linux_dsm_epyc7002/fs/partitions
Yasuaki Ishimatsu 7681bfeecc block: fix accounting bug on cross partition merges
/proc/diskstats would display a strange output as follows.

$ cat /proc/diskstats |grep sda
   8       0 sda 90524 7579 102154 20464 0 0 0 0 0 14096 20089
   8       1 sda1 19085 1352 21841 4209 0 0 0 0 4294967064 15689 4293424691
                                                ~~~~~~~~~~
   8       2 sda2 71252 3624 74891 15950 0 0 0 0 232 23995 1562390
   8       3 sda3 54 487 2188 92 0 0 0 0 0 88 92
   8       4 sda4 4 0 8 0 0 0 0 0 0 0 0
   8       5 sda5 81 2027 2130 138 0 0 0 0 0 87 137

Its reason is the wrong way of accounting hd_struct->in_flight. When a bio is
merged into a request belongs to different partition by ELEVATOR_FRONT_MERGE.

The detailed root cause is as follows.

Assuming that there are two partition, sda1 and sda2.

1. A request for sda2 is in request_queue. Hence sda1's hd_struct->in_flight
   is 0 and sda2's one is 1.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

2. A bio belongs to sda1 is issued and is merged into the request mentioned on
   step1 by ELEVATOR_BACK_MERGE. The first sector of the request is changed
   from sda2 region to sda1 region. However the two partition's
   hd_struct->in_flight are not changed.

        | hd_struct->in_flight
   ---------------------------
   sda1 |          0
   sda2 |          1
   ---------------------------

3. The request is finished and blk_account_io_done() is called. In this case,
   sda2's hd_struct->in_flight, not a sda1's one, is decremented.

        | hd_struct->in_flight
   ---------------------------
   sda1 |         -1
   sda2 |          1
   ---------------------------

The patch fixes the problem by caching the partition lookup
inside the request structure, hence making sure that the increment
and decrement will always happen on the same partition struct. This
also speeds up IO with accounting enabled, since it cuts down on
the number of lookups we have to do.

When reloading partition tables, quiesce IO to ensure that no
request references to the partition struct exists. When it is safe
to free the partition table, the IO for that device is restarted
again.

Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: stable@kernel.org
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-10-19 09:07:02 +02:00
..
acorn.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
acorn.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
amiga.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
amiga.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
atari.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
atari.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
check.c block: fix accounting bug on cross partition merges 2010-10-19 09:07:02 +02:00
check.h block, partition: add partition_meta_info to hd_struct 2010-09-15 16:13:18 +02:00
efi.c genhd, efi: add efi partition metadata to hd_structs 2010-09-15 16:13:28 +02:00
efi.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
ibm.c [S390] partitions: fix build error in ibm partition detection code 2010-08-13 10:06:55 +02:00
ibm.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
karma.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
karma.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
Kconfig partition: use DEFAULT_SGI_PARTITION for SGI_PARTION default 2008-02-06 10:41:08 -08:00
ldm.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
ldm.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
mac.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
mac.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
Makefile partition: add support for sysv68 partitions 2007-05-08 11:15:09 -07:00
msdos.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
msdos.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
osf.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
osf.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
sgi.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
sgi.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
sun.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
sun.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
sysv68.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
sysv68.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00
ultrix.c partitions: fix sometimes unreadable partition strings 2010-08-11 08:59:20 -07:00
ultrix.h block: use struct parsed_partitions *state universally in partition check code 2010-05-21 20:01:02 +02:00