk_release(struct device *dev)blk_put_queue (disk->queue);kfree (disk); } + +static int disk_uevent (struct device *dev,struct kobj_uevent_env *env) +{ +struct gendisk *disk =dev_to_disk (dev); +struct disk_part_iter piter; +struct hd_struct *part; +int cnt =0 ; + +disk_part_iter_init (&piter, disk,0 ); +while ((part =disk_part_iter_next (&piter))) + cnt++; +disk_part_iter_exit (&piter); +add_uevent_var (env,"NPARTS=%u" , cnt); +return 0 ; +} +struct class block_class = { .name ="block" , };
@@ -1134,6 +1150,7 @@ static struct device_type disk_type = { .groups = disk_attr_groups, .release = disk_release, .devnode = block_devnode, + .uevent = disk_uevent, };#ifdef CONFIG_PROC_FS
block/partition-generic.c +11 -0
@@ -216,10 +216,21 @@ static void part_release(struct device *dev)kfree (p); } +static int part_uevent (struct device *dev,struct kobj_uevent_env *env) +{ +struct hd_struct *part =dev_to_part (dev); + +add_uevent_var (env,"PARTN=%u" , part->partno); +if (part->info && part->info->volname[0 ]) +add_uevent_var (env,"PARTNAME=%s" , part->info->volname); +return 0 ; +} +struct device_type part_type = { .name ="partition" , .groups = part_attr_groups, .release = part_release, + .uevent = part_uevent, };static void delete_partition_rcu_cb (struct rcu_head *head)
Signature · 1 line
1.9.1
Quoted 19 lines · John Stultz
From: San Mehat <[email protected]>
This patch has been carried in the Android tree for quite some time and is one of the few patches required to get a mainline kernel up and running with an exsiting Android userspace. So I wanted to submit it for review and consideration if it should be merged.
For disk devices, a new uevent parameter 'NPARTS' specifies the number of partitions detected by the kernel. Partition devices get 'PARTN' which specifies the partitions index in the table, and 'PARTNAME', which specifies PARTNAME specifices the partition name of a partition device
Android's userspace uses this for creating device node links from the partition name and number: ie:
/dev/block/platform/soc/by-name/system
or
/dev/block/platform/soc/by-num/p1
Quoted 6 lines
One can see its usage here:
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#355
and
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#494Thoughts and feedback would be greatly appreciated!
Ping? Any thoughts or objections to this?
thanks
-john
Hi John,
Quoted 19 lines · John Stultz
From: San Mehat <[email protected]>
This patch has been carried in the Android tree for quite some time and is one of the few patches required to get a mainline kernel up and running with an exsiting Android userspace. So I wanted to submit it for review and consideration if it should be merged.
For disk devices, a new uevent parameter 'NPARTS' specifies the number of partitions detected by the kernel. Partition devices get 'PARTN' which specifies the partitions index in the table, and 'PARTNAME', which specifies PARTNAME specifices the partition name of a partition device
Android's userspace uses this for creating device node links from the partition name and number: ie:
/dev/block/platform/soc/by-name/system
or
/dev/block/platform/soc/by-num/p1
Quoted 4 lines
One can see its usage here:
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#355
and
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#494
Those links don't appear to use the NPARTS uevent parameter, unless I missed it.
Quoted 1 line
Thoughts and feedback would be greatly appreciated!
I can't say I have strong feelings one way or the other. I've CC'd Kay and Harald, hoping they might have an opinion.
Cheers,
Jeff
Quoted 73 lines
Cc Jens Axboe <[email protected]>Cc Rom Lemarchand <[email protected]>Cc Android Kernel Team <[email protected]>Signed-off-by Dima Zavin <[email protected]>[jstultz: Added more context to commit message & whitespace fix]Signed-off-by John Stultz <[email protected]>--- block/genhd.c | 17 +++++++++++++++++ block/partition-generic.c | 11 +++++++++++ 2 files changed, 28 insertions(+)block/genhd.c +17
-0@@ -1115,6 +1115,22 @@ static void disk_release(struct device *dev)blk_put_queue (disk->queue);kfree (disk); } + +static int disk_uevent (struct device *dev,struct kobj_uevent_env *env) +{ +struct gendisk *disk =dev_to_disk (dev); +struct disk_part_iter piter; +struct hd_struct *part; +int cnt =0 ; + +disk_part_iter_init (&piter, disk,0 ); +while ((part =disk_part_iter_next (&piter))) + cnt++; +disk_part_iter_exit (&piter); +add_uevent_var (env,"NPARTS=%u" , cnt); +return 0 ; +} +struct class block_class = { .name ="block" , };@@ -1134,6 +1150,7 @@ static struct device_type disk_type = { .groups = disk_attr_groups, .release = disk_release, .devnode = block_devnode, + .uevent = disk_uevent, };#ifdef CONFIG_PROC_FS block/partition-generic.c +11
-0@@ -216,10 +216,21 @@ static void part_release(struct device *dev)kfree (p); } +static int part_uevent (struct device *dev,struct kobj_uevent_env *env) +{ +struct hd_struct *part =dev_to_part (dev); + +add_uevent_var (env,"PARTN=%u" , part->partno); +if (part->info && part->info->volname[0 ]) +add_uevent_var (env,"PARTNAME=%s" , part->info->volname); +return 0 ; +} +struct device_type part_type = { .name ="partition" , .groups = part_attr_groups, .release = part_release, + .uevent = part_uevent, };static void delete_partition_rcu_cb (struct rcu_head *head)
Quoted 23 lines · Jeff Moyer
Hi John,
Quoted 19 lines · John Stultz
From: San Mehat <[email protected]>
This patch has been carried in the Android tree for quite some time and is one of the few patches required to get a mainline kernel up and running with an exsiting Android userspace. So I wanted to submit it for review and consideration if it should be merged.
For disk devices, a new uevent parameter 'NPARTS' specifies the number of partitions detected by the kernel. Partition devices get 'PARTN' which specifies the partitions index in the table, and 'PARTNAME', which specifies PARTNAME specifices the partition name of a partition device
Android's userspace uses this for creating device node links from the partition name and number: ie:
/dev/block/platform/soc/by-name/system
or
/dev/block/platform/soc/by-num/p1
Quoted 7 lines
Quoted 4 lines
One can see its usage here:
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#355
and
https://android.googlesource.com/platform/system/core/+/master/init/devices.cpp#494Those links don't appear to use the NPARTS uevent parameter, unless I missed it.
So good point! It looks like with Android Marshmallow they dropped using NPARTS. For context, here is where it was used with Lollipop:
http://androidxref.com/5.1.1_r6/xref/system/vold/DirectVolume.cpp#192
Unless Rom or someone on the Android team objects, I'll remove that chunk from the patch and re-submit.
Thanks for the review!
-john