From mboxrd@z Thu Jan 1 00:00:00 1970 From: NeilBrown Subject: Re: [PATCH v4] super1: fix sb->max_dev when adding a new disk in linear array Date: Wed, 24 May 2017 11:57:02 +1000 Message-ID: <87k257hv9t.fsf@notabene.neil.brown.name> References: <20170522061612.29410-1-lzhong@suse.com> <87d1b1jgju.fsf@notabene.neil.brown.name> <20e71aa6-e9e0-312f-747a-d25b872d2af8@suse.com> Mime-Version: 1.0 Content-Type: multipart/signed; boundary="=-=-="; micalg=pgp-sha256; protocol="application/pgp-signature" Return-path: In-Reply-To: <20e71aa6-e9e0-312f-747a-d25b872d2af8@suse.com> Sender: linux-raid-owner@vger.kernel.org To: Lidong Zhong , linux-raid@vger.kernel.org Cc: colyli@suse.com, Jes.Sorensen@gmail.com List-Id: linux-raid.ids --=-=-= Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Tue, May 23 2017, Lidong Zhong wrote: > On 05/22/2017 07:07 PM, NeilBrown wrote: >> On Mon, May 22 2017, Lidong Zhong wrote: >> >>> The value of sb->max_dev will always be increased by 1 when adding >>> a new disk in linear array. It causes an inconsistence between each >>> disk in the array and the "Array State" value of "mdadm --examine DISK" >>> is wrong. For example, when adding the first new disk into linear array >>> it will be: >>> >>> Array State : RAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA >>> ('A' =3D=3D active, '.' =3D=3D missing, 'R' =3D=3D replacing) >>> >>> Adding the second disk into linear array it will be >>> >>> Array State : .AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA >>> ('A' =3D=3D active, '.' =3D=3D missing, 'R' =3D=3D replacing) >>> >>> Signed-off-by: Lidong Zhong >>> --- >>> super1.c | 9 ++++++++- >>> 1 file changed, 8 insertions(+), 1 deletion(-) >>> >>> diff --git a/super1.c b/super1.c >>> index 2fcb814..03cea72 100644 >>> --- a/super1.c >>> +++ b/super1.c >>> @@ -1267,8 +1267,10 @@ static int update_super1(struct supertype *st, s= truct mdinfo *info, >>> break; >>> sb->dev_number =3D __cpu_to_le32(i); >>> info->disk.number =3D i; >>> - if (max >=3D __le32_to_cpu(sb->max_dev)) >>> + if (i >=3D max) { >>> + sb->dev_roles[max] =3D __cpu_to_le16(MD_DISK_ROLE_SPARE); >> > > Hi Neil, > >> Why do you assign to dev_roles[max]? > > I meant to assure there will always be a spare spot in dev_roles[], > that is sb->max_dev at least is at lease 1 more than raid_disks. > Now I see what you mean in your reply to my last version patch. > >> max must equal i here, and a few lines later: >> sb->dev_roles[i] =3D __cpu_to_le16(info->disk.raid_disk); >> >> your assignment is over-written. So it is pointless. >> If i was greater than max (which should be impossible), you assignment >> here would corrupt the dev_roles table. >> >> Please drop this assignment. > > Yes, just increase the max_dev value is enough. > >> >>> sb->max_dev =3D __cpu_to_le32(max+1); >>> + } >>> >>> random_uuid(sb->device_uuid); >>> >>> @@ -1293,9 +1295,14 @@ static int update_super1(struct supertype *st, s= truct mdinfo *info, >>> } >>> } >>> } else if (strcmp(update, "linear-grow-update") =3D=3D 0) { >>> + unsigned int max =3D __le32_to_cpu(sb->max_dev); >>> sb->raid_disks =3D __cpu_to_le32(info->array.raid_disks); >>> sb->dev_roles[info->disk.number] =3D >>> __cpu_to_le16(info->disk.raid_disk); >>> + if (info->array.raid_disks >=3D max) { >> >> if raid_disks =3D=3D max there is no need to change anything. >> It is only when raid_disks > max that you need to increase max. >> > > Yes, the max_dev should only be updated when raid_disks > max. > >>> + sb->dev_roles[max] =3D __cpu_to_le16(MD_DISK_ROLE_SPARE); >> >> When you increase max, you do need to assign MD_DISK_ROLE_SPARE to the >> new element, but you need to do that *before* disk.raid_disk is >> assigned, in case info->disk.number =3D=3D max (as it could be for the >> recently added device). >> > I think it's also pointless to assign MD_DISK_ROLE_SPARE > since there is no SPARE in dev_roles when we need to update > sb->max_dev. The newly added device will not meet the condition > as max_dev has already been updated, that's saying, we only > need to update the max_dev value for original disks. > The following code should work > > 1297 } else if (strcmp(update, "linear-grow-update") =3D=3D 0) { > 1298 unsigned int max =3D __le32_to_cpu(sb->max_dev); > 1299 sb->raid_disks =3D __cpu_to_le32(info->array.raid_disks); > 1300 sb->dev_roles[info->disk.number] =3D > 1301 __cpu_to_le16(info->disk.raid_disk); > 1302 if (info->array.raid_disks > max) {=20 >=20=20 > > 1303 sb->max_dev =3D __cpu_to_le32(max+1); > 1304 } Increasing max_dev and not initializing will leave the last entry in dev_roles[] uninitialised. That isn't good. MD_DISK_ROLE_SPARE doesn't mean there is a spare device in that slot. It means that if there is a device in that slot, it must be spare. If you leave it uninitialised, it will probably be zero, and then you will get "?" in the mdadm output again. NeilBrown > > Thank you for your patient review. > > Lidong > >> NeilBrown >> >> >>> + sb->max_dev =3D __cpu_to_le32(max+1); >>> + } >>> } else if (strcmp(update, "resync") =3D=3D 0) { >>> /* make sure resync happens */ >>> sb->resync_offset =3D 0ULL; >>> -- >>> 2.12.0 --=-=-= Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEG8Yp69OQ2HB7X0l6Oeye3VZigbkFAlkk6G4ACgkQOeye3VZi gbnsWg/8D+Q1VsOACdx+9HLqLBcICfxXorSRUPDPL+nSkBA2GiyRHJq3tCkqFVsl zlVcpGZMuadgLQ2mNMVtR3mLV2QHvt46q2h2j4/wcNDXXMQ3LVMKwG2v3WjdPJa/ NyCQWYdzdZwoCuPb8xdix5NuGSRdQrDTOeRiHSDCMcyQBBo82sxzSmlad+sApZw8 rmsOm/+4jdlkZZQeW2k3/0vxw4Q9NUu0GdY7lAB3N+rs+Y1dTYplVto+oXXSzhVi 77fDMaOrHeaNsWrsmonOkHBrzWAHGngo/ndY8LzcYDTpBJiB92supIQuGHkGL7Jh zDhrWOHfDuK4Uf1sqSEczKm3bvyOlALsqSFd3+m71EtiVMxajz7JtgZKfg6TgKY8 6t5UaKru2XxS+gNVnt9yjfezeZeZEjFQsXoz4NJqktfYiaBSBt2acrKrV5XP/PjQ PN9Ns0MwOvGZ92MXRQd5HEnyUYSoCSt2AemsJqdQmxMyRCrfbhEU8xIWTtkw5i9q kVc8dFO+LkMyyjntU+w0Xy3DutK3tMce0dAa4TS7Ahl2Qv9Q5rUXIM3LH0Sqi+54 7v5LVp0MwOXpbeeOTk60ilbbjCmy+DaYc874ljH9n2ge1wv7yb+PH3YMR/qpO1M8 9GbKhSigCquLkPAxZvS5oXpXWGO2JJBh8ZN4RWKuiAD6okUR23o= =+ZkQ -----END PGP SIGNATURE----- --=-=-=--