From: Adam Kwolek <adam.kwolek@intel.com>
To: neilb@suse.de
Cc: linux-raid@vger.kernel.org, dan.j.williams@intel.com,
ed.ciechanowski@intel.com, wojciech.neubauer@intel.com
Subject: [PATCH 0/9] Grow_continue() - single array.
Date: Wed, 02 Mar 2011 14:28:56 +0100 [thread overview]
Message-ID: <20110302132426.24771.99191.stgit@gklab-128-013.igk.intel.com> (raw)
The following series implements reshape restart from checkpoint.
It supports single array reshape.
Container operations are not fully supported (some code preparation is introduced only /fork()/).
As workaround for container operation multiple container assembly can be used.
During each assembly one array will be reshaped. mdmon switches next array metadata in to reshape state.
This information will be used during next container assembly process.
This series contains some reviewed patches that you not accept last time asn few new.
1. Calling start_reshape() I've changed, but passed flag calculation has to be a little different
that I've pointed in the emails (external and restart flags cannot be used together).
2. I'm resending patches for fork() and backup file check in Grow_continue().
This is begin of support for container operations. It is somehow similar to reshape_container() idea,
but adapted to assembly code.
Assembled array processing is forked. This will allow later (during detected container operation)
to wait when mdmon set this particular array for reshape.
3. I've added fixes for imsm:
a.Checkpoint calculation is wrong (per disk instead per array units).
It was visible during restart from checkpoint. Md reports it
b.Array during reshape was marked as 'dirty'.
This makes mdadm to start reshape from 0 on restart
4. Code for setting 'new' geometry is currently very simple. It sets new disks number only.
I've got a little problem where to place it. You pointed sysfs_set_array() as place to do this.
Inside sysfs_set_array() is placed a comment to set those parameters outside this function.
I've decided to follow suggestion from comment as they seams reasonable.
This code is narrowed to expansion case,
but I've put 'ToDo' information in to code comments, how/where it should be changed (I hope shortly).
I think code follows direction you give me in last email (except above exclusions).
BR
Adam
---
Adam Kwolek (9):
imsm: FIX: Variables declaration cleanup
FIX: Block array monitoring when assembling reshaped array
FIX: Set 'new' geometry when assembling reshaped array
FIX: Verify Backup file name before reshape
FIX: Continue reshape in the background
imsm: FIX: After checkpoint mark array have to be clean
imsm: FIX: Return blocks_per unit for general migration
FIX: Array during reshape cannot be configured
FIX: Do not configure and start, already started reshape
Assemble.c | 26 +++++++++++++++++++++++++-
Grow.c | 47 +++++++++++++++++++++++++++++++++++++++++------
super-intel.c | 7 +++++--
3 files changed, 71 insertions(+), 9 deletions(-)
--
Signature
next reply other threads:[~2011-03-02 13:28 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-02 13:28 Adam Kwolek [this message]
2011-03-02 13:29 ` [PATCH 1/9] FIX: Do not configure and start, already started reshape Adam Kwolek
2011-03-02 13:29 ` [PATCH 2/9] FIX: Array during reshape cannot be configured Adam Kwolek
2011-03-02 13:29 ` [PATCH 3/9] imsm: FIX: Return blocks_per unit for general migration Adam Kwolek
2011-03-02 13:29 ` [PATCH 4/9] imsm: FIX: After checkpoint mark array have to be clean Adam Kwolek
2011-03-08 5:07 ` NeilBrown
2011-03-08 8:52 ` Kwolek, Adam
2011-03-08 22:14 ` NeilBrown
2011-03-09 7:31 ` Kwolek, Adam
2011-03-02 13:29 ` [PATCH 5/9] FIX: Continue reshape in the background Adam Kwolek
2011-03-02 13:29 ` [PATCH 6/9] FIX: Verify Backup file name before reshape Adam Kwolek
2011-03-02 13:29 ` [PATCH 7/9] FIX: Set 'new' geometry when assembling reshaped array Adam Kwolek
2011-03-02 13:29 ` [PATCH 8/9] FIX: Block array monitoring " Adam Kwolek
2011-03-02 13:30 ` [PATCH 9/9] imsm: FIX: Variables declaration cleanup Adam Kwolek
2011-03-08 6:43 ` [PATCH 0/9] Grow_continue() - single array NeilBrown
2011-03-08 11:18 ` Kwolek, Adam
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110302132426.24771.99191.stgit@gklab-128-013.igk.intel.com \
--to=adam.kwolek@intel.com \
--cc=dan.j.williams@intel.com \
--cc=ed.ciechanowski@intel.com \
--cc=linux-raid@vger.kernel.org \
--cc=neilb@suse.de \
--cc=wojciech.neubauer@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox