Commit Graph

66 Commits

Author SHA1 Message Date
Alexander Motin
3ab0187add Remove request sorting from GEOM_MIRROR and GEOM_RAID.
When CPU is not busy, those queues are typically empty.  When CPU is busy,
then one more extra sorting is the last thing it needs.  If specific device
(HDD) really needs sorting, then it will be done later by CAM.

This supposed to fix livelock reported for mirror of two SSDs, when UFS
fires zillion of BIO_DELETE requests, that totally blocks I/O subsystem by
pointless sorting of requests and responses under single mutex lock.

MFC after:	2 weeks
2015-03-27 12:44:28 +00:00
Alexander Motin
0b1b7c2cec Replace constant with proper sizeof().
Submitted by:	Dmitry Luhtionov <dmitryluhtionov@gmail.com>
MFC after:	2 weeks
2015-02-25 10:18:11 +00:00
Alexander Motin
1e68fe9c33 Avoid unneeded malloc/memcpy/free if there is no metadata on disk.
Submitted by:	Dmitry Luhtionov <dmitryluhtionov@gmail.com>
MFC after:	2 weeks
2014-12-05 10:23:18 +00:00
Alexander Motin
26f0f92fa2 Decode some binary fields of Intel metadata.
Submitted by:	Dmitry Luhtionov <dmitryluhtionov@gmail.com>
MFC after:	2 weeks
2014-12-04 15:54:45 +00:00
Davide Italiano
2be111bf7d Follow up to r225617. In order to maximize the re-usability of kernel code
in userland rename in-kernel getenv()/setenv() to kern_setenv()/kern_getenv().
This fixes a namespace collision with libc symbols.

Submitted by:   kmacy
Tested by:      make universe
2014-10-16 18:04:43 +00:00
Hans Petter Selasky
af3b2549c4 Pull in r267961 and r267973 again. Fix for issues reported will follow. 2014-06-28 03:56:17 +00:00
Glen Barber
37a107a407 Revert r267961, r267973:
These changes prevent sysctl(8) from returning proper output,
such as:

 1) no output from sysctl(8)
 2) erroneously returning ENOMEM with tools like truss(1)
    or uname(1)
 truss: can not get etype: Cannot allocate memory
2014-06-27 22:05:21 +00:00
Hans Petter Selasky
3da1cf1e88 Extend the meaning of the CTLFLAG_TUN flag to automatically check if
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.

Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.

MFC after:	2 weeks
Sponsored by:	Mellanox Technologies
2014-06-27 16:33:43 +00:00
Alexander Motin
dea1e22600 Reduce number of opens by REOM RAID during provider taste.
Instead opening/closing provider by each of metadata classes, do it only
once in core code.  Since for SCSI disks open/close means sending some
SCSI commands to the device, this change reduces taste time.

MFC after:	2 weeks
Sponsored by:	iXsystems, Inc.
2014-04-28 15:03:52 +00:00
Alexander Motin
1229e83d2b Fix wrong sizes used to access PD_Type and PD_State DDF metadata fields.
This caused incorrect behavior of arrays with big-endian DDF metadata.
Little-endian (like used by Adaptec controllers) should not be harmed.
Add workaround should be enough to manage compatibility.

MFC after:	2 weeks
2014-04-10 16:00:33 +00:00
Eitan Adler
7a22215c53 Fix undefined behavior: (1 << 31) is not defined as 1 is an int and this
shifts into the sign bit.  Instead use (1U << 31) which gets the
expected result.

This fix is not ideal as it assumes a 32 bit int, but does fix the issue
for most cases.

A similar change was made in OpenBSD.

Discussed with:	-arch, rdivacky
Reviewed by:	cperciva
2013-11-30 22:17:27 +00:00
Alexander Motin
40ea77a036 Merge GEOM direct dispatch changes from the projects/camlock branch.
When safety requirements are met, it allows to avoid passing I/O requests
to GEOM g_up/g_down thread, executing them directly in the caller context.
That allows to avoid CPU bottlenecks in g_up/g_down threads, plus avoid
several context switches per I/O.

The defined now safety requirements are:
 - caller should not hold any locks and should be reenterable;
 - callee should not depend on GEOM dual-threaded concurency semantics;
 - on the way down, if request is unmapped while callee doesn't support it,
   the context should be sleepable;
 - kernel thread stack usage should be below 50%.

To keep compatibility with GEOM classes not meeting above requirements
new provider and consumer flags added:
 - G_CF_DIRECT_SEND -- consumer code meets caller requirements (request);
 - G_CF_DIRECT_RECEIVE -- consumer code meets callee requirements (done);
 - G_PF_DIRECT_SEND -- provider code meets caller requirements (done);
 - G_PF_DIRECT_RECEIVE -- provider code meets callee requirements (request).
Capable GEOM class can set them, allowing direct dispatch in cases where
it is safe.  If any of requirements are not met, request is queued to
g_up or g_down thread same as before.

Such GEOM classes were reviewed and updated to support direct dispatch:
CONCAT, DEV, DISK, GATE, MD, MIRROR, MULTIPATH, NOP, PART, RAID, STRIPE,
VFS, ZERO, ZFS::VDEV, ZFS::ZVOL, all classes based on g_slice KPI (LABEL,
MAP, FLASHMAP, etc).

To declare direct completion capability disk(9) KPI got new flag equivalent
to G_PF_DIRECT_SEND -- DISKFLAG_DIRECT_COMPLETION.  da(4) and ada(4) disk
drivers got it set now thanks to earlier CAM locking work.

This change more then twice increases peak block storage performance on
systems with manu CPUs, together with earlier CAM locking changes reaching
more then 1 million IOPS (512 byte raw reads from 16 SATA SSDs on 4 HBAs to
256 user-level threads).

Sponsored by:	iXsystems, Inc.
MFC after:	2 months
2013-10-22 08:22:19 +00:00
Alexander Motin
b43560ab19 MFprojects/camlock r256445:
Add unmapped I/O support to GEOM RAID.
2013-10-16 09:33:23 +00:00
Alexander Motin
0f0b2fd889 Return error when opening read-only volumes (like RAID4/5/...) for writing.
Previously opens succeeded, but actual write operations returned errors.

Requested by:	peter
MFC after:	2 weeks
2013-08-13 07:56:40 +00:00
Alexander Motin
db8645f05e Oops, wrong constant at r254269. 2013-08-13 06:25:34 +00:00
Alexander Motin
e70b565ba4 Fix reasonable but safe Clang warnings. 2013-08-13 06:21:36 +00:00
Alexander Motin
8531bb3f0c Introduce 3 seconds timeout on graid stop command (mostly with -f flag).
Since completion waiting goes in g_event thread, it may cause GEOM deadlock
if consumer on top (for example, ZFS) uses g_event thread for closing.
2013-07-27 15:02:19 +00:00
Alexander Motin
57eed4a86f Fix vdc->Secondary_Element_Count metadata field access from 16 to 8 bit.
In some cases it could cause kernel panic during failed drive replacement.

Reported by:	trasz
MFC after:	1 week
2013-05-20 00:33:54 +00:00
Alexander Motin
bcb6ad36f2 Return "descr" field alike to "Intel RAID1 volume" for GEOM RAID to make
it look better in bsdinstall.
2013-04-27 06:57:39 +00:00
Alexander Motin
a93c0ed463 Remove extra bio_data and bio_length copying to child request after calling
g_clone_bio(), that already copied them.
2013-03-26 05:42:12 +00:00
Sean Bruno
bd9fba0cfe Add legacy support to geom raid to create a /dev/arX device for support
of upgrading older machines using ataraid(4) to newer releases.

This optional parameter is controlled via kern.geom.raid.legacy_aliases
and will create a /dev/ar0 device that will point at /dev/raid/r0 for
example.

Tested on Dell SC 1425 DDF-1 format software raid controllers installing from
stable/7 and upgrading to stable/9 without having to adjust /etc/fstab

Reviewed by:	mav
Obtained from:	Yahoo!
MFC after:	2 Weeks
2013-03-08 20:07:32 +00:00
Alexander Motin
34d3281c57 Fix panic when Secondary_Element_Count == 1 and Secondary_Element_Seq
is not set (255).

Reported by:	sbruno
MFC after:	1 week
2013-03-07 18:55:37 +00:00
Alexander Motin
c3ec009a97 - Fix rebuild position broken at r245522.
- Identify one more metadata field.
2013-01-17 03:27:08 +00:00
Alexander Motin
821a0f639e For Promise/AMD metadata add support for disks with capacity above 2TiB
and for volumes with sector size above 512 bytes.
2013-01-17 00:50:25 +00:00
Alexander Motin
ed8180e665 Recalculate volume size only for real CONCATs. For SINGLE trust volume
size given by metadata, as it should be correct and in some cases can be
smaller then subdisk size.
2013-01-17 00:09:50 +00:00
Alexander Motin
4c10c25e33 Keep value of orig_config_id metadata field. Windows driver writes there
previous value of config_id when it is changed in some cases.  I guess it
may be used do avoid some split-brain conditions.
2013-01-14 20:31:45 +00:00
Alexander Motin
eb84fc957c Small cosmetic tuning of the IRRT status constants. 2013-01-14 16:38:43 +00:00
Alexander Motin
511c69d9ce Print some more metadata fields. 2013-01-14 13:06:35 +00:00
Alexander Motin
898a4b74f4 Windows driver writes relative volume IDs to metadata field. Use that value
as a hint for raid/rX device number to make it persistent across reboots.
2013-01-14 00:38:51 +00:00
Alexander Motin
f9462b9bbe - Add checks for Intel metadata version and attributes. Ignore disks with
unsupported metadata types like Intel Smart Response to not corrupt them.
 - Improve setting of these things during metadata writing to protect from
incapable BIOS'es and other implementations.
2013-01-13 23:00:40 +00:00
Alexander Motin
b99586c25f Improve support for disabled disks. If disabled disk disconnected and then
reconnected back, leave it as disconnected. If new disk inserted instead of
disabled, rebuild it and leave as enabled.
2013-01-13 14:30:37 +00:00
Alexander Motin
865aea63c3 Windows handles INIT and VERIFY as array-wide and it doesn't specify which
disks should be rebuilt. Our rebuild code is same time disk-centric.  To
handle this situation  properly check all disks for RBLD flags, and if no
disk specified try rebuild/resync all of them except newly inserted.
2013-01-12 21:51:49 +00:00
Alexander Motin
4c95a24141 Implement migration from single disk to RAID1/IRRT for Intel metadata.
Windows driver uses such migration when it creates new arrays.  While GEOM
RAID has no mechanism to implement migration in general case, this specifc
case still can be handled easily via degraded RAID1 creation followed by
regular rebuild.
2013-01-12 18:25:48 +00:00
Alexander Motin
26c538bc0b Add basic support for Intel Rapid Recover Technology (Intel RRT).
It is alike to RAID1, but with dedicating master and recovery disks and
providing manual control over synchronization.  It allows to use recovery
disk as snapshot of the master disk from the time of the last sync.

This implementation is not functionaly complete comparing to Windows,
but it is better then silent conversion to RAID1 on first boot.
2013-01-12 09:35:44 +00:00
Alexander Motin
650e245ebf Minor addition to r242323:
Alike to BIO_WRITE, report success if at least one subdisk succeeded with
BIO_DELETE.  But unlike BIO_WRITE don't fail disk on BIO_DELETE error.

Sponsored by:	iXsystems, Inc.
MFC after:	1 month
2012-10-29 21:08:06 +00:00
Alexander Motin
609a74746a Add basic BIO_DELETE support to GEOM RAID class for all RAID levels.
If at least one subdisk in the volume supports it, BIO_DELETE requests
will be propagated down.  Unfortunatelly, for RAID levels with redundancy
unmapped blocks will be mapped back during first rebuild/resync process.

Sponsored by:	iXsystems, Inc.
MFC after:	1 month
2012-10-29 18:04:38 +00:00
Alexander Motin
a479c51be3 Make GEOM RAID more aggressive in marking volumes as clean on shutdown
and move that action from shutdown_pre_sync to shutdown_post_sync stage
to avoid extra flapping.

ZFS tends to not close devices on shutdown, that doesn't allow GEOM RAID
to shutdown gracefully.  To handle that, mark volume as clean just when
shutdown time comes and there are no active writes.

MFC after:	2 weeks
2012-10-29 14:18:54 +00:00
Alexander Motin
c6f0cd57e3 NULL-ify last previously used pointer instead of last possible pointer.
This should be only a cosmetic change.

Found by:	Clang Static Analyzer
2012-10-10 20:41:37 +00:00
Alexander Motin
6871a543f9 Make graid command line a bit more friendly by allowing volume name or
provider name to be specified instead of geom name (first argument in all
subcommands except label).  In most cases there is only one array used
any way, so it is not really useful to make user type ugly geom names like
Intel-f0bdf223 or SiI-732c2b9448cf.  Though they can be used in some cases.

Sponsored by:	iXsystems, Inc.
MFC after:	1 month
2012-10-07 19:30:16 +00:00
Alexander Motin
c89d2fbe18 Add global and per-module sysctls/tunables to enable/disable metadata taste.
That should help to handle some cases when disk has some RAID metadata that
should be ignored, especially during boot.

MFC after:	3 days
2012-09-13 13:27:09 +00:00
Alexander Motin
d9d6849693 Add missing FAILED event to g_raid_subdisk_event2str() to print it properly
in debug messages.

Submitted by:	Dmitry Luhtionov <dmitryluhtionov@gmail.com>
2012-08-10 13:36:33 +00:00
Alexander Motin
ef844ef76f - Prevent error status leak if write to some of the RAID1/1E volume disks
failed while write to some other succeeded. Instead mark disk as failed.
- Make RAID1E less aggressive in failing disks to avoid volume breakage.

MFC after:	2 weeks
2012-05-11 13:20:17 +00:00
Alexander Motin
14f9f25ba0 Remove some hardcoded constants from code. 2012-05-06 16:41:27 +00:00
Alexander Motin
eb3b1cd0de Plug small memory leaks. 2012-05-06 12:55:20 +00:00
Alexander Motin
8f12ca2ee1 Add support for RAID5R. Slightly improve support for RAIDMDF. 2012-05-06 11:32:36 +00:00
Alexander Motin
86b0366909 Fix bug causing memory corruption and panics with big-endian metadata. 2012-05-04 08:59:19 +00:00
Alexander Motin
4b97ff6137 Implement read-only support for volumes in optimal state (without using
redundancy) for the following RAID levels: RAID4/5E/5EE/6/MDF.
2012-05-04 07:32:57 +00:00
Alexander Motin
8df8e26adc Add optional -o argument to the graid label to specify some metadata
format options. Use it for specifying byte order for the DDF metadata:
big-endian defined by specification and little-endian used by Adaptec.
2012-05-03 05:32:56 +00:00
Alexander Motin
d525d87560 Improve spare disks support. Unluckily, for some reason Adaptec 1430SA
RAID BIOS doesn't want to understand spare disks created by graid. But
at least spares created by BIOS are working fine now.
2012-05-01 18:00:31 +00:00
Alexander Motin
2b9c925ff0 Implement volume deletion if disk has more then one partition. 2012-05-01 09:21:21 +00:00