Add tests/05r1-internalbitmap-v1a

Passes all tests, nearly ready for release.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Getting ready for 2.0 release...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

remove ANNOUNCE-2.0-devel-?

remove ANNOUNCE-1*

Add raid10 doco to mdadm.8

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Lots of little manpage updates.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add RAID10 and other stuff to md.4

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Support internal bitmaps with format-1 superblocks.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Check-in swap_super.c

and update a test that uses it.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Support cross-compiling in Makefile

Thanks to "Anthony Russello" <russello@iomega.com>

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Print descriptive level when warning about pre-existing arrays.

Use map_num to get the name, so 'linear' doens't appear as '-1'.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix bug on new --detail code

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add test for multipath and make it work.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Assorted fixes

Support "--build"ing arrays with bitmaps.
hot-removal of bitmaps
--re-add of drives recently removed.
assorted extra tests

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix --detail output for version 1 superblocks.

As raid_disk no longer matches number, we do things differently.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix but when assembling v1 arrays

Because raid_disk and disk.number are not longer in-sync,
it needed some work.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add 'quite' option and tidy up some tests.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add a test.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Support nameing of version-1 arrays.

--name is recognised in --create and --assemble
name= is recognised in config file.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add write-behind support

Currently this includes
  --write-behind  to set level of write-behind supported
  --write-mostly  to flag devices as write-mostly.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Support fixing of byte-swapped superblocks.

Good for moving between little-endian and big-endian.
Still needs documentation.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add ANNOUNCE-2.0-devel-3

Add ANNOUNCE to inventory

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add test suite and release

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Assorted Fixes for multiple bugs.

Assemble would crash, or just not work.
A few other problem found by a new test-suite.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Redhat 6.2 seems to require sys/time.h as well as time.h

Thanks to  "Pieter D.J. Krul" <p at krul.cc>

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Don't for a chunk-size when "build"ing a raid1

as raid1 doesn't need chunksize

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make hot-add work with v 1.0.1 superblocks

Just some carelessness

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Document this...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add ANNOUNCE-2.0-devel-2

Release devel-2

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make sure we try hard enough to find a component device..

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix compiling of mdassemble

(again).

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix compilation with tinyc

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add man page for mdassemble

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Replace sprintf calls with snprintf

To quiet diet-libc

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add device files created with --auto to list of known device files.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Allow --force flag with --grow

.. so that you can 'grow' a raid1 to only one device.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Don't list device= in --examine --scan output.

As the device list isn't stable, recording it should be avoided.
The device= list is still available if --verbose is given (once).

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

When finding a /dev name for a device, prefer shorter names

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix silly example

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix some rounding errors in human_size and generally clean up the code

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix parsing of /dev/md/N in is_standard

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix a few typecasts

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Allow --auto to still be meaningful when --scan is given

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Initialise mis.next to NULL

Probably not needed, but safer.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix typo in --grow --help - don't claim '-F'

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make type of uuid always int[], rather than sometimes __u32[]

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Retry --stop --scan until all stoppable devices have been stopped

This is needed to reliably stop stacked arrays

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Document SparesMissing and DeviceDisappeared/WrongLevel

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Document --size=max option for --grow

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Avoid using lstat, it isn't right.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

From paul.clements@steeleye.com Thu May 19 11:09:55 2005
X-VM-v5-Data: ([nil nil nil nil t nil nil nil nil]
["2589" "Wednesday" "18" "May" "2005" "21:09:47" "-0400" "Paul Clements" "paul.clements@steeleye.com" nil "93" "[PATCH 1/1] mdadm 2.0 devel: fix bitmap size calculation and allow \"missing\" device in Build command" "^From:" "neilb@cse.unsw.edu.au" "neilb@cse.unsw.edu.au" "5" nil nil nil nil nil nil nil nil nil]
nil)
Return-Path: <paul.clements@steeleye.com>
Received: From
smtp-dist.unsw.edu.au ([149.171.97.16] == smtp-dist-01.services.comms.unsw.EDU.AU)
(for <neilb@cse.unsw.edu.au>) By note With Smtp ;
Thu, 19 May 2005 11:09:55 +1000
Received: from localhost (avspam-03.services.comms.unsw.edu.au [149.171.100.18])
by smtp-dist.unsw.edu.au (8.13.1/8.13.1) with ESMTP id j4J19t05017496
for <VRT+EFNJ=JQ=cse.unsw.edu.au=neilb@cse.unsw.edu.au>; Thu, 19 May 2005 11:09:55 +1000 (EST)
Received: from note.orchestra.cse.unsw.EDU.AU (note.orchestra.cse.unsw.EDU.AU [129.94.242.24])
by smtp.unsw.edu.au (8.13.1/8.13.1) with ESMTP id j4J19qEL010258
for <VRT+EFNJ=JQ=cse.unsw.edu.au=neilb@cse.unsw.edu.au>; Thu, 19 May 2005 11:09:53 +1000 (EST)
Received: From hancock.sc.steeleye.com ([209.192.50.48] == stat16.steeleye.com)
(for <neilb@cse.unsw.edu.au>) By note With Smtp ;
Thu, 19 May 2005 11:09:52 +1000
Received: from steeleye.com (clements.sc.steeleye.com [172.17.6.4])
by hancock.sc.steeleye.com (8.11.6/8.11.6) with ESMTP id j4J19lA29968
for <neilb@cse.unsw.edu.au>; Wed, 18 May 2005 21:09:48 -0400
Message-ID: <428BE75B.20803@steeleye.com>
User-Agent: Mozilla Thunderbird 0.5 (X11/20040208)
X-Accept-Language: en-us, en
MIME-Version: 1.0
References: <428A496D.6000303@steeleye.com>
In-Reply-To: <428A496D.6000303@steeleye.com>
From: Paul Clements <paul.clements@steeleye.com>
To: neilb@cse.unsw.edu.au
Subject: [PATCH 1/1] mdadm 2.0 devel: fix bitmap size calculation and allow
 "missing" device in Build command
Date: Wed, 18 May 2005 21:09:47 -0400
Content-Type: multipart/mixed;
 boundary="------------090601020206040308050709"

This is a multi-part message in MIME format.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Stuff

Description...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Stuff

Description...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make '-x nnn' spares work for type-1 superblocks.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Move the test for validity of raid_disk later.

This allows the check to be done after we know what
metadata format is being used, and so the max number of raiddisks
is known.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make --zero-superblock work for version 1 superblocks.

This requires passing the supertype to store_super

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix type - MD_SB_CLEAN is a bit number, not a bitmask.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Increase max-devs on type-1 superblocks

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Assorted fixes for bitmap related stuff

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Support --grow --bitmap=internal

Adding a filebased bitmap is not yet supported, and
this code is still under development.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add support for internal bitmaps

For version 0.90 superblocks, an internal bitmap can be specified at create.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

2.0-devel-1 release

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix a test that will allow arrays with bitmaps to be assembled.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Make sure to seed the random number generator for uuids

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Use ADD_NEW_DISK to hot-add to non-version-0 arrays

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix create bug which initialised superblocks wrongly.

Description...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Initial bitmap support

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Change "dirty" status to "active"

Description...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Wasn't initialising spare disks on create.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

A couple off minor fixed in multi-format management.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Release 1.8.1 after some man page updates and other fixes.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Only in mdadm-inter/: ANNOUNCE-2.0-devel-1
Only in mdadm-inter/: ANNOUNCE-2.0-devel-2
Only in mdadm-inter/: Create.c.orig
Only in mdadm-inter/: bitmap.c
Only in mdadm-inter/: bitmap.h
Only in mdadm-inter/: mdassemble.8
Only in mdadm-inter/: mkd1
diff -ru mdadm-pre/mdadm-1.11.0/super1.c mdadm-inter/super1.c
Signed-off-by: Neil Brown <neilb@suse.de>

super1

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Fix raid5 creation with new code.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add a 'super-switch' so that different format superblocks can be used.

This includes:
  adding --metadata= option to choose metadata format
  adding metadata= word to config file.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Separate sueprblock handling into separate file

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Release 1.11.0

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix embarassing bug in 1.10.0 which broke --add

Use a different variable to old result of open

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Release 1.10.0

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Call 'flush' after writing a superblock to disk.

Just in case...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Use O_EXCL when opening component devices to be assembled into an array

In 2.6, this will fail if the device is already in use, so we can detect this error
more easily.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix "--config=partitions" which was broken.

"devline" wants a list of words, not a whole line...

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add ANNOUNCE-1.8.1

Release 1.9.0

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Change handling of devices names in args

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix uninitialised variable compile warning.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Improve error message when mode is set twice.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Correct error message printed when the 'chosen_drive' cannot be added.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Reverse order of arrays  when --stop --scan

so that stacked arrays are stopped properly.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Reread partitions file for each array being assembled.

This makes assembling arrays from other arrays work better.
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Add some missing closes of mdfd

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Document new --auto behaviour

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Change MAJOR() etc to major() etc

This allows for larger device number if glibc supports
it (requires 2.3.3).
Also fail before creating larger device number if glibc
support isn't present.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

When assembling with --auto, honour common device names

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Change "dirty" to "active" in array status

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix some initialisations of ident

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix manpage typo

Thanks : Jérôme Petazzoni <jp@enix.org>

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>

Fix rpmbuild error

reported by  Stephan van Hienen <raid@a2000.nu>
Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>